summary |
shortlog |
changelog |
graph |
tags |
branches |
files |
changeset |
file |
revisions |
annotate |
diff |
raw

NEWS

author | haftmann |

Mon Jul 23 09:28:03 2012 +0200 (2012-07-23 ago) | |

changeset 48431 | 6efff142bb54 |

parent 48428 | ffa0618cc4d4 |

child 48574 | 4af9f3122138 |

permissions | -rw-r--r-- |

restrict unqualified imports from Haskell Prelude to a small set of fundamental operations

1 Isabelle NEWS -- history user-relevant changes

2 ==============================================

4 New in this Isabelle version

5 ----------------------------

7 *** General ***

9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which

10 is called fastforce / fast_force_tac already since Isabelle2011-1.

12 * Updated and extended "isar-ref" manual, reduced remaining material

13 in old "ref" manual.

16 *** Pure ***

18 * Code generation for Haskell: restrict unqualified imports from

19 Haskell Prelude to a small set of fundamental operations.

21 * Command "export_code": relative file names are interpreted

22 relatively to master directory of current theory rather than

23 the rather arbitrary current working directory.

24 INCOMPATIBILITY.

26 * Discontinued obsolete attribute "COMP". Potential INCOMPATIBILITY,

27 use regular rule composition via "OF" / "THEN", or explicit proof

28 structure instead. Note that Isabelle/ML provides a variety of

29 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied

30 with some care where this is really required.

33 *** HOL ***

35 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel

36 execution for code generated towards Isabelle/ML.

38 * Simproc "finite_Collect" rewrites set comprehensions into pointfree

39 expressions.

41 * Quickcheck:

43 - added an optimisation for equality premises.

44 It is switched on by default, and can be switched off by setting

45 the configuration quickcheck_optimise_equality to false.

47 * The SMT solver Z3 has now by default a restricted set of directly

48 supported features. For the full set of features (div/mod, nonlinear

49 arithmetic, datatypes/records) with potential proof reconstruction

50 failures, enable the configuration option "z3_with_extensions".

51 Minor INCOMPATIBILITY.

53 * Sledgehammer:

55 - Added MaSh relevance filter based on machine-learning; see the

56 Sledgehammer manual for details.

57 - Rationalized type encodings ("type_enc" option).

58 - Renamed options:

59 max_relevant ~> max_facts

60 relevance_thresholds ~> fact_thresholds

63 *** Document preparation ***

65 * Default for \<euro> is now based on eurosym package, instead of

66 slightly exotic babel/greek.

69 *** System ***

71 * Discontinued support for Poly/ML 5.2.1, which was the last version

72 without exception positions and advanced ML compiler/toplevel

73 configuration.

77 New in Isabelle2012 (May 2012)

78 ------------------------------

80 *** General ***

82 * Prover IDE (PIDE) improvements:

84 - more robust Sledgehammer integration (as before the sledgehammer

85 command-line needs to be typed into the source buffer)

86 - markup for bound variables

87 - markup for types of term variables (displayed as tooltips)

88 - support for user-defined Isar commands within the running session

89 - improved support for Unicode outside original 16bit range

90 e.g. glyph for \<A> (thanks to jEdit 4.5.1)

92 * Forward declaration of outer syntax keywords within the theory

93 header -- minor INCOMPATIBILITY for user-defined commands. Allow new

94 commands to be used in the same theory where defined.

96 * Auxiliary contexts indicate block structure for specifications with

97 additional parameters and assumptions. Such unnamed contexts may be

98 nested within other targets, like 'theory', 'locale', 'class',

99 'instantiation' etc. Results from the local context are generalized

100 accordingly and applied to the enclosing target context. Example:

102 context

103 fixes x y z :: 'a

104 assumes xy: "x = y" and yz: "y = z"

105 begin

107 lemma my_trans: "x = z" using xy yz by simp

109 end

111 thm my_trans

113 The most basic application is to factor-out context elements of

114 several fixes/assumes/shows theorem statements, e.g. see

115 ~~/src/HOL/Isar_Examples/Group_Context.thy

117 Any other local theory specification element works within the "context

118 ... begin ... end" block as well.

120 * Bundled declarations associate attributed fact expressions with a

121 given name in the context. These may be later included in other

122 contexts. This allows to manage context extensions casually, without

123 the logical dependencies of locales and locale interpretation. See

124 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.

126 * Commands 'lemmas' and 'theorems' allow local variables using 'for'

127 declaration, and results are standardized before being stored. Thus

128 old-style "standard" after instantiation or composition of facts

129 becomes obsolete. Minor INCOMPATIBILITY, due to potential change of

130 indices of schematic variables.

132 * Rule attributes in local theory declarations (e.g. locale or class)

133 are now statically evaluated: the resulting theorem is stored instead

134 of the original expression. INCOMPATIBILITY in rare situations, where

135 the historic accident of dynamic re-evaluation in interpretations

136 etc. was exploited.

138 * New tutorial "Programming and Proving in Isabelle/HOL"

139 ("prog-prove"). It completely supersedes "A Tutorial Introduction to

140 Structured Isar Proofs" ("isar-overview"), which has been removed. It

141 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order

142 Logic" as the recommended beginners tutorial, but does not cover all

143 of the material of that old tutorial.

145 * Updated and extended reference manuals: "isar-ref",

146 "implementation", "system"; reduced remaining material in old "ref"

147 manual.

150 *** Pure ***

152 * Command 'definition' no longer exports the foundational "raw_def"

153 into the user context. Minor INCOMPATIBILITY, may use the regular

154 "def" result with attribute "abs_def" to imitate the old version.

156 * Attribute "abs_def" turns an equation of the form "f x y == t" into

157 "f == %x y. t", which ensures that "simp" or "unfold" steps always

158 expand it. This also works for object-logic equality. (Formerly

159 undocumented feature.)

161 * Sort constraints are now propagated in simultaneous statements, just

162 like type constraints. INCOMPATIBILITY in rare situations, where

163 distinct sorts used to be assigned accidentally. For example:

165 lemma "P (x::'a::foo)" and "Q (y::'a::bar)" -- "now illegal"

167 lemma "P (x::'a)" and "Q (y::'a::bar)"

168 -- "now uniform 'a::bar instead of default sort for first occurrence (!)"

170 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more

171 tolerant against multiple unifiers, as long as the final result is

172 unique. (As before, rules are composed in canonical right-to-left

173 order to accommodate newly introduced premises.)

175 * Renamed some inner syntax categories:

177 num ~> num_token

178 xnum ~> xnum_token

179 xstr ~> str_token

181 Minor INCOMPATIBILITY. Note that in practice "num_const" or

182 "num_position" etc. are mainly used instead (which also include

183 position information via constraints).

185 * Simplified configuration options for syntax ambiguity: see

186 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref

187 manual. Minor INCOMPATIBILITY.

189 * Discontinued configuration option "syntax_positions": atomic terms

190 in parse trees are always annotated by position constraints.

192 * Old code generator for SML and its commands 'code_module',

193 'code_library', 'consts_code', 'types_code' have been discontinued.

194 Use commands of the generic code generator instead. INCOMPATIBILITY.

196 * Redundant attribute "code_inline" has been discontinued. Use

197 "code_unfold" instead. INCOMPATIBILITY.

199 * Dropped attribute "code_unfold_post" in favor of the its dual

200 "code_abbrev", which yields a common pattern in definitions like

202 definition [code_abbrev]: "f = t"

204 INCOMPATIBILITY.

206 * Obsolete 'types' command has been discontinued. Use 'type_synonym'

207 instead. INCOMPATIBILITY.

209 * Discontinued old "prems" fact, which used to refer to the accidental

210 collection of foundational premises in the context (already marked as

211 legacy since Isabelle2011).

214 *** HOL ***

216 * Type 'a set is now a proper type constructor (just as before

217 Isabelle2008). Definitions mem_def and Collect_def have disappeared.

218 Non-trivial INCOMPATIBILITY. For developments keeping predicates and

219 sets separate, it is often sufficient to rephrase some set S that has

220 been accidentally used as predicates by "%x. x : S", and some

221 predicate P that has been accidentally used as set by "{x. P x}".

222 Corresponding proofs in a first step should be pruned from any

223 tinkering with former theorems mem_def and Collect_def as far as

224 possible.

226 For developments which deliberately mix predicates and sets, a

227 planning step is necessary to determine what should become a predicate

228 and what a set. It can be helpful to carry out that step in

229 Isabelle2011-1 before jumping right into the current release.

231 * Code generation by default implements sets as container type rather

232 than predicates. INCOMPATIBILITY.

234 * New type synonym 'a rel = ('a * 'a) set

236 * The representation of numerals has changed. Datatype "num"

237 represents strictly positive binary numerals, along with functions

238 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent

239 positive and negated numeric literals, respectively. See also

240 definitions in ~~/src/HOL/Num.thy. Potential INCOMPATIBILITY, some

241 user theories may require adaptations as follows:

243 - Theorems with number_ring or number_semiring constraints: These

244 classes are gone; use comm_ring_1 or comm_semiring_1 instead.

246 - Theories defining numeric types: Remove number, number_semiring,

247 and number_ring instances. Defer all theorems about numerals until

248 after classes one and semigroup_add have been instantiated.

250 - Numeral-only simp rules: Replace each rule having a "number_of v"

251 pattern with two copies, one for numeral and one for neg_numeral.

253 - Theorems about subclasses of semiring_1 or ring_1: These classes

254 automatically support numerals now, so more simp rules and

255 simprocs may now apply within the proof.

257 - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:

258 Redefine using other integer operations.

260 * Transfer: New package intended to generalize the existing

261 "descending" method and related theorem attributes from the Quotient

262 package. (Not all functionality is implemented yet, but future

263 development will focus on Transfer as an eventual replacement for the

264 corresponding parts of the Quotient package.)

266 - transfer_rule attribute: Maintains a collection of transfer rules,

267 which relate constants at two different types. Transfer rules may

268 relate different type instances of the same polymorphic constant,

269 or they may relate an operation on a raw type to a corresponding

270 operation on an abstract type (quotient or subtype). For example:

272 ((A ===> B) ===> list_all2 A ===> list_all2 B) map map

273 (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int

275 - transfer method: Replaces a subgoal on abstract types with an

276 equivalent subgoal on the corresponding raw types. Constants are

277 replaced with corresponding ones according to the transfer rules.

278 Goals are generalized over all free variables by default; this is

279 necessary for variables whose types change, but can be overridden

280 for specific variables with e.g. "transfer fixing: x y z". The

281 variant transfer' method allows replacing a subgoal with one that

282 is logically stronger (rather than equivalent).

284 - relator_eq attribute: Collects identity laws for relators of

285 various type constructors, e.g. "list_all2 (op =) = (op =)". The

286 transfer method uses these lemmas to infer transfer rules for

287 non-polymorphic constants on the fly.

289 - transfer_prover method: Assists with proving a transfer rule for a

290 new constant, provided the constant is defined in terms of other

291 constants that already have transfer rules. It should be applied

292 after unfolding the constant definitions.

294 - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer

295 from type nat to type int.

297 * Lifting: New package intended to generalize the quotient_definition

298 facility of the Quotient package; designed to work with Transfer.

300 - lift_definition command: Defines operations on an abstract type in

301 terms of a corresponding operation on a representation

302 type. Example syntax:

304 lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"

305 is List.insert

307 Users must discharge a respectfulness proof obligation when each

308 constant is defined. (For a type copy, i.e. a typedef with UNIV,

309 the proof is discharged automatically.) The obligation is

310 presented in a user-friendly, readable form; a respectfulness

311 theorem in the standard format and a transfer rule are generated

312 by the package.

314 - Integration with code_abstype: For typedefs (e.g. subtypes

315 corresponding to a datatype invariant, such as dlist),

316 lift_definition generates a code certificate theorem and sets up

317 code generation for each constant.

319 - setup_lifting command: Sets up the Lifting package to work with a

320 user-defined type. The user must provide either a quotient theorem

321 or a type_definition theorem. The package configures transfer

322 rules for equality and quantifiers on the type, and sets up the

323 lift_definition command to work with the type.

325 - Usage examples: See Quotient_Examples/Lift_DList.thy,

326 Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,

327 Word/Word.thy and Library/Float.thy.

329 * Quotient package:

331 - The 'quotient_type' command now supports a 'morphisms' option with

332 rep and abs functions, similar to typedef.

334 - 'quotient_type' sets up new types to work with the Lifting and

335 Transfer packages, as with 'setup_lifting'.

337 - The 'quotient_definition' command now requires the user to prove a

338 respectfulness property at the point where the constant is

339 defined, similar to lift_definition; INCOMPATIBILITY.

341 - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems

342 accordingly, INCOMPATIBILITY.

344 * New diagnostic command 'find_unused_assms' to find potentially

345 superfluous assumptions in theorems using Quickcheck.

347 * Quickcheck:

349 - Quickcheck returns variable assignments as counterexamples, which

350 allows to reveal the underspecification of functions under test.

351 For example, refuting "hd xs = x", it presents the variable

352 assignment xs = [] and x = a1 as a counterexample, assuming that

353 any property is false whenever "hd []" occurs in it.

355 These counterexample are marked as potentially spurious, as

356 Quickcheck also returns "xs = []" as a counterexample to the

357 obvious theorem "hd xs = hd xs".

359 After finding a potentially spurious counterexample, Quickcheck

360 continues searching for genuine ones.

362 By default, Quickcheck shows potentially spurious and genuine

363 counterexamples. The option "genuine_only" sets quickcheck to only

364 show genuine counterexamples.

366 - The command 'quickcheck_generator' creates random and exhaustive

367 value generators for a given type and operations.

369 It generates values by using the operations as if they were

370 constructors of that type.

372 - Support for multisets.

374 - Added "use_subtype" options.

376 - Added "quickcheck_locale" configuration to specify how to process

377 conjectures in a locale context.

379 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option

380 and affecting 'rat' and 'real'.

382 * Sledgehammer:

383 - Integrated more tightly with SPASS, as described in the ITP 2012

384 paper "More SPASS with Isabelle".

385 - Made it try "smt" as a fallback if "metis" fails or times out.

386 - Added support for the following provers: Alt-Ergo (via Why3 and

387 TFF1), iProver, iProver-Eq.

388 - Sped up the minimizer.

389 - Added "lam_trans", "uncurry_aliases", and "minimize" options.

390 - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").

391 - Renamed "sound" option to "strict".

393 * Metis: Added possibility to specify lambda translations scheme as a

394 parenthesized argument (e.g., "by (metis (lifting) ...)").

396 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".

398 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.

400 * New "case_product" attribute to generate a case rule doing multiple

401 case distinctions at the same time. E.g.

403 list.exhaust [case_product nat.exhaust]

405 produces a rule which can be used to perform case distinction on both

406 a list and a nat.

408 * New "eventually_elim" method as a generalized variant of the

409 eventually_elim* rules. Supports structured proofs.

411 * Typedef with implicit set definition is considered legacy. Use

412 "typedef (open)" form instead, which will eventually become the

413 default.

415 * Record: code generation can be switched off manually with

417 declare [[record_coden = false]] -- "default true"

419 * Datatype: type parameters allow explicit sort constraints.

421 * Concrete syntax for case expressions includes constraints for source

422 positions, and thus produces Prover IDE markup for its bindings.

423 INCOMPATIBILITY for old-style syntax translations that augment the

424 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of

425 one_case.

427 * Clarified attribute "mono_set": pure declaration without modifying

428 the result of the fact expression.

430 * More default pred/set conversions on a couple of relation operations

431 and predicates. Added powers of predicate relations. Consolidation

432 of some relation theorems:

434 converse_def ~> converse_unfold

435 rel_comp_def ~> relcomp_unfold

436 symp_def ~> (modified, use symp_def and sym_def instead)

437 transp_def ~> transp_trans

438 Domain_def ~> Domain_unfold

439 Range_def ~> Domain_converse [symmetric]

441 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.

443 See theory "Relation" for examples for making use of pred/set

444 conversions by means of attributes "to_set" and "to_pred".

446 INCOMPATIBILITY.

448 * Renamed facts about the power operation on relations, i.e., relpow

449 to match the constant's name:

451 rel_pow_1 ~> relpow_1

452 rel_pow_0_I ~> relpow_0_I

453 rel_pow_Suc_I ~> relpow_Suc_I

454 rel_pow_Suc_I2 ~> relpow_Suc_I2

455 rel_pow_0_E ~> relpow_0_E

456 rel_pow_Suc_E ~> relpow_Suc_E

457 rel_pow_E ~> relpow_E

458 rel_pow_Suc_D2 ~> relpow_Suc_D2

459 rel_pow_Suc_E2 ~> relpow_Suc_E2

460 rel_pow_Suc_D2' ~> relpow_Suc_D2'

461 rel_pow_E2 ~> relpow_E2

462 rel_pow_add ~> relpow_add

463 rel_pow_commute ~> relpow

464 rel_pow_empty ~> relpow_empty:

465 rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow

466 rel_pow_imp_rtrancl ~> relpow_imp_rtrancl

467 rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow

468 rtrancl_imp_rel_pow ~> rtrancl_imp_relpow

469 rel_pow_fun_conv ~> relpow_fun_conv

470 rel_pow_finite_bounded1 ~> relpow_finite_bounded1

471 rel_pow_finite_bounded ~> relpow_finite_bounded

472 rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow

473 trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow

474 single_valued_rel_pow ~> single_valued_relpow

476 INCOMPATIBILITY.

478 * Theory Relation: Consolidated constant name for relation composition

479 and corresponding theorem names:

481 - Renamed constant rel_comp to relcomp.

483 - Dropped abbreviation pred_comp. Use relcompp instead.

485 - Renamed theorems:

487 rel_compI ~> relcompI

488 rel_compEpair ~> relcompEpair

489 rel_compE ~> relcompE

490 pred_comp_rel_comp_eq ~> relcompp_relcomp_eq

491 rel_comp_empty1 ~> relcomp_empty1

492 rel_comp_mono ~> relcomp_mono

493 rel_comp_subset_Sigma ~> relcomp_subset_Sigma

494 rel_comp_distrib ~> relcomp_distrib

495 rel_comp_distrib2 ~> relcomp_distrib2

496 rel_comp_UNION_distrib ~> relcomp_UNION_distrib

497 rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2

498 single_valued_rel_comp ~> single_valued_relcomp

499 rel_comp_def ~> relcomp_unfold

500 converse_rel_comp ~> converse_relcomp

501 pred_compI ~> relcomppI

502 pred_compE ~> relcomppE

503 pred_comp_bot1 ~> relcompp_bot1

504 pred_comp_bot2 ~> relcompp_bot2

505 transp_pred_comp_less_eq ~> transp_relcompp_less_eq

506 pred_comp_mono ~> relcompp_mono

507 pred_comp_distrib ~> relcompp_distrib

508 pred_comp_distrib2 ~> relcompp_distrib2

509 converse_pred_comp ~> converse_relcompp

511 finite_rel_comp ~> finite_relcomp

513 set_rel_comp ~> set_relcomp

515 INCOMPATIBILITY.

517 * Theory Divides: Discontinued redundant theorems about div and mod.

518 INCOMPATIBILITY, use the corresponding generic theorems instead.

520 DIVISION_BY_ZERO ~> div_by_0, mod_by_0

521 zdiv_self ~> div_self

522 zmod_self ~> mod_self

523 zdiv_zero ~> div_0

524 zmod_zero ~> mod_0

525 zdiv_zmod_equality ~> div_mod_equality2

526 zdiv_zmod_equality2 ~> div_mod_equality

527 zmod_zdiv_trivial ~> mod_div_trivial

528 zdiv_zminus_zminus ~> div_minus_minus

529 zmod_zminus_zminus ~> mod_minus_minus

530 zdiv_zminus2 ~> div_minus_right

531 zmod_zminus2 ~> mod_minus_right

532 zdiv_minus1_right ~> div_minus1_right

533 zmod_minus1_right ~> mod_minus1_right

534 zdvd_mult_div_cancel ~> dvd_mult_div_cancel

535 zmod_zmult1_eq ~> mod_mult_right_eq

536 zpower_zmod ~> power_mod

537 zdvd_zmod ~> dvd_mod

538 zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd

539 mod_mult_distrib ~> mult_mod_left

540 mod_mult_distrib2 ~> mult_mod_right

542 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use

543 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.

545 * Finite_Set.fold now qualified. INCOMPATIBILITY.

547 * Consolidated theorem names concerning fold combinators:

549 inf_INFI_fold_inf ~> inf_INF_fold_inf

550 sup_SUPR_fold_sup ~> sup_SUP_fold_sup

551 INFI_fold_inf ~> INF_fold_inf

552 SUPR_fold_sup ~> SUP_fold_sup

553 union_set ~> union_set_fold

554 minus_set ~> minus_set_fold

555 INFI_set_fold ~> INF_set_fold

556 SUPR_set_fold ~> SUP_set_fold

557 INF_code ~> INF_set_foldr

558 SUP_code ~> SUP_set_foldr

559 foldr.simps ~> foldr.simps (in point-free formulation)

560 foldr_fold_rev ~> foldr_conv_fold

561 foldl_fold ~> foldl_conv_fold

562 foldr_foldr ~> foldr_conv_foldl

563 foldl_foldr ~> foldl_conv_foldr

564 fold_set_remdups ~> fold_set_fold_remdups

565 fold_set ~> fold_set_fold

566 fold1_set ~> fold1_set_fold

568 INCOMPATIBILITY.

570 * Dropped rarely useful theorems concerning fold combinators:

571 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,

572 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,

573 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,

574 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,

575 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,

576 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.

577 INCOMPATIBILITY. For the common phrases "%xs. List.foldr plus xs 0"

578 and "List.foldl plus 0", prefer "List.listsum". Otherwise it can be

579 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by

580 unfolding "foldr_conv_fold" and "foldl_conv_fold".

582 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,

583 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,

584 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,

585 INF_set_foldr, SUP_set_foldr. INCOMPATIBILITY. Prefer corresponding

586 lemmas over fold rather than foldr, or make use of lemmas

587 fold_conv_foldr and fold_rev.

589 * Congruence rules Option.map_cong and Option.bind_cong for recursion

590 through option types.

592 * "Transitive_Closure.ntrancl": bounded transitive closure on

593 relations.

595 * Constant "Set.not_member" now qualified. INCOMPATIBILITY.

597 * Theory Int: Discontinued many legacy theorems specific to type int.

598 INCOMPATIBILITY, use the corresponding generic theorems instead.

600 zminus_zminus ~> minus_minus

601 zminus_0 ~> minus_zero

602 zminus_zadd_distrib ~> minus_add_distrib

603 zadd_commute ~> add_commute

604 zadd_assoc ~> add_assoc

605 zadd_left_commute ~> add_left_commute

606 zadd_ac ~> add_ac

607 zmult_ac ~> mult_ac

608 zadd_0 ~> add_0_left

609 zadd_0_right ~> add_0_right

610 zadd_zminus_inverse2 ~> left_minus

611 zmult_zminus ~> mult_minus_left

612 zmult_commute ~> mult_commute

613 zmult_assoc ~> mult_assoc

614 zadd_zmult_distrib ~> left_distrib

615 zadd_zmult_distrib2 ~> right_distrib

616 zdiff_zmult_distrib ~> left_diff_distrib

617 zdiff_zmult_distrib2 ~> right_diff_distrib

618 zmult_1 ~> mult_1_left

619 zmult_1_right ~> mult_1_right

620 zle_refl ~> order_refl

621 zle_trans ~> order_trans

622 zle_antisym ~> order_antisym

623 zle_linear ~> linorder_linear

624 zless_linear ~> linorder_less_linear

625 zadd_left_mono ~> add_left_mono

626 zadd_strict_right_mono ~> add_strict_right_mono

627 zadd_zless_mono ~> add_less_le_mono

628 int_0_less_1 ~> zero_less_one

629 int_0_neq_1 ~> zero_neq_one

630 zless_le ~> less_le

631 zpower_zadd_distrib ~> power_add

632 zero_less_zpower_abs_iff ~> zero_less_power_abs_iff

633 zero_le_zpower_abs ~> zero_le_power_abs

635 * Theory Deriv: Renamed

637 DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing

639 * Theory Library/Multiset: Improved code generation of multisets.

641 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets

642 are expressed via type classes again. The special syntax

643 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant

644 setsum_set, which is now subsumed by Big_Operators.setsum.

645 INCOMPATIBILITY.

647 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,

648 use theory HOL/Library/Nat_Bijection instead.

650 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black

651 trees is now inside a type class context. Names of affected

652 operations and lemmas have been prefixed by rbt_. INCOMPATIBILITY for

653 theories working directly with raw red-black trees, adapt the names as

654 follows:

656 Operations:

657 bulkload -> rbt_bulkload

658 del_from_left -> rbt_del_from_left

659 del_from_right -> rbt_del_from_right

660 del -> rbt_del

661 delete -> rbt_delete

662 ins -> rbt_ins

663 insert -> rbt_insert

664 insertw -> rbt_insert_with

665 insert_with_key -> rbt_insert_with_key

666 map_entry -> rbt_map_entry

667 lookup -> rbt_lookup

668 sorted -> rbt_sorted

669 tree_greater -> rbt_greater

670 tree_less -> rbt_less

671 tree_less_symbol -> rbt_less_symbol

672 union -> rbt_union

673 union_with -> rbt_union_with

674 union_with_key -> rbt_union_with_key

676 Lemmas:

677 balance_left_sorted -> balance_left_rbt_sorted

678 balance_left_tree_greater -> balance_left_rbt_greater

679 balance_left_tree_less -> balance_left_rbt_less

680 balance_right_sorted -> balance_right_rbt_sorted

681 balance_right_tree_greater -> balance_right_rbt_greater

682 balance_right_tree_less -> balance_right_rbt_less

683 balance_sorted -> balance_rbt_sorted

684 balance_tree_greater -> balance_rbt_greater

685 balance_tree_less -> balance_rbt_less

686 bulkload_is_rbt -> rbt_bulkload_is_rbt

687 combine_sorted -> combine_rbt_sorted

688 combine_tree_greater -> combine_rbt_greater

689 combine_tree_less -> combine_rbt_less

690 delete_in_tree -> rbt_delete_in_tree

691 delete_is_rbt -> rbt_delete_is_rbt

692 del_from_left_tree_greater -> rbt_del_from_left_rbt_greater

693 del_from_left_tree_less -> rbt_del_from_left_rbt_less

694 del_from_right_tree_greater -> rbt_del_from_right_rbt_greater

695 del_from_right_tree_less -> rbt_del_from_right_rbt_less

696 del_in_tree -> rbt_del_in_tree

697 del_inv1_inv2 -> rbt_del_inv1_inv2

698 del_sorted -> rbt_del_rbt_sorted

699 del_tree_greater -> rbt_del_rbt_greater

700 del_tree_less -> rbt_del_rbt_less

701 dom_lookup_Branch -> dom_rbt_lookup_Branch

702 entries_lookup -> entries_rbt_lookup

703 finite_dom_lookup -> finite_dom_rbt_lookup

704 insert_sorted -> rbt_insert_rbt_sorted

705 insertw_is_rbt -> rbt_insertw_is_rbt

706 insertwk_is_rbt -> rbt_insertwk_is_rbt

707 insertwk_sorted -> rbt_insertwk_rbt_sorted

708 insertw_sorted -> rbt_insertw_rbt_sorted

709 ins_sorted -> ins_rbt_sorted

710 ins_tree_greater -> ins_rbt_greater

711 ins_tree_less -> ins_rbt_less

712 is_rbt_sorted -> is_rbt_rbt_sorted

713 lookup_balance -> rbt_lookup_balance

714 lookup_bulkload -> rbt_lookup_rbt_bulkload

715 lookup_delete -> rbt_lookup_rbt_delete

716 lookup_Empty -> rbt_lookup_Empty

717 lookup_from_in_tree -> rbt_lookup_from_in_tree

718 lookup_in_tree -> rbt_lookup_in_tree

719 lookup_ins -> rbt_lookup_ins

720 lookup_insert -> rbt_lookup_rbt_insert

721 lookup_insertw -> rbt_lookup_rbt_insertw

722 lookup_insertwk -> rbt_lookup_rbt_insertwk

723 lookup_keys -> rbt_lookup_keys

724 lookup_map -> rbt_lookup_map

725 lookup_map_entry -> rbt_lookup_rbt_map_entry

726 lookup_tree_greater -> rbt_lookup_rbt_greater

727 lookup_tree_less -> rbt_lookup_rbt_less

728 lookup_union -> rbt_lookup_rbt_union

729 map_entry_color_of -> rbt_map_entry_color_of

730 map_entry_inv1 -> rbt_map_entry_inv1

731 map_entry_inv2 -> rbt_map_entry_inv2

732 map_entry_is_rbt -> rbt_map_entry_is_rbt

733 map_entry_sorted -> rbt_map_entry_rbt_sorted

734 map_entry_tree_greater -> rbt_map_entry_rbt_greater

735 map_entry_tree_less -> rbt_map_entry_rbt_less

736 map_tree_greater -> map_rbt_greater

737 map_tree_less -> map_rbt_less

738 map_sorted -> map_rbt_sorted

739 paint_sorted -> paint_rbt_sorted

740 paint_lookup -> paint_rbt_lookup

741 paint_tree_greater -> paint_rbt_greater

742 paint_tree_less -> paint_rbt_less

743 sorted_entries -> rbt_sorted_entries

744 tree_greater_eq_trans -> rbt_greater_eq_trans

745 tree_greater_nit -> rbt_greater_nit

746 tree_greater_prop -> rbt_greater_prop

747 tree_greater_simps -> rbt_greater_simps

748 tree_greater_trans -> rbt_greater_trans

749 tree_less_eq_trans -> rbt_less_eq_trans

750 tree_less_nit -> rbt_less_nit

751 tree_less_prop -> rbt_less_prop

752 tree_less_simps -> rbt_less_simps

753 tree_less_trans -> rbt_less_trans

754 tree_ord_props -> rbt_ord_props

755 union_Branch -> rbt_union_Branch

756 union_is_rbt -> rbt_union_is_rbt

757 unionw_is_rbt -> rbt_unionw_is_rbt

758 unionwk_is_rbt -> rbt_unionwk_is_rbt

759 unionwk_sorted -> rbt_unionwk_rbt_sorted

761 * Theory HOL/Library/Float: Floating point numbers are now defined as

762 a subset of the real numbers. All operations are defined using the

763 lifing-framework and proofs use the transfer method. INCOMPATIBILITY.

765 Changed Operations:

766 float_abs -> abs

767 float_nprt -> nprt

768 float_pprt -> pprt

769 pow2 -> use powr

770 round_down -> float_round_down

771 round_up -> float_round_up

772 scale -> exponent

774 Removed Operations:

775 ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod

777 Renamed Lemmas:

778 abs_float_def -> Float.compute_float_abs

779 bitlen_ge0 -> bitlen_nonneg

780 bitlen.simps -> Float.compute_bitlen

781 float_components -> Float_mantissa_exponent

782 float_divl.simps -> Float.compute_float_divl

783 float_divr.simps -> Float.compute_float_divr

784 float_eq_odd -> mult_powr_eq_mult_powr_iff

785 float_power -> real_of_float_power

786 lapprox_posrat_def -> Float.compute_lapprox_posrat

787 lapprox_rat.simps -> Float.compute_lapprox_rat

788 le_float_def' -> Float.compute_float_le

789 le_float_def -> less_eq_float.rep_eq

790 less_float_def' -> Float.compute_float_less

791 less_float_def -> less_float.rep_eq

792 normfloat_def -> Float.compute_normfloat

793 normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0

794 normfloat -> normfloat_def

795 normfloat_unique -> use normfloat_def

796 number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral

797 one_float_def -> Float.compute_float_one

798 plus_float_def -> Float.compute_float_plus

799 rapprox_posrat_def -> Float.compute_rapprox_posrat

800 rapprox_rat.simps -> Float.compute_rapprox_rat

801 real_of_float_0 -> zero_float.rep_eq

802 real_of_float_1 -> one_float.rep_eq

803 real_of_float_abs -> abs_float.rep_eq

804 real_of_float_add -> plus_float.rep_eq

805 real_of_float_minus -> uminus_float.rep_eq

806 real_of_float_mult -> times_float.rep_eq

807 real_of_float_simp -> Float.rep_eq

808 real_of_float_sub -> minus_float.rep_eq

809 round_down.simps -> Float.compute_float_round_down

810 round_up.simps -> Float.compute_float_round_up

811 times_float_def -> Float.compute_float_times

812 uminus_float_def -> Float.compute_float_uminus

813 zero_float_def -> Float.compute_float_zero

815 Lemmas not necessary anymore, use the transfer method:

816 bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,

817 float_divr, float_le_simp, float_less1_mantissa_bound,

818 float_less_simp, float_less_zero, float_le_zero,

819 float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,

820 floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,

821 lapprox_rat_bottom, normalized_float, rapprox_posrat,

822 rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,

823 real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,

824 round_up, zero_le_float, zero_less_float

826 * New theory HOL/Library/DAList provides an abstract type for

827 association lists with distinct keys.

829 * Session HOL/IMP: Added new theory of abstract interpretation of

830 annotated commands.

832 * Session HOL-Import: Re-implementation from scratch is faster,

833 simpler, and more scalable. Requires a proof bundle, which is

834 available as an external component. Discontinued old (and mostly

835 dead) Importer for HOL4 and HOL Light. INCOMPATIBILITY.

837 * Session HOL-Word: Discontinued many redundant theorems specific to

838 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems

839 instead.

841 word_sub_alt ~> word_sub_wi

842 word_add_alt ~> word_add_def

843 word_mult_alt ~> word_mult_def

844 word_minus_alt ~> word_minus_def

845 word_0_alt ~> word_0_wi

846 word_1_alt ~> word_1_wi

847 word_add_0 ~> add_0_left

848 word_add_0_right ~> add_0_right

849 word_mult_1 ~> mult_1_left

850 word_mult_1_right ~> mult_1_right

851 word_add_commute ~> add_commute

852 word_add_assoc ~> add_assoc

853 word_add_left_commute ~> add_left_commute

854 word_mult_commute ~> mult_commute

855 word_mult_assoc ~> mult_assoc

856 word_mult_left_commute ~> mult_left_commute

857 word_left_distrib ~> left_distrib

858 word_right_distrib ~> right_distrib

859 word_left_minus ~> left_minus

860 word_diff_0_right ~> diff_0_right

861 word_diff_self ~> diff_self

862 word_sub_def ~> diff_minus

863 word_diff_minus ~> diff_minus

864 word_add_ac ~> add_ac

865 word_mult_ac ~> mult_ac

866 word_plus_ac0 ~> add_0_left add_0_right add_ac

867 word_times_ac1 ~> mult_1_left mult_1_right mult_ac

868 word_order_trans ~> order_trans

869 word_order_refl ~> order_refl

870 word_order_antisym ~> order_antisym

871 word_order_linear ~> linorder_linear

872 lenw1_zero_neq_one ~> zero_neq_one

873 word_number_of_eq ~> number_of_eq

874 word_of_int_add_hom ~> wi_hom_add

875 word_of_int_sub_hom ~> wi_hom_sub

876 word_of_int_mult_hom ~> wi_hom_mult

877 word_of_int_minus_hom ~> wi_hom_neg

878 word_of_int_succ_hom ~> wi_hom_succ

879 word_of_int_pred_hom ~> wi_hom_pred

880 word_of_int_0_hom ~> word_0_wi

881 word_of_int_1_hom ~> word_1_wi

883 * Session HOL-Word: New proof method "word_bitwise" for splitting

884 machine word equalities and inequalities into logical circuits,

885 defined in HOL/Word/WordBitwise.thy. Supports addition, subtraction,

886 multiplication, shifting by constants, bitwise operators and numeric

887 constants. Requires fixed-length word types, not 'a word. Solves

888 many standard word identities outright and converts more into first

889 order problems amenable to blast or similar. See also examples in

890 HOL/Word/Examples/WordExamples.thy.

892 * Session HOL-Probability: Introduced the type "'a measure" to

893 represent measures, this replaces the records 'a algebra and 'a

894 measure_space. The locales based on subset_class now have two

895 locale-parameters the space \<Omega> and the set of measurable sets M.

896 The product of probability spaces uses now the same constant as the

897 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)

898 measure". Most constants are defined now outside of locales and gain

899 an additional parameter, like null_sets, almost_eventually or \<mu>'.

900 Measure space constructions for distributions and densities now got

901 their own constants distr and density. Instead of using locales to

902 describe measure spaces with a finite space, the measure count_space

903 and point_measure is introduced. INCOMPATIBILITY.

905 Renamed constants:

906 measure -> emeasure

907 finite_measure.\<mu>' -> measure

908 product_algebra_generator -> prod_algebra

909 product_prob_space.emb -> prod_emb

910 product_prob_space.infprod_algebra -> PiM

912 Removed locales:

913 completeable_measure_space

914 finite_measure_space

915 finite_prob_space

916 finite_product_finite_prob_space

917 finite_product_sigma_algebra

918 finite_sigma_algebra

919 measure_space

920 pair_finite_prob_space

921 pair_finite_sigma_algebra

922 pair_finite_space

923 pair_sigma_algebra

924 product_sigma_algebra

926 Removed constants:

927 conditional_space

928 distribution -> use distr measure, or distributed predicate

929 image_space

930 joint_distribution -> use distr measure, or distributed predicate

931 pair_measure_generator

932 product_prob_space.infprod_algebra -> use PiM

933 subvimage

935 Replacement theorems:

936 finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite

937 finite_measure.empty_measure -> measure_empty

938 finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq

939 finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq

940 finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably

941 finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure

942 finite_measure.finite_measure -> finite_measure.emeasure_finite

943 finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton

944 finite_measure.positive_measure' -> measure_nonneg

945 finite_measure.real_measure -> finite_measure.emeasure_real

946 finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb

947 finite_product_sigma_algebra.in_P -> sets_PiM_I_finite

948 finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty

949 information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed

950 information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple

951 information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple

952 information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple

953 information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple

954 information_space.entropy_commute -> information_space.entropy_commute_simple

955 information_space.entropy_eq -> information_space.entropy_simple_distributed

956 information_space.entropy_generic_eq -> information_space.entropy_simple_distributed

957 information_space.entropy_positive -> information_space.entropy_nonneg_simple

958 information_space.entropy_uniform_max -> information_space.entropy_uniform

959 information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq

960 information_space.KL_eq_0 -> information_space.KL_same_eq_0

961 information_space.KL_ge_0 -> information_space.KL_nonneg

962 information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed

963 information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple

964 Int_stable_cuboids -> Int_stable_atLeastAtMost

965 Int_stable_product_algebra_generator -> positive_integral

966 measure_preserving -> equality "distr M N f = N" "f : measurable M N"

967 measure_space.additive -> emeasure_additive

968 measure_space.AE_iff_null_set -> AE_iff_null

969 measure_space.almost_everywhere_def -> eventually_ae_filter

970 measure_space.almost_everywhere_vimage -> AE_distrD

971 measure_space.continuity_from_above -> INF_emeasure_decseq

972 measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq

973 measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq

974 measure_space.continuity_from_below -> SUP_emeasure_incseq

975 measure_space_density -> emeasure_density

976 measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density

977 measure_space.integrable_vimage -> integrable_distr

978 measure_space.integral_translated_density -> integral_density

979 measure_space.integral_vimage -> integral_distr

980 measure_space.measure_additive -> plus_emeasure

981 measure_space.measure_compl -> emeasure_compl

982 measure_space.measure_countable_increasing -> emeasure_countable_increasing

983 measure_space.measure_countably_subadditive -> emeasure_subadditive_countably

984 measure_space.measure_decseq -> decseq_emeasure

985 measure_space.measure_Diff -> emeasure_Diff

986 measure_space.measure_Diff_null_set -> emeasure_Diff_null_set

987 measure_space.measure_eq_0 -> emeasure_eq_0

988 measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite

989 measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton

990 measure_space.measure_incseq -> incseq_emeasure

991 measure_space.measure_insert -> emeasure_insert

992 measure_space.measure_mono -> emeasure_mono

993 measure_space.measure_not_negative -> emeasure_not_MInf

994 measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq

995 measure_space.measure_setsum -> setsum_emeasure

996 measure_space.measure_setsum_split -> setsum_emeasure_cover

997 measure_space.measure_space_vimage -> emeasure_distr

998 measure_space.measure_subadditive_finite -> emeasure_subadditive_finite

999 measure_space.measure_subadditive -> subadditive

1000 measure_space.measure_top -> emeasure_space

1001 measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0

1002 measure_space.measure_Un_null_set -> emeasure_Un_null_set

1003 measure_space.positive_integral_translated_density -> positive_integral_density

1004 measure_space.positive_integral_vimage -> positive_integral_distr

1005 measure_space.real_continuity_from_above -> Lim_measure_decseq

1006 measure_space.real_continuity_from_below -> Lim_measure_incseq

1007 measure_space.real_measure_countably_subadditive -> measure_subadditive_countably

1008 measure_space.real_measure_Diff -> measure_Diff

1009 measure_space.real_measure_finite_Union -> measure_finite_Union

1010 measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton

1011 measure_space.real_measure_subadditive -> measure_subadditive

1012 measure_space.real_measure_Union -> measure_Union

1013 measure_space.real_measure_UNION -> measure_UNION

1014 measure_space.simple_function_vimage -> simple_function_comp

1015 measure_space.simple_integral_vimage -> simple_integral_distr

1016 measure_space.simple_integral_vimage -> simple_integral_distr

1017 measure_unique_Int_stable -> measure_eqI_generator_eq

1018 measure_unique_Int_stable_vimage -> measure_eqI_generator_eq

1019 pair_sigma_algebra.measurable_cut_fst -> sets_Pair1

1020 pair_sigma_algebra.measurable_cut_snd -> sets_Pair2

1021 pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1

1022 pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2

1023 pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff

1024 pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap

1025 pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'

1026 pair_sigma_algebra.sets_swap -> sets_pair_swap

1027 pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1

1028 pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2

1029 pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap

1030 pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2

1031 pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt

1032 pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times

1033 prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM

1034 prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq

1035 prob_space.measure_space_1 -> prob_space.emeasure_space_1

1036 prob_space.prob_space_vimage -> prob_space_distr

1037 prob_space.random_variable_restrict -> measurable_restrict

1038 prob_space_unique_Int_stable -> measure_eqI_prob_space

1039 product_algebraE -> prod_algebraE_all

1040 product_algebra_generator_der -> prod_algebra_eq_finite

1041 product_algebra_generator_into_space -> prod_algebra_sets_into_space

1042 product_algebraI -> sets_PiM_I_finite

1043 product_measure_exists -> product_sigma_finite.sigma_finite

1044 product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator

1045 product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb

1046 product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty

1047 product_prob_space.measurable_component -> measurable_component_singleton

1048 product_prob_space.measurable_emb -> measurable_prod_emb

1049 product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single

1050 product_prob_space.measurable_singleton_infprod -> measurable_component_singleton

1051 product_prob_space.measure_emb -> emeasure_prod_emb

1052 product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict

1053 product_sigma_algebra.product_algebra_into_space -> space_closed

1054 product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge

1055 product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton

1056 product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge

1057 sequence_space.measure_infprod -> sequence_space.measure_PiM_countable

1058 sets_product_algebra -> sets_PiM

1059 sigma_algebra.measurable_sigma -> measurable_measure_of

1060 sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint

1061 sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr

1062 sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq

1063 space_product_algebra -> space_PiM

1065 * Session HOL-TPTP: support to parse and import TPTP problems (all

1066 languages) into Isabelle/HOL.

1069 *** FOL ***

1071 * New "case_product" attribute (see HOL).

1074 *** ZF ***

1076 * Greater support for structured proofs involving induction or case

1077 analysis.

1079 * Much greater use of mathematical symbols.

1081 * Removal of many ML theorem bindings. INCOMPATIBILITY.

1084 *** ML ***

1086 * Antiquotation @{keyword "name"} produces a parser for outer syntax

1087 from a minor keyword introduced via theory header declaration.

1089 * Antiquotation @{command_spec "name"} produces the

1090 Outer_Syntax.command_spec from a major keyword introduced via theory

1091 header declaration; it can be passed to Outer_Syntax.command etc.

1093 * Local_Theory.define no longer hard-wires default theorem name

1094 "foo_def", but retains the binding as given. If that is Binding.empty

1095 / Attrib.empty_binding, the result is not registered as user-level

1096 fact. The Local_Theory.define_internal variant allows to specify a

1097 non-empty name (used for the foundation in the background theory),

1098 while omitting the fact binding in the user-context. Potential

1099 INCOMPATIBILITY for derived definitional packages: need to specify

1100 naming policy for primitive definitions more explicitly.

1102 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in

1103 conformance with similar operations in structure Term and Logic.

1105 * Antiquotation @{attributes [...]} embeds attribute source

1106 representation into the ML text, which is particularly useful with

1107 declarations like Local_Theory.note.

1109 * Structure Proof_Context follows standard naming scheme. Old

1110 ProofContext has been discontinued. INCOMPATIBILITY.

1112 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle

1113 change of semantics: update is applied to auxiliary local theory

1114 context as well.

1116 * Modernized some old-style infix operations:

1118 addeqcongs ~> Simplifier.add_eqcong

1119 deleqcongs ~> Simplifier.del_eqcong

1120 addcongs ~> Simplifier.add_cong

1121 delcongs ~> Simplifier.del_cong

1122 setmksimps ~> Simplifier.set_mksimps

1123 setmkcong ~> Simplifier.set_mkcong

1124 setmksym ~> Simplifier.set_mksym

1125 setmkeqTrue ~> Simplifier.set_mkeqTrue

1126 settermless ~> Simplifier.set_termless

1127 setsubgoaler ~> Simplifier.set_subgoaler

1128 addsplits ~> Splitter.add_split

1129 delsplits ~> Splitter.del_split

1132 *** System ***

1134 * USER_HOME settings variable points to cross-platform user home

1135 directory, which coincides with HOME on POSIX systems only. Likewise,

1136 the Isabelle path specification "~" now expands to $USER_HOME, instead

1137 of former $HOME. A different default for USER_HOME may be set

1138 explicitly in shell environment, before Isabelle settings are

1139 evaluated. Minor INCOMPATIBILITY: need to adapt Isabelle path where

1140 the generic user home was intended.

1142 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name

1143 notation, which is useful for the jEdit file browser, for example.

1145 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar

1146 (not just JRE).

1150 New in Isabelle2011-1 (October 2011)

1151 ------------------------------------

1153 *** General ***

1155 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as

1156 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.

1158 - Management of multiple theory files directly from the editor

1159 buffer store -- bypassing the file-system (no requirement to save

1160 files for checking).

1162 - Markup of formal entities within the text buffer, with semantic

1163 highlighting, tooltips and hyperlinks to jump to defining source

1164 positions.

1166 - Improved text rendering, with sub/superscripts in the source

1167 buffer (including support for copy/paste wrt. output panel, HTML

1168 theory output and other non-Isabelle text boxes).

1170 - Refined scheduling of proof checking and printing of results,

1171 based on interactive editor view. (Note: jEdit folding and

1172 narrowing allows to restrict buffer perspectives explicitly.)

1174 - Reduced CPU performance requirements, usable on machines with few

1175 cores.

1177 - Reduced memory requirements due to pruning of unused document

1178 versions (garbage collection).

1180 See also ~~/src/Tools/jEdit/README.html for further information,

1181 including some remaining limitations.

1183 * Theory loader: source files are exclusively located via the master

1184 directory of each theory node (where the .thy file itself resides).

1185 The global load path (such as src/HOL/Library) has been discontinued.

1186 Note that the path element ~~ may be used to reference theories in the

1187 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".

1188 INCOMPATIBILITY.

1190 * Theory loader: source files are identified by content via SHA1

1191 digests. Discontinued former path/modtime identification and optional

1192 ISABELLE_FILE_IDENT plugin scripts.

1194 * Parallelization of nested Isar proofs is subject to

1195 Goal.parallel_proofs_threshold (default 100). See also isabelle

1196 usedir option -Q.

1198 * Name space: former unsynchronized references are now proper

1199 configuration options, with more conventional names:

1201 long_names ~> names_long

1202 short_names ~> names_short

1203 unique_names ~> names_unique

1205 Minor INCOMPATIBILITY, need to declare options in context like this:

1207 declare [[names_unique = false]]

1209 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`. Note

1210 that the result needs to be unique, which means fact specifications

1211 may have to be refined after enriching a proof context.

1213 * Attribute "case_names" has been refined: the assumptions in each case

1214 can be named now by following the case name with [name1 name2 ...].

1216 * Isabelle/Isar reference manual has been updated and extended:

1217 - "Synopsis" provides a catalog of main Isar language concepts.

1218 - Formal references in syntax diagrams, via @{rail} antiquotation.

1219 - Updated material from classic "ref" manual, notably about

1220 "Classical Reasoner".

1223 *** HOL ***

1225 * Class bot and top require underlying partial order rather than

1226 preorder: uniqueness of bot and top is guaranteed. INCOMPATIBILITY.

1228 * Class complete_lattice: generalized a couple of lemmas from sets;

1229 generalized theorems INF_cong and SUP_cong. New type classes for

1230 complete boolean algebras and complete linear orders. Lemmas

1231 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in

1232 class complete_linorder.

1234 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,

1235 Sup_fun_def, Inf_apply, Sup_apply.

1237 Removed redundant lemmas (the right hand side gives hints how to

1238 replace them for (metis ...), or (simp only: ...) proofs):

1240 Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]

1241 Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]

1242 Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right

1243 Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right

1244 Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right

1245 Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right

1246 Inter_def ~> INF_def, image_def

1247 Union_def ~> SUP_def, image_def

1248 INT_eq ~> INF_def, and image_def

1249 UN_eq ~> SUP_def, and image_def

1250 INF_subset ~> INF_superset_mono [OF _ order_refl]

1252 More consistent and comprehensive names:

1254 INTER_eq_Inter_image ~> INF_def

1255 UNION_eq_Union_image ~> SUP_def

1256 INFI_def ~> INF_def

1257 SUPR_def ~> SUP_def

1258 INF_leI ~> INF_lower

1259 INF_leI2 ~> INF_lower2

1260 le_INFI ~> INF_greatest

1261 le_SUPI ~> SUP_upper

1262 le_SUPI2 ~> SUP_upper2

1263 SUP_leI ~> SUP_least

1264 INFI_bool_eq ~> INF_bool_eq

1265 SUPR_bool_eq ~> SUP_bool_eq

1266 INFI_apply ~> INF_apply

1267 SUPR_apply ~> SUP_apply

1268 INTER_def ~> INTER_eq

1269 UNION_def ~> UNION_eq

1271 INCOMPATIBILITY.

1273 * Renamed theory Complete_Lattice to Complete_Lattices.

1274 INCOMPATIBILITY.

1276 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,

1277 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,

1278 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,

1279 Sup_insert are now declared as [simp]. INCOMPATIBILITY.

1281 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,

1282 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,

1283 sup_inf_absob, sup_left_idem are now declared as [simp]. Minor

1284 INCOMPATIBILITY.

1286 * Added syntactic classes "inf" and "sup" for the respective

1287 constants. INCOMPATIBILITY: Changes in the argument order of the

1288 (mostly internal) locale predicates for some derived classes.

1290 * Theorem collections ball_simps and bex_simps do not contain theorems

1291 referring to UNION any longer; these have been moved to collection

1292 UN_ball_bex_simps. INCOMPATIBILITY.

1294 * Theory Archimedean_Field: floor now is defined as parameter of a

1295 separate type class floor_ceiling.

1297 * Theory Finite_Set: more coherent development of fold_set locales:

1299 locale fun_left_comm ~> locale comp_fun_commute

1300 locale fun_left_comm_idem ~> locale comp_fun_idem

1302 Both use point-free characterization; interpretation proofs may need

1303 adjustment. INCOMPATIBILITY.

1305 * Theory Limits: Type "'a net" has been renamed to "'a filter", in

1306 accordance with standard mathematical terminology. INCOMPATIBILITY.

1308 * Theory Complex_Main: The locale interpretations for the

1309 bounded_linear and bounded_bilinear locales have been removed, in

1310 order to reduce the number of duplicate lemmas. Users must use the

1311 original names for distributivity theorems, potential INCOMPATIBILITY.

1313 divide.add ~> add_divide_distrib

1314 divide.diff ~> diff_divide_distrib

1315 divide.setsum ~> setsum_divide_distrib

1316 mult.add_right ~> right_distrib

1317 mult.diff_right ~> right_diff_distrib

1318 mult_right.setsum ~> setsum_right_distrib

1319 mult_left.diff ~> left_diff_distrib

1321 * Theory Complex_Main: Several redundant theorems have been removed or

1322 replaced by more general versions. INCOMPATIBILITY.

1324 real_diff_def ~> minus_real_def

1325 real_divide_def ~> divide_real_def

1326 real_less_def ~> less_le

1327 real_abs_def ~> abs_real_def

1328 real_sgn_def ~> sgn_real_def

1329 real_mult_commute ~> mult_commute

1330 real_mult_assoc ~> mult_assoc

1331 real_mult_1 ~> mult_1_left

1332 real_add_mult_distrib ~> left_distrib

1333 real_zero_not_eq_one ~> zero_neq_one

1334 real_mult_inverse_left ~> left_inverse

1335 INVERSE_ZERO ~> inverse_zero

1336 real_le_refl ~> order_refl

1337 real_le_antisym ~> order_antisym

1338 real_le_trans ~> order_trans

1339 real_le_linear ~> linear

1340 real_le_eq_diff ~> le_iff_diff_le_0

1341 real_add_left_mono ~> add_left_mono

1342 real_mult_order ~> mult_pos_pos

1343 real_mult_less_mono2 ~> mult_strict_left_mono

1344 real_of_int_real_of_nat ~> real_of_int_of_nat_eq

1345 real_0_le_divide_iff ~> zero_le_divide_iff

1346 realpow_two_disj ~> power2_eq_iff

1347 real_squared_diff_one_factored ~> square_diff_one_factored

1348 realpow_two_diff ~> square_diff_square_factored

1349 reals_complete2 ~> complete_real

1350 real_sum_squared_expand ~> power2_sum

1351 exp_ln_eq ~> ln_unique

1352 expi_add ~> exp_add

1353 expi_zero ~> exp_zero

1354 lemma_DERIV_subst ~> DERIV_cong

1355 LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff

1356 LIMSEQ_const ~> tendsto_const

1357 LIMSEQ_norm ~> tendsto_norm

1358 LIMSEQ_add ~> tendsto_add

1359 LIMSEQ_minus ~> tendsto_minus

1360 LIMSEQ_minus_cancel ~> tendsto_minus_cancel

1361 LIMSEQ_diff ~> tendsto_diff

1362 bounded_linear.LIMSEQ ~> bounded_linear.tendsto

1363 bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto

1364 LIMSEQ_mult ~> tendsto_mult

1365 LIMSEQ_inverse ~> tendsto_inverse

1366 LIMSEQ_divide ~> tendsto_divide

1367 LIMSEQ_pow ~> tendsto_power

1368 LIMSEQ_setsum ~> tendsto_setsum

1369 LIMSEQ_setprod ~> tendsto_setprod

1370 LIMSEQ_norm_zero ~> tendsto_norm_zero_iff

1371 LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff

1372 LIMSEQ_imp_rabs ~> tendsto_rabs

1373 LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]

1374 LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]

1375 LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]

1376 LIMSEQ_Complex ~> tendsto_Complex

1377 LIM_ident ~> tendsto_ident_at

1378 LIM_const ~> tendsto_const

1379 LIM_add ~> tendsto_add

1380 LIM_add_zero ~> tendsto_add_zero

1381 LIM_minus ~> tendsto_minus

1382 LIM_diff ~> tendsto_diff

1383 LIM_norm ~> tendsto_norm

1384 LIM_norm_zero ~> tendsto_norm_zero

1385 LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel

1386 LIM_norm_zero_iff ~> tendsto_norm_zero_iff

1387 LIM_rabs ~> tendsto_rabs

1388 LIM_rabs_zero ~> tendsto_rabs_zero

1389 LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel

1390 LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff

1391 LIM_compose ~> tendsto_compose

1392 LIM_mult ~> tendsto_mult

1393 LIM_scaleR ~> tendsto_scaleR

1394 LIM_of_real ~> tendsto_of_real

1395 LIM_power ~> tendsto_power

1396 LIM_inverse ~> tendsto_inverse

1397 LIM_sgn ~> tendsto_sgn

1398 isCont_LIM_compose ~> isCont_tendsto_compose

1399 bounded_linear.LIM ~> bounded_linear.tendsto

1400 bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero

1401 bounded_bilinear.LIM ~> bounded_bilinear.tendsto

1402 bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero

1403 bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero

1404 bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero

1405 LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]

1407 * Theory Complex_Main: The definition of infinite series was

1408 generalized. Now it is defined on the type class {topological_space,

1409 comm_monoid_add}. Hence it is useable also for extended real numbers.

1411 * Theory Complex_Main: The complex exponential function "expi" is now

1412 a type-constrained abbreviation for "exp :: complex => complex"; thus

1413 several polymorphic lemmas about "exp" are now applicable to "expi".

1415 * Code generation:

1417 - Theory Library/Code_Char_ord provides native ordering of

1418 characters in the target language.

1420 - Commands code_module and code_library are legacy, use export_code

1421 instead.

1423 - Method "evaluation" is legacy, use method "eval" instead.

1425 - Legacy evaluator "SML" is deactivated by default. May be

1426 reactivated by the following theory command:

1428 setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}

1430 * Declare ext [intro] by default. Rare INCOMPATIBILITY.

1432 * New proof method "induction" that gives induction hypotheses the

1433 name "IH", thus distinguishing them from further hypotheses that come

1434 from rule induction. The latter are still called "hyps". Method

1435 "induction" is a thin wrapper around "induct" and follows the same

1436 syntax.

1438 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is

1439 still available as a legacy feature for some time.

1441 * Nitpick:

1442 - Added "need" and "total_consts" options.

1443 - Reintroduced "show_skolems" option by popular demand.

1444 - Renamed attribute: nitpick_def ~> nitpick_unfold.

1445 INCOMPATIBILITY.

1447 * Sledgehammer:

1448 - Use quasi-sound (and efficient) translations by default.

1449 - Added support for the following provers: E-ToFoF, LEO-II,

1450 Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.

1451 - Automatically preplay and minimize proofs before showing them if

1452 this can be done within reasonable time.

1453 - sledgehammer available_provers ~> sledgehammer supported_provers.

1454 INCOMPATIBILITY.

1455 - Added "preplay_timeout", "slicing", "type_enc", "sound",

1456 "max_mono_iters", and "max_new_mono_instances" options.

1457 - Removed "explicit_apply" and "full_types" options as well as "Full

1458 Types" Proof General menu item. INCOMPATIBILITY.

1460 * Metis:

1461 - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.

1462 - Obsoleted "metisFT" -- use "metis (full_types)" instead.

1463 INCOMPATIBILITY.

1465 * Command 'try':

1466 - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and

1467 "elim:" options. INCOMPATIBILITY.

1468 - Introduced 'try' that not only runs 'try_methods' but also

1469 'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.

1471 * Quickcheck:

1472 - Added "eval" option to evaluate terms for the found counterexample

1473 (currently only supported by the default (exhaustive) tester).

1474 - Added post-processing of terms to obtain readable counterexamples

1475 (currently only supported by the default (exhaustive) tester).

1476 - New counterexample generator quickcheck[narrowing] enables

1477 narrowing-based testing. Requires the Glasgow Haskell compiler

1478 with its installation location defined in the Isabelle settings

1479 environment as ISABELLE_GHC.

1480 - Removed quickcheck tester "SML" based on the SML code generator

1481 (formly in HOL/Library).

1483 * Function package: discontinued option "tailrec". INCOMPATIBILITY,

1484 use 'partial_function' instead.

1486 * Theory Library/Extended_Reals replaces now the positive extended

1487 reals found in probability theory. This file is extended by

1488 Multivariate_Analysis/Extended_Real_Limits.

1490 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,

1491 from where it must be imported explicitly if it is really required.

1492 INCOMPATIBILITY.

1494 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has

1495 been moved here. INCOMPATIBILITY.

1497 * Theory Library/Saturated provides type of numbers with saturated

1498 arithmetic.

1500 * Theory Library/Product_Lattice defines a pointwise ordering for the

1501 product type 'a * 'b, and provides instance proofs for various order

1502 and lattice type classes.

1504 * Theory Library/Countable now provides the "countable_datatype" proof

1505 method for proving "countable" class instances for datatypes.

1507 * Theory Library/Cset_Monad allows do notation for computable sets

1508 (cset) via the generic monad ad-hoc overloading facility.

1510 * Library: Theories of common data structures are split into theories

1511 for implementation, an invariant-ensuring type, and connection to an

1512 abstract type. INCOMPATIBILITY.

1514 - RBT is split into RBT and RBT_Mapping.

1515 - AssocList is split and renamed into AList and AList_Mapping.

1516 - DList is split into DList_Impl, DList, and DList_Cset.

1517 - Cset is split into Cset and List_Cset.

1519 * Theory Library/Nat_Infinity has been renamed to

1520 Library/Extended_Nat, with name changes of the following types and

1521 constants:

1523 type inat ~> type enat

1524 Fin ~> enat

1525 Infty ~> infinity (overloaded)

1526 iSuc ~> eSuc

1527 the_Fin ~> the_enat

1529 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has

1530 been renamed accordingly. INCOMPATIBILITY.

1532 * Session Multivariate_Analysis: The euclidean_space type class now

1533 fixes a constant "Basis :: 'a set" consisting of the standard

1534 orthonormal basis for the type. Users now have the option of

1535 quantifying over this set instead of using the "basis" function, e.g.

1536 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".

1538 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed

1539 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants

1540 "Cart_nth" and "Cart_lambda" have been respectively renamed to

1541 "vec_nth" and "vec_lambda"; theorems mentioning those names have

1542 changed to match. Definition theorems for overloaded constants now use

1543 the standard "foo_vec_def" naming scheme. A few other theorems have

1544 been renamed as follows (INCOMPATIBILITY):

1546 Cart_eq ~> vec_eq_iff

1547 dist_nth_le_cart ~> dist_vec_nth_le

1548 tendsto_vector ~> vec_tendstoI

1549 Cauchy_vector ~> vec_CauchyI

1551 * Session Multivariate_Analysis: Several duplicate theorems have been

1552 removed, and other theorems have been renamed or replaced with more

1553 general versions. INCOMPATIBILITY.

1555 finite_choice ~> finite_set_choice

1556 eventually_conjI ~> eventually_conj

1557 eventually_and ~> eventually_conj_iff

1558 eventually_false ~> eventually_False

1559 setsum_norm ~> norm_setsum

1560 Lim_sequentially ~> LIMSEQ_def

1561 Lim_ident_at ~> LIM_ident

1562 Lim_const ~> tendsto_const

1563 Lim_cmul ~> tendsto_scaleR [OF tendsto_const]

1564 Lim_neg ~> tendsto_minus

1565 Lim_add ~> tendsto_add

1566 Lim_sub ~> tendsto_diff

1567 Lim_mul ~> tendsto_scaleR

1568 Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]

1569 Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]

1570 Lim_linear ~> bounded_linear.tendsto

1571 Lim_component ~> tendsto_euclidean_component

1572 Lim_component_cart ~> tendsto_vec_nth

1573 Lim_inner ~> tendsto_inner [OF tendsto_const]

1574 dot_lsum ~> inner_setsum_left

1575 dot_rsum ~> inner_setsum_right

1576 continuous_cmul ~> continuous_scaleR [OF continuous_const]

1577 continuous_neg ~> continuous_minus

1578 continuous_sub ~> continuous_diff

1579 continuous_vmul ~> continuous_scaleR [OF _ continuous_const]

1580 continuous_mul ~> continuous_scaleR

1581 continuous_inv ~> continuous_inverse

1582 continuous_at_within_inv ~> continuous_at_within_inverse

1583 continuous_at_inv ~> continuous_at_inverse

1584 continuous_at_norm ~> continuous_norm [OF continuous_at_id]

1585 continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]

1586 continuous_at_component ~> continuous_component [OF continuous_at_id]

1587 continuous_on_neg ~> continuous_on_minus

1588 continuous_on_sub ~> continuous_on_diff

1589 continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]

1590 continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]

1591 continuous_on_mul ~> continuous_on_scaleR

1592 continuous_on_mul_real ~> continuous_on_mult

1593 continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]

1594 continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]

1595 continuous_on_inverse ~> continuous_on_inv

1596 uniformly_continuous_on_neg ~> uniformly_continuous_on_minus

1597 uniformly_continuous_on_sub ~> uniformly_continuous_on_diff

1598 subset_interior ~> interior_mono

1599 subset_closure ~> closure_mono

1600 closure_univ ~> closure_UNIV

1601 real_arch_lt ~> reals_Archimedean2

1602 real_arch ~> reals_Archimedean3

1603 real_abs_norm ~> abs_norm_cancel

1604 real_abs_sub_norm ~> norm_triangle_ineq3

1605 norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2

1607 * Session HOL-Probability:

1608 - Caratheodory's extension lemma is now proved for ring_of_sets.

1609 - Infinite products of probability measures are now available.

1610 - Sigma closure is independent, if the generator is independent

1611 - Use extended reals instead of positive extended

1612 reals. INCOMPATIBILITY.

1614 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.

1616 expand_fun_below ~> fun_below_iff

1617 below_fun_ext ~> fun_belowI

1618 expand_cfun_eq ~> cfun_eq_iff

1619 ext_cfun ~> cfun_eqI

1620 expand_cfun_below ~> cfun_below_iff

1621 below_cfun_ext ~> cfun_belowI

1622 monofun_fun_fun ~> fun_belowD

1623 monofun_fun_arg ~> monofunE

1624 monofun_lub_fun ~> adm_monofun [THEN admD]

1625 cont_lub_fun ~> adm_cont [THEN admD]

1626 cont2cont_Rep_CFun ~> cont2cont_APP

1627 cont_Rep_CFun_app ~> cont_APP_app

1628 cont_Rep_CFun_app_app ~> cont_APP_app_app

1629 cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]

1630 cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]

1631 contlub_cfun ~> lub_APP [symmetric]

1632 contlub_LAM ~> lub_LAM [symmetric]

1633 thelubI ~> lub_eqI

1634 UU_I ~> bottomI

1635 lift_distinct1 ~> lift.distinct(1)

1636 lift_distinct2 ~> lift.distinct(2)

1637 Def_not_UU ~> lift.distinct(2)

1638 Def_inject ~> lift.inject

1639 below_UU_iff ~> below_bottom_iff

1640 eq_UU_iff ~> eq_bottom_iff

1643 *** Document preparation ***

1645 * Antiquotation @{rail} layouts railroad syntax diagrams, see also

1646 isar-ref manual, both for description and actual application of the

1647 same.

1649 * Antiquotation @{value} evaluates the given term and presents its

1650 result.

1652 * Antiquotations: term style "isub" provides ad-hoc conversion of

1653 variables x1, y23 into subscripted form x\<^isub>1,

1654 y\<^isub>2\<^isub>3.

1656 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>

1657 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).

1659 * Localized \isabellestyle switch can be used within blocks or groups

1660 like this:

1662 \isabellestyle{it} %preferred default

1663 {\isabellestylett @{text "typewriter stuff"}}

1665 * Discontinued special treatment of hard tabulators. Implicit

1666 tab-width is now defined as 1. Potential INCOMPATIBILITY for visual

1667 layouts.

1670 *** ML ***

1672 * The inner syntax of sort/type/term/prop supports inlined YXML

1673 representations within quoted string tokens. By encoding logical

1674 entities via Term_XML (in ML or Scala) concrete syntax can be

1675 bypassed, which is particularly useful for producing bits of text

1676 under external program control.

1678 * Antiquotations for ML and document preparation are managed as theory

1679 data, which requires explicit setup.

1681 * Isabelle_Process.is_active allows tools to check if the official

1682 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop

1683 (better known as Proof General).

1685 * Structure Proof_Context follows standard naming scheme. Old

1686 ProofContext is still available for some time as legacy alias.

1688 * Structure Timing provides various operations for timing; supersedes

1689 former start_timing/end_timing etc.

1691 * Path.print is the official way to show file-system paths to users

1692 (including quotes etc.).

1694 * Inner syntax: identifiers in parse trees of generic categories

1695 "logic", "aprop", "idt" etc. carry position information (disguised as

1696 type constraints). Occasional INCOMPATIBILITY with non-compliant

1697 translations that choke on unexpected type constraints. Positions can

1698 be stripped in ML translations via Syntax.strip_positions /

1699 Syntax.strip_positions_ast, or via the syntax constant

1700 "_strip_positions" within parse trees. As last resort, positions can

1701 be disabled via the configuration option Syntax.positions, which is

1702 called "syntax_positions" in Isar attribute syntax.

1704 * Discontinued special status of various ML structures that contribute

1705 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less

1706 pervasive content, no inclusion in structure Syntax. INCOMPATIBILITY,

1707 refer directly to Ast.Constant, Lexicon.is_identifier,

1708 Syntax_Trans.mk_binder_tr etc.

1710 * Typed print translation: discontinued show_sorts argument, which is

1711 already available via context of "advanced" translation.

1713 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic

1714 goal states; body tactic needs to address all subgoals uniformly.

1716 * Slightly more special eq_list/eq_set, with shortcut involving

1717 pointer equality (assumes that eq relation is reflexive).

1719 * Classical tactics use proper Proof.context instead of historic types

1720 claset/clasimpset. Old-style declarations like addIs, addEs, addDs

1721 operate directly on Proof.context. Raw type claset retains its use as

1722 snapshot of the classical context, which can be recovered via

1723 (put_claset HOL_cs) etc. Type clasimpset has been discontinued.

1724 INCOMPATIBILITY, classical tactics and derived proof methods require

1725 proper Proof.context.

1728 *** System ***

1730 * Discontinued support for Poly/ML 5.2, which was the last version

1731 without proper multithreading and TimeLimit implementation.

1733 * Discontinued old lib/scripts/polyml-platform, which has been

1734 obsolete since Isabelle2009-2.

1736 * Various optional external tools are referenced more robustly and

1737 uniformly by explicit Isabelle settings as follows:

1739 ISABELLE_CSDP (formerly CSDP_EXE)

1740 ISABELLE_GHC (formerly EXEC_GHC or GHC_PATH)

1741 ISABELLE_OCAML (formerly EXEC_OCAML)

1742 ISABELLE_SWIPL (formerly EXEC_SWIPL)

1743 ISABELLE_YAP (formerly EXEC_YAP)

1745 Note that automated detection from the file-system or search path has

1746 been discontinued. INCOMPATIBILITY.

1748 * Scala layer provides JVM method invocation service for static

1749 methods of type (String)String, see Invoke_Scala.method in ML. For

1750 example:

1752 Invoke_Scala.method "java.lang.System.getProperty" "java.home"

1754 Together with YXML.string_of_body/parse_body and XML.Encode/Decode

1755 this allows to pass structured values between ML and Scala.

1757 * The IsabelleText fonts includes some further glyphs to support the

1758 Prover IDE. Potential INCOMPATIBILITY: users who happen to have

1759 installed a local copy (which is normally *not* required) need to

1760 delete or update it from ~~/lib/fonts/.

1764 New in Isabelle2011 (January 2011)

1765 ----------------------------------

1767 *** General ***

1769 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see

1770 src/Tools/jEdit). This also serves as IDE for Isabelle/ML, with

1771 useful tooltips and hyperlinks produced from its static analysis. The

1772 bundled component provides an executable Isabelle tool that can be run

1773 like this:

1775 Isabelle2011/bin/isabelle jedit

1777 * Significantly improved Isabelle/Isar implementation manual.

1779 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER

1780 (and thus refers to something like $HOME/.isabelle/Isabelle2011),

1781 while the default heap location within that directory lacks that extra

1782 suffix. This isolates multiple Isabelle installations from each

1783 other, avoiding problems with old settings in new versions.

1784 INCOMPATIBILITY, need to copy/upgrade old user settings manually.

1786 * Source files are always encoded as UTF-8, instead of old-fashioned

1787 ISO-Latin-1. INCOMPATIBILITY. Isabelle LaTeX documents might require

1788 the following package declarations:

1790 \usepackage[utf8]{inputenc}

1791 \usepackage{textcomp}

1793 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that

1794 a Unicode character is treated as a single symbol, not a sequence of

1795 non-ASCII bytes as before. Since Isabelle/ML string literals may

1796 contain symbols without further backslash escapes, Unicode can now be

1797 used here as well. Recall that Symbol.explode in ML provides a

1798 consistent view on symbols, while raw explode (or String.explode)

1799 merely give a byte-oriented representation.

1801 * Theory loader: source files are primarily located via the master

1802 directory of each theory node (where the .thy file itself resides).

1803 The global load path is still partially available as legacy feature.

1804 Minor INCOMPATIBILITY due to subtle change in file lookup: use

1805 explicit paths, relatively to the theory.

1807 * Special treatment of ML file names has been discontinued.

1808 Historically, optional extensions .ML or .sml were added on demand --

1809 at the cost of clarity of file dependencies. Recall that Isabelle/ML

1810 files exclusively use the .ML extension. Minor INCOMPATIBILTY.

1812 * Various options that affect pretty printing etc. are now properly

1813 handled within the context via configuration options, instead of

1814 unsynchronized references or print modes. There are both ML Config.T

1815 entities and Isar declaration attributes to access these.

1817 ML (Config.T) Isar (attribute)

1819 eta_contract eta_contract

1820 show_brackets show_brackets

1821 show_sorts show_sorts

1822 show_types show_types

1823 show_question_marks show_question_marks

1824 show_consts show_consts

1825 show_abbrevs show_abbrevs

1827 Syntax.ast_trace syntax_ast_trace

1828 Syntax.ast_stat syntax_ast_stat

1829 Syntax.ambiguity_level syntax_ambiguity_level

1831 Goal_Display.goals_limit goals_limit

1832 Goal_Display.show_main_goal show_main_goal

1834 Method.rule_trace rule_trace

1836 Thy_Output.display thy_output_display

1837 Thy_Output.quotes thy_output_quotes

1838 Thy_Output.indent thy_output_indent

1839 Thy_Output.source thy_output_source

1840 Thy_Output.break thy_output_break

1842 Note that corresponding "..._default" references in ML may only be

1843 changed globally at the ROOT session setup, but *not* within a theory.

1844 The option "show_abbrevs" supersedes the former print mode

1845 "no_abbrevs" with inverted meaning.

1847 * More systematic naming of some configuration options.

1848 INCOMPATIBILITY.

1850 trace_simp ~> simp_trace

1851 debug_simp ~> simp_debug

1853 * Support for real valued configuration options, using simplistic

1854 floating-point notation that coincides with the inner syntax for

1855 float_token.

1857 * Support for real valued preferences (with approximative PGIP type):

1858 front-ends need to accept "pgint" values in float notation.

1859 INCOMPATIBILITY.

1861 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from

1862 DejaVu Sans.

1864 * Discontinued support for Poly/ML 5.0 and 5.1 versions.

1867 *** Pure ***

1869 * Command 'type_synonym' (with single argument) replaces somewhat

1870 outdated 'types', which is still available as legacy feature for some

1871 time.

1873 * Command 'nonterminal' (with 'and' separated list of arguments)

1874 replaces somewhat outdated 'nonterminals'. INCOMPATIBILITY.

1876 * Command 'notepad' replaces former 'example_proof' for

1877 experimentation in Isar without any result. INCOMPATIBILITY.

1879 * Locale interpretation commands 'interpret' and 'sublocale' accept

1880 lists of equations to map definitions in a locale to appropriate

1881 entities in the context of the interpretation. The 'interpretation'

1882 command already provided this functionality.

1884 * Diagnostic command 'print_dependencies' prints the locale instances

1885 that would be activated if the specified expression was interpreted in

1886 the current context. Variant "print_dependencies!" assumes a context

1887 without interpretations.

1889 * Diagnostic command 'print_interps' prints interpretations in proofs

1890 in addition to interpretations in theories.

1892 * Discontinued obsolete 'global' and 'local' commands to manipulate

1893 the theory name space. Rare INCOMPATIBILITY. The ML functions

1894 Sign.root_path and Sign.local_path may be applied directly where this

1895 feature is still required for historical reasons.

1897 * Discontinued obsolete 'constdefs' command. INCOMPATIBILITY, use

1898 'definition' instead.

1900 * The "prems" fact, which refers to the accidental collection of

1901 foundational premises in the context, is now explicitly marked as

1902 legacy feature and will be discontinued soon. Consider using "assms"

1903 of the head statement or reference facts by explicit names.

1905 * Document antiquotations @{class} and @{type} print classes and type

1906 constructors.

1908 * Document antiquotation @{file} checks file/directory entries within

1909 the local file system.

1912 *** HOL ***

1914 * Coercive subtyping: functions can be declared as coercions and type

1915 inference will add them as necessary upon input of a term. Theory

1916 Complex_Main declares real :: nat => real and real :: int => real as

1917 coercions. A coercion function f is declared like this:

1919 declare [[coercion f]]

1921 To lift coercions through type constructors (e.g. from nat => real to

1922 nat list => real list), map functions can be declared, e.g.

1924 declare [[coercion_map map]]

1926 Currently coercion inference is activated only in theories including

1927 real numbers, i.e. descendants of Complex_Main. This is controlled by

1928 the configuration option "coercion_enabled", e.g. it can be enabled in

1929 other theories like this:

1931 declare [[coercion_enabled]]

1933 * Command 'partial_function' provides basic support for recursive

1934 function definitions over complete partial orders. Concrete instances

1935 are provided for i) the option type, ii) tail recursion on arbitrary

1936 types, and iii) the heap monad of Imperative_HOL. See

1937 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy

1938 for examples.

1940 * Function package: f.psimps rules are no longer implicitly declared

1941 as [simp]. INCOMPATIBILITY.

1943 * Datatype package: theorems generated for executable equality (class

1944 "eq") carry proper names and are treated as default code equations.

1946 * Inductive package: now offers command 'inductive_simps' to

1947 automatically derive instantiated and simplified equations for

1948 inductive predicates, similar to 'inductive_cases'.

1950 * Command 'enriched_type' allows to register properties of the

1951 functorial structure of types.

1953 * Improved infrastructure for term evaluation using code generator

1954 techniques, in particular static evaluation conversions.

1956 * Code generator: Scala (2.8 or higher) has been added to the target

1957 languages.

1959 * Code generator: globbing constant expressions "*" and "Theory.*"

1960 have been replaced by the more idiomatic "_" and "Theory._".

1961 INCOMPATIBILITY.

1963 * Code generator: export_code without explicit file declaration prints

1964 to standard output. INCOMPATIBILITY.

1966 * Code generator: do not print function definitions for case

1967 combinators any longer.

1969 * Code generator: simplification with rules determined with

1970 src/Tools/Code/code_simp.ML and method "code_simp".

1972 * Code generator for records: more idiomatic representation of record

1973 types. Warning: records are not covered by ancient SML code

1974 generation any longer. INCOMPATIBILITY. In cases of need, a suitable

1975 rep_datatype declaration helps to succeed then:

1977 record 'a foo = ...

1978 ...

1979 rep_datatype foo_ext ...

1981 * Records: logical foundation type for records does not carry a

1982 '_type' suffix any longer (obsolete due to authentic syntax).

1983 INCOMPATIBILITY.

1985 * Quickcheck now by default uses exhaustive testing instead of random

1986 testing. Random testing can be invoked by "quickcheck [random]",

1987 exhaustive testing by "quickcheck [exhaustive]".

1989 * Quickcheck instantiates polymorphic types with small finite

1990 datatypes by default. This enables a simple execution mechanism to

1991 handle quantifiers and function equality over the finite datatypes.

1993 * Quickcheck random generator has been renamed from "code" to

1994 "random". INCOMPATIBILITY.

1996 * Quickcheck now has a configurable time limit which is set to 30

1997 seconds by default. This can be changed by adding [timeout = n] to the

1998 quickcheck command. The time limit for Auto Quickcheck is still set

1999 independently.

2001 * Quickcheck in locales considers interpretations of that locale for

2002 counter example search.

2004 * Sledgehammer:

2005 - Added "smt" and "remote_smt" provers based on the "smt" proof

2006 method. See the Sledgehammer manual for details ("isabelle doc

2007 sledgehammer").

2008 - Renamed commands:

2009 sledgehammer atp_info ~> sledgehammer running_provers

2010 sledgehammer atp_kill ~> sledgehammer kill_provers

2011 sledgehammer available_atps ~> sledgehammer available_provers

2012 INCOMPATIBILITY.

2013 - Renamed options:

2014 sledgehammer [atps = ...] ~> sledgehammer [provers = ...]

2015 sledgehammer [atp = ...] ~> sledgehammer [prover = ...]

2016 sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]

2017 (and "ms" and "min" are no longer supported)

2018 INCOMPATIBILITY.

2020 * Nitpick:

2021 - Renamed options:

2022 nitpick [timeout = 77 s] ~> nitpick [timeout = 77]

2023 nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]

2024 INCOMPATIBILITY.

2025 - Added support for partial quotient types.

2026 - Added local versions of the "Nitpick.register_xxx" functions.

2027 - Added "whack" option.

2028 - Allow registration of quotient types as codatatypes.

2029 - Improved "merge_type_vars" option to merge more types.

2030 - Removed unsound "fast_descrs" option.

2031 - Added custom symmetry breaking for datatypes, making it possible to reach

2032 higher cardinalities.

2033 - Prevent the expansion of too large definitions.

2035 * Proof methods "metis" and "meson" now have configuration options

2036 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled

2037 to diagnose these tools. E.g.

2039 using [[metis_trace = true]]

2041 * Auto Solve: Renamed "Auto Solve Direct". The tool is now available

2042 manually as command 'solve_direct'.

2044 * The default SMT solver Z3 must be enabled explicitly (due to

2045 licensing issues) by setting the environment variable

2046 Z3_NON_COMMERCIAL in etc/settings of the component, for example. For

2047 commercial applications, the SMT solver CVC3 is provided as fall-back;

2048 changing the SMT solver is done via the configuration option

2049 "smt_solver".

2051 * Remote SMT solvers need to be referred to by the "remote_" prefix,

2052 i.e. "remote_cvc3" and "remote_z3".

2054 * Added basic SMT support for datatypes, records, and typedefs using

2055 the oracle mode (no proofs). Direct support of pairs has been dropped

2056 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT

2057 support for a similar behavior). Minor INCOMPATIBILITY.

2059 * Changed SMT configuration options:

2060 - Renamed:

2061 z3_proofs ~> smt_oracle (with inverted meaning)

2062 z3_trace_assms ~> smt_trace_used_facts

2063 INCOMPATIBILITY.

2064 - Added:

2065 smt_verbose

2066 smt_random_seed

2067 smt_datatypes

2068 smt_infer_triggers

2069 smt_monomorph_limit

2070 cvc3_options

2071 remote_cvc3_options

2072 remote_z3_options

2073 yices_options

2075 * Boogie output files (.b2i files) need to be declared in the theory

2076 header.

2078 * Simplification procedure "list_to_set_comprehension" rewrites list

2079 comprehensions applied to List.set to set comprehensions. Occasional

2080 INCOMPATIBILITY, may be deactivated like this:

2082 declare [[simproc del: list_to_set_comprehension]]

2084 * Removed old version of primrec package. INCOMPATIBILITY.

2086 * Removed simplifier congruence rule of "prod_case", as has for long

2087 been the case with "split". INCOMPATIBILITY.

2089 * String.literal is a type, but not a datatype. INCOMPATIBILITY.

2091 * Removed [split_format ... and ... and ...] version of

2092 [split_format]. Potential INCOMPATIBILITY.

2094 * Predicate "sorted" now defined inductively, with nice induction

2095 rules. INCOMPATIBILITY: former sorted.simps now named sorted_simps.

2097 * Constant "contents" renamed to "the_elem", to free the generic name

2098 contents for other uses. INCOMPATIBILITY.

2100 * Renamed class eq and constant eq (for code generation) to class

2101 equal and constant equal, plus renaming of related facts and various

2102 tuning. INCOMPATIBILITY.

2104 * Dropped type classes mult_mono and mult_mono1. INCOMPATIBILITY.

2106 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".

2107 INCOMPATIBILITY.

2109 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to

2110 avoid confusion with finite sets. INCOMPATIBILITY.

2112 * Abandoned locales equiv, congruent and congruent2 for equivalence

2113 relations. INCOMPATIBILITY: use equivI rather than equiv_intro (same

2114 for congruent(2)).

2116 * Some previously unqualified names have been qualified:

2118 types

2119 bool ~> HOL.bool

2120 nat ~> Nat.nat

2122 constants

2123 Trueprop ~> HOL.Trueprop

2124 True ~> HOL.True

2125 False ~> HOL.False

2126 op & ~> HOL.conj

2127 op | ~> HOL.disj

2128 op --> ~> HOL.implies

2129 op = ~> HOL.eq

2130 Not ~> HOL.Not

2131 The ~> HOL.The

2132 All ~> HOL.All

2133 Ex ~> HOL.Ex

2134 Ex1 ~> HOL.Ex1

2135 Let ~> HOL.Let

2136 If ~> HOL.If

2137 Ball ~> Set.Ball

2138 Bex ~> Set.Bex

2139 Suc ~> Nat.Suc

2140 Pair ~> Product_Type.Pair

2141 fst ~> Product_Type.fst

2142 snd ~> Product_Type.snd

2143 curry ~> Product_Type.curry

2144 op : ~> Set.member

2145 Collect ~> Set.Collect

2147 INCOMPATIBILITY.

2149 * More canonical naming convention for some fundamental definitions:

2151 bot_bool_eq ~> bot_bool_def

2152 top_bool_eq ~> top_bool_def

2153 inf_bool_eq ~> inf_bool_def

2154 sup_bool_eq ~> sup_bool_def

2155 bot_fun_eq ~> bot_fun_def

2156 top_fun_eq ~> top_fun_def

2157 inf_fun_eq ~> inf_fun_def

2158 sup_fun_eq ~> sup_fun_def

2160 INCOMPATIBILITY.

2162 * More stylized fact names:

2164 expand_fun_eq ~> fun_eq_iff

2165 expand_set_eq ~> set_eq_iff

2166 set_ext ~> set_eqI

2167 nat_number ~> eval_nat_numeral

2169 INCOMPATIBILITY.

2171 * Refactoring of code-generation specific operations in theory List:

2173 constants

2174 null ~> List.null

2176 facts

2177 mem_iff ~> member_def

2178 null_empty ~> null_def

2180 INCOMPATIBILITY. Note that these were not supposed to be used

2181 regularly unless for striking reasons; their main purpose was code

2182 generation.

2184 Various operations from the Haskell prelude are used for generating

2185 Haskell code.

2187 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV". Term

2188 "surj f" is now an abbreviation of "range f = UNIV". The theorems

2189 bij_def and surj_def are unchanged. INCOMPATIBILITY.

2191 * Abolished some non-alphabetic type names: "prod" and "sum" replace

2192 "*" and "+" respectively. INCOMPATIBILITY.

2194 * Name "Plus" of disjoint sum operator "<+>" is now hidden. Write

2195 "Sum_Type.Plus" instead.

2197 * Constant "split" has been merged with constant "prod_case"; names of

2198 ML functions, facts etc. involving split have been retained so far,

2199 though. INCOMPATIBILITY.

2201 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"

2202 instead. INCOMPATIBILITY.

2204 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".

2205 INCOMPATIBILITY.

2207 * Former theory Library/Enum is now part of the HOL-Main image.

2208 INCOMPATIBILITY: all constants of the Enum theory now have to be

2209 referred to by its qualified name.

2211 enum ~> Enum.enum

2212 nlists ~> Enum.nlists

2213 product ~> Enum.product

2215 * Theory Library/Monad_Syntax provides do-syntax for monad types.

2216 Syntax in Library/State_Monad has been changed to avoid ambiguities.

2217 INCOMPATIBILITY.

2219 * Theory Library/SetsAndFunctions has been split into

2220 Library/Function_Algebras and Library/Set_Algebras; canonical names

2221 for instance definitions for functions; various improvements.

2222 INCOMPATIBILITY.

2224 * Theory Library/Multiset provides stable quicksort implementation of

2225 sort_key.

2227 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.

2228 INCOMPATIBILITY.

2230 * Session Multivariate_Analysis: introduced a type class for euclidean

2231 space. Most theorems are now stated in terms of euclidean spaces

2232 instead of finite cartesian products.

2234 types

2235 real ^ 'n ~> 'a::real_vector

2236 ~> 'a::euclidean_space

2237 ~> 'a::ordered_euclidean_space

2238 (depends on your needs)

2240 constants

2241 _ $ _ ~> _ $$ _

2242 \<chi> x. _ ~> \<chi>\<chi> x. _

2243 CARD('n) ~> DIM('a)

2245 Also note that the indices are now natural numbers and not from some

2246 finite type. Finite cartesian products of euclidean spaces, products

2247 of euclidean spaces the real and complex numbers are instantiated to

2248 be euclidean_spaces. INCOMPATIBILITY.

2250 * Session Probability: introduced pextreal as positive extended real

2251 numbers. Use pextreal as value for measures. Introduce the

2252 Radon-Nikodym derivative, product spaces and Fubini's theorem for

2253 arbitrary sigma finite measures. Introduces Lebesgue measure based on

2254 the integral in Multivariate Analysis. INCOMPATIBILITY.

2256 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.

2257 INCOMPATIBILITY.

2259 * Session SPARK (with image HOL-SPARK) provides commands to load and

2260 prove verification conditions generated by the SPARK Ada program

2261 verifier. See also src/HOL/SPARK and src/HOL/SPARK/Examples.

2264 *** HOL-Algebra ***

2266 * Theorems for additive ring operations (locale abelian_monoid and

2267 descendants) are generated by interpretation from their multiplicative

2268 counterparts. Names (in particular theorem names) have the mandatory

2269 qualifier 'add'. Previous theorem names are redeclared for

2270 compatibility.

2272 * Structure "int_ring" is now an abbreviation (previously a

2273 definition). This fits more natural with advanced interpretations.

2276 *** HOLCF ***

2278 * The domain package now runs in definitional mode by default: The

2279 former command 'new_domain' is now called 'domain'. To use the domain

2280 package in its original axiomatic mode, use 'domain (unsafe)'.

2281 INCOMPATIBILITY.

2283 * The new class "domain" is now the default sort. Class "predomain"

2284 is an unpointed version of "domain". Theories can be updated by

2285 replacing sort annotations as shown below. INCOMPATIBILITY.

2287 'a::type ~> 'a::countable

2288 'a::cpo ~> 'a::predomain

2289 'a::pcpo ~> 'a::domain

2291 * The old type class "rep" has been superseded by class "domain".

2292 Accordingly, users of the definitional package must remove any

2293 "default_sort rep" declarations. INCOMPATIBILITY.

2295 * The domain package (definitional mode) now supports unpointed

2296 predomain argument types, as long as they are marked 'lazy'. (Strict

2297 arguments must be in class "domain".) For example, the following

2298 domain definition now works:

2300 domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")

2302 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class

2303 instances for types from main HOL: bool, nat, int, char, 'a + 'b,

2304 'a option, and 'a list. Additionally, it configures fixrec and the

2305 domain package to work with these types. For example:

2307 fixrec isInl :: "('a + 'b) u -> tr"

2308 where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"

2310 domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")

2312 * The "(permissive)" option of fixrec has been replaced with a

2313 per-equation "(unchecked)" option. See

2314 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.

2316 * The "bifinite" class no longer fixes a constant "approx"; the class

2317 now just asserts that such a function exists. INCOMPATIBILITY.

2319 * Former type "alg_defl" has been renamed to "defl". HOLCF no longer

2320 defines an embedding of type 'a defl into udom by default; instances

2321 of "bifinite" and "domain" classes are available in

2322 src/HOL/HOLCF/Library/Defl_Bifinite.thy.

2324 * The syntax "REP('a)" has been replaced with "DEFL('a)".

2326 * The predicate "directed" has been removed. INCOMPATIBILITY.

2328 * The type class "finite_po" has been removed. INCOMPATIBILITY.

2330 * The function "cprod_map" has been renamed to "prod_map".

2331 INCOMPATIBILITY.

2333 * The monadic bind operator on each powerdomain has new binder syntax

2334 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents

2335 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".

2337 * The infix syntax for binary union on each powerdomain has changed

2338 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set

2339 syntax. INCOMPATIBILITY.

2341 * The constant "UU" has been renamed to "bottom". The syntax "UU" is

2342 still supported as an input translation.

2344 * Renamed some theorems (the original names are also still available).

2346 expand_fun_below ~> fun_below_iff

2347 below_fun_ext ~> fun_belowI

2348 expand_cfun_eq ~> cfun_eq_iff

2349 ext_cfun ~> cfun_eqI

2350 expand_cfun_below ~> cfun_below_iff

2351 below_cfun_ext ~> cfun_belowI

2352 cont2cont_Rep_CFun ~> cont2cont_APP

2354 * The Abs and Rep functions for various types have changed names.

2355 Related theorem names have also changed to match. INCOMPATIBILITY.

2357 Rep_CFun ~> Rep_cfun

2358 Abs_CFun ~> Abs_cfun

2359 Rep_Sprod ~> Rep_sprod

2360 Abs_Sprod ~> Abs_sprod

2361 Rep_Ssum ~> Rep_ssum

2362 Abs_Ssum ~> Abs_ssum

2364 * Lemmas with names of the form *_defined_iff or *_strict_iff have

2365 been renamed to *_bottom_iff. INCOMPATIBILITY.

2367 * Various changes to bisimulation/coinduction with domain package:

2369 - Definitions of "bisim" constants no longer mention definedness.

2370 - With mutual recursion, "bisim" predicate is now curried.

2371 - With mutual recursion, each type gets a separate coind theorem.

2372 - Variable names in bisim_def and coinduct rules have changed.

2374 INCOMPATIBILITY.

2376 * Case combinators generated by the domain package for type "foo" are

2377 now named "foo_case" instead of "foo_when". INCOMPATIBILITY.

2379 * Several theorems have been renamed to more accurately reflect the

2380 names of constants and types involved. INCOMPATIBILITY.

2382 thelub_const ~> lub_const

2383 lub_const ~> is_lub_const

2384 thelubI ~> lub_eqI

2385 is_lub_lub ~> is_lubD2

2386 lubI ~> is_lub_lub

2387 unique_lub ~> is_lub_unique

2388 is_ub_lub ~> is_lub_rangeD1

2389 lub_bin_chain ~> is_lub_bin_chain

2390 lub_fun ~> is_lub_fun

2391 thelub_fun ~> lub_fun

2392 thelub_cfun ~> lub_cfun

2393 thelub_Pair ~> lub_Pair

2394 lub_cprod ~> is_lub_prod

2395 thelub_cprod ~> lub_prod

2396 minimal_cprod ~> minimal_prod

2397 inst_cprod_pcpo ~> inst_prod_pcpo

2398 UU_I ~> bottomI

2399 compact_UU ~> compact_bottom

2400 deflation_UU ~> deflation_bottom

2401 finite_deflation_UU ~> finite_deflation_bottom

2403 * Many legacy theorem names have been discontinued. INCOMPATIBILITY.

2405 sq_ord_less_eq_trans ~> below_eq_trans

2406 sq_ord_eq_less_trans ~> eq_below_trans

2407 refl_less ~> below_refl

2408 trans_less ~> below_trans

2409 antisym_less ~> below_antisym

2410 antisym_less_inverse ~> po_eq_conv [THEN iffD1]

2411 box_less ~> box_below

2412 rev_trans_less ~> rev_below_trans

2413 not_less2not_eq ~> not_below2not_eq

2414 less_UU_iff ~> below_UU_iff

2415 flat_less_iff ~> flat_below_iff

2416 adm_less ~> adm_below

2417 adm_not_less ~> adm_not_below

2418 adm_compact_not_less ~> adm_compact_not_below

2419 less_fun_def ~> below_fun_def

2420 expand_fun_less ~> fun_below_iff

2421 less_fun_ext ~> fun_belowI

2422 less_discr_def ~> below_discr_def

2423 discr_less_eq ~> discr_below_eq

2424 less_unit_def ~> below_unit_def

2425 less_cprod_def ~> below_prod_def

2426 prod_lessI ~> prod_belowI

2427 Pair_less_iff ~> Pair_below_iff

2428 fst_less_iff ~> fst_below_iff

2429 snd_less_iff ~> snd_below_iff

2430 expand_cfun_less ~> cfun_below_iff

2431 less_cfun_ext ~> cfun_belowI

2432 injection_less ~> injection_below

2433 less_up_def ~> below_up_def

2434 not_Iup_less ~> not_Iup_below

2435 Iup_less ~> Iup_below

2436 up_less ~> up_below

2437 Def_inject_less_eq ~> Def_below_Def

2438 Def_less_is_eq ~> Def_below_iff

2439 spair_less_iff ~> spair_below_iff

2440 less_sprod ~> below_sprod

2441 spair_less ~> spair_below

2442 sfst_less_iff ~> sfst_below_iff

2443 ssnd_less_iff ~> ssnd_below_iff

2444 fix_least_less ~> fix_least_below

2445 dist_less_one ~> dist_below_one

2446 less_ONE ~> below_ONE

2447 ONE_less_iff ~> ONE_below_iff

2448 less_sinlD ~> below_sinlD

2449 less_sinrD ~> below_sinrD

2452 *** FOL and ZF ***

2454 * All constant names are now qualified internally and use proper

2455 identifiers, e.g. "IFOL.eq" instead of "op =". INCOMPATIBILITY.

2458 *** ML ***

2460 * Antiquotation @{assert} inlines a function bool -> unit that raises

2461 Fail if the argument is false. Due to inlining the source position of

2462 failed assertions is included in the error output.

2464 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML

2465 text is in practice always evaluated with a stable theory checkpoint.

2466 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.

2468 * Antiquotation @{theory A} refers to theory A from the ancestry of

2469 the current context, not any accidental theory loader state as before.

2470 Potential INCOMPATIBILITY, subtle change in semantics.

2472 * Syntax.pretty_priority (default 0) configures the required priority

2473 of pretty-printed output and thus affects insertion of parentheses.

2475 * Syntax.default_root (default "any") configures the inner syntax

2476 category (nonterminal symbol) for parsing of terms.

2478 * Former exception Library.UnequalLengths now coincides with

2479 ListPair.UnequalLengths.

2481 * Renamed structure MetaSimplifier to Raw_Simplifier. Note that the

2482 main functionality is provided by structure Simplifier.

2484 * Renamed raw "explode" function to "raw_explode" to emphasize its

2485 meaning. Note that internally to Isabelle, Symbol.explode is used in

2486 almost all situations.

2488 * Discontinued obsolete function sys_error and exception SYS_ERROR.

2489 See implementation manual for further details on exceptions in

2490 Isabelle/ML.

2492 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its

2493 meaning.

2495 * Renamed structure PureThy to Pure_Thy and moved most of its

2496 operations to structure Global_Theory, to emphasize that this is

2497 rarely-used global-only stuff.

2499 * Discontinued Output.debug. Minor INCOMPATIBILITY, use plain writeln

2500 instead (or tracing for high-volume output).

2502 * Configuration option show_question_marks only affects regular pretty

2503 printing of types and terms, not raw Term.string_of_vname.

2505 * ML_Context.thm and ML_Context.thms are no longer pervasive. Rare

2506 INCOMPATIBILITY, superseded by static antiquotations @{thm} and

2507 @{thms} for most purposes.

2509 * ML structure Unsynchronized is never opened, not even in Isar

2510 interaction mode as before. Old Unsynchronized.set etc. have been

2511 discontinued -- use plain := instead. This should be *rare* anyway,

2512 since modern tools always work via official context data, notably

2513 configuration options.

2515 * Parallel and asynchronous execution requires special care concerning

2516 interrupts. Structure Exn provides some convenience functions that

2517 avoid working directly with raw Interrupt. User code must not absorb

2518 interrupts -- intermediate handling (for cleanup etc.) needs to be

2519 followed by re-raising of the original exception. Another common

2520 source of mistakes are "handle _" patterns, which make the meaning of

2521 the program subject to physical effects of the environment.

2525 New in Isabelle2009-2 (June 2010)

2526 ---------------------------------

2528 *** General ***

2530 * Authentic syntax for *all* logical entities (type classes, type

2531 constructors, term constants): provides simple and robust

2532 correspondence between formal entities and concrete syntax. Within

2533 the parse tree / AST representations, "constants" are decorated by

2534 their category (class, type, const) and spelled out explicitly with

2535 their full internal name.

2537 Substantial INCOMPATIBILITY concerning low-level syntax declarations

2538 and translations (translation rules and translation functions in ML).

2539 Some hints on upgrading:

2541 - Many existing uses of 'syntax' and 'translations' can be replaced

2542 by more modern 'type_notation', 'notation' and 'abbreviation',

2543 which are independent of this issue.

2545 - 'translations' require markup within the AST; the term syntax

2546 provides the following special forms:

2548 CONST c -- produces syntax version of constant c from context

2549 XCONST c -- literally c, checked as constant from context

2550 c -- literally c, if declared by 'syntax'

2552 Plain identifiers are treated as AST variables -- occasionally the

2553 system indicates accidental variables via the error "rhs contains

2554 extra variables".

2556 Type classes and type constructors are marked according to their

2557 concrete syntax. Some old translations rules need to be written

2558 for the "type" category, using type constructor application

2559 instead of pseudo-term application of the default category

2560 "logic".

2562 - 'parse_translation' etc. in ML may use the following

2563 antiquotations:

2565 @{class_syntax c} -- type class c within parse tree / AST

2566 @{term_syntax c} -- type constructor c within parse tree / AST

2567 @{const_syntax c} -- ML version of "CONST c" above

2568 @{syntax_const c} -- literally c (checked wrt. 'syntax' declarations)

2570 - Literal types within 'typed_print_translations', i.e. those *not*

2571 represented as pseudo-terms are represented verbatim. Use @{class

2572 c} or @{type_name c} here instead of the above syntax

2573 antiquotations.

2575 Note that old non-authentic syntax was based on unqualified base

2576 names, so all of the above "constant" names would coincide. Recall

2577 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to

2578 diagnose syntax problems.

2580 * Type constructors admit general mixfix syntax, not just infix.

2582 * Concrete syntax may be attached to local entities without a proof

2583 body, too. This works via regular mixfix annotations for 'fix',

2584 'def', 'obtain' etc. or via the explicit 'write' command, which is

2585 similar to the 'notation' command in theory specifications.

2587 * Discontinued unnamed infix syntax (legacy feature for many years) --

2588 need to specify constant name and syntax separately. Internal ML

2589 datatype constructors have been renamed from InfixName to Infix etc.

2590 Minor INCOMPATIBILITY.

2592 * Schematic theorem statements need to be explicitly markup as such,

2593 via commands 'schematic_lemma', 'schematic_theorem',

2594 'schematic_corollary'. Thus the relevance of the proof is made

2595 syntactically clear, which impacts performance in a parallel or

2596 asynchronous interactive environment. Minor INCOMPATIBILITY.

2598 * Use of cumulative prems via "!" in some proof methods has been

2599 discontinued (old legacy feature).

2601 * References 'trace_simp' and 'debug_simp' have been replaced by

2602 configuration options stored in the context. Enabling tracing (the

2603 case of debugging is similar) in proofs works via

2605 using [[trace_simp = true]]

2607 Tracing is then active for all invocations of the simplifier in

2608 subsequent goal refinement steps. Tracing may also still be enabled or

2609 disabled via the ProofGeneral settings menu.

2611 * Separate commands 'hide_class', 'hide_type', 'hide_const',

2612 'hide_fact' replace the former 'hide' KIND command. Minor

2613 INCOMPATIBILITY.

2615 * Improved parallelism of proof term normalization: usedir -p2 -q0 is

2616 more efficient than combinations with -q1 or -q2.

2619 *** Pure ***

2621 * Proofterms record type-class reasoning explicitly, using the

2622 "unconstrain" operation internally. This eliminates all sort

2623 constraints from a theorem and proof, introducing explicit

2624 OFCLASS-premises. On the proof term level, this operation is

2625 automatically applied at theorem boundaries, such that closed proofs

2626 are always free of sort constraints. INCOMPATIBILITY for tools that

2627 inspect proof terms.

2629 * Local theory specifications may depend on extra type variables that

2630 are not present in the result type -- arguments TYPE('a) :: 'a itself

2631 are added internally. For example:

2633 definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"

2635 * Predicates of locales introduced by classes carry a mandatory

2636 "class" prefix. INCOMPATIBILITY.

2638 * Vacuous class specifications observe default sort. INCOMPATIBILITY.

2640 * Old 'axclass' command has been discontinued. INCOMPATIBILITY, use

2641 'class' instead.

2643 * Command 'code_reflect' allows to incorporate generated ML code into

2644 runtime environment; replaces immature code_datatype antiquotation.

2645 INCOMPATIBILITY.

2647 * Code generator: simple concept for abstract datatypes obeying

2648 invariants.

2650 * Code generator: details of internal data cache have no impact on the

2651 user space functionality any longer.

2653 * Methods "unfold_locales" and "intro_locales" ignore non-locale

2654 subgoals. This is more appropriate for interpretations with 'where'.

2655 INCOMPATIBILITY.

2657 * Command 'example_proof' opens an empty proof body. This allows to

2658 experiment with Isar, without producing any persistent result.

2660 * Commands 'type_notation' and 'no_type_notation' declare type syntax

2661 within a local theory context, with explicit checking of the

2662 constructors involved (in contrast to the raw 'syntax' versions).

2664 * Commands 'types' and 'typedecl' now work within a local theory

2665 context -- without introducing dependencies on parameters or

2666 assumptions, which is not possible in Isabelle/Pure.

2668 * Command 'defaultsort' has been renamed to 'default_sort', it works

2669 within a local theory context. Minor INCOMPATIBILITY.

2672 *** HOL ***

2674 * Command 'typedef' now works within a local theory context -- without

2675 introducing dependencies on parameters or assumptions, which is not

2676 possible in Isabelle/Pure/HOL. Note that the logical environment may

2677 contain multiple interpretations of local typedefs (with different

2678 non-emptiness proofs), even in a global theory context.

2680 * New package for quotient types. Commands 'quotient_type' and

2681 'quotient_definition' may be used for defining types and constants by

2682 quotient constructions. An example is the type of integers created by

2683 quotienting pairs of natural numbers:

2685 fun

2686 intrel :: "(nat * nat) => (nat * nat) => bool"

2687 where

2688 "intrel (x, y) (u, v) = (x + v = u + y)"

2690 quotient_type int = "nat * nat" / intrel

2691 by (auto simp add: equivp_def expand_fun_eq)

2693 quotient_definition

2694 "0::int" is "(0::nat, 0::nat)"

2696 The method "lifting" can be used to lift of theorems from the

2697 underlying "raw" type to the quotient type. The example

2698 src/HOL/Quotient_Examples/FSet.thy includes such a quotient

2699 construction and provides a reasoning infrastructure for finite sets.

2701 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid

2702 clash with new theory Quotient in Main HOL.

2704 * Moved the SMT binding into the main HOL session, eliminating

2705 separate HOL-SMT session.

2707 * List membership infix mem operation is only an input abbreviation.

2708 INCOMPATIBILITY.

2710 * Theory Library/Word.thy has been removed. Use library Word/Word.thy

2711 for future developements; former Library/Word.thy is still present in

2712 the AFP entry RSAPPS.

2714 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no

2715 longer shadowed. INCOMPATIBILITY.

2717 * Dropped theorem duplicate comp_arith; use semiring_norm instead.

2718 INCOMPATIBILITY.

2720 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.

2721 INCOMPATIBILITY.

2723 * Dropped normalizing_semiring etc; use the facts in semiring classes

2724 instead. INCOMPATIBILITY.

2726 * Dropped several real-specific versions of lemmas about floor and

2727 ceiling; use the generic lemmas from theory "Archimedean_Field"

2728 instead. INCOMPATIBILITY.

2730 floor_number_of_eq ~> floor_number_of

2731 le_floor_eq_number_of ~> number_of_le_floor

2732 le_floor_eq_zero ~> zero_le_floor

2733 le_floor_eq_one ~> one_le_floor

2734 floor_less_eq_number_of ~> floor_less_number_of

2735 floor_less_eq_zero ~> floor_less_zero

2736 floor_less_eq_one ~> floor_less_one

2737 less_floor_eq_number_of ~> number_of_less_floor

2738 less_floor_eq_zero ~> zero_less_floor

2739 less_floor_eq_one ~> one_less_floor

2740 floor_le_eq_number_of ~> floor_le_number_of

2741 floor_le_eq_zero ~> floor_le_zero

2742 floor_le_eq_one ~> floor_le_one

2743 floor_subtract_number_of ~> floor_diff_number_of

2744 floor_subtract_one ~> floor_diff_one

2745 ceiling_number_of_eq ~> ceiling_number_of

2746 ceiling_le_eq_number_of ~> ceiling_le_number_of

2747 ceiling_le_zero_eq ~> ceiling_le_zero

2748 ceiling_le_eq_one ~> ceiling_le_one

2749 less_ceiling_eq_number_of ~> number_of_less_ceiling

2750 less_ceiling_eq_zero ~> zero_less_ceiling

2751 less_ceiling_eq_one ~> one_less_ceiling

2752 ceiling_less_eq_number_of ~> ceiling_less_number_of

2753 ceiling_less_eq_zero ~> ceiling_less_zero

2754 ceiling_less_eq_one ~> ceiling_less_one

2755 le_ceiling_eq_number_of ~> number_of_le_ceiling

2756 le_ceiling_eq_zero ~> zero_le_ceiling

2757 le_ceiling_eq_one ~> one_le_ceiling

2758 ceiling_subtract_number_of ~> ceiling_diff_number_of

2759 ceiling_subtract_one ~> ceiling_diff_one

2761 * Theory "Finite_Set": various folding_XXX locales facilitate the

2762 application of the various fold combinators on finite sets.

2764 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"

2765 provides abstract red-black tree type which is backed by "RBT_Impl" as

2766 implementation. INCOMPATIBILTY.

2768 * Theory Library/Coinductive_List has been removed -- superseded by

2769 AFP/thys/Coinductive.

2771 * Theory PReal, including the type "preal" and related operations, has

2772 been removed. INCOMPATIBILITY.

2774 * Real: new development using Cauchy Sequences.

2776 * Split off theory "Big_Operators" containing setsum, setprod,

2777 Inf_fin, Sup_fin, Min, Max from theory Finite_Set. INCOMPATIBILITY.

2779 * Theory "Rational" renamed to "Rat", for consistency with "Nat",

2780 "Int" etc. INCOMPATIBILITY.

2782 * Constant Rat.normalize needs to be qualified. INCOMPATIBILITY.

2784 * New set of rules "ac_simps" provides combined assoc / commute

2785 rewrites for all interpretations of the appropriate generic locales.

2787 * Renamed theory "OrderedGroup" to "Groups" and split theory

2788 "Ring_and_Field" into theories "Rings" and "Fields"; for more

2789 appropriate and more consistent names suitable for name prefixes

2790 within the HOL theories. INCOMPATIBILITY.

2792 * Some generic constants have been put to appropriate theories:

2793 - less_eq, less: Orderings

2794 - zero, one, plus, minus, uminus, times, abs, sgn: Groups

2795 - inverse, divide: Rings

2796 INCOMPATIBILITY.

2798 * More consistent naming of type classes involving orderings (and

2799 lattices):

2801 lower_semilattice ~> semilattice_inf

2802 upper_semilattice ~> semilattice_sup

2804 dense_linear_order ~> dense_linorder

2806 pordered_ab_group_add ~> ordered_ab_group_add

2807 pordered_ab_group_add_abs ~> ordered_ab_group_add_abs

2808 pordered_ab_semigroup_add ~> ordered_ab_semigroup_add

2809 pordered_ab_semigroup_add_imp_le ~> ordered_ab_semigroup_add_imp_le

2810 pordered_cancel_ab_semigroup_add ~> ordered_cancel_ab_semigroup_add

2811 pordered_cancel_comm_semiring ~> ordered_cancel_comm_semiring

2812 pordered_cancel_semiring ~> ordered_cancel_semiring

2813 pordered_comm_monoid_add ~> ordered_comm_monoid_add

2814 pordered_comm_ring ~> ordered_comm_ring

2815 pordered_comm_semiring ~> ordered_comm_semiring

2816 pordered_ring ~> ordered_ring

2817 pordered_ring_abs ~> ordered_ring_abs

2818 pordered_semiring ~> ordered_semiring

2820 ordered_ab_group_add ~> linordered_ab_group_add

2821 ordered_ab_semigroup_add ~> linordered_ab_semigroup_add

2822 ordered_cancel_ab_semigroup_add ~> linordered_cancel_ab_semigroup_add

2823 ordered_comm_semiring_strict ~> linordered_comm_semiring_strict

2824 ordered_field ~> linordered_field

2825 ordered_field_no_lb ~> linordered_field_no_lb

2826 ordered_field_no_ub ~> linordered_field_no_ub

2827 ordered_field_dense_linear_order ~> dense_linordered_field

2828 ordered_idom ~> linordered_idom

2829 ordered_ring ~> linordered_ring

2830 ordered_ring_le_cancel_factor ~> linordered_ring_le_cancel_factor

2831 ordered_ring_less_cancel_factor ~> linordered_ring_less_cancel_factor

2832 ordered_ring_strict ~> linordered_ring_strict

2833 ordered_semidom ~> linordered_semidom

2834 ordered_semiring ~> linordered_semiring

2835 ordered_semiring_1 ~> linordered_semiring_1

2836 ordered_semiring_1_strict ~> linordered_semiring_1_strict

2837 ordered_semiring_strict ~> linordered_semiring_strict

2839 The following slightly odd type classes have been moved to a

2840 separate theory Library/Lattice_Algebras:

2842 lordered_ab_group_add ~> lattice_ab_group_add

2843 lordered_ab_group_add_abs ~> lattice_ab_group_add_abs

2844 lordered_ab_group_add_meet ~> semilattice_inf_ab_group_add

2845 lordered_ab_group_add_join ~> semilattice_sup_ab_group_add

2846 lordered_ring ~> lattice_ring

2848 INCOMPATIBILITY.

2850 * Refined field classes:

2851 - classes division_ring_inverse_zero, field_inverse_zero,

2852 linordered_field_inverse_zero include rule inverse 0 = 0 --

2853 subsumes former division_by_zero class;

2854 - numerous lemmas have been ported from field to division_ring.

2855 INCOMPATIBILITY.

2857 * Refined algebra theorem collections:

2858 - dropped theorem group group_simps, use algebra_simps instead;

2859 - dropped theorem group ring_simps, use field_simps instead;

2860 - proper theorem collection field_simps subsumes former theorem

2861 groups field_eq_simps and field_simps;

2862 - dropped lemma eq_minus_self_iff which is a duplicate for

2863 equal_neg_zero.

2864 INCOMPATIBILITY.

2866 * Theory Finite_Set and List: some lemmas have been generalized from

2867 sets to lattices:

2869 fun_left_comm_idem_inter ~> fun_left_comm_idem_inf

2870 fun_left_comm_idem_union ~> fun_left_comm_idem_sup

2871 inter_Inter_fold_inter ~> inf_Inf_fold_inf

2872 union_Union_fold_union ~> sup_Sup_fold_sup

2873 Inter_fold_inter ~> Inf_fold_inf

2874 Union_fold_union ~> Sup_fold_sup

2875 inter_INTER_fold_inter ~> inf_INFI_fold_inf

2876 union_UNION_fold_union ~> sup_SUPR_fold_sup

2877 INTER_fold_inter ~> INFI_fold_inf

2878 UNION_fold_union ~> SUPR_fold_sup

2880 * Theory "Complete_Lattice": lemmas top_def and bot_def have been

2881 replaced by the more convenient lemmas Inf_empty and Sup_empty.

2882 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed

2883 by Inf_insert and Sup_insert. Lemmas Inf_UNIV and Sup_UNIV replace

2884 former Inf_Univ and Sup_Univ. Lemmas inf_top_right and sup_bot_right

2885 subsume inf_top and sup_bot respectively. INCOMPATIBILITY.

2887 * Reorganized theory Multiset: swapped notation of pointwise and

2888 multiset order:

2890 - pointwise ordering is instance of class order with standard syntax

2891 <= and <;

2892 - multiset ordering has syntax <=# and <#; partial order properties

2893 are provided by means of interpretation with prefix

2894 multiset_order;

2895 - less duplication, less historical organization of sections,

2896 conversion from associations lists to multisets, rudimentary code

2897 generation;

2898 - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,

2899 if needed.

2901 Renamed:

2903 multiset_eq_conv_count_eq ~> multiset_ext_iff

2904 multi_count_ext ~> multiset_ext

2905 diff_union_inverse2 ~> diff_union_cancelR

2907 INCOMPATIBILITY.

2909 * Theory Permutation: replaced local "remove" by List.remove1.

2911 * Code generation: ML and OCaml code is decorated with signatures.

2913 * Theory List: added transpose.

2915 * Library/Nat_Bijection.thy is a collection of bijective functions

2916 between nat and other types, which supersedes the older libraries

2917 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy. INCOMPATIBILITY.

2919 Constants:

2920 Nat_Int_Bij.nat2_to_nat ~> prod_encode

2921 Nat_Int_Bij.nat_to_nat2 ~> prod_decode

2922 Nat_Int_Bij.int_to_nat_bij ~> int_encode

2923 Nat_Int_Bij.nat_to_int_bij ~> int_decode

2924 Countable.pair_encode ~> prod_encode

2925 NatIso.prod2nat ~> prod_encode

2926 NatIso.nat2prod ~> prod_decode

2927 NatIso.sum2nat ~> sum_encode

2928 NatIso.nat2sum ~> sum_decode

2929 NatIso.list2nat ~> list_encode

2930 NatIso.nat2list ~> list_decode

2931 NatIso.set2nat ~> set_encode

2932 NatIso.nat2set ~> set_decode

2934 Lemmas:

2935 Nat_Int_Bij.bij_nat_to_int_bij ~> bij_int_decode

2936 Nat_Int_Bij.nat2_to_nat_inj ~> inj_prod_encode

2937 Nat_Int_Bij.nat2_to_nat_surj ~> surj_prod_encode

2938 Nat_Int_Bij.nat_to_nat2_inj ~> inj_prod_decode

2939 Nat_Int_Bij.nat_to_nat2_surj ~> surj_prod_decode

2940 Nat_Int_Bij.i2n_n2i_id ~> int_encode_inverse

2941 Nat_Int_Bij.n2i_i2n_id ~> int_decode_inverse

2942 Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode

2943 Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode

2944 Nat_Int_Bij.inj_nat_to_int_bij ~> inj_int_encode

2945 Nat_Int_Bij.inj_int_to_nat_bij ~> inj_int_decode

2946 Nat_Int_Bij.bij_nat_to_int_bij ~> bij_int_encode

2947 Nat_Int_Bij.bij_int_to_nat_bij ~> bij_int_decode

2949 * Sledgehammer:

2950 - Renamed ATP commands:

2951 atp_info ~> sledgehammer running_atps

2952 atp_kill ~> sledgehammer kill_atps

2953 atp_messages ~> sledgehammer messages

2954 atp_minimize ~> sledgehammer minimize

2955 print_atps ~> sledgehammer available_atps

2956 INCOMPATIBILITY.

2957 - Added user's manual ("isabelle doc sledgehammer").

2958 - Added option syntax and "sledgehammer_params" to customize

2959 Sledgehammer's behavior. See the manual for details.

2960 - Modified the Isar proof reconstruction code so that it produces

2961 direct proofs rather than proofs by contradiction. (This feature

2962 is still experimental.)

2963 - Made Isar proof reconstruction work for SPASS, remote ATPs, and in

2964 full-typed mode.

2965 - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.

2967 * Nitpick:

2968 - Added and implemented "binary_ints" and "bits" options.

2969 - Added "std" option and implemented support for nonstandard models.

2970 - Added and implemented "finitize" option to improve the precision

2971 of infinite datatypes based on a monotonicity analysis.

2972 - Added support for quotient types.

2973 - Added support for "specification" and "ax_specification"

2974 constructs.

2975 - Added support for local definitions (for "function" and

2976 "termination" proofs).

2977 - Added support for term postprocessors.

2978 - Optimized "Multiset.multiset" and "FinFun.finfun".

2979 - Improved efficiency of "destroy_constrs" optimization.

2980 - Fixed soundness bugs related to "destroy_constrs" optimization and

2981 record getters.

2982 - Fixed soundness bug related to higher-order constructors.

2983 - Fixed soundness bug when "full_descrs" is enabled.

2984 - Improved precision of set constructs.

2985 - Added "atoms" option.

2986 - Added cache to speed up repeated Kodkod invocations on the same

2987 problems.

2988 - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and

2989 "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and

2990 "SAT4J_Light". INCOMPATIBILITY.

2991 - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",

2992 "sharing_depth", and "show_skolems" options. INCOMPATIBILITY.

2993 - Removed "nitpick_intro" attribute. INCOMPATIBILITY.

2995 * Method "induct" now takes instantiations of the form t, where t is not

2996 a variable, as a shorthand for "x == t", where x is a fresh variable.

2997 If this is not intended, t has to be enclosed in parentheses.

2998 By default, the equalities generated by definitional instantiations

2999 are pre-simplified, which may cause parameters of inductive cases

3000 to disappear, or may even delete some of the inductive cases.

3001 Use "induct (no_simp)" instead of "induct" to restore the old

3002 behaviour. The (no_simp) option is also understood by the "cases"

3003 and "nominal_induct" methods, which now perform pre-simplification, too.

3004 INCOMPATIBILITY.

3007 *** HOLCF ***

3009 * Variable names in lemmas generated by the domain package have

3010 changed; the naming scheme is now consistent with the HOL datatype

3011 package. Some proof scripts may be affected, INCOMPATIBILITY.

3013 * The domain package no longer defines the function "foo_copy" for

3014 recursive domain "foo". The reach lemma is now stated directly in

3015 terms of "foo_take". Lemmas and proofs that mention "foo_copy" must

3016 be reformulated in terms of "foo_take", INCOMPATIBILITY.

3018 * Most definedness lemmas generated by the domain package (previously

3019 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form

3020 like "foo$x = UU <-> x = UU", which works better as a simp rule.

3021 Proofs that used definedness lemmas as intro rules may break,

3022 potential INCOMPATIBILITY.

3024 * Induction and casedist rules generated by the domain package now

3025 declare proper case_names (one called "bottom", and one named for each

3026 constructor). INCOMPATIBILITY.

3028 * For mutually-recursive domains, separate "reach" and "take_lemma"

3029 rules are generated for each domain, INCOMPATIBILITY.

3031 foo_bar.reach ~> foo.reach bar.reach

3032 foo_bar.take_lemmas ~> foo.take_lemma bar.take_lemma

3034 * Some lemmas generated by the domain package have been renamed for

3035 consistency with the datatype package, INCOMPATIBILITY.

3037 foo.ind ~> foo.induct

3038 foo.finite_ind ~> foo.finite_induct

3039 foo.coind ~> foo.coinduct

3040 foo.casedist ~> foo.exhaust

3041 foo.exhaust ~> foo.nchotomy

3043 * For consistency with other definition packages, the fixrec package

3044 now generates qualified theorem names, INCOMPATIBILITY.

3046 foo_simps ~> foo.simps

3047 foo_unfold ~> foo.unfold

3048 foo_induct ~> foo.induct

3050 * The "fixrec_simp" attribute has been removed. The "fixrec_simp"

3051 method and internal fixrec proofs now use the default simpset instead.

3052 INCOMPATIBILITY.

3054 * The "contlub" predicate has been removed. Proof scripts should use

3055 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.

3057 * The "admw" predicate has been removed, INCOMPATIBILITY.

3059 * The constants cpair, cfst, and csnd have been removed in favor of

3060 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.

3063 *** ML ***

3065 * Antiquotations for basic formal entities:

3067 @{class NAME} -- type class

3068 @{class_syntax NAME} -- syntax representation of the above

3070 @{type_name NAME} -- logical type

3071 @{type_abbrev NAME} -- type abbreviation

3072 @{nonterminal NAME} -- type of concrete syntactic category

3073 @{type_syntax NAME} -- syntax representation of any of the above

3075 @{const_name NAME} -- logical constant (INCOMPATIBILITY)

3076 @{const_abbrev NAME} -- abbreviated constant

3077 @{const_syntax NAME} -- syntax representation of any of the above

3079 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw

3080 syntax constant (cf. 'syntax' command).

3082 * Antiquotation @{make_string} inlines a function to print arbitrary

3083 values similar to the ML toplevel. The result is compiler dependent

3084 and may fall back on "?" in certain situations.

3086 * Diagnostic commands 'ML_val' and 'ML_command' may refer to

3087 antiquotations @{Isar.state} and @{Isar.goal}. This replaces impure

3088 Isar.state() and Isar.goal(), which belong to the old TTY loop and do

3089 not work with the asynchronous Isar document model.

3091 * Configuration options now admit dynamic default values, depending on

3092 the context or even global references.

3094 * SHA1.digest digests strings according to SHA-1 (see RFC 3174). It

3095 uses an efficient external library if available (for Poly/ML).

3097 * Renamed some important ML structures, while keeping the old names

3098 for some time as aliases within the structure Legacy:

3100 OuterKeyword ~> Keyword

3101 OuterLex ~> Token

3102 OuterParse ~> Parse

3103 OuterSyntax ~> Outer_Syntax

3104 PrintMode ~> Print_Mode

3105 SpecParse ~> Parse_Spec

3106 ThyInfo ~> Thy_Info

3107 ThyLoad ~> Thy_Load

3108 ThyOutput ~> Thy_Output

3109 TypeInfer ~> Type_Infer

3111 Note that "open Legacy" simplifies porting of sources, but forgetting

3112 to remove it again will complicate porting again in the future.

3114 * Most operations that refer to a global context are named

3115 accordingly, e.g. Simplifier.global_context or

3116 ProofContext.init_global. There are some situations where a global

3117 context actually works, but under normal circumstances one needs to

3118 pass the proper local context through the code!

3120 * Discontinued old TheoryDataFun with its copy/init operation -- data

3121 needs to be pure. Functor Theory_Data_PP retains the traditional

3122 Pretty.pp argument to merge, which is absent in the standard

3123 Theory_Data version.

3125 * Sorts.certify_sort and derived "cert" operations for types and terms

3126 no longer minimize sorts. Thus certification at the boundary of the

3127 inference kernel becomes invariant under addition of class relations,

3128 which is an important monotonicity principle. Sorts are now minimized

3129 in the syntax layer only, at the boundary between the end-user and the

3130 system. Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort

3131 explicitly in rare situations.

3133 * Renamed old-style Drule.standard to Drule.export_without_context, to

3134 emphasize that this is in no way a standard operation.

3135 INCOMPATIBILITY.

3137 * Subgoal.FOCUS (and variants): resulting goal state is normalized as

3138 usual for resolution. Rare INCOMPATIBILITY.

3140 * Renamed varify/unvarify operations to varify_global/unvarify_global

3141 to emphasize that these only work in a global situation (which is

3142 quite rare).

3144 * Curried take and drop in library.ML; negative length is interpreted

3145 as infinity (as in chop). Subtle INCOMPATIBILITY.

3147 * Proof terms: type substitutions on proof constants now use canonical

3148 order of type variables. INCOMPATIBILITY for tools working with proof

3149 terms.

3151 * Raw axioms/defs may no longer carry sort constraints, and raw defs

3152 may no longer carry premises. User-level specifications are

3153 transformed accordingly by Thm.add_axiom/add_def.

3156 *** System ***

3158 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;

3159 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions. Note that

3160 proof terms are enabled unconditionally in the new HOL-Proofs image.

3162 * Discontinued old ISABELLE and ISATOOL environment settings (legacy

3163 feature since Isabelle2009). Use ISABELLE_PROCESS and ISABELLE_TOOL,

3164 respectively.

3166 * Old lib/scripts/polyml-platform is superseded by the

3167 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit

3168 variant, even on a 64 bit machine. The following example setting

3169 prefers 64 bit if available:

3171 ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"

3173 * The preliminary Isabelle/jEdit application demonstrates the emerging

3174 Isabelle/Scala layer for advanced prover interaction and integration.

3175 See src/Tools/jEdit or "isabelle jedit" provided by the properly built

3176 component.

3178 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono

3179 and Bluesky TeX fonts. It provides the usual Isabelle symbols,

3180 similar to the default assignment of the document preparation system

3181 (cf. isabellesym.sty). The Isabelle/Scala class Isabelle_System

3182 provides some operations for direct access to the font without asking

3183 the user for manual installation.

3187 New in Isabelle2009-1 (December 2009)

3188 -------------------------------------

3190 *** General ***

3192 * Discontinued old form of "escaped symbols" such as \\<forall>. Only

3193 one backslash should be used, even in ML sources.

3196 *** Pure ***

3198 * Locale interpretation propagates mixins along the locale hierarchy.

3199 The currently only available mixins are the equations used to map

3200 local definitions to terms of the target domain of an interpretation.

3202 * Reactivated diagnostic command 'print_interps'. Use "print_interps

3203 loc" to print all interpretations of locale "loc" in the theory.

3204 Interpretations in proofs are not shown.

3206 * Thoroughly revised locales tutorial. New section on conditional

3207 interpretation.

3209 * On instantiation of classes, remaining undefined class parameters

3210 are formally declared. INCOMPATIBILITY.

3213 *** Document preparation ***

3215 * New generalized style concept for printing terms: @{foo (style) ...}

3216 instead of @{foo_style style ...} (old form is still retained for

3217 backward compatibility). Styles can be also applied for

3218 antiquotations prop, term_type and typeof.

3221 *** HOL ***

3223 * New proof method "smt" for a combination of first-order logic with

3224 equality, linear and nonlinear (natural/integer/real) arithmetic, and

3225 fixed-size bitvectors; there is also basic support for higher-order

3226 features (esp. lambda abstractions). It is an incomplete decision

3227 procedure based on external SMT solvers using the oracle mechanism;

3228 for the SMT solver Z3, this method is proof-producing. Certificates

3229 are provided to avoid calling the external solvers solely for

3230 re-checking proofs. Due to a remote SMT service there is no need for

3231 installing SMT solvers locally. See src/HOL/SMT.

3233 * New commands to load and prove verification conditions generated by

3234 the Boogie program verifier or derived systems (e.g. the Verifying C

3235 Compiler (VCC) or Spec#). See src/HOL/Boogie.

3237 * New counterexample generator tool 'nitpick' based on the Kodkod

3238 relational model finder. See src/HOL/Tools/Nitpick and

3239 src/HOL/Nitpick_Examples.

3241 * New commands 'code_pred' and 'values' to invoke the predicate

3242 compiler and to enumerate values of inductive predicates.

3244 * A tabled implementation of the reflexive transitive closure.

3246 * New implementation of quickcheck uses generic code generator;

3247 default generators are provided for all suitable HOL types, records

3248 and datatypes. Old quickcheck can be re-activated importing theory

3249 Library/SML_Quickcheck.

3251 * New testing tool Mirabelle for automated proof tools. Applies

3252 several tools and tactics like sledgehammer, metis, or quickcheck, to

3253 every proof step in a theory. To be used in batch mode via the

3254 "mirabelle" utility.

3256 * New proof method "sos" (sum of squares) for nonlinear real

3257 arithmetic (originally due to John Harison). It requires theory

3258 Library/Sum_Of_Squares. It is not a complete decision procedure but

3259 works well in practice on quantifier-free real arithmetic with +, -,

3260 *, ^, =, <= and <, i.e. boolean combinations of equalities and

3261 inequalities between polynomials. It makes use of external

3262 semidefinite programming solvers. Method "sos" generates a

3263 certificate that can be pasted into the proof thus avoiding the need

3264 to call an external tool every time the proof is checked. See

3265 src/HOL/Library/Sum_Of_Squares.

3267 * New method "linarith" invokes existing linear arithmetic decision

3268 procedure only.

3270 * New command 'atp_minimal' reduces result produced by Sledgehammer.

3272 * New Sledgehammer option "Full Types" in Proof General settings menu.

3273 Causes full type information to be output to the ATPs. This slows

3274 ATPs down considerably but eliminates a source of unsound "proofs"

3275 that fail later.

3277 * New method "metisFT": A version of metis that uses full type

3278 information in order to avoid failures of proof reconstruction.

3280 * New evaluator "approximate" approximates an real valued term using

3281 the same method as the approximation method.

3283 * Method "approximate" now supports arithmetic expressions as

3284 boundaries of intervals and implements interval splitting and Taylor

3285 series expansion.

3287 * ML antiquotation @{code_datatype} inserts definition of a datatype

3288 generated by the code generator; e.g. see src/HOL/Predicate.thy.

3290 * New theory SupInf of the supremum and infimum operators for sets of

3291 reals.

3293 * New theory Probability, which contains a development of measure

3294 theory, eventually leading to Lebesgue integration and probability.

3296 * Extended Multivariate Analysis to include derivation and Brouwer's

3297 fixpoint theorem.

3299 * Reorganization of number theory, INCOMPATIBILITY:

3300 - new number theory development for nat and int, in theories Divides

3301 and GCD as well as in new session Number_Theory

3302 - some constants and facts now suffixed with _nat and _int

3303 accordingly

3304 - former session NumberTheory now named Old_Number_Theory, including

3305 theories Legacy_GCD and Primes (prefer Number_Theory if possible)

3306 - moved theory Pocklington from src/HOL/Library to

3307 src/HOL/Old_Number_Theory

3309 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and

3310 lcm of finite and infinite sets. It is shown that they form a complete

3311 lattice.

3313 * Class semiring_div requires superclass no_zero_divisors and proof of

3314 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,

3315 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been

3316 generalized to class semiring_div, subsuming former theorems

3317 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and

3318 zdiv_zmult_zmult2. div_mult_mult1 is now [simp] by default.

3319 INCOMPATIBILITY.

3321 * Refinements to lattice classes and sets:

3322 - less default intro/elim rules in locale variant, more default

3323 intro/elim rules in class variant: more uniformity

3324 - lemma ge_sup_conv renamed to le_sup_iff, in accordance with

3325 le_inf_iff

3326 - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and

3327 sup_aci)

3328 - renamed ACI to inf_sup_aci

3329 - new class "boolean_algebra"

3330 - class "complete_lattice" moved to separate theory

3331 "Complete_Lattice"; corresponding constants (and abbreviations)

3332 renamed and with authentic syntax:

3333 Set.Inf ~> Complete_Lattice.Inf

3334 Set.Sup ~> Complete_Lattice.Sup

3335 Set.INFI ~> Complete_Lattice.INFI

3336 Set.SUPR ~> Complete_Lattice.SUPR

3337 Set.Inter ~> Complete_Lattice.Inter

3338 Set.Union ~> Complete_Lattice.Union

3339 Set.INTER ~> Complete_Lattice.INTER

3340 Set.UNION ~> Complete_Lattice.UNION

3341 - authentic syntax for

3342 Set.Pow

3343 Set.image

3344 - mere abbreviations:

3345 Set.empty (for bot)

3346 Set.UNIV (for top)

3347 Set.inter (for inf, formerly Set.Int)

3348 Set.union (for sup, formerly Set.Un)

3349 Complete_Lattice.Inter (for Inf)

3350 Complete_Lattice.Union (for Sup)

3351 Complete_Lattice.INTER (for INFI)

3352 Complete_Lattice.UNION (for SUPR)

3353 - object-logic definitions as far as appropriate

3355 INCOMPATIBILITY. Care is required when theorems Int_subset_iff or

3356 Un_subset_iff are explicitly deleted as default simp rules; then also

3357 their lattice counterparts le_inf_iff and le_sup_iff have to be

3358 deleted to achieve the desired effect.

3360 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp

3361 rules by default any longer; the same applies to min_max.inf_absorb1

3362 etc. INCOMPATIBILITY.

3364 * Rules sup_Int_eq and sup_Un_eq are no longer declared as

3365 pred_set_conv by default. INCOMPATIBILITY.

3367 * Power operations on relations and functions are now one dedicated

3368 constant "compow" with infix syntax "^^". Power operation on

3369 multiplicative monoids retains syntax "^" and is now defined generic

3370 in class power. INCOMPATIBILITY.

3372 * Relation composition "R O S" now has a more standard argument order:

3373 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}". INCOMPATIBILITY,

3374 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally

3375 break, since the O_assoc rule was not rewritten like this. Fix using

3376 O_assoc[symmetric]. The same applies to the curried version "R OO S".

3378 * Function "Inv" is renamed to "inv_into" and function "inv" is now an

3379 abbreviation for "inv_into UNIV". Lemmas are renamed accordingly.

3380 INCOMPATIBILITY.

3382 * Most rules produced by inductive and datatype package have mandatory

3383 prefixes. INCOMPATIBILITY.

3385 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by

3386 the attribute of the same name. Each of the theorems in the list

3387 DERIV_intros assumes composition with an additional function and

3388 matches a variable to the derivative, which has to be solved by the

3389 Simplifier. Hence (auto intro!: DERIV_intros) computes the derivative

3390 of most elementary terms. Former Maclauren.DERIV_tac and

3391 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).

3392 INCOMPATIBILITY.

3394 * Code generator attributes follow the usual underscore convention:

3395 code_unfold replaces code unfold

3396 code_post replaces code post

3397 etc.

3398 INCOMPATIBILITY.

3400 * Renamed methods:

3401 sizechange -> size_change

3402 induct_scheme -> induction_schema

3403 INCOMPATIBILITY.

3405 * Discontinued abbreviation "arbitrary" of constant "undefined".

3406 INCOMPATIBILITY, use "undefined" directly.

3408 * Renamed theorems:

3409 Suc_eq_add_numeral_1 -> Suc_eq_plus1

3410 Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left

3411 Suc_plus1 -> Suc_eq_plus1

3412 *anti_sym -> *antisym*

3413 vector_less_eq_def -> vector_le_def

3414 INCOMPATIBILITY.

3416 * Added theorem List.map_map as [simp]. Removed List.map_compose.

3417 INCOMPATIBILITY.

3419 * Removed predicate "M hassize n" (<--> card M = n & finite M).

3420 INCOMPATIBILITY.

3423 *** HOLCF ***

3425 * Theory Representable defines a class "rep" of domains that are

3426 representable (via an ep-pair) in the universal domain type "udom".

3427 Instances are provided for all type constructors defined in HOLCF.

3429 * The 'new_domain' command is a purely definitional version of the

3430 domain package, for representable domains. Syntax is identical to the

3431 old domain package. The 'new_domain' package also supports indirect

3432 recursion using previously-defined type constructors. See

3433 src/HOLCF/ex/New_Domain.thy for examples.

3435 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant

3436 on the left-hand side of an equation, and then performs

3437 simplification. Rewriting is done using rules declared with the

3438 "fixrec_simp" attribute. The "fixrec_simp" method is intended as a

3439 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.

3441 * The pattern-match compiler in 'fixrec' can now handle constructors

3442 with HOL function types. Pattern-match combinators for the Pair

3443 constructor are pre-configured.

3445 * The 'fixrec' package now produces better fixed-point induction rules

3446 for mutually-recursive definitions: Induction rules have conclusions

3447 of the form "P foo bar" instead of "P <foo, bar>".

3449 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has

3450 been renamed to "below". The name "below" now replaces "less" in many

3451 theorem names. (Legacy theorem names using "less" are still supported

3452 as well.)

3454 * The 'fixrec' package now supports "bottom patterns". Bottom

3455 patterns can be used to generate strictness rules, or to make

3456 functions more strict (much like the bang-patterns supported by the

3457 Glasgow Haskell Compiler). See src/HOLCF/ex/Fixrec_ex.thy for

3458 examples.

3461 *** ML ***

3463 * Support for Poly/ML 5.3.0, with improved reporting of compiler

3464 errors and run-time exceptions, including detailed source positions.

3466 * Structure Name_Space (formerly NameSpace) now manages uniquely

3467 identified entries, with some additional information such as source

3468 position, logical grouping etc.

3470 * Theory and context data is now introduced by the simplified and

3471 modernized functors Theory_Data, Proof_Data, Generic_Data. Data needs

3472 to be pure, but the old TheoryDataFun for mutable data (with explicit

3473 copy operation) is still available for some time.

3475 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)

3476 provides a high-level programming interface to synchronized state

3477 variables with atomic update. This works via pure function

3478 application within a critical section -- its runtime should be as

3479 short as possible; beware of deadlocks if critical code is nested,

3480 either directly or indirectly via other synchronized variables!

3482 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)

3483 wraps raw ML references, explicitly indicating their non-thread-safe

3484 behaviour. The Isar toplevel keeps this structure open, to

3485 accommodate Proof General as well as quick and dirty interactive

3486 experiments with references.

3488 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for

3489 parallel tactical reasoning.

3491 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS

3492 are similar to SUBPROOF, but are slightly more flexible: only the

3493 specified parts of the subgoal are imported into the context, and the

3494 body tactic may introduce new subgoals and schematic variables.

3496 * Old tactical METAHYPS, which does not observe the proof context, has

3497 been renamed to Old_Goals.METAHYPS and awaits deletion. Use SUBPROOF

3498 or Subgoal.FOCUS etc.

3500 * Renamed functor TableFun to Table, and GraphFun to Graph. (Since

3501 functors have their own ML name space there is no point to mark them

3502 separately.) Minor INCOMPATIBILITY.

3504 * Renamed NamedThmsFun to Named_Thms. INCOMPATIBILITY.

3506 * Renamed several structures FooBar to Foo_Bar. Occasional,

3507 INCOMPATIBILITY.

3509 * Operations of structure Skip_Proof no longer require quick_and_dirty

3510 mode, which avoids critical setmp.

3512 * Eliminated old Attrib.add_attributes, Method.add_methods and related

3513 combinators for "args". INCOMPATIBILITY, need to use simplified

3514 Attrib/Method.setup introduced in Isabelle2009.

3516 * Proper context for simpset_of, claset_of, clasimpset_of. May fall

3517 back on global_simpset_of, global_claset_of, global_clasimpset_of as

3518 last resort. INCOMPATIBILITY.

3520 * Display.pretty_thm now requires a proper context (cf. former

3521 ProofContext.pretty_thm). May fall back on Display.pretty_thm_global

3522 or even Display.pretty_thm_without_context as last resort.

3523 INCOMPATIBILITY.

3525 * Discontinued Display.pretty_ctyp/cterm etc. INCOMPATIBILITY, use

3526 Syntax.pretty_typ/term directly, preferably with proper context

3527 instead of global theory.

3530 *** System ***

3532 * Further fine tuning of parallel proof checking, scales up to 8 cores

3533 (max. speedup factor 5.0). See also Goal.parallel_proofs in ML and

3534 usedir option -q.

3536 * Support for additional "Isabelle components" via etc/components, see

3537 also the system manual.

3539 * The isabelle makeall tool now operates on all components with

3540 IsaMakefile, not just hardwired "logics".

3542 * Removed "compress" option from isabelle-process and isabelle usedir;

3543 this is always enabled.

3545 * Discontinued support for Poly/ML 4.x versions.

3547 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'

3548 on a given logic image. This requires the lighttpd webserver and is

3549 currently supported on Linux only.

3553 New in Isabelle2009 (April 2009)

3554 --------------------------------

3556 *** General ***

3558 * Simplified main Isabelle executables, with less surprises on

3559 case-insensitive file-systems (such as Mac OS).

3561 - The main Isabelle tool wrapper is now called "isabelle" instead of

3562 "isatool."

3564 - The former "isabelle" alias for "isabelle-process" has been

3565 removed (should rarely occur to regular users).

3567 - The former "isabelle-interface" and its alias "Isabelle" have been

3568 removed (interfaces are now regular Isabelle tools).

3570 Within scripts and make files, the Isabelle environment variables

3571 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,

3572 respectively. (The latter are still available as legacy feature.)

3574 The old isabelle-interface wrapper could react in confusing ways if

3575 the interface was uninstalled or changed otherwise. Individual

3576 interface tool configuration is now more explicit, see also the

3577 Isabelle system manual. In particular, Proof General is now available

3578 via "isabelle emacs".

3580 INCOMPATIBILITY, need to adapt derivative scripts. Users may need to

3581 purge installed copies of Isabelle executables and re-run "isabelle

3582 install -p ...", or use symlinks.

3584 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the

3585 old ~/isabelle, which was slightly non-standard and apt to cause

3586 surprises on case-insensitive file-systems (such as Mac OS).

3588 INCOMPATIBILITY, need to move existing ~/isabelle/etc,

3589 ~/isabelle/heaps, ~/isabelle/browser_info to the new place. Special

3590 care is required when using older releases of Isabelle. Note that

3591 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any

3592 Isabelle distribution, in order to use the new ~/.isabelle uniformly.

3594 * Proofs of fully specified statements are run in parallel on

3595 multi-core systems. A speedup factor of 2.5 to 3.2 can be expected on

3596 a regular 4-core machine, if the initial heap space is made reasonably

3597 large (cf. Poly/ML option -H). (Requires Poly/ML 5.2.1 or later.)

3599 * The main reference manuals ("isar-ref", "implementation", and

3600 "system") have been updated and extended. Formally checked references

3601 as hyperlinks are now available uniformly.

3604 *** Pure ***

3606 * Complete re-implementation of locales. INCOMPATIBILITY in several

3607 respects. The most important changes are listed below. See the

3608 Tutorial on Locales ("locales" manual) for details.

3610 - In locale expressions, instantiation replaces renaming. Parameters

3611 must be declared in a for clause. To aid compatibility with previous

3612 parameter inheritance, in locale declarations, parameters that are not

3613 'touched' (instantiation position "_" or omitted) are implicitly added

3614 with their syntax at the beginning of the for clause.

3616 - Syntax from abbreviations and definitions in locales is available in

3617 locale expressions and context elements. The latter is particularly

3618 useful in locale declarations.

3620 - More flexible mechanisms to qualify names generated by locale

3621 expressions. Qualifiers (prefixes) may be specified in locale

3622 expressions, and can be marked as mandatory (syntax: "name!:") or

3623 optional (syntax "name?:"). The default depends for plain "name:"

3624 depends on the situation where a locale expression is used: in

3625 commands 'locale' and 'sublocale' prefixes are optional, in

3626 'interpretation' and 'interpret' prefixes are mandatory. The old

3627 implicit qualifiers derived from the parameter names of a locale are

3628 no longer generated.

3630 - Command "sublocale l < e" replaces "interpretation l < e". The

3631 instantiation clause in "interpretation" and "interpret" (square

3632 brackets) is no longer available. Use locale expressions.

3634 - When converting proof scripts, mandatory qualifiers in

3635 'interpretation' and 'interpret' should be retained by default, even

3636 if this is an INCOMPATIBILITY compared to former behavior. In the

3637 worst case, use the "name?:" form for non-mandatory ones. Qualifiers

3638 in locale expressions range over a single locale instance only.

3640 - Dropped locale element "includes". This is a major INCOMPATIBILITY.

3641 In existing theorem specifications replace the includes element by the

3642 respective context elements of the included locale, omitting those

3643 that are already present in the theorem specification. Multiple

3644 assume elements of a locale should be replaced by a single one

3645 involving the locale predicate. In the proof body, declarations (most

3646 notably theorems) may be regained by interpreting the respective

3647 locales in the proof context as required (command "interpret").

3649 If using "includes" in replacement of a target solely because the

3650 parameter types in the theorem are not as general as in the target,

3651 consider declaring a new locale with additional type constraints on

3652 the parameters (context element "constrains").

3654 - Discontinued "locale (open)". INCOMPATIBILITY.

3656 - Locale interpretation commands no longer attempt to simplify goal.

3657 INCOMPATIBILITY: in rare situations the generated goal differs. Use

3658 methods intro_locales and unfold_locales to clarify.

3660 - Locale interpretation commands no longer accept interpretation

3661 attributes. INCOMPATIBILITY.

3663 * Class declaration: so-called "base sort" must not be given in import

3664 list any longer, but is inferred from the specification. Particularly

3665 in HOL, write

3667 class foo = ...

3669 instead of

3671 class foo = type + ...

3673 * Class target: global versions of theorems stemming do not carry a

3674 parameter prefix any longer. INCOMPATIBILITY.

3676 * Class 'instance' command no longer accepts attached definitions.

3677 INCOMPATIBILITY, use proper 'instantiation' target instead.

3679 * Recovered hiding of consts, which was accidentally broken in

3680 Isabelle2007. Potential INCOMPATIBILITY, ``hide const c'' really

3681 makes c inaccessible; consider using ``hide (open) const c'' instead.

3683 * Slightly more coherent Pure syntax, with updated documentation in

3684 isar-ref manual. Removed locales meta_term_syntax and

3685 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,

3686 INCOMPATIBILITY in rare situations. Note that &&& should not be used

3687 directly in regular applications.

3689 * There is a new syntactic category "float_const" for signed decimal

3690 fractions (e.g. 123.45 or -123.45).

3692 * Removed exotic 'token_translation' command. INCOMPATIBILITY, use ML

3693 interface with 'setup' command instead.

3695 * Command 'local_setup' is similar to 'setup', but operates on a local

3696 theory context.

3698 * The 'axiomatization' command now only works within a global theory

3699 context. INCOMPATIBILITY.

3701 * Goal-directed proof now enforces strict proof irrelevance wrt. sort

3702 hypotheses. Sorts required in the course of reasoning need to be

3703 covered by the constraints in the initial statement, completed by the

3704 type instance information of the background theory. Non-trivial sort

3705 hypotheses, which rarely occur in practice, may be specified via

3706 vacuous propositions of the form SORT_CONSTRAINT('a::c). For example:

3708 lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...

3710 The result contains an implicit sort hypotheses as before --

3711 SORT_CONSTRAINT premises are eliminated as part of the canonical rule

3712 normalization.

3714 * Generalized Isar history, with support for linear undo, direct state

3715 addressing etc.

3717 * Changed defaults for unify configuration options:

3719 unify_trace_bound = 50 (formerly 25)

3720 unify_search_bound = 60 (formerly 30)

3722 * Different bookkeeping for code equations (INCOMPATIBILITY):

3724 a) On theory merge, the last set of code equations for a particular

3725 constant is taken (in accordance with the policy applied by other

3726 parts of the code generator framework).

3728 b) Code equations stemming from explicit declarations (e.g. code

3729 attribute) gain priority over default code equations stemming

3730 from definition, primrec, fun etc.

3732 * Keyword 'code_exception' now named 'code_abort'. INCOMPATIBILITY.

3734 * Unified theorem tables for both code generators. Thus [code

3735 func] has disappeared and only [code] remains. INCOMPATIBILITY.

3737 * Command 'find_consts' searches for constants based on type and name

3738 patterns, e.g.

3740 find_consts "_ => bool"

3742 By default, matching is against subtypes, but it may be restricted to

3743 the whole type. Searching by name is possible. Multiple queries are

3744 conjunctive and queries may be negated by prefixing them with a

3745 hyphen:

3747 find_consts strict: "_ => bool" name: "Int" -"int => int"

3749 * New 'find_theorems' criterion "solves" matches theorems that

3750 directly solve the current goal (modulo higher-order unification).

3752 * Auto solve feature for main theorem statements: whenever a new goal

3753 is stated, "find_theorems solves" is called; any theorems that could

3754 solve the lemma directly are listed as part of the goal state.

3755 Cf. associated options in Proof General Isabelle settings menu,

3756 enabled by default, with reasonable timeout for pathological cases of

3757 higher-order unification.

3760 *** Document preparation ***

3762 * Antiquotation @{lemma} now imitates a regular terminal proof,

3763 demanding keyword 'by' and supporting the full method expression

3764 syntax just like the Isar command 'by'.

3767 *** HOL ***

3769 * Integrated main parts of former image HOL-Complex with HOL. Entry

3770 points Main and Complex_Main remain as before.

3772 * Logic image HOL-Plain provides a minimal HOL with the most important

3773 tools available (inductive, datatype, primrec, ...). This facilitates

3774 experimentation and tool development. Note that user applications

3775 (and library theories) should never refer to anything below theory

3776 Main, as before.

3778 * Logic image HOL-Main stops at theory Main, and thus facilitates

3779 experimentation due to shorter build times.

3781 * Logic image HOL-NSA contains theories of nonstandard analysis which

3782 were previously part of former HOL-Complex. Entry point Hyperreal

3783 remains valid, but theories formerly using Complex_Main should now use

3784 new entry point Hypercomplex.

3786 * Generic ATP manager for Sledgehammer, based on ML threads instead of

3787 Posix processes. Avoids potentially expensive forking of the ML

3788 process. New thread-based implementation also works on non-Unix

3789 platforms (Cygwin). Provers are no longer hardwired, but defined

3790 within the theory via plain ML wrapper functions. Basic Sledgehammer

3791 commands are covered in the isar-ref manual.

3793 * Wrapper scripts for remote SystemOnTPTP service allows to use

3794 sledgehammer without local ATP installation (Vampire etc.). Other

3795 provers may be included via suitable ML wrappers, see also

3796 src/HOL/ATP_Linkup.thy.

3798 * ATP selection (E/Vampire/Spass) is now via Proof General's settings

3799 menu.

3801 * The metis method no longer fails because the theorem is too trivial

3802 (contains the empty clause).

3804 * The metis method now fails in the usual manner, rather than raising

3805 an exception, if it determines that it cannot prove the theorem.

3807 * Method "coherent" implements a prover for coherent logic (see also

3808 src/Tools/coherent.ML).

3810 * Constants "undefined" and "default" replace "arbitrary". Usually

3811 "undefined" is the right choice to replace "arbitrary", though

3812 logically there is no difference. INCOMPATIBILITY.

3814 * Command "value" now integrates different evaluation mechanisms. The

3815 result of the first successful evaluation mechanism is printed. In

3816 square brackets a particular named evaluation mechanisms may be

3817 specified (currently, [SML], [code] or [nbe]). See further

3818 src/HOL/ex/Eval_Examples.thy.

3820 * Normalization by evaluation now allows non-leftlinear equations.

3821 Declare with attribute [code nbe].

3823 * Methods "case_tac" and "induct_tac" now refer to the very same rules

3824 as the structured Isar versions "cases" and "induct", cf. the

3825 corresponding "cases" and "induct" attributes. Mutual induction rules

3826 are now presented as a list of individual projections

3827 (e.g. foo_bar.inducts for types foo and bar); the old format with

3828 explicit HOL conjunction is no longer supported. INCOMPATIBILITY, in

3829 rare situations a different rule is selected --- notably nested tuple

3830 elimination instead of former prod.exhaust: use explicit (case_tac t

3831 rule: prod.exhaust) here.

3833 * Attributes "cases", "induct", "coinduct" support "del" option.

3835 * Removed fact "case_split_thm", which duplicates "case_split".

3837 * The option datatype has been moved to a new theory Option. Renamed

3838 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.

3840 * New predicate "strict_mono" classifies strict functions on partial

3841 orders. With strict functions on linear orders, reasoning about

3842 (in)equalities is facilitated by theorems "strict_mono_eq",

3843 "strict_mono_less_eq" and "strict_mono_less".

3845 * Some set operations are now proper qualified constants with

3846 authentic syntax. INCOMPATIBILITY:

3848 op Int ~> Set.Int

3849 op Un ~> Set.Un

3850 INTER ~> Set.INTER

3851 UNION ~> Set.UNION

3852 Inter ~> Set.Inter

3853 Union ~> Set.Union

3854 {} ~> Set.empty

3855 UNIV ~> Set.UNIV

3857 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in

3858 theory Set.

3860 * Auxiliary class "itself" has disappeared -- classes without any

3861 parameter are treated as expected by the 'class' command.

3863 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.

3865 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,

3866 Dense_Linear_Order) are now in directory HOL/Decision_Procs.

3868 * Theory src/HOL/Decision_Procs/Approximation provides the new proof

3869 method "approximation". It proves formulas on real values by using

3870 interval arithmetic. In the formulas are also the transcendental

3871 functions sin, cos, tan, atan, ln, exp and the constant pi are

3872 allowed. For examples see

3873 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.

3875 * Theory "Reflection" now resides in HOL/Library.

3877 * Entry point to Word library now simply named "Word".

3878 INCOMPATIBILITY.

3880 * Made source layout more coherent with logical distribution

3881 structure:

3883 src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy

3884 src/HOL/Library/Code_Message.thy ~> src/HOL/

3885 src/HOL/Library/GCD.thy ~> src/HOL/

3886 src/HOL/Library/Order_Relation.thy ~> src/HOL/

3887 src/HOL/Library/Parity.thy ~> src/HOL/

3888 src/HOL/Library/Univ_Poly.thy ~> src/HOL/

3889 src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/

3890 src/HOL/Real/Lubs.thy ~> src/HOL/

3891 src/HOL/Real/PReal.thy ~> src/HOL/

3892 src/HOL/Real/Rational.thy ~> src/HOL/

3893 src/HOL/Real/RComplete.thy ~> src/HOL/

3894 src/HOL/Real/RealDef.thy ~> src/HOL/

3895 src/HOL/Real/RealPow.thy ~> src/HOL/

3896 src/HOL/Real/Real.thy ~> src/HOL/

3897 src/HOL/Complex/Complex_Main.thy ~> src/HOL/

3898 src/HOL/Complex/Complex.thy ~> src/HOL/

3899 src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/

3900 src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/

3901 src/HOL/Hyperreal/Deriv.thy ~> src/HOL/

3902 src/HOL/Hyperreal/Fact.thy ~> src/HOL/

3903 src/HOL/Hyperreal/Integration.thy ~> src/HOL/

3904 src/HOL/Hyperreal/Lim.thy ~> src/HOL/

3905 src/HOL/Hyperreal/Ln.thy ~> src/HOL/

3906 src/HOL/Hyperreal/Log.thy ~> src/HOL/

3907 src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/

3908 src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/

3909 src/HOL/Hyperreal/Series.thy ~> src/HOL/

3910 src/HOL/Hyperreal/SEQ.thy ~> src/HOL/

3911 src/HOL/Hyperreal/Taylor.thy ~> src/HOL/

3912 src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/

3913 src/HOL/Real/Float ~> src/HOL/Library/

3914 src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach

3915 src/HOL/Real/RealVector.thy ~> src/HOL/

3917 src/HOL/arith_data.ML ~> src/HOL/Tools

3918 src/HOL/hologic.ML ~> src/HOL/Tools

3919 src/HOL/simpdata.ML ~> src/HOL/Tools

3920 src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML

3921 src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools

3922 src/HOL/nat_simprocs.ML ~> src/HOL/Tools

3923 src/HOL/Real/float_arith.ML ~> src/HOL/Tools

3924 src/HOL/Real/float_syntax.ML ~> src/HOL/Tools

3925 src/HOL/Real/rat_arith.ML ~> src/HOL/Tools

3926 src/HOL/Real/real_arith.ML ~> src/HOL/Tools

3928 src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL

3929 src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL

3930 src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL

3931 src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL

3932 src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL

3933 src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL

3935 * If methods "eval" and "evaluation" encounter a structured proof

3936 state with !!/==>, only the conclusion is evaluated to True (if

3937 possible), avoiding strange error messages.

3939 * Method "sizechange" automates termination proofs using (a

3940 modification of) the size-change principle. Requires SAT solver. See

3941 src/HOL/ex/Termination.thy for examples.

3943 * Simplifier: simproc for let expressions now unfolds if bound

3944 variable occurs at most once in let expression body. INCOMPATIBILITY.

3946 * Method "arith": Linear arithmetic now ignores all inequalities when

3947 fast_arith_neq_limit is exceeded, instead of giving up entirely.

3949 * New attribute "arith" for facts that should always be used

3950 automatically by arithmetic. It is intended to be used locally in

3951 proofs, e.g.

3953 assumes [arith]: "x > 0"

3955 Global usage is discouraged because of possible performance impact.

3957 * New classes "top" and "bot" with corresponding operations "top" and

3958 "bot" in theory Orderings; instantiation of class "complete_lattice"

3959 requires instantiation of classes "top" and "bot". INCOMPATIBILITY.

3961 * Changed definition lemma "less_fun_def" in order to provide an

3962 instance for preorders on functions; use lemma "less_le" instead.

3963 INCOMPATIBILITY.

3965 * Theory Orderings: class "wellorder" moved here, with explicit

3966 induction rule "less_induct" as assumption. For instantiation of

3967 "wellorder" by means of predicate "wf", use rule wf_wellorderI.

3968 INCOMPATIBILITY.

3970 * Theory Orderings: added class "preorder" as superclass of "order".

3971 INCOMPATIBILITY: Instantiation proofs for order, linorder

3972 etc. slightly changed. Some theorems named order_class.* now named

3973 preorder_class.*.

3975 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,

3976 "diag" to "Id_on".

3978 * Theory Finite_Set: added a new fold combinator of type

3980 ('a => 'b => 'b) => 'b => 'a set => 'b

3982 Occasionally this is more convenient than the old fold combinator

3983 which is now defined in terms of the new one and renamed to

3984 fold_image.

3986 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"

3987 and "ring_simps" have been replaced by "algebra_simps" (which can be

3988 extended with further lemmas!). At the moment both still exist but

3989 the former will disappear at some point.

3991 * Theory Power: Lemma power_Suc is now declared as a simp rule in

3992 class recpower. Type-specific simp rules for various recpower types

3993 have been removed. INCOMPATIBILITY, rename old lemmas as follows:

3995 rat_power_0 -> power_0

3996 rat_power_Suc -> power_Suc

3997 realpow_0 -> power_0

3998 realpow_Suc -> power_Suc

3999 complexpow_0 -> power_0

4000 complexpow_Suc -> power_Suc

4001 power_poly_0 -> power_0

4002 power_poly_Suc -> power_Suc

4004 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been

4005 moved to separate class dvd in Ring_and_Field; a couple of lemmas on

4006 dvd has been generalized to class comm_semiring_1. Likewise a bunch

4007 of lemmas from Divides has been generalized from nat to class

4008 semiring_div. INCOMPATIBILITY. This involves the following theorem

4009 renames resulting from duplicate elimination:

4011 dvd_def_mod ~> dvd_eq_mod_eq_0

4012 zero_dvd_iff ~> dvd_0_left_iff

4013 dvd_0 ~> dvd_0_right

4014 DIVISION_BY_ZERO_DIV ~> div_by_0

4015 DIVISION_BY_ZERO_MOD ~> mod_by_0

4016 mult_div ~> div_mult_self2_is_id

4017 mult_mod ~> mod_mult_self2_is_0

4019 * Theory IntDiv: removed many lemmas that are instances of class-based

4020 generalizations (from Divides and Ring_and_Field). INCOMPATIBILITY,

4021 rename old lemmas as follows:

4023 dvd_diff -> nat_dvd_diff

4024 dvd_zminus_iff -> dvd_minus_iff

4025 mod_add1_eq -> mod_add_eq

4026 mod_mult1_eq -> mod_mult_right_eq

4027 mod_mult1_eq' -> mod_mult_left_eq

4028 mod_mult_distrib_mod -> mod_mult_eq

4029 nat_mod_add_left_eq -> mod_add_left_eq

4030 nat_mod_add_right_eq -> mod_add_right_eq

4031 nat_mod_div_trivial -> mod_div_trivial

4032 nat_mod_mod_trivial -> mod_mod_trivial

4033 zdiv_zadd_self1 -> div_add_self1

4034 zdiv_zadd_self2 -> div_add_self2

4035 zdiv_zmult_self1 -> div_mult_self2_is_id

4036 zdiv_zmult_self2 -> div_mult_self1_is_id

4037 zdvd_triv_left -> dvd_triv_left

4038 zdvd_triv_right -> dvd_triv_right

4039 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left

4040 zmod_eq0_zdvd_iff -> dvd_eq_mod_eq_0[symmetric]

4041 zmod_zadd_left_eq -> mod_add_left_eq

4042 zmod_zadd_right_eq -> mod_add_right_eq

4043 zmod_zadd_self1 -> mod_add_self1

4044 zmod_zadd_self2 -> mod_add_self2

4045 zmod_zadd1_eq -> mod_add_eq

4046 zmod_zdiff1_eq -> mod_diff_eq

4047 zmod_zdvd_zmod -> mod_mod_cancel

4048 zmod_zmod_cancel -> mod_mod_cancel

4049 zmod_zmult_self1 -> mod_mult_self2_is_0

4050 zmod_zmult_self2 -> mod_mult_self1_is_0

4051 zmod_1 -> mod_by_1

4052 zdiv_1 -> div_by_1

4053 zdvd_abs1 -> abs_dvd_iff

4054 zdvd_abs2 -> dvd_abs_iff

4055 zdvd_refl -> dvd_refl

4056 zdvd_trans -> dvd_trans

4057 zdvd_zadd -> dvd_add

4058 zdvd_zdiff -> dvd_diff

4059 zdvd_zminus_iff -> dvd_minus_iff

4060 zdvd_zminus2_iff -> minus_dvd_iff

4061 zdvd_zmultD -> dvd_mult_right

4062 zdvd_zmultD2 -> dvd_mult_left

4063 zdvd_zmult_mono -> mult_dvd_mono

4064 zdvd_0_right -> dvd_0_right

4065 zdvd_0_left -> dvd_0_left_iff

4066 zdvd_1_left -> one_dvd

4067 zminus_dvd_iff -> minus_dvd_iff

4069 * Theory Rational: 'Fract k 0' now equals '0'. INCOMPATIBILITY.

4071 * The real numbers offer decimal input syntax: 12.34 is translated

4072 into 1234/10^2. This translation is not reversed upon output.

4074 * Theory Library/Polynomial defines an abstract type 'a poly of

4075 univariate polynomials with coefficients of type 'a. In addition to

4076 the standard ring operations, it also supports div and mod. Code

4077 generation is also supported, using list-style constructors.

4079 * Theory Library/Inner_Product defines a class of real_inner for real

4080 inner product spaces, with an overloaded operation inner :: 'a => 'a

4081 => real. Class real_inner is a subclass of real_normed_vector from

4082 theory RealVector.

4084 * Theory Library/Product_Vector provides instances for the product

4085 type 'a * 'b of several classes from RealVector and Inner_Product.

4086 Definitions of addition, subtraction, scalar multiplication, norms,

4087 and inner products are included.

4089 * Theory Library/Bit defines the field "bit" of integers modulo 2. In

4090 addition to the field operations, numerals and case syntax are also

4091 supported.

4093 * Theory Library/Diagonalize provides constructive version of Cantor's

4094 first diagonalization argument.

4096 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,

4097 zlcm (for int); carried together from various gcd/lcm developements in

4098 the HOL Distribution. Constants zgcd and zlcm replace former igcd and

4099 ilcm; corresponding theorems renamed accordingly. INCOMPATIBILITY,

4100 may recover tupled syntax as follows:

4102 hide (open) const gcd

4103 abbreviation gcd where

4104 "gcd == (%(a, b). GCD.gcd a b)"

4105 notation (output)

4106 GCD.gcd ("gcd '(_, _')")

4108 The same works for lcm, zgcd, zlcm.

4110 * Theory Library/Nat_Infinity: added addition, numeral syntax and more

4111 instantiations for algebraic structures. Removed some duplicate

4112 theorems. Changes in simp rules. INCOMPATIBILITY.

4114 * ML antiquotation @{code} takes a constant as argument and generates

4115 corresponding code in background and inserts name of the corresponding

4116 resulting ML value/function/datatype constructor binding in place.

4117 All occurrences of @{code} with a single ML block are generated

4118 simultaneously. Provides a generic and safe interface for

4119 instrumentalizing code generation. See

4120 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.

4121 In future you ought to refrain from ad-hoc compiling generated SML

4122 code on the ML toplevel. Note that (for technical reasons) @{code}

4123 cannot refer to constants for which user-defined serializations are

4124 set. Refer to the corresponding ML counterpart directly in that

4125 cases.

4127 * Command 'rep_datatype': instead of theorem names the command now

4128 takes a list of terms denoting the constructors of the type to be

4129 represented as datatype. The characteristic theorems have to be

4130 proven. INCOMPATIBILITY. Also observe that the following theorems

4131 have disappeared in favour of existing ones:

4133 unit_induct ~> unit.induct

4134 prod_induct ~> prod.induct

4135 sum_induct ~> sum.induct

4136 Suc_Suc_eq ~> nat.inject

4137 Suc_not_Zero Zero_not_Suc ~> nat.distinct

4140 *** HOL-Algebra ***

4142 * New locales for orders and lattices where the equivalence relation

4143 is not restricted to equality. INCOMPATIBILITY: all order and lattice

4144 locales use a record structure with field eq for the equivalence.

4146 * New theory of factorial domains.

4148 * Units_l_inv and Units_r_inv are now simp rules by default.

4149 INCOMPATIBILITY. Simplifier proof that require deletion of l_inv

4150 and/or r_inv will now also require deletion of these lemmas.

4152 * Renamed the following theorems, INCOMPATIBILITY:

4154 UpperD ~> Upper_memD

4155 LowerD ~> Lower_memD

4156 least_carrier ~> least_closed

4157 greatest_carrier ~> greatest_closed

4158 greatest_Lower_above ~> greatest_Lower_below

4159 one_zero ~> carrier_one_zero

4160 one_not_zero ~> carrier_one_not_zero (collision with assumption)

4163 *** HOL-Nominal ***

4165 * Nominal datatypes can now contain type-variables.

4167 * Commands 'nominal_inductive' and 'equivariance' work with local

4168 theory targets.

4170 * Nominal primrec can now works with local theory targets and its

4171 specification syntax now conforms to the general format as seen in

4172 'inductive' etc.

4174 * Method "perm_simp" honours the standard simplifier attributes

4175 (no_asm), (no_asm_use) etc.

4177 * The new predicate #* is defined like freshness, except that on the

4178 left hand side can be a set or list of atoms.

4180 * Experimental command 'nominal_inductive2' derives strong induction

4181 principles for inductive definitions. In contrast to

4182 'nominal_inductive', which can only deal with a fixed number of

4183 binders, it can deal with arbitrary expressions standing for sets of

4184 atoms to be avoided. The only inductive definition we have at the

4185 moment that needs this generalisation is the typing rule for Lets in

4186 the algorithm W:

4188 Gamma |- t1 : T1 (x,close Gamma T1)::Gamma |- t2 : T2 x#Gamma

4189 -----------------------------------------------------------------

4190 Gamma |- Let x be t1 in t2 : T2

4192 In this rule one wants to avoid all the binders that are introduced by

4193 "close Gamma T1". We are looking for other examples where this

4194 feature might be useful. Please let us know.

4197 *** HOLCF ***

4199 * Reimplemented the simplification procedure for proving continuity

4200 subgoals. The new simproc is extensible; users can declare additional

4201 continuity introduction rules with the attribute [cont2cont].

4203 * The continuity simproc now uses a different introduction rule for

4204 solving continuity subgoals on terms with lambda abstractions. In

4205 some rare cases the new simproc may fail to solve subgoals that the

4206 old one could solve, and "simp add: cont2cont_LAM" may be necessary.

4207 Potential INCOMPATIBILITY.

4209 * Command 'fixrec': specification syntax now conforms to the general

4210 format as seen in 'inductive' etc. See src/HOLCF/ex/Fixrec_ex.thy for

4211 examples. INCOMPATIBILITY.

4214 *** ZF ***

4216 * Proof of Zorn's Lemma for partial orders.

4219 *** ML ***

4221 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for

4222 Poly/ML 5.2.1 or later. Important note: the TimeLimit facility

4223 depends on multithreading, so timouts will not work before Poly/ML

4224 5.2.1!

4226 * High-level support for concurrent ML programming, see

4227 src/Pure/Cuncurrent. The data-oriented model of "future values" is

4228 particularly convenient to organize independent functional

4229 computations. The concept of "synchronized variables" provides a

4230 higher-order interface for components with shared state, avoiding the

4231 delicate details of mutexes and condition variables. (Requires

4232 Poly/ML 5.2.1 or later.)

4234 * ML bindings produced via Isar commands are stored within the Isar

4235 context (theory or proof). Consequently, commands like 'use' and 'ML'

4236 become thread-safe and work with undo as expected (concerning

4237 top-level bindings, not side-effects on global references).

4238 INCOMPATIBILITY, need to provide proper Isar context when invoking the

4239 compiler at runtime; really global bindings need to be given outside a

4240 theory. (Requires Poly/ML 5.2 or later.)

4242 * Command 'ML_prf' is analogous to 'ML' but works within a proof

4243 context. Top-level ML bindings are stored within the proof context in

4244 a purely sequential fashion, disregarding the nested proof structure.

4245 ML bindings introduced by 'ML_prf' are discarded at the end of the

4246 proof. (Requires Poly/ML 5.2 or later.)

4248 * Simplified ML attribute and method setup, cf. functions Attrib.setup

4249 and Method.setup, as well as Isar commands 'attribute_setup' and

4250 'method_setup'. INCOMPATIBILITY for 'method_setup', need to simplify

4251 existing code accordingly, or use plain 'setup' together with old

4252 Method.add_method.

4254 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm

4255 to 'a -> thm, while results are always tagged with an authentic oracle

4256 name. The Isar command 'oracle' is now polymorphic, no argument type

4257 is specified. INCOMPATIBILITY, need to simplify existing oracle code

4258 accordingly. Note that extra performance may be gained by producing

4259 the cterm carefully, avoiding slow Thm.cterm_of.

4261 * Simplified interface for defining document antiquotations via

4262 ThyOutput.antiquotation, ThyOutput.output, and optionally

4263 ThyOutput.maybe_pretty_source. INCOMPATIBILITY, need to simplify user

4264 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common

4265 examples.

4267 * More systematic treatment of long names, abstract name bindings, and

4268 name space operations. Basic operations on qualified names have been

4269 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,

4270 Long_Name.append. Old type bstring has been mostly replaced by

4271 abstract type binding (see structure Binding), which supports precise

4272 qualification by packages and local theory targets, as well as proper

4273 tracking of source positions. INCOMPATIBILITY, need to wrap old

4274 bstring values into Binding.name, or better pass through abstract

4275 bindings everywhere. See further src/Pure/General/long_name.ML,

4276 src/Pure/General/binding.ML and src/Pure/General/name_space.ML

4278 * Result facts (from PureThy.note_thms, ProofContext.note_thms,

4279 LocalTheory.note etc.) now refer to the *full* internal name, not the

4280 bstring as before. INCOMPATIBILITY, not detected by ML type-checking!

4282 * Disposed old type and term read functions (Sign.read_def_typ,

4283 Sign.read_typ, Sign.read_def_terms, Sign.read_term,

4284 Thm.read_def_cterms, Thm.read_cterm etc.). INCOMPATIBILITY, should

4285 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,

4286 Syntax.read_term_global etc.; see also OldGoals.read_term as last

4287 resort for legacy applications.

4289 * Disposed old declarations, tactics, tactic combinators that refer to

4290 the simpset or claset of an implicit theory (such as Addsimps,

4291 Simp_tac, SIMPSET). INCOMPATIBILITY, should use @{simpset} etc. in

4292 embedded ML text, or local_simpset_of with a proper context passed as

4293 explicit runtime argument.

4295 * Rules and tactics that read instantiations (read_instantiate,

4296 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof

4297 context, which is required for parsing and type-checking. Moreover,

4298 the variables are specified as plain indexnames, not string encodings

4299 thereof. INCOMPATIBILITY.

4301 * Generic Toplevel.add_hook interface allows to analyze the result of

4302 transactions. E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML

4303 for theorem dependency output of transactions resulting in a new

4304 theory state.

4306 * ML antiquotations: block-structured compilation context indicated by

4307 \<lbrace> ... \<rbrace>; additional antiquotation forms:

4309 @{binding name} - basic name binding

4310 @{let ?pat = term} - term abbreviation (HO matching)

4311 @{note name = fact} - fact abbreviation

4312 @{thm fact} - singleton fact (with attributes)

4313 @{thms fact} - general fact (with attributes)

4314 @{lemma prop by method} - singleton goal

4315 @{lemma prop by meth1 meth2} - singleton goal

4316 @{lemma prop1 ... propN by method} - general goal

4317 @{lemma prop1 ... propN by meth1 meth2} - general goal

4318 @{lemma (open) ...} - open derivation

4321 *** System ***

4323 * The Isabelle "emacs" tool provides a specific interface to invoke

4324 Proof General / Emacs, with more explicit failure if that is not

4325 installed (the old isabelle-interface script silently falls back on

4326 isabelle-process). The PROOFGENERAL_HOME setting determines the

4327 installation location of the Proof General distribution.

4329 * Isabelle/lib/classes/Pure.jar provides basic support to integrate

4330 the Isabelle process into a JVM/Scala application. See

4331 Isabelle/lib/jedit/plugin for a minimal example. (The obsolete Java

4332 process wrapper has been discontinued.)

4334 * Added homegrown Isabelle font with unicode layout, see lib/fonts.

4336 * Various status messages (with exact source position information) are

4337 emitted, if proper markup print mode is enabled. This allows

4338 user-interface components to provide detailed feedback on internal

4339 prover operations.

4343 New in Isabelle2008 (June 2008)

4344 -------------------------------

4346 *** General ***

4348 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized

4349 and updated, with formally checked references as hyperlinks.

4351 * Theory loader: use_thy (and similar operations) no longer set the

4352 implicit ML context, which was occasionally hard to predict and in

4353 conflict with concurrency. INCOMPATIBILITY, use ML within Isar which

4354 provides a proper context already.

4356 * Theory loader: old-style ML proof scripts being *attached* to a thy

4357 file are no longer supported. INCOMPATIBILITY, regular 'uses' and

4358 'use' within a theory file will do the job.

4360 * Name space merge now observes canonical order, i.e. the second space

4361 is inserted into the first one, while existing entries in the first

4362 space take precedence. INCOMPATIBILITY in rare situations, may try to

4363 swap theory imports.

4365 * Syntax: symbol \<chi> is now considered a letter. Potential

4366 INCOMPATIBILITY in identifier syntax etc.

4368 * Outer syntax: string tokens no longer admit escaped white space,

4369 which was an accidental (undocumented) feature. INCOMPATIBILITY, use

4370 white space without escapes.

4372 * Outer syntax: string tokens may contain arbitrary character codes

4373 specified via 3 decimal digits (as in SML). E.g. "foo\095bar" for

4374 "foo_bar".

4377 *** Pure ***

4379 * Context-dependent token translations. Default setup reverts locally

4380 fixed variables, and adds hilite markup for undeclared frees.

4382 * Unused theorems can be found using the new command 'unused_thms'.

4383 There are three ways of invoking it:

4385 (1) unused_thms

4386 Only finds unused theorems in the current theory.

4388 (2) unused_thms thy_1 ... thy_n -

4389 Finds unused theorems in the current theory and all of its ancestors,

4390 excluding the theories thy_1 ... thy_n and all of their ancestors.

4392 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m

4393 Finds unused theorems in the theories thy'_1 ... thy'_m and all of

4394 their ancestors, excluding the theories thy_1 ... thy_n and all of

4395 their ancestors.

4397 In order to increase the readability of the list produced by

4398 unused_thms, theorems that have been created by a particular instance

4399 of a theory command such as 'inductive' or 'function' are considered

4400 to belong to the same "group", meaning that if at least one theorem in

4401 this group is used, the other theorems in the same group are no longer

4402 reported as unused. Moreover, if all theorems in the group are

4403 unused, only one theorem in the group is displayed.

4405 Note that proof objects have to be switched on in order for

4406 unused_thms to work properly (i.e. !proofs must be >= 1, which is

4407 usually the case when using Proof General with the default settings).

4409 * Authentic naming of facts disallows ad-hoc overwriting of previous

4410 theorems within the same name space. INCOMPATIBILITY, need to remove

4411 duplicate fact bindings, or even accidental fact duplications. Note

4412 that tools may maintain dynamically scoped facts systematically, using

4413 PureThy.add_thms_dynamic.

4415 * Command 'hide' now allows to hide from "fact" name space as well.

4417 * Eliminated destructive theorem database, simpset, claset, and

4418 clasimpset. Potential INCOMPATIBILITY, really need to observe linear

4419 update of theories within ML code.

4421 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.

4422 INCOMPATIBILITY, object-logics depending on former Pure require

4423 additional setup PureThy.old_appl_syntax_setup; object-logics

4424 depending on former CPure need to refer to Pure.

4426 * Commands 'use' and 'ML' are now purely functional, operating on

4427 theory/local_theory. Removed former 'ML_setup' (on theory), use 'ML'

4428 instead. Added 'ML_val' as mere diagnostic replacement for 'ML'.

4429 INCOMPATIBILITY.

4431 * Command 'setup': discontinued implicit version with ML reference.

4433 * Instantiation target allows for simultaneous specification of class

4434 instance operations together with an instantiation proof.

4435 Type-checking phase allows to refer to class operations uniformly.

4436 See src/HOL/Complex/Complex.thy for an Isar example and

4437 src/HOL/Library/Eval.thy for an ML example.

4439 * Indexing of literal facts: be more serious about including only

4440 facts from the visible specification/proof context, but not the

4441 background context (locale etc.). Affects `prop` notation and method

4442 "fact". INCOMPATIBILITY: need to name facts explicitly in rare

4443 situations.

4445 * Method "cases", "induct", "coinduct": removed obsolete/undocumented

4446 "(open)" option, which used to expose internal bound variables to the

4447 proof text.

4449 * Isar statements: removed obsolete case "rule_context".

4450 INCOMPATIBILITY, better use explicit fixes/assumes.

4452 * Locale proofs: default proof step now includes 'unfold_locales';

4453 hence 'proof' without argument may be used to unfold locale

4454 predicates.

4457 *** Document preparation ***

4459 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for

4460 both pdf and dvi (hyperlinks usually work in xdvi as well); removed

4461 obsolete thumbpdf setup (contemporary PDF viewers do this on the

4462 spot); renamed link color from "darkblue" to "linkcolor" (default

4463 value unchanged, can be redefined via \definecolor); no longer sets

4464 "a4paper" option (unnecessary or even intrusive).

4466 * Antiquotation @{lemma A method} proves proposition A by the given

4467 method (either a method name or a method name plus (optional) method

4468 arguments in parentheses) and prints A just like @{prop A}.

4471 *** HOL ***

4473 * New primrec package. Specification syntax conforms in style to

4474 definition/function/.... No separate induction rule is provided. The

4475 "primrec" command distinguishes old-style and new-style specifications

4476 by syntax. The former primrec package is now named OldPrimrecPackage.

4477 When adjusting theories, beware: constants stemming from new-style

4478 primrec specifications have authentic syntax.

4480 * Metis prover is now an order of magnitude faster, and also works

4481 with multithreading.

4483 * Metis: the maximum number of clauses that can be produced from a

4484 theorem is now given by the attribute max_clauses. Theorems that

4485 exceed this number are ignored, with a warning printed.

4487 * Sledgehammer no longer produces structured proofs by default. To

4488 enable, declare [[sledgehammer_full = true]]. Attributes

4489 reconstruction_modulus, reconstruction_sorts renamed

4490 sledgehammer_modulus, sledgehammer_sorts. INCOMPATIBILITY.

4492 * Method "induct_scheme" derives user-specified induction rules

4493 from well-founded induction and completeness of patterns. This factors

4494 out some operations that are done internally by the function package

4495 and makes them available separately. See

4496 src/HOL/ex/Induction_Scheme.thy for examples.

4498 * More flexible generation of measure functions for termination

4499 proofs: Measure functions can be declared by proving a rule of the

4500 form "is_measure f" and giving it the [measure_function] attribute.

4501 The "is_measure" predicate is logically meaningless (always true), and

4502 just guides the heuristic. To find suitable measure functions, the

4503 termination prover sets up the goal "is_measure ?f" of the appropriate

4504 type and generates all solutions by prolog-style backwards proof using

4505 the declared rules.

4507 This setup also deals with rules like

4509 "is_measure f ==> is_measure (list_size f)"

4511 which accommodates nested datatypes that recurse through lists.

4512 Similar rules are predeclared for products and option types.

4514 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"

4516 INCOMPATIBILITIES:

4518 - Definitions of overloaded constants on sets have to be replaced by

4519 definitions on => and bool.

4521 - Some definitions of overloaded operators on sets can now be proved

4522 using the definitions of the operators on => and bool. Therefore,

4523 the following theorems have been renamed:

4525 subset_def -> subset_eq

4526 psubset_def -> psubset_eq

4527 set_diff_def -> set_diff_eq

4528 Compl_def -> Compl_eq

4529 Sup_set_def -> Sup_set_eq

4530 Inf_set_def -> Inf_set_eq

4531 sup_set_def -> sup_set_eq

4532 inf_set_def -> inf_set_eq

4534 - Due to the incompleteness of the HO unification algorithm, some

4535 rules such as subst may require manual instantiation, if some of

4536 the unknowns in the rule is a set.

4538 - Higher order unification and forward proofs:

4539 The proof pattern

4541 have "P (S::'a set)" <...>

4542 then have "EX S. P S" ..

4544 no longer works (due to the incompleteness of the HO unification

4545 algorithm) and must be replaced by the pattern

4547 have "EX S. P S"

4548 proof

4549 show "P S" <...>

4550 qed

4552 - Calculational reasoning with subst (or similar rules):

4553 The proof pattern

4555 have "P (S::'a set)" <...>

4556 also have "S = T" <...>

4557 finally have "P T" .

4559 no longer works (for similar reasons as the previous example) and

4560 must be replaced by something like

4562 have "P (S::'a set)" <...>

4563 moreover have "S = T" <...>

4564 ultimately have "P T" by simp

4566 - Tactics or packages written in ML code:

4567 Code performing pattern matching on types via

4569 Type ("set", [T]) => ...

4571 must be rewritten. Moreover, functions like strip_type or

4572 binder_types no longer return the right value when applied to a

4573 type of the form

4575 T1 => ... => Tn => U => bool

4577 rather than

4579 T1 => ... => Tn => U set

4581 * Merged theories Wellfounded_Recursion, Accessible_Part and

4582 Wellfounded_Relations to theory Wellfounded.

4584 * Explicit class "eq" for executable equality. INCOMPATIBILITY.

4586 * Class finite no longer treats UNIV as class parameter. Use class

4587 enum from theory Library/Enum instead to achieve a similar effect.

4588 INCOMPATIBILITY.

4590 * Theory List: rule list_induct2 now has explicitly named cases "Nil"

4591 and "Cons". INCOMPATIBILITY.

4593 * HOL (and FOL): renamed variables in rules imp_elim and swap.

4594 Potential INCOMPATIBILITY.

4596 * Theory Product_Type: duplicated lemmas split_Pair_apply and

4597 injective_fst_snd removed, use split_eta and prod_eqI instead.

4598 Renamed upd_fst to apfst and upd_snd to apsnd. INCOMPATIBILITY.

4600 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of

4601 the same name in theory Orderings:

4603 less_trans

4604 less_linear

4605 le_imp_less_or_eq

4606 le_less_trans

4607 less_le_trans

4608 less_not_sym

4609 less_asym

4611 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to

4612 less_irrefl_nat. Potential INCOMPATIBILITY due to more general types

4613 and different variable names.

4615 * Library/Option_ord.thy: Canonical order on option type.

4617 * Library/RBT.thy: Red-black trees, an efficient implementation of

4618 finite maps.

4620 * Library/Countable.thy: Type class for countable types.

4622 * Theory Int: The representation of numerals has changed. The infix

4623 operator BIT and the bit datatype with constructors B0 and B1 have

4624 disappeared. INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in

4625 place of "x BIT bit.B0" and "y BIT bit.B1", respectively. Theorems

4626 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"

4627 accordingly.

4629 * Theory Nat: definition of <= and < on natural numbers no longer

4630 depend on well-founded relations. INCOMPATIBILITY. Definitions

4631 le_def and less_def have disappeared. Consider lemmas not_less

4632 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.

4634 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin

4635 (whose purpose mainly is for various fold_set functionals) have been

4636 abandoned in favor of the existing algebraic classes

4637 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,

4638 lower_semilattice (resp. upper_semilattice) and linorder.

4639 INCOMPATIBILITY.

4641 * Theory Transitive_Closure: induct and cases rules now declare proper

4642 case_names ("base" and "step"). INCOMPATIBILITY.

4644 * Theorem Inductive.lfp_ordinal_induct generalized to complete

4645 lattices. The form set-specific version is available as

4646 Inductive.lfp_ordinal_induct_set.

4648 * Renamed theorems "power.simps" to "power_int.simps".

4649 INCOMPATIBILITY.

4651 * Class semiring_div provides basic abstract properties of semirings

4652 with division and modulo operations. Subsumes former class dvd_mod.

4654 * Merged theories IntDef, Numeral and IntArith into unified theory

4655 Int. INCOMPATIBILITY.

4657 * Theory Library/Code_Index: type "index" now represents natural

4658 numbers rather than integers. INCOMPATIBILITY.

4660 * New class "uminus" with operation "uminus" (split of from class

4661 "minus" which now only has operation "minus", binary).

4662 INCOMPATIBILITY.

4664 * Constants "card", "internal_split", "option_map" now with authentic

4665 syntax. INCOMPATIBILITY.

4667 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,

4668 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,

4669 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,

4670 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,

4671 Sup_set_def, le_def, less_def, option_map_def now with object

4672 equality. INCOMPATIBILITY.

4674 * Records. Removed K_record, and replaced it by pure lambda term

4675 %x. c. The simplifier setup is now more robust against eta expansion.

4676 INCOMPATIBILITY: in cases explicitly referring to K_record.

4678 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.

4680 * Library/ListVector: new theory of arithmetic vector operations.

4682 * Library/Order_Relation: new theory of various orderings as sets of

4683 pairs. Defines preorders, partial orders, linear orders and

4684 well-orders on sets and on types.

4687 *** ZF ***

4689 * Renamed some theories to allow to loading both ZF and HOL in the

4690 same session:

4692 Datatype -> Datatype_ZF

4693 Inductive -> Inductive_ZF

4694 Int -> Int_ZF

4695 IntDiv -> IntDiv_ZF

4696 Nat -> Nat_ZF

4697 List -> List_ZF

4698 Main -> Main_ZF

4700 INCOMPATIBILITY: ZF theories that import individual theories below

4701 Main might need to be adapted. Regular theory Main is still

4702 available, as trivial extension of Main_ZF.

4705 *** ML ***

4707 * ML within Isar: antiquotation @{const name} or @{const

4708 name(typargs)} produces statically-checked Const term.

4710 * Functor NamedThmsFun: data is available to the user as dynamic fact

4711 (of the same name). Removed obsolete print command.

4713 * Removed obsolete "use_legacy_bindings" function.

4715 * The ``print mode'' is now a thread-local value derived from a global

4716 template (the former print_mode reference), thus access becomes

4717 non-critical. The global print_mode reference is for session

4718 management only; user-code should use print_mode_value,

4719 print_mode_active, PrintMode.setmp etc. INCOMPATIBILITY.

4721 * Functions system/system_out provide a robust way to invoke external

4722 shell commands, with propagation of interrupts (requires Poly/ML

4723 5.2.1). Do not use OS.Process.system etc. from the basis library!

4726 *** System ***

4728 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---

4729 in accordance with Proof General 3.7, which prefers GNU emacs.

4731 * isatool tty runs Isabelle process with plain tty interaction;

4732 optional line editor may be specified via ISABELLE_LINE_EDITOR

4733 setting, the default settings attempt to locate "ledit" and "rlwrap".

4735 * isatool browser now works with Cygwin as well, using general

4736 "javapath" function defined in Isabelle process environment.

4738 * YXML notation provides a simple and efficient alternative to

4739 standard XML transfer syntax. See src/Pure/General/yxml.ML and

4740 isatool yxml as described in the Isabelle system manual.

4742 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)

4743 provides general wrapper for managing an Isabelle process in a robust

4744 fashion, with ``cooked'' output from stdin/stderr.

4746 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),

4747 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).

4749 * Removed obsolete THIS_IS_ISABELLE_BUILD feature. NB: the documented

4750 way of changing the user's settings is via

4751 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash

4752 script.

4754 * Multithreading.max_threads := 0 refers to the number of actual CPU

4755 cores of the underlying machine, which is a good starting point for

4756 optimal performance tuning. The corresponding usedir option -M allows

4757 "max" as an alias for "0". WARNING: does not work on certain versions

4758 of Mac OS (with Poly/ML 5.1).

4760 * isabelle-process: non-ML sessions are run with "nice", to reduce the

4761 adverse effect of Isabelle flooding interactive front-ends (notably

4762 ProofGeneral / XEmacs).

4766 New in Isabelle2007 (November 2007)

4767 -----------------------------------

4769 *** General ***

4771 * More uniform information about legacy features, notably a

4772 warning/error of "Legacy feature: ...", depending on the state of the

4773 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:

4774 legacy features will disappear eventually.

4776 * Theory syntax: the header format ``theory A = B + C:'' has been

4777 discontinued in favour of ``theory A imports B C begin''. Use isatool

4778 fixheaders to convert existing theory files. INCOMPATIBILITY.

4780 * Theory syntax: the old non-Isar theory file format has been

4781 discontinued altogether. Note that ML proof scripts may still be used

4782 with Isar theories; migration is usually quite simple with the ML

4783 function use_legacy_bindings. INCOMPATIBILITY.

4785 * Theory syntax: some popular names (e.g. 'class', 'declaration',

4786 'fun', 'help', 'if') are now keywords. INCOMPATIBILITY, use double

4787 quotes.

4789 * Theory loader: be more serious about observing the static theory

4790 header specifications (including optional directories), but not the

4791 accidental file locations of previously successful loads. The strict

4792 update policy of former update_thy is now already performed by

4793 use_thy, so the former has been removed; use_thys updates several

4794 theories simultaneously, just as 'imports' within a theory header

4795 specification, but without merging the results. Potential

4796 INCOMPATIBILITY: may need to refine theory headers and commands

4797 ROOT.ML which depend on load order.

4799 * Theory loader: optional support for content-based file

4800 identification, instead of the traditional scheme of full physical

4801 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting

4802 (cf. the system manual). The new scheme allows to work with

4803 non-finished theories in persistent session images, such that source

4804 files may be moved later on without requiring reloads.

4806 * Theory loader: old-style ML proof scripts being *attached* to a thy

4807 file (with the same base name as the theory) are considered a legacy

4808 feature, which will disappear eventually. Even now, the theory loader

4809 no longer maintains dependencies on such files.

4811 * Syntax: the scope for resolving ambiguities via type-inference is

4812 now limited to individual terms, instead of whole simultaneous

4813 specifications as before. This greatly reduces the complexity of the

4814 syntax module and improves flexibility by separating parsing and

4815 type-checking. INCOMPATIBILITY: additional type-constraints (explicit

4816 'fixes' etc.) are required in rare situations.

4818 * Syntax: constants introduced by new-style packages ('definition',

4819 'abbreviation' etc.) are passed through the syntax module in

4820 ``authentic mode''. This means that associated mixfix annotations

4821 really stick to such constants, independently of potential name space

4822 ambiguities introduced later on. INCOMPATIBILITY: constants in parse

4823 trees are represented slightly differently, may need to adapt syntax

4824 translations accordingly. Use CONST marker in 'translations' and

4825 @{const_syntax} antiquotation in 'parse_translation' etc.

4827 * Legacy goal package: reduced interface to the bare minimum required

4828 to keep existing proof scripts running. Most other user-level

4829 functions are now part of the OldGoals structure, which is *not* open

4830 by default (consider isatool expandshort before open OldGoals).

4831 Removed top_sg, prin, printyp, pprint_term/typ altogether, because

4832 these tend to cause confusion about the actual goal (!) context being

4833 used here, which is not necessarily the same as the_context().

4835 * Command 'find_theorems': supports "*" wild-card in "name:"

4836 criterion; "with_dups" option. Certain ProofGeneral versions might

4837 support a specific search form (see ProofGeneral/CHANGES).

4839 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1

4840 by default, which means that "prems" (and also "fixed variables") are

4841 suppressed from proof state output. Note that the ProofGeneral

4842 settings mechanism allows to change and save options persistently, but

4843 older versions of Isabelle will fail to start up if a negative prems

4844 limit is imposed.

4846 * Local theory targets may be specified by non-nested blocks of

4847 ``context/locale/class ... begin'' followed by ``end''. The body may

4848 contain definitions, theorems etc., including any derived mechanism

4849 that has been implemented on top of these primitives. This concept

4850 generalizes the existing ``theorem (in ...)'' towards more versatility

4851 and scalability.

4853 * Proof General interface: proper undo of final 'end' command;

4854 discontinued Isabelle/classic mode (ML proof scripts).

4857 *** Document preparation ***

4859 * Added antiquotation @{theory name} which prints the given name,

4860 after checking that it refers to a valid ancestor theory in the

4861 current context.

4863 * Added antiquotations @{ML_type text} and @{ML_struct text} which

4864 check the given source text as ML type/structure, printing verbatim.

4866 * Added antiquotation @{abbrev "c args"} which prints the abbreviation

4867 "c args == rhs" given in the current context. (Any number of

4868 arguments may be given on the LHS.)

4871 *** Pure ***

4873 * The 'class' package offers a combination of axclass and locale to

4874 achieve Haskell-like type classes in Isabelle. Definitions and

4875 theorems within a class context produce both relative results (with

4876 implicit parameters according to the locale context), and polymorphic

4877 constants with qualified polymorphism (according to the class

4878 context). Within the body context of a 'class' target, a separate

4879 syntax layer ("user space type system") takes care of converting

4880 between global polymorphic consts and internal locale representation.

4881 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).

4882 "isatool doc classes" provides a tutorial.

4884 * Generic code generator framework allows to generate executable

4885 code for ML and Haskell (including Isabelle classes). A short usage

4886 sketch:

4888 internal compilation:

4889 export_code <list of constants (term syntax)> in SML

4890 writing SML code to a file:

4891 export_code <list of constants (term syntax)> in SML <filename>

4892 writing OCaml code to a file:

4893 export_code <list of constants (term syntax)> in OCaml <filename>

4894 writing Haskell code to a bunch of files:

4895 export_code <list of constants (term syntax)> in Haskell <filename>

4897 evaluating closed propositions to True/False using code generation:

4898 method ``eval''

4900 Reasonable default setup of framework in HOL.

4902 Theorem attributs for selecting and transforming function equations theorems:

4904 [code fun]: select a theorem as function equation for a specific constant

4905 [code fun del]: deselect a theorem as function equation for a specific constant

4906 [code inline]: select an equation theorem for unfolding (inlining) in place

4907 [code inline del]: deselect an equation theorem for unfolding (inlining) in place

4909 User-defined serializations (target in {SML, OCaml, Haskell}):

4911 code_const <and-list of constants (term syntax)>

4912 {(target) <and-list of const target syntax>}+

4914 code_type <and-list of type constructors>

4915 {(target) <and-list of type target syntax>}+

4917 code_instance <and-list of instances>

4918 {(target)}+

4919 where instance ::= <type constructor> :: <class>

4921 code_class <and_list of classes>

4922 {(target) <and-list of class target syntax>}+

4923 where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?

4925 code_instance and code_class only are effective to target Haskell.

4927 For example usage see src/HOL/ex/Codegenerator.thy and

4928 src/HOL/ex/Codegenerator_Pretty.thy. A separate tutorial on code

4929 generation from Isabelle/HOL theories is available via "isatool doc

4930 codegen".

4932 * Code generator: consts in 'consts_code' Isar commands are now

4933 referred to by usual term syntax (including optional type

4934 annotations).

4936 * Command 'no_translations' removes translation rules from theory

4937 syntax.

4939 * Overloaded definitions are now actually checked for acyclic

4940 dependencies. The overloading scheme is slightly more general than

4941 that of Haskell98, although Isabelle does not demand an exact

4942 correspondence to type class and instance declarations.

4943 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more

4944 exotic versions of overloading -- at the discretion of the user!

4946 Polymorphic constants are represented via type arguments, i.e. the

4947 instantiation that matches an instance against the most general

4948 declaration given in the signature. For example, with the declaration

4949 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented

4950 as c(nat). Overloading is essentially simultaneous structural

4951 recursion over such type arguments. Incomplete specification patterns

4952 impose global constraints on all occurrences, e.g. c('a * 'a) on the

4953 LHS means that more general c('a * 'b) will be disallowed on any RHS.

4954 Command 'print_theory' outputs the normalized system of recursive

4955 equations, see section "definitions".

4957 * Configuration options are maintained within the theory or proof

4958 context (with name and type bool/int/string), providing a very simple

4959 interface to a poor-man's version of general context data. Tools may

4960 declare options in ML (e.g. using Attrib.config_int) and then refer to

4961 these values using Config.get etc. Users may change options via an

4962 associated attribute of the same name. This form of context

4963 declaration works particularly well with commands 'declare' or

4964 'using', for example ``declare [[foo = 42]]''. Thus it has become

4965 very easy to avoid global references, which would not observe Isar

4966 toplevel undo/redo and fail to work with multithreading.

4968 Various global ML references of Pure and HOL have been turned into

4969 configuration options:

4971 Unify.search_bound unify_search_bound

4972 Unify.trace_bound unify_trace_bound

4973 Unify.trace_simp unify_trace_simp

4974 Unify.trace_types unify_trace_types

4975 Simplifier.simp_depth_limit simp_depth_limit

4976 Blast.depth_limit blast_depth_limit

4977 DatatypeProp.dtK datatype_distinctness_limit

4978 fast_arith_neq_limit fast_arith_neq_limit

4979 fast_arith_split_limit fast_arith_split_limit

4981 * Named collections of theorems may be easily installed as context

4982 data using the functor NamedThmsFun (see also

4983 src/Pure/Tools/named_thms.ML). The user may add or delete facts via

4984 attributes; there is also a toplevel print command. This facility is

4985 just a common case of general context data, which is the preferred way

4986 for anything more complex than just a list of facts in canonical

4987 order.

4989 * Isar: command 'declaration' augments a local theory by generic

4990 declaration functions written in ML. This enables arbitrary content

4991 being added to the context, depending on a morphism that tells the

4992 difference of the original declaration context wrt. the application

4993 context encountered later on.

4995 * Isar: proper interfaces for simplification procedures. Command

4996 'simproc_setup' declares named simprocs (with match patterns, and body

4997 text in ML). Attribute "simproc" adds/deletes simprocs in the current

4998 context. ML antiquotation @{simproc name} retrieves named simprocs.

5000 * Isar: an extra pair of brackets around attribute declarations

5001 abbreviates a theorem reference involving an internal dummy fact,

5002 which will be ignored later --- only the effect of the attribute on

5003 the background context will persist. This form of in-place

5004 declarations is particularly useful with commands like 'declare' and

5005 'using', for example ``have A using [[simproc a]] by simp''.

5007 * Isar: method "assumption" (and implicit closing of subproofs) now

5008 takes simple non-atomic goal assumptions into account: after applying

5009 an assumption as a rule the resulting subgoals are solved by atomic

5010 assumption steps. This is particularly useful to finish 'obtain'

5011 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",

5012 without referring to the original premise "!!x. P x ==> thesis" in the

5013 Isar proof context. POTENTIAL INCOMPATIBILITY: method "assumption" is

5014 more permissive.

5016 * Isar: implicit use of prems from the Isar proof context is

5017 considered a legacy feature. Common applications like ``have A .''

5018 may be replaced by ``have A by fact'' or ``note `A`''. In general,

5019 referencing facts explicitly here improves readability and

5020 maintainability of proof texts.

5022 * Isar: improper proof element 'guess' is like 'obtain', but derives

5023 the obtained context from the course of reasoning! For example:

5025 assume "EX x y. A x & B y" -- "any previous fact"

5026 then guess x and y by clarify

5028 This technique is potentially adventurous, depending on the facts and

5029 proof tools being involved here.

5031 * Isar: known facts from the proof context may be specified as literal

5032 propositions, using ASCII back-quote syntax. This works wherever

5033 named facts used to be allowed so far, in proof commands, proof

5034 methods, attributes etc. Literal facts are retrieved from the context

5035 according to unification of type and term parameters. For example,

5036 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known

5037 theorems in the current context, then these are valid literal facts:

5038 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.

5040 There is also a proof method "fact" which does the same composition

5041 for explicit goal states, e.g. the following proof texts coincide with

5042 certain special cases of literal facts:

5044 have "A" by fact == note `A`

5045 have "A ==> B" by fact == note `A ==> B`

5046 have "!!x. P x ==> Q x" by fact == note `!!x. P x ==> Q x`

5047 have "P a ==> Q a" by fact == note `P a ==> Q a`

5049 * Isar: ":" (colon) is no longer a symbolic identifier character in

5050 outer syntax. Thus symbolic identifiers may be used without

5051 additional white space in declarations like this: ``assume *: A''.

5053 * Isar: 'print_facts' prints all local facts of the current context,

5054 both named and unnamed ones.

5056 * Isar: 'def' now admits simultaneous definitions, e.g.:

5058 def x == "t" and y == "u"

5060 * Isar: added command 'unfolding', which is structurally similar to

5061 'using', but affects both the goal state and facts by unfolding given

5062 rewrite rules. Thus many occurrences of the 'unfold' method or

5063 'unfolded' attribute may be replaced by first-class proof text.

5065 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',

5066 and command 'unfolding' now all support object-level equalities

5067 (potentially conditional). The underlying notion of rewrite rule is

5068 analogous to the 'rule_format' attribute, but *not* that of the

5069 Simplifier (which is usually more generous).

5071 * Isar: the new attribute [rotated n] (default n = 1) rotates the

5072 premises of a theorem by n. Useful in conjunction with drule.

5074 * Isar: the goal restriction operator [N] (default N = 1) evaluates a

5075 method expression within a sandbox consisting of the first N

5076 sub-goals, which need to exist. For example, ``simp_all [3]''

5077 simplifies the first three sub-goals, while (rule foo, simp_all)[]

5078 simplifies all new goals that emerge from applying rule foo to the

5079 originally first one.

5081 * Isar: schematic goals are no longer restricted to higher-order

5082 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as

5083 expected.

5085 * Isar: the conclusion of a long theorem statement is now either

5086 'shows' (a simultaneous conjunction, as before), or 'obtains'

5087 (essentially a disjunction of cases with local parameters and

5088 assumptions). The latter allows to express general elimination rules

5089 adequately; in this notation common elimination rules look like this:

5091 lemma exE: -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"

5092 assumes "EX x. P x"

5093 obtains x where "P x"

5095 lemma conjE: -- "A & B ==> (A ==> B ==> thesis) ==> thesis"

5096 assumes "A & B"

5097 obtains A and B

5099 lemma disjE: -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"

5100 assumes "A | B"

5101 obtains

5102 A

5103 | B

5105 The subsequent classical rules even refer to the formal "thesis"

5106 explicitly:

5108 lemma classical: -- "(~ thesis ==> thesis) ==> thesis"

5109 obtains "~ thesis"

5111 lemma Peirce's_Law: -- "((thesis ==> something) ==> thesis) ==> thesis"

5112 obtains "thesis ==> something"

5114 The actual proof of an 'obtains' statement is analogous to that of the

5115 Isar proof element 'obtain', only that there may be several cases.

5116 Optional case names may be specified in parentheses; these will be

5117 available both in the present proof and as annotations in the

5118 resulting rule, for later use with the 'cases' method (cf. attribute

5119 case_names).

5121 * Isar: the assumptions of a long theorem statement are available as

5122 "assms" fact in the proof context. This is more appropriate than the

5123 (historical) "prems", which refers to all assumptions of the current

5124 context, including those from the target locale, proof body etc.

5126 * Isar: 'print_statement' prints theorems from the current theory or

5127 proof context in long statement form, according to the syntax of a

5128 top-level lemma.

5130 * Isar: 'obtain' takes an optional case name for the local context

5131 introduction rule (default "that").

5133 * Isar: removed obsolete 'concl is' patterns. INCOMPATIBILITY, use

5134 explicit (is "_ ==> ?foo") in the rare cases where this still happens

5135 to occur.

5137 * Pure: syntax "CONST name" produces a fully internalized constant

5138 according to the current context. This is particularly useful for

5139 syntax translations that should refer to internal constant

5140 representations independently of name spaces.

5142 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"

5143 instead of "FOO ". This allows multiple binder declarations to coexist

5144 in the same context. INCOMPATIBILITY.

5146 * Isar/locales: 'notation' provides a robust interface to the 'syntax'

5147 primitive that also works in a locale context (both for constants and

5148 fixed variables). Type declaration and internal syntactic representation

5149 of given constants retrieved from the context. Likewise, the

5150 'no_notation' command allows to remove given syntax annotations from the

5151 current context.

5153 * Isar/locales: new derived specification elements 'axiomatization',

5154 'definition', 'abbreviation', which support type-inference, admit

5155 object-level specifications (equality, equivalence). See also the

5156 isar-ref manual. Examples:

5158 axiomatization

5159 eq (infix "===" 50) where

5160 eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"

5162 definition "f x y = x + y + 1"

5163 definition g where "g x = f x x"

5165 abbreviation

5166 neq (infix "=!=" 50) where

5167 "x =!= y == ~ (x === y)"

5169 These specifications may be also used in a locale context. Then the

5170 constants being introduced depend on certain fixed parameters, and the

5171 constant name is qualified by the locale base name. An internal

5172 abbreviation takes care for convenient input and output, making the

5173 parameters implicit and using the original short name. See also

5174 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic

5175 entities from a monomorphic theory.

5177 Presently, abbreviations are only available 'in' a target locale, but

5178 not inherited by general import expressions. Also note that

5179 'abbreviation' may be used as a type-safe replacement for 'syntax' +

5180 'translations' in common applications. The "no_abbrevs" print mode

5181 prevents folding of abbreviations in term output.

5183 Concrete syntax is attached to specified constants in internal form,

5184 independently of name spaces. The parse tree representation is

5185 slightly different -- use 'notation' instead of raw 'syntax', and

5186 'translations' with explicit "CONST" markup to accommodate this.

5188 * Pure/Isar: unified syntax for new-style specification mechanisms

5189 (e.g. 'definition', 'abbreviation', or 'inductive' in HOL) admits

5190 full type inference and dummy patterns ("_"). For example:

5192 definition "K x _ = x"

5194 inductive conj for A B

5195 where "A ==> B ==> conj A B"

5197 * Pure: command 'print_abbrevs' prints all constant abbreviations of

5198 the current context. Print mode "no_abbrevs" prevents inversion of

5199 abbreviations on output.

5201 * Isar/locales: improved parameter handling: use of locales "var" and

5202 "struct" no longer necessary; - parameter renamings are no longer

5203 required to be injective. For example, this allows to define

5204 endomorphisms as locale endom = homom mult mult h.

5206 * Isar/locales: changed the way locales with predicates are defined.

5207 Instead of accumulating the specification, the imported expression is

5208 now an interpretation. INCOMPATIBILITY: different normal form of

5209 locale expressions. In particular, in interpretations of locales with

5210 predicates, goals repesenting already interpreted fragments are not

5211 removed automatically. Use methods `intro_locales' and

5212 `unfold_locales'; see below.

5214 * Isar/locales: new methods `intro_locales' and `unfold_locales'

5215 provide backward reasoning on locales predicates. The methods are

5216 aware of interpretations and discharge corresponding goals.

5217 `intro_locales' is less aggressive then `unfold_locales' and does not

5218 unfold predicates to assumptions.

5220 * Isar/locales: the order in which locale fragments are accumulated

5221 has changed. This enables to override declarations from fragments due

5222 to interpretations -- for example, unwanted simp rules.

5224 * Isar/locales: interpretation in theories and proof contexts has been

5225 extended. One may now specify (and prove) equations, which are

5226 unfolded in interpreted theorems. This is useful for replacing

5227 defined concepts (constants depending on locale parameters) by

5228 concepts already existing in the target context. Example:

5230 interpretation partial_order ["op <= :: [int, int] => bool"]

5231 where "partial_order.less (op <=) (x::int) y = (x < y)"

5233 Typically, the constant `partial_order.less' is created by a

5234 definition specification element in the context of locale

5235 partial_order.

5237 * Method "induct": improved internal context management to support

5238 local fixes and defines on-the-fly. Thus explicit meta-level

5239 connectives !! and ==> are rarely required anymore in inductive goals

5240 (using object-logic connectives for this purpose has been long

5241 obsolete anyway). Common proof patterns are explained in

5242 src/HOL/Induct/Common_Patterns.thy, see also

5243 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic

5244 examples.

5246 * Method "induct": improved handling of simultaneous goals. Instead of

5247 introducing object-level conjunction, the statement is now split into

5248 several conclusions, while the corresponding symbolic cases are nested

5249 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,

5250 see src/HOL/Induct/Common_Patterns.thy, for example.

5252 * Method "induct": mutual induction rules are now specified as a list

5253 of rule sharing the same induction cases. HOL packages usually provide

5254 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive

5255 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify

5256 mutual induction rules differently, i.e. like this:

5258 (induct rule: foo_bar.inducts)

5259 (induct set: foo bar)

5260 (induct pred: foo bar)

5261 (induct type: foo bar)

5263 The ML function ProjectRule.projections turns old-style rules into the

5264 new format.

5266 * Method "coinduct": dual of induction, see

5267 src/HOL/Library/Coinductive_List.thy for various examples.

5269 * Method "cases", "induct", "coinduct": the ``(open)'' option is

5270 considered a legacy feature.

5272 * Attribute "symmetric" produces result with standardized schematic

5273 variables (index 0). Potential INCOMPATIBILITY.

5275 * Simplifier: by default the simplifier trace only shows top level

5276 rewrites now. That is, trace_simp_depth_limit is set to 1 by

5277 default. Thus there is less danger of being flooded by the trace. The

5278 trace indicates where parts have been suppressed.

5280 * Provers/classical: removed obsolete classical version of elim_format

5281 attribute; classical elim/dest rules are now treated uniformly when

5282 manipulating the claset.

5284 * Provers/classical: stricter checks to ensure that supplied intro,

5285 dest and elim rules are well-formed; dest and elim rules must have at

5286 least one premise.

5288 * Provers/classical: attributes dest/elim/intro take an optional

5289 weight argument for the rule (just as the Pure versions). Weights are

5290 ignored by automated tools, but determine the search order of single

5291 rule steps.

5293 * Syntax: input syntax now supports dummy variable binding "%_. b",

5294 where the body does not mention the bound variable. Note that dummy

5295 patterns implicitly depend on their context of bounds, which makes

5296 "{_. _}" match any set comprehension as expected. Potential

5297 INCOMPATIBILITY -- parse translations need to cope with syntactic

5298 constant "_idtdummy" in the binding position.

5300 * Syntax: removed obsolete syntactic constant "_K" and its associated

5301 parse translation. INCOMPATIBILITY -- use dummy abstraction instead,

5302 for example "A -> B" => "Pi A (%_. B)".

5304 * Pure: 'class_deps' command visualizes the subclass relation, using

5305 the graph browser tool.

5307 * Pure: 'print_theory' now suppresses certain internal declarations by

5308 default; use '!' option for full details.

5311 *** HOL ***

5313 * Method "metis" proves goals by applying the Metis general-purpose

5314 resolution prover (see also http://gilith.com/software/metis/).

5315 Examples are in the directory MetisExamples. WARNING: the

5316 Isabelle/HOL-Metis integration does not yet work properly with

5317 multi-threading.

5319 * Command 'sledgehammer' invokes external automatic theorem provers as

5320 background processes. It generates calls to the "metis" method if

5321 successful. These can be pasted into the proof. Users do not have to

5322 wait for the automatic provers to return. WARNING: does not really

5323 work with multi-threading.

5325 * New "auto_quickcheck" feature tests outermost goal statements for

5326 potential counter-examples. Controlled by ML references

5327 auto_quickcheck (default true) and auto_quickcheck_time_limit (default

5328 5000 milliseconds). Fails silently if statements is outside of

5329 executable fragment, or any other codgenerator problem occurs.

5331 * New constant "undefined" with axiom "undefined x = undefined".

5333 * Added class "HOL.eq", allowing for code generation with polymorphic

5334 equality.

5336 * Some renaming of class constants due to canonical name prefixing in

5337 the new 'class' package:

5339 HOL.abs ~> HOL.abs_class.abs

5340 HOL.divide ~> HOL.divide_class.divide

5341 0 ~> HOL.zero_class.zero

5342 1 ~> HOL.one_class.one

5343 op + ~> HOL.plus_class.plus

5344 op - ~> HOL.minus_class.minus

5345 uminus ~> HOL.minus_class.uminus

5346 op * ~> HOL.times_class.times

5347 op < ~> HOL.ord_class.less

5348 op <= > HOL.ord_class.less_eq

5349 Nat.power ~> Power.power_class.power

5350 Nat.size ~> Nat.size_class.size

5351 Numeral.number_of ~> Numeral.number_class.number_of

5352 FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf

5353 FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup

5354 Orderings.min ~> Orderings.ord_class.min

5355 Orderings.max ~> Orderings.ord_class.max

5356 Divides.op div ~> Divides.div_class.div

5357 Divides.op mod ~> Divides.div_class.mod

5358 Divides.op dvd ~> Divides.div_class.dvd

5360 INCOMPATIBILITY. Adaptions may be required in the following cases:

5362 a) User-defined constants using any of the names "plus", "minus",

5363 "times", "less" or "less_eq". The standard syntax translations for

5364 "+", "-" and "*" may go wrong. INCOMPATIBILITY: use more specific

5365 names.

5367 b) Variables named "plus", "minus", "times", "less", "less_eq"

5368 INCOMPATIBILITY: use more specific names.

5370 c) Permutative equations (e.g. "a + b = b + a")

5371 Since the change of names also changes the order of terms, permutative

5372 rewrite rules may get applied in a different order. Experience shows

5373 that this is rarely the case (only two adaptions in the whole Isabelle

5374 distribution). INCOMPATIBILITY: rewrite proofs

5376 d) ML code directly refering to constant names

5377 This in general only affects hand-written proof tactics, simprocs and

5378 so on. INCOMPATIBILITY: grep your sourcecode and replace names.

5379 Consider using @{const_name} antiquotation.

5381 * New class "default" with associated constant "default".

5383 * Function "sgn" is now overloaded and available on int, real, complex

5384 (and other numeric types), using class "sgn". Two possible defs of

5385 sgn are given as equational assumptions in the classes sgn_if and

5386 sgn_div_norm; ordered_idom now also inherits from sgn_if.

5387 INCOMPATIBILITY.

5389 * Locale "partial_order" now unified with class "order" (cf. theory

5390 Orderings), added parameter "less". INCOMPATIBILITY.

5392 * Renamings in classes "order" and "linorder": facts "refl", "trans" and

5393 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid

5394 clashes with HOL "refl" and "trans". INCOMPATIBILITY.

5396 * Classes "order" and "linorder": potential INCOMPATIBILITY due to

5397 changed order of proof goals in instance proofs.

5399 * The transitivity reasoner for partial and linear orders is set up

5400 for classes "order" and "linorder". Instances of the reasoner are available

5401 in all contexts importing or interpreting the corresponding locales.

5402 Method "order" invokes the reasoner separately; the reasoner

5403 is also integrated with the Simplifier as a solver. Diagnostic

5404 command 'print_orders' shows the available instances of the reasoner

5405 in the current context.

5407 * Localized monotonicity predicate in theory "Orderings"; integrated

5408 lemmas max_of_mono and min_of_mono with this predicate.

5409 INCOMPATIBILITY.

5411 * Formulation of theorem "dense" changed slightly due to integration

5412 with new class dense_linear_order.

5414 * Uniform lattice theory development in HOL.

5416 constants "meet" and "join" now named "inf" and "sup"

5417 constant "Meet" now named "Inf"

5419 classes "meet_semilorder" and "join_semilorder" now named

5420 "lower_semilattice" and "upper_semilattice"

5421 class "lorder" now named "lattice"

5422 class "comp_lat" now named "complete_lattice"

5424 Instantiation of lattice classes allows explicit definitions

5425 for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).

5427 INCOMPATIBILITY. Theorem renames:

5429 meet_left_le ~> inf_le1

5430 meet_right_le ~> inf_le2

5431 join_left_le ~> sup_ge1

5432 join_right_le ~> sup_ge2

5433 meet_join_le ~> inf_sup_ord

5434 le_meetI ~> le_infI

5435 join_leI ~> le_supI

5436 le_meet ~> le_inf_iff

5437 le_join ~> ge_sup_conv

5438 meet_idempotent ~> inf_idem

5439 join_idempotent ~> sup_idem

5440 meet_comm ~> inf_commute

5441 join_comm ~> sup_commute

5442 meet_leI1 ~> le_infI1

5443 meet_leI2 ~> le_infI2

5444 le_joinI1 ~> le_supI1

5445 le_joinI2 ~> le_supI2

5446 meet_assoc ~> inf_assoc

5447 join_assoc ~> sup_assoc

5448 meet_left_comm ~> inf_left_commute

5449 meet_left_idempotent ~> inf_left_idem

5450 join_left_comm ~> sup_left_commute

5451 join_left_idempotent ~> sup_left_idem

5452 meet_aci ~> inf_aci

5453 join_aci ~> sup_aci

5454 le_def_meet ~> le_iff_inf

5455 le_def_join ~> le_iff_sup

5456 join_absorp2 ~> sup_absorb2

5457 join_absorp1 ~> sup_absorb1

5458 meet_absorp1 ~> inf_absorb1

5459 meet_absorp2 ~> inf_absorb2

5460 meet_join_absorp ~> inf_sup_absorb

5461 join_meet_absorp ~> sup_inf_absorb

5462 distrib_join_le ~> distrib_sup_le

5463 distrib_meet_le ~> distrib_inf_le

5465 add_meet_distrib_left ~> add_inf_distrib_left

5466 add_join_distrib_left ~> add_sup_distrib_left

5467 is_join_neg_meet ~> is_join_neg_inf

5468 is_meet_neg_join ~> is_meet_neg_sup

5469 add_meet_distrib_right ~> add_inf_distrib_right

5470 add_join_distrib_right ~> add_sup_distrib_right

5471 add_meet_join_distribs ~> add_sup_inf_distribs

5472 join_eq_neg_meet ~> sup_eq_neg_inf

5473 meet_eq_neg_join ~> inf_eq_neg_sup

5474 add_eq_meet_join ~> add_eq_inf_sup

5475 meet_0_imp_0 ~> inf_0_imp_0

5476 join_0_imp_0 ~> sup_0_imp_0

5477 meet_0_eq_0 ~> inf_0_eq_0

5478 join_0_eq_0 ~> sup_0_eq_0

5479 neg_meet_eq_join ~> neg_inf_eq_sup

5480 neg_join_eq_meet ~> neg_sup_eq_inf

5481 join_eq_if ~> sup_eq_if

5483 mono_meet ~> mono_inf

5484 mono_join ~> mono_sup

5485 meet_bool_eq ~> inf_bool_eq

5486 join_bool_eq ~> sup_bool_eq

5487 meet_fun_eq ~> inf_fun_eq

5488 join_fun_eq ~> sup_fun_eq

5489 meet_set_eq ~> inf_set_eq

5490 join_set_eq ~> sup_set_eq

5491 meet1_iff ~> inf1_iff

5492 meet2_iff ~> inf2_iff

5493 meet1I ~> inf1I

5494 meet2I ~> inf2I

5495 meet1D1 ~> inf1D1

5496 meet2D1 ~> inf2D1

5497 meet1D2 ~> inf1D2

5498 meet2D2 ~> inf2D2

5499 meet1E ~> inf1E

5500 meet2E ~> inf2E

5501 join1_iff ~> sup1_iff

5502 join2_iff ~> sup2_iff

5503 join1I1 ~> sup1I1

5504 join2I1 ~> sup2I1

5505 join1I1 ~> sup1I1

5506 join2I2 ~> sup1I2

5507 join1CI ~> sup1CI

5508 join2CI ~> sup2CI

5509 join1E ~> sup1E

5510 join2E ~> sup2E

5512 is_meet_Meet ~> is_meet_Inf

5513 Meet_bool_def ~> Inf_bool_def

5514 Meet_fun_def ~> Inf_fun_def

5515 Meet_greatest ~> Inf_greatest

5516 Meet_lower ~> Inf_lower

5517 Meet_set_def ~> Inf_set_def

5519 Sup_def ~> Sup_Inf

5520 Sup_bool_eq ~> Sup_bool_def

5521 Sup_fun_eq ~> Sup_fun_def

5522 Sup_set_eq ~> Sup_set_def

5524 listsp_meetI ~> listsp_infI

5525 listsp_meet_eq ~> listsp_inf_eq

5527 meet_min ~> inf_min

5528 join_max ~> sup_max

5530 * Added syntactic class "size"; overloaded constant "size" now has

5531 type "'a::size ==> bool"

5533 * Internal reorganisation of `size' of datatypes: size theorems

5534 "foo.size" are no longer subsumed by "foo.simps" (but are still

5535 simplification rules by default!); theorems "prod.size" now named

5536 "*.size".

5538 * Class "div" now inherits from class "times" rather than "type".

5539 INCOMPATIBILITY.

5541 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,

5542 Linorder etc. have disappeared; operations defined in terms of

5543 fold_set now are named Inf_fin, Sup_fin. INCOMPATIBILITY.

5545 * HOL/Nat: neq0_conv no longer declared as iff. INCOMPATIBILITY.

5547 * HOL-Word: New extensive library and type for generic, fixed size

5548 machine words, with arithmetic, bit-wise, shifting and rotating

5549 operations, reflection into int, nat, and bool lists, automation for

5550 linear arithmetic (by automatic reflection into nat or int), including

5551 lemmas on overflow and monotonicity. Instantiated to all appropriate

5552 arithmetic type classes, supporting automatic simplification of

5553 numerals on all operations.

5555 * Library/Boolean_Algebra: locales for abstract boolean algebras.

5557 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).

5559 * Code generator library theories:

5560 - Code_Integer represents HOL integers by big integer literals in target

5561 languages.

5562 - Code_Char represents HOL characters by character literals in target

5563 languages.

5564 - Code_Char_chr like Code_Char, but also offers treatment of character

5565 codes; includes Code_Integer.

5566 - Executable_Set allows to generate code for finite sets using lists.

5567 - Executable_Rat implements rational numbers as triples (sign, enumerator,

5568 denominator).

5569 - Executable_Real implements a subset of real numbers, namly those

5570 representable by rational numbers.

5571 - Efficient_Nat implements natural numbers by integers, which in general will

5572 result in higher efficency; pattern matching with 0/Suc is eliminated;

5573 includes Code_Integer.

5574 - Code_Index provides an additional datatype index which is mapped to

5575 target-language built-in integers.

5576 - Code_Message provides an additional datatype message_string which is isomorphic to

5577 strings; messages are mapped to target-language strings.

5579 * New package for inductive predicates

5581 An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via

5583 inductive

5584 p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"

5585 for z_1 :: U_1 and ... and z_n :: U_m

5586 where

5587 rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"

5588 | ...

5590 with full support for type-inference, rather than

5592 consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"

5594 abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"

5595 where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"

5597 inductive "s z_1 ... z_m"

5598 intros

5599 rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"

5600 ...

5602 For backward compatibility, there is a wrapper allowing inductive

5603 sets to be defined with the new package via

5605 inductive_set

5606 s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"

5607 for z_1 :: U_1 and ... and z_n :: U_m

5608 where

5609 rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"

5610 | ...

5612 or

5614 inductive_set

5615 s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"

5616 and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"

5617 for z_1 :: U_1 and ... and z_n :: U_m

5618 where

5619 "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"

5620 | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"

5621 | ...

5623 if the additional syntax "p ..." is required.

5625 Numerous examples can be found in the subdirectories src/HOL/Auth,

5626 src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.

5628 INCOMPATIBILITIES:

5630 - Since declaration and definition of inductive sets or predicates

5631 is no longer separated, abbreviations involving the newly

5632 introduced sets or predicates must be specified together with the

5633 introduction rules after the 'where' keyword (see above), rather

5634 than before the actual inductive definition.

5636 - The variables in induction and elimination rules are now

5637 quantified in the order of their occurrence in the introduction

5638 rules, rather than in alphabetical order. Since this may break

5639 some proofs, these proofs either have to be repaired, e.g. by

5640 reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'

5641 statements of the form

5643 case (rule_i a_i_1 ... a_i_{k_i})

5645 or the old order of quantification has to be restored by explicitly adding

5646 meta-level quantifiers in the introduction rules, i.e.

5648 | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"

5650 - The format of the elimination rules is now

5652 p z_1 ... z_m x_1 ... x_n ==>

5653 (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)

5654 ==> ... ==> P

5656 for predicates and

5658 (x_1, ..., x_n) : s z_1 ... z_m ==>

5659 (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)

5660 ==> ... ==> P

5662 for sets rather than

5664 x : s z_1 ... z_m ==>

5665 (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)

5666 ==> ... ==> P

5668 This may require terms in goals to be expanded to n-tuples

5669 (e.g. using case_tac or simplification with the split_paired_all

5670 rule) before the above elimination rule is applicable.

5672 - The elimination or case analysis rules for (mutually) inductive

5673 sets or predicates are now called "p_1.cases" ... "p_k.cases". The

5674 list of rules "p_1_..._p_k.elims" is no longer available.

5676 * New package "function"/"fun" for general recursive functions,

5677 supporting mutual and nested recursion, definitions in local contexts,

5678 more general pattern matching and partiality. See HOL/ex/Fundefs.thy

5679 for small examples, and the separate tutorial on the function

5680 package. The old recdef "package" is still available as before, but

5681 users are encouraged to use the new package.

5683 * Method "lexicographic_order" automatically synthesizes termination

5684 relations as lexicographic combinations of size measures.

5686 * Case-expressions allow arbitrary constructor-patterns (including

5687 "_") and take their order into account, like in functional

5688 programming. Internally, this is translated into nested

5689 case-expressions; missing cases are added and mapped to the predefined

5690 constant "undefined". In complicated cases printing may no longer show

5691 the original input but the internal form. Lambda-abstractions allow

5692 the same form of pattern matching: "% pat1 => e1 | ..." is an

5693 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new

5694 variable.

5696 * IntDef: The constant "int :: nat => int" has been removed; now "int"

5697 is an abbreviation for "of_nat :: nat => int". The simplification

5698 rules for "of_nat" have been changed to work like "int" did

5699 previously. Potential INCOMPATIBILITY:

5700 - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"

5701 - of_nat_diff and of_nat_mult are no longer default simp rules

5703 * Method "algebra" solves polynomial equations over (semi)rings using

5704 Groebner bases. The (semi)ring structure is defined by locales and the

5705 tool setup depends on that generic context. Installing the method for

5706 a specific type involves instantiating the locale and possibly adding

5707 declarations for computation on the coefficients. The method is

5708 already instantiated for natural numbers and for the axiomatic class

5709 of idoms with numerals. See also the paper by Chaieb and Wenzel at

5710 CALCULEMUS 2007 for the general principles underlying this

5711 architecture of context-aware proof-tools.

5713 * Method "ferrack" implements quantifier elimination over

5714 special-purpose dense linear orders using locales (analogous to

5715 "algebra"). The method is already installed for class

5716 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,

5717 rat, etc.

5719 * Former constant "List.op @" now named "List.append". Use ML

5720 antiquotations @{const_name List.append} or @{term " ... @ ... "} to

5721 circumvent possible incompatibilities when working on ML level.

5723 * primrec: missing cases mapped to "undefined" instead of "arbitrary".

5725 * New function listsum :: 'a list => 'a for arbitrary monoids.

5726 Special syntax: "SUM x <- xs. f x" (and latex variants)

5728 * New syntax for Haskell-like list comprehension (input only), eg.

5729 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.

5731 * The special syntax for function "filter" has changed from [x :

5732 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list

5733 comprehension syntax, and for uniformity. INCOMPATIBILITY.

5735 * [a..b] is now defined for arbitrary linear orders. It used to be

5736 defined on nat only, as an abbreviation for [a..<Suc b]

5737 INCOMPATIBILITY.

5739 * Renamed lemma "set_take_whileD" to "set_takeWhileD".

5741 * New functions "sorted" and "sort" in src/HOL/List.thy.

5743 * New lemma collection field_simps (an extension of ring_simps) for

5744 manipulating (in)equations involving division. Multiplies with all

5745 denominators that can be proved to be non-zero (in equations) or

5746 positive/negative (in inequations).

5748 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib

5749 have been improved and renamed to ring_simps, group_simps and

5750 ring_distribs. Removed lemmas field_xyz in theory Ring_and_Field

5751 because they were subsumed by lemmas xyz. INCOMPATIBILITY.

5753 * Theory Library/Commutative_Ring: switched from recdef to function

5754 package; constants add, mul, pow now curried. Infix syntax for

5755 algebraic operations.

5757 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.

5758 INCOMPATIBILITY.

5760 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.

5761 INCOMPATIBILITY.

5763 * HOL/records: generalised field-update to take a function on the

5764 field rather than the new value: r(|A := x|) is translated to A_update

5765 (K x) r The K-combinator that is internally used is called K_record.

5766 INCOMPATIBILITY: Usage of the plain update functions has to be

5767 adapted.

5769 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0

5770 * x = 0, which are required for a semiring. Richer structures do not

5771 inherit from semiring_0 anymore, because this property is a theorem

5772 there, not an axiom. INCOMPATIBILITY: In instances of semiring_0,

5773 there is more to prove, but this is mostly trivial.

5775 * Class "recpower" is generalized to arbitrary monoids, not just

5776 commutative semirings. INCOMPATIBILITY: may need to incorporate

5777 commutativity or semiring properties additionally.

5779 * Constant "List.list_all2" in List.thy now uses authentic syntax.

5780 INCOMPATIBILITY: translations containing list_all2 may go wrong,

5781 better use 'abbreviation'.

5783 * Renamed constant "List.op mem" to "List.member". INCOMPATIBILITY.

5785 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has

5786 been abandoned in favour of plain 'int'. INCOMPATIBILITY --

5787 significant changes for setting up numeral syntax for types:

5788 - New constants Numeral.pred and Numeral.succ instead

5789 of former Numeral.bin_pred and Numeral.bin_succ.

5790 - Use integer operations instead of bin_add, bin_mult and so on.

5791 - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.

5792 - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.

5794 See src/HOL/Integ/IntArith.thy for an example setup.

5796 * Command 'normal_form' computes the normal form of a term that may

5797 contain free variables. For example ``normal_form "rev [a, b, c]"''

5798 produces ``[b, c, a]'' (without proof). This command is suitable for

5799 heavy-duty computations because the functions are compiled to ML

5800 first. Correspondingly, a method "normalization" is provided. See

5801 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.

5803 * Alternative iff syntax "A <-> B" for equality on bool (with priority

5804 25 like -->); output depends on the "iff" print_mode, the default is

5805 "A = B" (with priority 50).

5807 * Relations less (<) and less_eq (<=) are also available on type bool.

5808 Modified syntax to disallow nesting without explicit parentheses,

5809 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z". Potential

5810 INCOMPATIBILITY.

5812 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).

5814 * Relation composition operator "op O" now has precedence 75 and binds

5815 stronger than union and intersection. INCOMPATIBILITY.

5817 * The old set interval syntax "{m..n(}" (and relatives) has been

5818 removed. Use "{m..<n}" (and relatives) instead.

5820 * In the context of the assumption "~(s = t)" the Simplifier rewrites

5821 "t = s" to False (by simproc "neq"). INCOMPATIBILITY, consider using

5822 ``declare [[simproc del: neq]]''.

5824 * Simplifier: "m dvd n" where m and n are numbers is evaluated to

5825 True/False.

5827 * Theorem Cons_eq_map_conv no longer declared as "simp".

5829 * Theorem setsum_mult renamed to setsum_right_distrib.

5831 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the

5832 ``rule'' method.

5834 * Reimplemented methods "sat" and "satx", with several improvements:

5835 goals no longer need to be stated as "<prems> ==> False", equivalences

5836 (i.e. "=" on type bool) are handled, variable names of the form

5837 "lit_<n>" are no longer reserved, significant speedup.

5839 * Methods "sat" and "satx" can now replay MiniSat proof traces.

5840 zChaff is still supported as well.

5842 * 'inductive' and 'datatype': provide projections of mutual rules,

5843 bundled as foo_bar.inducts;

5845 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to

5846 Library.

5848 * Library: moved theory Accessible_Part to main HOL.

5850 * Library: added theory Coinductive_List of potentially infinite lists

5851 as greatest fixed-point.

5853 * Library: added theory AssocList which implements (finite) maps as

5854 association lists.

5856 * Method "evaluation" solves goals (i.e. a boolean expression)

5857 efficiently by compiling it to ML. The goal is "proved" (via an

5858 oracle) if it evaluates to True.

5860 * Linear arithmetic now splits certain operators (e.g. min, max, abs)

5861 also when invoked by the simplifier. This results in the Simplifier

5862 being more powerful on arithmetic goals. INCOMPATIBILITY.

5863 Configuration option fast_arith_split_limit=0 recovers the old

5864 behavior.

5866 * Support for hex (0x20) and binary (0b1001) numerals.

5868 * New method: reify eqs (t), where eqs are equations for an

5869 interpretation I :: 'a list => 'b => 'c and t::'c is an optional

5870 parameter, computes a term s::'b and a list xs::'a list and proves the

5871 theorem I xs s = t. This is also known as reification or quoting. The

5872 resulting theorem is applied to the subgoal to substitute t with I xs

5873 s. If t is omitted, the subgoal itself is reified.

5875 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)

5876 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,

5877 where f is supposed to be a computable function (in the sense of code

5878 generattion). The method uses reify to compute s and xs as above then

5879 applies corr_thm and uses normalization by evaluation to "prove" f s =

5880 r and finally gets the theorem t = r, which is again applied to the

5881 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.

5883 * Reflection: Automatic reification now handels binding, an example is

5884 available in src/HOL/ex/ReflectionEx.thy

5886 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a

5887 command 'statespace' that is similar to 'record', but introduces an

5888 abstract specification based on the locale infrastructure instead of

5889 HOL types. This leads to extra flexibility in composing state spaces,

5890 in particular multiple inheritance and renaming of components.

5893 *** HOL-Complex ***

5895 * Hyperreal: Functions root and sqrt are now defined on negative real

5896 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.

5897 Nonnegativity side conditions have been removed from many lemmas, so

5898 that more subgoals may now be solved by simplification; potential

5899 INCOMPATIBILITY.

5901 * Real: new type classes formalize real normed vector spaces and

5902 algebras, using new overloaded constants scaleR :: real => 'a => 'a

5903 and norm :: 'a => real.

5905 * Real: constant of_real :: real => 'a::real_algebra_1 injects from

5906 reals into other types. The overloaded constant Reals :: 'a set is now

5907 defined as range of_real; potential INCOMPATIBILITY.

5909 * Real: proper support for ML code generation, including 'quickcheck'.

5910 Reals are implemented as arbitrary precision rationals.

5912 * Hyperreal: Several constants that previously worked only for the

5913 reals have been generalized, so they now work over arbitrary vector

5914 spaces. Type annotations may need to be added in some cases; potential

5915 INCOMPATIBILITY.

5917 Infinitesimal :: ('a::real_normed_vector) star set

5918 HFinite :: ('a::real_normed_vector) star set

5919 HInfinite :: ('a::real_normed_vector) star set

5920 approx :: ('a::real_normed_vector) star => 'a star => bool

5921 monad :: ('a::real_normed_vector) star => 'a star set

5922 galaxy :: ('a::real_normed_vector) star => 'a star set

5923 (NS)LIMSEQ :: [nat => 'a::real_normed_vector, 'a] => bool

5924 (NS)convergent :: (nat => 'a::real_normed_vector) => bool

5925 (NS)Bseq :: (nat => 'a::real_normed_vector) => bool

5926 (NS)Cauchy :: (nat => 'a::real_normed_vector) => bool

5927 (NS)LIM :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool

5928 is(NS)Cont :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool

5929 deriv :: ['a::real_normed_field => 'a, 'a, 'a] => bool

5930 sgn :: 'a::real_normed_vector => 'a

5931 exp :: 'a::{recpower,real_normed_field,banach} => 'a

5933 * Complex: Some complex-specific constants are now abbreviations for

5934 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =

5935 hnorm. Other constants have been entirely removed in favor of the

5936 polymorphic versions (INCOMPATIBILITY):

5938 approx <-- capprox

5939 HFinite <-- CFinite

5940 HInfinite <-- CInfinite

5941 Infinitesimal <-- CInfinitesimal

5942 monad <-- cmonad

5943 galaxy <-- cgalaxy

5944 (NS)LIM <-- (NS)CLIM, (NS)CRLIM

5945 is(NS)Cont <-- is(NS)Contc, is(NS)contCR

5946 (ns)deriv <-- (ns)cderiv

5949 *** HOL-Algebra ***

5951 * Formalisation of ideals and the quotient construction over rings.

5953 * Order and lattice theory no longer based on records.

5954 INCOMPATIBILITY.

5956 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->

5957 greatest_closed. INCOMPATIBILITY.

5959 * Method algebra is now set up via an attribute. For examples see

5960 Ring.thy. INCOMPATIBILITY: the method is now weaker on combinations

5961 of algebraic structures.

5963 * Renamed theory CRing to Ring.

5966 *** HOL-Nominal ***

5968 * Substantial, yet incomplete support for nominal datatypes (binding

5969 structures) based on HOL-Nominal logic. See src/HOL/Nominal and

5970 src/HOL/Nominal/Examples. Prospective users should consult

5971 http://isabelle.in.tum.de/nominal/

5974 *** ML ***

5976 * ML basics: just one true type int, which coincides with IntInf.int

5977 (even on SML/NJ).

5979 * ML within Isar: antiquotations allow to embed statically-checked

5980 formal entities in the source, referring to the context available at

5981 compile-time. For example:

5983 ML {* @{sort "{zero,one}"} *}

5984 ML {* @{typ "'a => 'b"} *}

5985 ML {* @{term "%x. x"} *}

5986 ML {* @{prop "x == y"} *}

5987 ML {* @{ctyp "'a => 'b"} *}

5988 ML {* @{cterm "%x. x"} *}

5989 ML {* @{cprop "x == y"} *}

5990 ML {* @{thm asm_rl} *}

5991 ML {* @{thms asm_rl} *}

5992 ML {* @{type_name c} *}

5993 ML {* @{type_syntax c} *}

5994 ML {* @{const_name c} *}

5995 ML {* @{const_syntax c} *}

5996 ML {* @{context} *}

5997 ML {* @{theory} *}

5998 ML {* @{theory Pure} *}

5999 ML {* @{theory_ref} *}

6000 ML {* @{theory_ref Pure} *}

6001 ML {* @{simpset} *}

6002 ML {* @{claset} *}

6003 ML {* @{clasimpset} *}

6005 The same works for sources being ``used'' within an Isar context.

6007 * ML in Isar: improved error reporting; extra verbosity with

6008 ML_Context.trace enabled.

6010 * Pure/General/table.ML: the join operations now works via exceptions

6011 DUP/SAME instead of type option. This is simpler in simple cases, and

6012 admits slightly more efficient complex applications.

6014 * Pure: 'advanced' translation functions (parse_translation etc.) now

6015 use Context.generic instead of just theory.

6017 * Pure: datatype Context.generic joins theory/Proof.context and

6018 provides some facilities for code that works in either kind of

6019 context, notably GenericDataFun for uniform theory and proof data.

6021 * Pure: simplified internal attribute type, which is now always

6022 Context.generic * thm -> Context.generic * thm. Global (theory) vs.

6023 local (Proof.context) attributes have been discontinued, while

6024 minimizing code duplication. Thm.rule_attribute and

6025 Thm.declaration_attribute build canonical attributes; see also structure

6026 Context for further operations on Context.generic, notably

6027 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type

6028 declarations and definitions.

6030 * Context data interfaces (Theory/Proof/GenericDataFun): removed

6031 name/print, uninitialized data defaults to ad-hoc copy of empty value,

6032 init only required for impure data. INCOMPATIBILITY: empty really need

6033 to be empty (no dependencies on theory content!)

6035 * Pure/kernel: consts certification ignores sort constraints given in

6036 signature declarations. (This information is not relevant to the

6037 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,

6038 potential INCOMPATIBILITY.

6040 * Pure: axiomatic type classes are now purely definitional, with

6041 explicit proofs of class axioms and super class relations performed

6042 internally. See Pure/axclass.ML for the main internal interfaces --

6043 notably AxClass.define_class supercedes AxClass.add_axclass, and

6044 AxClass.axiomatize_class/classrel/arity supersede

6045 Sign.add_classes/classrel/arities.

6047 * Pure/Isar: Args/Attrib parsers operate on Context.generic --

6048 global/local versions on theory vs. Proof.context have been

6049 discontinued; Attrib.syntax and Method.syntax have been adapted

6050 accordingly. INCOMPATIBILITY, need to adapt parser expressions for

6051 attributes, methods, etc.

6053 * Pure: several functions of signature "... -> theory -> theory * ..."

6054 have been reoriented to "... -> theory -> ... * theory" in order to

6055 allow natural usage in combination with the ||>, ||>>, |-> and

6056 fold_map combinators.

6058 * Pure: official theorem names (closed derivations) and additional

6059 comments (tags) are now strictly separate. Name hints -- which are

6060 maintained as tags -- may be attached any time without affecting the

6061 derivation.

6063 * Pure: primitive rule lift_rule now takes goal cterm instead of an

6064 actual goal state (thm). Use Thm.lift_rule (Thm.cprem_of st i) to

6065 achieve the old behaviour.

6067 * Pure: the "Goal" constant is now called "prop", supporting a

6068 slightly more general idea of ``protecting'' meta-level rule

6069 statements.

6071 * Pure: Logic.(un)varify only works in a global context, which is now

6072 enforced instead of silently assumed. INCOMPATIBILITY, may use

6073 Logic.legacy_(un)varify as temporary workaround.

6075 * Pure: structure Name provides scalable operations for generating

6076 internal variable names, notably Name.variants etc. This replaces

6077 some popular functions from term.ML:

6079 Term.variant -> Name.variant

6080 Term.variantlist -> Name.variant_list

6081 Term.invent_names -> Name.invent_list

6083 Note that low-level renaming rarely occurs in new code -- operations

6084 from structure Variable are used instead (see below).

6086 * Pure: structure Variable provides fundamental operations for proper

6087 treatment of fixed/schematic variables in a context. For example,

6088 Variable.import introduces fixes for schematics of given facts and

6089 Variable.export reverses the effect (up to renaming) -- this replaces

6090 various freeze_thaw operations.

6092 * Pure: structure Goal provides simple interfaces for

6093 init/conclude/finish and tactical prove operations (replacing former

6094 Tactic.prove). Goal.prove is the canonical way to prove results

6095 within a given context; Goal.prove_global is a degraded version for

6096 theory level goals, including a global Drule.standard. Note that

6097 OldGoals.prove_goalw_cterm has long been obsolete, since it is

6098 ill-behaved in a local proof context (e.g. with local fixes/assumes or

6099 in a locale context).

6101 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)

6102 and type checking (Syntax.check_term etc.), with common combinations

6103 (Syntax.read_term etc.). These supersede former Sign.read_term etc.

6104 which are considered legacy and await removal.

6106 * Pure/Syntax: generic interfaces for type unchecking

6107 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),

6108 with common combinations (Syntax.pretty_term, Syntax.string_of_term

6109 etc.). Former Sign.pretty_term, Sign.string_of_term etc. are still

6110 available for convenience, but refer to the very same operations using

6111 a mere theory instead of a full context.

6113 * Isar: simplified treatment of user-level errors, using exception

6114 ERROR of string uniformly. Function error now merely raises ERROR,

6115 without any side effect on output channels. The Isar toplevel takes

6116 care of proper display of ERROR exceptions. ML code may use plain

6117 handle/can/try; cat_error may be used to concatenate errors like this:

6119 ... handle ERROR msg => cat_error msg "..."

6121 Toplevel ML code (run directly or through the Isar toplevel) may be

6122 embedded into the Isar toplevel with exception display/debug like

6123 this:

6125 Isar.toplevel (fn () => ...)

6127 INCOMPATIBILITY, removed special transform_error facilities, removed

6128 obsolete variants of user-level exceptions (ERROR_MESSAGE,

6129 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)

6130 -- use plain ERROR instead.

6132 * Isar: theory setup now has type (theory -> theory), instead of a

6133 list. INCOMPATIBILITY, may use #> to compose setup functions.

6135 * Isar: ML toplevel pretty printer for type Proof.context, subject to

6136 ProofContext.debug/verbose flags.

6138 * Isar: Toplevel.theory_to_proof admits transactions that modify the

6139 theory before entering a proof state. Transactions now always see a

6140 quasi-functional intermediate checkpoint, both in interactive and

6141 batch mode.

6143 * Isar: simplified interfaces for outer syntax. Renamed

6144 OuterSyntax.add_keywords to OuterSyntax.keywords. Removed

6145 OuterSyntax.add_parsers -- this functionality is now included in

6146 OuterSyntax.command etc. INCOMPATIBILITY.

6148 * Simplifier: the simpset of a running simplification process now

6149 contains a proof context (cf. Simplifier.the_context), which is the

6150 very context that the initial simpset has been retrieved from (by

6151 simpset_of/local_simpset_of). Consequently, all plug-in components

6152 (solver, looper etc.) may depend on arbitrary proof data.

6154 * Simplifier.inherit_context inherits the proof context (plus the

6155 local bounds) of the current simplification process; any simproc

6156 etc. that calls the Simplifier recursively should do this! Removed

6157 former Simplifier.inherit_bounds, which is already included here --

6158 INCOMPATIBILITY. Tools based on low-level rewriting may even have to

6159 specify an explicit context using Simplifier.context/theory_context.

6161 * Simplifier/Classical Reasoner: more abstract interfaces

6162 change_simpset/claset for modifying the simpset/claset reference of a

6163 theory; raw versions simpset/claset_ref etc. have been discontinued --

6164 INCOMPATIBILITY.

6166 * Provers: more generic wrt. syntax of object-logics, avoid hardwired

6167 "Trueprop" etc.

6170 *** System ***

6172 * settings: the default heap location within ISABELLE_HOME_USER now

6173 includes ISABELLE_IDENTIFIER. This simplifies use of multiple

6174 Isabelle installations.

6176 * isabelle-process: option -S (secure mode) disables some critical

6177 operations, notably runtime compilation and evaluation of ML source

6178 code.

6180 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.

6182 * Support for parallel execution, using native multicore support of

6183 Poly/ML 5.1. The theory loader exploits parallelism when processing

6184 independent theories, according to the given theory header

6185 specifications. The maximum number of worker threads is specified via

6186 usedir option -M or the "max-threads" setting in Proof General. A

6187 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up

6188 to 6 on a 8-core machine. User-code needs to observe certain

6189 guidelines for thread-safe programming, see appendix A in the Isar

6190 Implementation manual.

6194 New in Isabelle2005 (October 2005)

6195 ----------------------------------

6197 *** General ***

6199 * Theory headers: the new header syntax for Isar theories is

6201 theory <name>

6202 imports <theory1> ... <theoryN>

6203 uses <file1> ... <fileM>

6204 begin

6206 where the 'uses' part is optional. The previous syntax

6208 theory <name> = <theory1> + ... + <theoryN>:

6210 will disappear in the next release. Use isatool fixheaders to convert

6211 existing theory files. Note that there is no change in ancient

6212 non-Isar theories now, but these will disappear soon.

6214 * Theory loader: parent theories can now also be referred to via

6215 relative and absolute paths.

6217 * Command 'find_theorems' searches for a list of criteria instead of a

6218 list of constants. Known criteria are: intro, elim, dest, name:string,

6219 simp:term, and any term. Criteria can be preceded by '-' to select

6220 theorems that do not match. Intro, elim, dest select theorems that

6221 match the current goal, name:s selects theorems whose fully qualified

6222 name contain s, and simp:term selects all simplification rules whose

6223 lhs match term. Any other term is interpreted as pattern and selects

6224 all theorems matching the pattern. Available in ProofGeneral under

6225 'ProofGeneral -> Find Theorems' or C-c C-f. Example:

6227 C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."

6229 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",

6230 matching the current goal as introduction rule and not having "HOL."

6231 in their name (i.e. not being defined in theory HOL).

6233 * Command 'thms_containing' has been discontinued in favour of

6234 'find_theorems'; INCOMPATIBILITY.

6236 * Communication with Proof General is now 8bit clean, which means that

6237 Unicode text in UTF-8 encoding may be used within theory texts (both

6238 formal and informal parts). Cf. option -U of the Isabelle Proof

6239 General interface. Here are some simple examples (cf. src/HOL/ex):