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

NEWS

author | wenzelm |

Sun Nov 18 16:04:13 2012 +0100 (2012-11-18) | |

changeset 50119 | 5c370a036de7 |

parent 50118 | 89a14e495526 |

child 50126 | 3dec88149176 |

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

more generous tracing_limit, with explicit system option;

1 Isabelle NEWS -- history user-relevant changes

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

4 New in this Isabelle version

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

7 *** General ***

9 * Configuration option show_markup controls direct inlining of markup

10 into the printed representation of formal entities --- notably type

11 and sort constraints. This enables Prover IDE users to retrieve that

12 information via tooltips in the output window, for example.

14 * Command 'ML_file' evaluates ML text from a file directly within the

15 theory, without any predeclaration via 'uses' in the theory header.

17 * Old command 'use' command and corresponding keyword 'uses' in the

18 theory header are legacy features and will be discontinued soon.

19 Tools that load their additional source files may imitate the

20 'ML_file' implementation, such that the system can take care of

21 dependencies properly.

23 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which

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

26 * Updated and extended "isar-ref" and "implementation" manual, reduced

27 remaining material in old "ref" manual.

29 * Improved support for auxiliary contexts indicate block structure for

30 specifications: nesting of "context fixes ... context assumes ..."

31 and "class ... context ...".

33 * More informative error messages for Isar proof commands involving

34 lazy enumerations (method applications etc.).

37 *** Prover IDE -- Isabelle/Scala/jEdit ***

39 * Parallel terminal proofs ('by') are enabled by default, likewise

40 proofs that are built into packages like 'datatype', 'function'. This

41 allows to "run ahead" checking the theory specifications on the

42 surface, while the prover is still crunching on internal

43 justifications. Unfinished / cancelled proofs are restarted as

44 required to complete full proof checking eventually.

46 * Improved output panel with tooltips, hyperlinks etc. based on the

47 same Rich_Text_Area as regular Isabelle/jEdit buffers. Activation of

48 tooltips leads to some window that supports the same recursively,

49 which can lead to stacks of tooltips as the semantic document content

50 is explored. ESCAPE closes the whole stack, individual windows may be

51 closed separately, or detached to become independent jEdit dockables.

53 * More robust incremental parsing of outer syntax (partial comments,

54 malformed symbols). Changing the balance of open/close quotes and

55 comment delimiters works more conveniently with unfinished situations

56 that frequently occur in user interaction.

58 * More efficient painting and improved reactivity when editing large

59 files. More scalable management of formal document content.

61 * Smarter handling of tracing messages: output window informs about

62 accumulated messages; prover transactions are limited to emit maximum

63 amount of output, before being canceled (cf. system option

64 "editor_tracing_limit"). This avoids swamping the front-end with

65 potentially infinite message streams.

67 * More plugin options and preferences, based on Isabelle/Scala. The

68 jEdit plugin option panel provides access to some Isabelle/Scala

69 options, including tuning parameters for editor reactivity and color

70 schemes.

72 * Uniform Java 7 platform on Linux, Mac OS X, Windows: recent updates

73 from Oracle provide better multi-platform experience. This version is

74 now bundled exclusively with Isabelle.

77 *** Pure ***

79 * Code generation for Haskell: restrict unqualified imports from

80 Haskell Prelude to a small set of fundamental operations.

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

83 relatively to master directory of current theory rather than

84 the rather arbitrary current working directory.

85 INCOMPATIBILITY.

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

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

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

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

91 with some care where this is really required.

93 * Command 'typ' supports an additional variant with explicit sort

94 constraint, to infer and check the most general type conforming to a

95 given given sort. Example (in HOL):

97 typ "_ * _ * bool * unit" :: finite

100 *** HOL ***

102 * Removed constant "chars". Prefer "Enum.enum" on type "char"

103 directly. INCOMPATIBILITY.

105 * Moved operation product, sublists and n_lists from Enum.thy

106 to List.thy. INCOMPATIBILITY.

108 * Simplified 'typedef' specifications: historical options for implicit

109 set definition and alternative name have been discontinued. The

110 former behavior of "typedef (open) t = A" is now the default, but

111 written just "typedef t = A". INCOMPATIBILITY, need to adapt theories

112 accordingly.

114 * Theory "Library/Multiset":

116 - Renamed constants

117 fold_mset ~> Multiset.fold -- for coherence with other fold combinators

119 - Renamed facts

120 fold_mset_commute ~> fold_mset_comm -- for coherence with fold_comm

122 INCOMPATIBILITY.

124 * Theorem UN_o generalized to SUP_comp. INCOMPATIBILITY.

126 * Class "comm_monoid_diff" formalises properties of bounded

127 subtraction, with natural numbers and multisets as typical instances.

129 * Theory "Library/Option_ord" provides instantiation of option type

130 to lattice type classes.

132 * New combinator "Option.these" with type "'a option set => 'a set".

134 * Renamed theory Library/List_Prefix to Library/Sublist.

135 INCOMPATIBILITY. Related changes are:

137 - Renamed constants:

139 prefix ~> prefixeq

140 strict_prefix ~> prefix

142 Renamed lemmas accordingly, INCOMPATIBILITY.

144 - Replaced constant "postfix" by "suffixeq" with swapped argument order

145 (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix

146 syntax "xs >>= ys"; use "suffixeq ys xs" instead. Renamed lemmas

147 accordingly. INCOMPATIBILITY.

149 - New constant "emb" for homeomorphic embedding on lists. New

150 abbreviation "sub" for special case "emb (op =)".

152 - Library/Sublist does no longer provide "order" and "bot" type class

153 instances for the prefix order (merely corresponding locale

154 interpretations). The type class instances are to be found in

155 Library/Prefix_Order. INCOMPATIBILITY.

157 - The sublist relation from Library/Sublist_Order is now based on

158 "Sublist.sub". Replaced lemmas:

160 le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff

161 le_list_append_mono ~> Sublist.emb_append_mono

162 le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2

163 le_list_Cons_EX ~> Sublist.emb_ConsD

164 le_list_drop_Cons2 ~> Sublist.sub_Cons2'

165 le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq

166 le_list_drop_Cons ~> Sublist.sub_Cons'

167 le_list_drop_many ~> Sublist.sub_drop_many

168 le_list_filter_left ~> Sublist.sub_filter_left

169 le_list_rev_drop_many ~> Sublist.sub_rev_drop_many

170 le_list_rev_take_iff ~> Sublist.sub_append

171 le_list_same_length ~> Sublist.sub_same_length

172 le_list_take_many_iff ~> Sublist.sub_append'

173 less_eq_list.drop ~> less_eq_list_drop

174 less_eq_list.induct ~> less_eq_list_induct

175 not_le_list_length ~> Sublist.not_sub_length

177 INCOMPATIBILITY.

179 * HOL/Rings: renamed lemmas

181 left_distrib ~> distrib_right

182 right_distrib ~> distrib_left

184 in class semiring. INCOMPATIBILITY.

186 * HOL/BNF: New (co)datatype package based on bounded natural

187 functors with support for mixed, nested recursion and interesting

188 non-free datatypes.

190 * HOL/Cardinals: Theories of ordinals and cardinals

191 (supersedes the AFP entry "Ordinals_and_Cardinals").

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

194 execution for code generated towards Isabelle/ML.

196 * Library/FinFun.thy: theory of almost everywhere constant functions

197 (supersedes the AFP entry "Code Generation for Functions as Data").

199 * Library/Phantom.thy: generic phantom type to make a type parameter

200 appear in a constant's type. This alternative to adding TYPE('a) as

201 another parameter avoids unnecessary closures in generated code.

203 * Library/RBT_Impl.thy: efficient construction of red-black trees

204 from sorted associative lists. Merging two trees with rbt_union may

205 return a structurally different tree than before. MINOR INCOMPATIBILITY.

207 * Simproc "finite_Collect" rewrites set comprehensions into pointfree

208 expressions.

210 * Preprocessing of the code generator rewrites set comprehensions into

211 pointfree expressions.

213 * Quickcheck:

215 - added an optimisation for equality premises.

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

217 the configuration quickcheck_optimise_equality to false.

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

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

221 arithmetic, datatypes/records) with potential proof reconstruction

222 failures, enable the configuration option "z3_with_extensions".

223 Minor INCOMPATIBILITY.

225 * Sledgehammer:

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

228 Sledgehammer manual for details.

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

230 - Renamed "kill_provers" subcommand to "kill"

231 - Renamed options:

232 isar_proof ~> isar_proofs

233 isar_shrink_factor ~> isar_shrink

234 max_relevant ~> max_facts

235 relevance_thresholds ~> fact_thresholds

238 *** Document preparation ***

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

241 slightly exotic babel/greek.

243 * Document variant NAME may use different LaTeX entry point

244 document/root_NAME.tex if that file exists, instead of the common

245 document/root.tex.

247 * Simplified custom document/build script, instead of old-style

248 document/IsaMakefile. Minor INCOMPATIBILITY.

251 *** ML ***

253 * Type Seq.results and related operations support embedded error

254 messages within lazy enumerations, and thus allow to provide

255 informative errors in the absence of any usable results.

257 * Renamed Position.str_of to Position.here to emphasize that this is a

258 formal device to inline positions into message text, but not

259 necessarily printing visible text.

262 *** System ***

264 * The default limit for maximum number of worker threads is now 8,

265 instead of 4.

267 * The ML system is configured as regular component, and no longer

268 picked up from some surrounding directory. Potential INCOMPATIBILITY

269 for home-made configurations.

271 * The "isabelle logo" tool produces EPS and PDF format simultaneously.

272 Minor INCOMPATIBILITY in command-line options.

274 * Advanced support for Isabelle sessions and build management, see

275 "system" manual for the chapter of that name, especially the "isabelle

276 build" tool and its examples. INCOMPATIBILITY, isabelle usedir /

277 mkdir / make are rendered obsolete.

279 * Discontinued obsolete "isabelle makeall".

281 * Discontinued obsolete IsaMakefile and ROOT.ML files from the

282 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that

283 provides some traditional targets that invoke "isabelle build". Note

284 that this is inefficient! Applications of Isabelle/HOL involving

285 "isabelle make" should be upgraded to use "isabelle build" directly.

287 * Discontinued obsolete Isabelle/build script, it is superseded by the

288 regular isabelle build tool. For example:

290 isabelle build -s -b HOLCF

292 * The "isabelle options" tool prints Isabelle system options, as

293 required for "isabelle build", for example.

295 * The "isabelle mkroot" tool prepares session root directories for use

296 with "isabelle build", similar to former "isabelle mkdir" for

297 "isabelle usedir".

299 * The "isabelle components" tool helps to resolve add-on components

300 that are not bundled, or referenced from a bare-bones repository

301 version of Isabelle.

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

304 without exception positions and advanced ML compiler/toplevel

305 configuration.

307 * Discontinued special treatment of Proof General -- no longer guess

308 PROOFGENERAL_HOME based on accidental file-system layout. Minor

309 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS

310 settings manually, or use a Proof General version that has been

311 bundled as Isabelle component.

314 New in Isabelle2012 (May 2012)

315 ------------------------------

317 *** General ***

319 * Prover IDE (PIDE) improvements:

321 - more robust Sledgehammer integration (as before the sledgehammer

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

323 - markup for bound variables

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

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

326 - improved support for Unicode outside original 16bit range

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

329 * Forward declaration of outer syntax keywords within the theory

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

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

333 * Auxiliary contexts indicate block structure for specifications with

334 additional parameters and assumptions. Such unnamed contexts may be

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

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

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

339 context

340 fixes x y z :: 'a

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

342 begin

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

346 end

348 thm my_trans

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

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

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

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

355 ... begin ... end" block as well.

357 * Bundled declarations associate attributed fact expressions with a

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

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

360 the logical dependencies of locales and locale interpretation. See

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

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

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

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

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

367 indices of schematic variables.

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

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

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

372 the historic accident of dynamic re-evaluation in interpretations

373 etc. was exploited.

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

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

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

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

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

380 of the material of that old tutorial.

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

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

384 manual.

387 *** Pure ***

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

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

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

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

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

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

396 undocumented feature.)

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

399 like type constraints. INCOMPATIBILITY in rare situations, where

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

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

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

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

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

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

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

410 order to accommodate newly introduced premises.)

412 * Renamed some inner syntax categories:

414 num ~> num_token

415 xnum ~> xnum_token

416 xstr ~> str_token

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

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

420 position information via constraints).

422 * Simplified configuration options for syntax ambiguity: see

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

424 manual. Minor INCOMPATIBILITY.

426 * Discontinued configuration option "syntax_positions": atomic terms

427 in parse trees are always annotated by position constraints.

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

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

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

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

434 "code_unfold" instead. INCOMPATIBILITY.

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

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

439 definition [code_abbrev]: "f = t"

441 INCOMPATIBILITY.

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

444 instead. INCOMPATIBILITY.

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

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

448 legacy since Isabelle2011).

451 *** HOL ***

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

454 Isabelle2008). Definitions mem_def and Collect_def have disappeared.

455 Non-trivial INCOMPATIBILITY. For developments keeping predicates and

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

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

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

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

460 tinkering with former theorems mem_def and Collect_def as far as

461 possible.

463 For developments which deliberately mix predicates and sets, a

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

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

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

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

469 than predicates. INCOMPATIBILITY.

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

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

474 represents strictly positive binary numerals, along with functions

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

476 positive and negated numeric literals, respectively. See also

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

478 user theories may require adaptations as follows:

480 - Theorems with number_ring or number_semiring constraints: These

481 classes are gone; use comm_ring_1 or comm_semiring_1 instead.

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

484 and number_ring instances. Defer all theorems about numerals until

485 after classes one and semigroup_add have been instantiated.

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

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

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

491 automatically support numerals now, so more simp rules and

492 simprocs may now apply within the proof.

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

495 Redefine using other integer operations.

497 * Transfer: New package intended to generalize the existing

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

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

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

501 corresponding parts of the Quotient package.)

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

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

505 relate different type instances of the same polymorphic constant,

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

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

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

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

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

513 equivalent subgoal on the corresponding raw types. Constants are

514 replaced with corresponding ones according to the transfer rules.

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

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

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

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

519 is logically stronger (rather than equivalent).

521 - relator_eq attribute: Collects identity laws for relators of

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

523 transfer method uses these lemmas to infer transfer rules for

524 non-polymorphic constants on the fly.

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

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

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

529 after unfolding the constant definitions.

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

532 from type nat to type int.

534 * Lifting: New package intended to generalize the quotient_definition

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

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

538 terms of a corresponding operation on a representation

539 type. Example syntax:

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

542 is List.insert

544 Users must discharge a respectfulness proof obligation when each

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

546 the proof is discharged automatically.) The obligation is

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

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

549 by the package.

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

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

553 lift_definition generates a code certificate theorem and sets up

554 code generation for each constant.

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

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

558 or a type_definition theorem. The package configures transfer

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

560 lift_definition command to work with the type.

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

563 Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,

564 Word/Word.thy and Library/Float.thy.

566 * Quotient package:

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

569 rep and abs functions, similar to typedef.

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

572 Transfer packages, as with 'setup_lifting'.

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

575 respectfulness property at the point where the constant is

576 defined, similar to lift_definition; INCOMPATIBILITY.

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

579 accordingly, INCOMPATIBILITY.

581 * New diagnostic command 'find_unused_assms' to find potentially

582 superfluous assumptions in theorems using Quickcheck.

584 * Quickcheck:

586 - Quickcheck returns variable assignments as counterexamples, which

587 allows to reveal the underspecification of functions under test.

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

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

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

592 These counterexample are marked as potentially spurious, as

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

594 obvious theorem "hd xs = hd xs".

596 After finding a potentially spurious counterexample, Quickcheck

597 continues searching for genuine ones.

599 By default, Quickcheck shows potentially spurious and genuine

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

601 show genuine counterexamples.

603 - The command 'quickcheck_generator' creates random and exhaustive

604 value generators for a given type and operations.

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

607 constructors of that type.

609 - Support for multisets.

611 - Added "use_subtype" options.

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

614 conjectures in a locale context.

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

617 and affecting 'rat' and 'real'.

619 * Sledgehammer:

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

621 paper "More SPASS with Isabelle".

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

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

624 TFF1), iProver, iProver-Eq.

625 - Sped up the minimizer.

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

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

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

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

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

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

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

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

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

640 list.exhaust [case_product nat.exhaust]

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

643 a list and a nat.

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

646 eventually_elim* rules. Supports structured proofs.

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

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

650 default.

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

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

656 * Datatype: type parameters allow explicit sort constraints.

658 * Concrete syntax for case expressions includes constraints for source

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

660 INCOMPATIBILITY for old-style syntax translations that augment the

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

662 one_case.

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

665 the result of the fact expression.

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

668 and predicates. Added powers of predicate relations. Consolidation

669 of some relation theorems:

671 converse_def ~> converse_unfold

672 rel_comp_def ~> relcomp_unfold

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

674 transp_def ~> transp_trans

675 Domain_def ~> Domain_unfold

676 Range_def ~> Domain_converse [symmetric]

678 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.

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

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

683 INCOMPATIBILITY.

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

686 to match the constant's name:

688 rel_pow_1 ~> relpow_1

689 rel_pow_0_I ~> relpow_0_I

690 rel_pow_Suc_I ~> relpow_Suc_I

691 rel_pow_Suc_I2 ~> relpow_Suc_I2

692 rel_pow_0_E ~> relpow_0_E

693 rel_pow_Suc_E ~> relpow_Suc_E

694 rel_pow_E ~> relpow_E

695 rel_pow_Suc_D2 ~> relpow_Suc_D2

696 rel_pow_Suc_E2 ~> relpow_Suc_E2

697 rel_pow_Suc_D2' ~> relpow_Suc_D2'

698 rel_pow_E2 ~> relpow_E2

699 rel_pow_add ~> relpow_add

700 rel_pow_commute ~> relpow

701 rel_pow_empty ~> relpow_empty:

702 rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow

703 rel_pow_imp_rtrancl ~> relpow_imp_rtrancl

704 rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow

705 rtrancl_imp_rel_pow ~> rtrancl_imp_relpow

706 rel_pow_fun_conv ~> relpow_fun_conv

707 rel_pow_finite_bounded1 ~> relpow_finite_bounded1

708 rel_pow_finite_bounded ~> relpow_finite_bounded

709 rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow

710 trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow

711 single_valued_rel_pow ~> single_valued_relpow

713 INCOMPATIBILITY.

715 * Theory Relation: Consolidated constant name for relation composition

716 and corresponding theorem names:

718 - Renamed constant rel_comp to relcomp.

720 - Dropped abbreviation pred_comp. Use relcompp instead.

722 - Renamed theorems:

724 rel_compI ~> relcompI

725 rel_compEpair ~> relcompEpair

726 rel_compE ~> relcompE

727 pred_comp_rel_comp_eq ~> relcompp_relcomp_eq

728 rel_comp_empty1 ~> relcomp_empty1

729 rel_comp_mono ~> relcomp_mono

730 rel_comp_subset_Sigma ~> relcomp_subset_Sigma

731 rel_comp_distrib ~> relcomp_distrib

732 rel_comp_distrib2 ~> relcomp_distrib2

733 rel_comp_UNION_distrib ~> relcomp_UNION_distrib

734 rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2

735 single_valued_rel_comp ~> single_valued_relcomp

736 rel_comp_def ~> relcomp_unfold

737 converse_rel_comp ~> converse_relcomp

738 pred_compI ~> relcomppI

739 pred_compE ~> relcomppE

740 pred_comp_bot1 ~> relcompp_bot1

741 pred_comp_bot2 ~> relcompp_bot2

742 transp_pred_comp_less_eq ~> transp_relcompp_less_eq

743 pred_comp_mono ~> relcompp_mono

744 pred_comp_distrib ~> relcompp_distrib

745 pred_comp_distrib2 ~> relcompp_distrib2

746 converse_pred_comp ~> converse_relcompp

748 finite_rel_comp ~> finite_relcomp

750 set_rel_comp ~> set_relcomp

752 INCOMPATIBILITY.

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

755 INCOMPATIBILITY, use the corresponding generic theorems instead.

757 DIVISION_BY_ZERO ~> div_by_0, mod_by_0

758 zdiv_self ~> div_self

759 zmod_self ~> mod_self

760 zdiv_zero ~> div_0

761 zmod_zero ~> mod_0

762 zdiv_zmod_equality ~> div_mod_equality2

763 zdiv_zmod_equality2 ~> div_mod_equality

764 zmod_zdiv_trivial ~> mod_div_trivial

765 zdiv_zminus_zminus ~> div_minus_minus

766 zmod_zminus_zminus ~> mod_minus_minus

767 zdiv_zminus2 ~> div_minus_right

768 zmod_zminus2 ~> mod_minus_right

769 zdiv_minus1_right ~> div_minus1_right

770 zmod_minus1_right ~> mod_minus1_right

771 zdvd_mult_div_cancel ~> dvd_mult_div_cancel

772 zmod_zmult1_eq ~> mod_mult_right_eq

773 zpower_zmod ~> power_mod

774 zdvd_zmod ~> dvd_mod

775 zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd

776 mod_mult_distrib ~> mult_mod_left

777 mod_mult_distrib2 ~> mult_mod_right

779 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use

780 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.

782 * Finite_Set.fold now qualified. INCOMPATIBILITY.

784 * Consolidated theorem names concerning fold combinators:

786 inf_INFI_fold_inf ~> inf_INF_fold_inf

787 sup_SUPR_fold_sup ~> sup_SUP_fold_sup

788 INFI_fold_inf ~> INF_fold_inf

789 SUPR_fold_sup ~> SUP_fold_sup

790 union_set ~> union_set_fold

791 minus_set ~> minus_set_fold

792 INFI_set_fold ~> INF_set_fold

793 SUPR_set_fold ~> SUP_set_fold

794 INF_code ~> INF_set_foldr

795 SUP_code ~> SUP_set_foldr

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

797 foldr_fold_rev ~> foldr_conv_fold

798 foldl_fold ~> foldl_conv_fold

799 foldr_foldr ~> foldr_conv_foldl

800 foldl_foldr ~> foldl_conv_foldr

801 fold_set_remdups ~> fold_set_fold_remdups

802 fold_set ~> fold_set_fold

803 fold1_set ~> fold1_set_fold

805 INCOMPATIBILITY.

807 * Dropped rarely useful theorems concerning fold combinators:

808 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,

809 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,

810 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,

811 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,

812 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,

813 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.

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

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

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

817 unfolding "foldr_conv_fold" and "foldl_conv_fold".

819 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,

820 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,

821 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,

822 INF_set_foldr, SUP_set_foldr. INCOMPATIBILITY. Prefer corresponding

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

824 fold_conv_foldr and fold_rev.

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

827 through option types.

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

830 relations.

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

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

835 INCOMPATIBILITY, use the corresponding generic theorems instead.

837 zminus_zminus ~> minus_minus

838 zminus_0 ~> minus_zero

839 zminus_zadd_distrib ~> minus_add_distrib

840 zadd_commute ~> add_commute

841 zadd_assoc ~> add_assoc

842 zadd_left_commute ~> add_left_commute

843 zadd_ac ~> add_ac

844 zmult_ac ~> mult_ac

845 zadd_0 ~> add_0_left

846 zadd_0_right ~> add_0_right

847 zadd_zminus_inverse2 ~> left_minus

848 zmult_zminus ~> mult_minus_left

849 zmult_commute ~> mult_commute

850 zmult_assoc ~> mult_assoc

851 zadd_zmult_distrib ~> left_distrib

852 zadd_zmult_distrib2 ~> right_distrib

853 zdiff_zmult_distrib ~> left_diff_distrib

854 zdiff_zmult_distrib2 ~> right_diff_distrib

855 zmult_1 ~> mult_1_left

856 zmult_1_right ~> mult_1_right

857 zle_refl ~> order_refl

858 zle_trans ~> order_trans

859 zle_antisym ~> order_antisym

860 zle_linear ~> linorder_linear

861 zless_linear ~> linorder_less_linear

862 zadd_left_mono ~> add_left_mono

863 zadd_strict_right_mono ~> add_strict_right_mono

864 zadd_zless_mono ~> add_less_le_mono

865 int_0_less_1 ~> zero_less_one

866 int_0_neq_1 ~> zero_neq_one

867 zless_le ~> less_le

868 zpower_zadd_distrib ~> power_add

869 zero_less_zpower_abs_iff ~> zero_less_power_abs_iff

870 zero_le_zpower_abs ~> zero_le_power_abs

872 * Theory Deriv: Renamed

874 DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing

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

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

879 are expressed via type classes again. The special syntax

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

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

882 INCOMPATIBILITY.

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

885 use theory HOL/Library/Nat_Bijection instead.

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

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

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

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

891 follows:

893 Operations:

894 bulkload -> rbt_bulkload

895 del_from_left -> rbt_del_from_left

896 del_from_right -> rbt_del_from_right

897 del -> rbt_del

898 delete -> rbt_delete

899 ins -> rbt_ins

900 insert -> rbt_insert

901 insertw -> rbt_insert_with

902 insert_with_key -> rbt_insert_with_key

903 map_entry -> rbt_map_entry

904 lookup -> rbt_lookup

905 sorted -> rbt_sorted

906 tree_greater -> rbt_greater

907 tree_less -> rbt_less

908 tree_less_symbol -> rbt_less_symbol

909 union -> rbt_union

910 union_with -> rbt_union_with

911 union_with_key -> rbt_union_with_key

913 Lemmas:

914 balance_left_sorted -> balance_left_rbt_sorted

915 balance_left_tree_greater -> balance_left_rbt_greater

916 balance_left_tree_less -> balance_left_rbt_less

917 balance_right_sorted -> balance_right_rbt_sorted

918 balance_right_tree_greater -> balance_right_rbt_greater

919 balance_right_tree_less -> balance_right_rbt_less

920 balance_sorted -> balance_rbt_sorted

921 balance_tree_greater -> balance_rbt_greater

922 balance_tree_less -> balance_rbt_less

923 bulkload_is_rbt -> rbt_bulkload_is_rbt

924 combine_sorted -> combine_rbt_sorted

925 combine_tree_greater -> combine_rbt_greater

926 combine_tree_less -> combine_rbt_less

927 delete_in_tree -> rbt_delete_in_tree

928 delete_is_rbt -> rbt_delete_is_rbt

929 del_from_left_tree_greater -> rbt_del_from_left_rbt_greater

930 del_from_left_tree_less -> rbt_del_from_left_rbt_less

931 del_from_right_tree_greater -> rbt_del_from_right_rbt_greater

932 del_from_right_tree_less -> rbt_del_from_right_rbt_less

933 del_in_tree -> rbt_del_in_tree

934 del_inv1_inv2 -> rbt_del_inv1_inv2

935 del_sorted -> rbt_del_rbt_sorted

936 del_tree_greater -> rbt_del_rbt_greater

937 del_tree_less -> rbt_del_rbt_less

938 dom_lookup_Branch -> dom_rbt_lookup_Branch

939 entries_lookup -> entries_rbt_lookup

940 finite_dom_lookup -> finite_dom_rbt_lookup

941 insert_sorted -> rbt_insert_rbt_sorted

942 insertw_is_rbt -> rbt_insertw_is_rbt

943 insertwk_is_rbt -> rbt_insertwk_is_rbt

944 insertwk_sorted -> rbt_insertwk_rbt_sorted

945 insertw_sorted -> rbt_insertw_rbt_sorted

946 ins_sorted -> ins_rbt_sorted

947 ins_tree_greater -> ins_rbt_greater

948 ins_tree_less -> ins_rbt_less

949 is_rbt_sorted -> is_rbt_rbt_sorted

950 lookup_balance -> rbt_lookup_balance

951 lookup_bulkload -> rbt_lookup_rbt_bulkload

952 lookup_delete -> rbt_lookup_rbt_delete

953 lookup_Empty -> rbt_lookup_Empty

954 lookup_from_in_tree -> rbt_lookup_from_in_tree

955 lookup_in_tree -> rbt_lookup_in_tree

956 lookup_ins -> rbt_lookup_ins

957 lookup_insert -> rbt_lookup_rbt_insert

958 lookup_insertw -> rbt_lookup_rbt_insertw

959 lookup_insertwk -> rbt_lookup_rbt_insertwk

960 lookup_keys -> rbt_lookup_keys

961 lookup_map -> rbt_lookup_map

962 lookup_map_entry -> rbt_lookup_rbt_map_entry

963 lookup_tree_greater -> rbt_lookup_rbt_greater

964 lookup_tree_less -> rbt_lookup_rbt_less

965 lookup_union -> rbt_lookup_rbt_union

966 map_entry_color_of -> rbt_map_entry_color_of

967 map_entry_inv1 -> rbt_map_entry_inv1

968 map_entry_inv2 -> rbt_map_entry_inv2

969 map_entry_is_rbt -> rbt_map_entry_is_rbt

970 map_entry_sorted -> rbt_map_entry_rbt_sorted

971 map_entry_tree_greater -> rbt_map_entry_rbt_greater

972 map_entry_tree_less -> rbt_map_entry_rbt_less

973 map_tree_greater -> map_rbt_greater

974 map_tree_less -> map_rbt_less

975 map_sorted -> map_rbt_sorted

976 paint_sorted -> paint_rbt_sorted

977 paint_lookup -> paint_rbt_lookup

978 paint_tree_greater -> paint_rbt_greater

979 paint_tree_less -> paint_rbt_less

980 sorted_entries -> rbt_sorted_entries

981 tree_greater_eq_trans -> rbt_greater_eq_trans

982 tree_greater_nit -> rbt_greater_nit

983 tree_greater_prop -> rbt_greater_prop

984 tree_greater_simps -> rbt_greater_simps

985 tree_greater_trans -> rbt_greater_trans

986 tree_less_eq_trans -> rbt_less_eq_trans

987 tree_less_nit -> rbt_less_nit

988 tree_less_prop -> rbt_less_prop

989 tree_less_simps -> rbt_less_simps

990 tree_less_trans -> rbt_less_trans

991 tree_ord_props -> rbt_ord_props

992 union_Branch -> rbt_union_Branch

993 union_is_rbt -> rbt_union_is_rbt

994 unionw_is_rbt -> rbt_unionw_is_rbt

995 unionwk_is_rbt -> rbt_unionwk_is_rbt

996 unionwk_sorted -> rbt_unionwk_rbt_sorted

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

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

1000 lifing-framework and proofs use the transfer method. INCOMPATIBILITY.

1002 Changed Operations:

1003 float_abs -> abs

1004 float_nprt -> nprt

1005 float_pprt -> pprt

1006 pow2 -> use powr

1007 round_down -> float_round_down

1008 round_up -> float_round_up

1009 scale -> exponent

1011 Removed Operations:

1012 ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod

1014 Renamed Lemmas:

1015 abs_float_def -> Float.compute_float_abs

1016 bitlen_ge0 -> bitlen_nonneg

1017 bitlen.simps -> Float.compute_bitlen

1018 float_components -> Float_mantissa_exponent

1019 float_divl.simps -> Float.compute_float_divl

1020 float_divr.simps -> Float.compute_float_divr

1021 float_eq_odd -> mult_powr_eq_mult_powr_iff

1022 float_power -> real_of_float_power

1023 lapprox_posrat_def -> Float.compute_lapprox_posrat

1024 lapprox_rat.simps -> Float.compute_lapprox_rat

1025 le_float_def' -> Float.compute_float_le

1026 le_float_def -> less_eq_float.rep_eq

1027 less_float_def' -> Float.compute_float_less

1028 less_float_def -> less_float.rep_eq

1029 normfloat_def -> Float.compute_normfloat

1030 normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0

1031 normfloat -> normfloat_def

1032 normfloat_unique -> use normfloat_def

1033 number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral

1034 one_float_def -> Float.compute_float_one

1035 plus_float_def -> Float.compute_float_plus

1036 rapprox_posrat_def -> Float.compute_rapprox_posrat

1037 rapprox_rat.simps -> Float.compute_rapprox_rat

1038 real_of_float_0 -> zero_float.rep_eq

1039 real_of_float_1 -> one_float.rep_eq

1040 real_of_float_abs -> abs_float.rep_eq

1041 real_of_float_add -> plus_float.rep_eq

1042 real_of_float_minus -> uminus_float.rep_eq

1043 real_of_float_mult -> times_float.rep_eq

1044 real_of_float_simp -> Float.rep_eq

1045 real_of_float_sub -> minus_float.rep_eq

1046 round_down.simps -> Float.compute_float_round_down

1047 round_up.simps -> Float.compute_float_round_up

1048 times_float_def -> Float.compute_float_times

1049 uminus_float_def -> Float.compute_float_uminus

1050 zero_float_def -> Float.compute_float_zero

1052 Lemmas not necessary anymore, use the transfer method:

1053 bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,

1054 float_divr, float_le_simp, float_less1_mantissa_bound,

1055 float_less_simp, float_less_zero, float_le_zero,

1056 float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,

1057 floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,

1058 lapprox_rat_bottom, normalized_float, rapprox_posrat,

1059 rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,

1060 real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,

1061 round_up, zero_le_float, zero_less_float

1063 * New theory HOL/Library/DAList provides an abstract type for

1064 association lists with distinct keys.

1066 * Session HOL/IMP: Added new theory of abstract interpretation of

1067 annotated commands.

1069 * Session HOL-Import: Re-implementation from scratch is faster,

1070 simpler, and more scalable. Requires a proof bundle, which is

1071 available as an external component. Discontinued old (and mostly

1072 dead) Importer for HOL4 and HOL Light. INCOMPATIBILITY.

1074 * Session HOL-Word: Discontinued many redundant theorems specific to

1075 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems

1076 instead.

1078 word_sub_alt ~> word_sub_wi

1079 word_add_alt ~> word_add_def

1080 word_mult_alt ~> word_mult_def

1081 word_minus_alt ~> word_minus_def

1082 word_0_alt ~> word_0_wi

1083 word_1_alt ~> word_1_wi

1084 word_add_0 ~> add_0_left

1085 word_add_0_right ~> add_0_right

1086 word_mult_1 ~> mult_1_left

1087 word_mult_1_right ~> mult_1_right

1088 word_add_commute ~> add_commute

1089 word_add_assoc ~> add_assoc

1090 word_add_left_commute ~> add_left_commute

1091 word_mult_commute ~> mult_commute

1092 word_mult_assoc ~> mult_assoc

1093 word_mult_left_commute ~> mult_left_commute

1094 word_left_distrib ~> left_distrib

1095 word_right_distrib ~> right_distrib

1096 word_left_minus ~> left_minus

1097 word_diff_0_right ~> diff_0_right

1098 word_diff_self ~> diff_self

1099 word_sub_def ~> diff_minus

1100 word_diff_minus ~> diff_minus

1101 word_add_ac ~> add_ac

1102 word_mult_ac ~> mult_ac

1103 word_plus_ac0 ~> add_0_left add_0_right add_ac

1104 word_times_ac1 ~> mult_1_left mult_1_right mult_ac

1105 word_order_trans ~> order_trans

1106 word_order_refl ~> order_refl

1107 word_order_antisym ~> order_antisym

1108 word_order_linear ~> linorder_linear

1109 lenw1_zero_neq_one ~> zero_neq_one

1110 word_number_of_eq ~> number_of_eq

1111 word_of_int_add_hom ~> wi_hom_add

1112 word_of_int_sub_hom ~> wi_hom_sub

1113 word_of_int_mult_hom ~> wi_hom_mult

1114 word_of_int_minus_hom ~> wi_hom_neg

1115 word_of_int_succ_hom ~> wi_hom_succ

1116 word_of_int_pred_hom ~> wi_hom_pred

1117 word_of_int_0_hom ~> word_0_wi

1118 word_of_int_1_hom ~> word_1_wi

1120 * Session HOL-Word: New proof method "word_bitwise" for splitting

1121 machine word equalities and inequalities into logical circuits,

1122 defined in HOL/Word/WordBitwise.thy. Supports addition, subtraction,

1123 multiplication, shifting by constants, bitwise operators and numeric

1124 constants. Requires fixed-length word types, not 'a word. Solves

1125 many standard word identities outright and converts more into first

1126 order problems amenable to blast or similar. See also examples in

1127 HOL/Word/Examples/WordExamples.thy.

1129 * Session HOL-Probability: Introduced the type "'a measure" to

1130 represent measures, this replaces the records 'a algebra and 'a

1131 measure_space. The locales based on subset_class now have two

1132 locale-parameters the space \<Omega> and the set of measurable sets M.

1133 The product of probability spaces uses now the same constant as the

1134 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)

1135 measure". Most constants are defined now outside of locales and gain

1136 an additional parameter, like null_sets, almost_eventually or \<mu>'.

1137 Measure space constructions for distributions and densities now got

1138 their own constants distr and density. Instead of using locales to

1139 describe measure spaces with a finite space, the measure count_space

1140 and point_measure is introduced. INCOMPATIBILITY.

1142 Renamed constants:

1143 measure -> emeasure

1144 finite_measure.\<mu>' -> measure

1145 product_algebra_generator -> prod_algebra

1146 product_prob_space.emb -> prod_emb

1147 product_prob_space.infprod_algebra -> PiM

1149 Removed locales:

1150 completeable_measure_space

1151 finite_measure_space

1152 finite_prob_space

1153 finite_product_finite_prob_space

1154 finite_product_sigma_algebra

1155 finite_sigma_algebra

1156 measure_space

1157 pair_finite_prob_space

1158 pair_finite_sigma_algebra

1159 pair_finite_space

1160 pair_sigma_algebra

1161 product_sigma_algebra

1163 Removed constants:

1164 conditional_space

1165 distribution -> use distr measure, or distributed predicate

1166 image_space

1167 joint_distribution -> use distr measure, or distributed predicate

1168 pair_measure_generator

1169 product_prob_space.infprod_algebra -> use PiM

1170 subvimage

1172 Replacement theorems:

1173 finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite

1174 finite_measure.empty_measure -> measure_empty

1175 finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq

1176 finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq

1177 finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably

1178 finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure

1179 finite_measure.finite_measure -> finite_measure.emeasure_finite

1180 finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton

1181 finite_measure.positive_measure' -> measure_nonneg

1182 finite_measure.real_measure -> finite_measure.emeasure_real

1183 finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb

1184 finite_product_sigma_algebra.in_P -> sets_PiM_I_finite

1185 finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty

1186 information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed

1187 information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple

1188 information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple

1189 information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple

1190 information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple

1191 information_space.entropy_commute -> information_space.entropy_commute_simple

1192 information_space.entropy_eq -> information_space.entropy_simple_distributed

1193 information_space.entropy_generic_eq -> information_space.entropy_simple_distributed

1194 information_space.entropy_positive -> information_space.entropy_nonneg_simple

1195 information_space.entropy_uniform_max -> information_space.entropy_uniform

1196 information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq

1197 information_space.KL_eq_0 -> information_space.KL_same_eq_0

1198 information_space.KL_ge_0 -> information_space.KL_nonneg

1199 information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed

1200 information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple

1201 Int_stable_cuboids -> Int_stable_atLeastAtMost

1202 Int_stable_product_algebra_generator -> positive_integral

1203 measure_preserving -> equality "distr M N f = N" "f : measurable M N"

1204 measure_space.additive -> emeasure_additive

1205 measure_space.AE_iff_null_set -> AE_iff_null

1206 measure_space.almost_everywhere_def -> eventually_ae_filter

1207 measure_space.almost_everywhere_vimage -> AE_distrD

1208 measure_space.continuity_from_above -> INF_emeasure_decseq

1209 measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq

1210 measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq

1211 measure_space.continuity_from_below -> SUP_emeasure_incseq

1212 measure_space_density -> emeasure_density

1213 measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density

1214 measure_space.integrable_vimage -> integrable_distr

1215 measure_space.integral_translated_density -> integral_density

1216 measure_space.integral_vimage -> integral_distr

1217 measure_space.measure_additive -> plus_emeasure

1218 measure_space.measure_compl -> emeasure_compl

1219 measure_space.measure_countable_increasing -> emeasure_countable_increasing

1220 measure_space.measure_countably_subadditive -> emeasure_subadditive_countably

1221 measure_space.measure_decseq -> decseq_emeasure

1222 measure_space.measure_Diff -> emeasure_Diff

1223 measure_space.measure_Diff_null_set -> emeasure_Diff_null_set

1224 measure_space.measure_eq_0 -> emeasure_eq_0

1225 measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite

1226 measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton

1227 measure_space.measure_incseq -> incseq_emeasure

1228 measure_space.measure_insert -> emeasure_insert

1229 measure_space.measure_mono -> emeasure_mono

1230 measure_space.measure_not_negative -> emeasure_not_MInf

1231 measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq

1232 measure_space.measure_setsum -> setsum_emeasure

1233 measure_space.measure_setsum_split -> setsum_emeasure_cover

1234 measure_space.measure_space_vimage -> emeasure_distr

1235 measure_space.measure_subadditive_finite -> emeasure_subadditive_finite

1236 measure_space.measure_subadditive -> subadditive

1237 measure_space.measure_top -> emeasure_space

1238 measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0

1239 measure_space.measure_Un_null_set -> emeasure_Un_null_set

1240 measure_space.positive_integral_translated_density -> positive_integral_density

1241 measure_space.positive_integral_vimage -> positive_integral_distr

1242 measure_space.real_continuity_from_above -> Lim_measure_decseq

1243 measure_space.real_continuity_from_below -> Lim_measure_incseq

1244 measure_space.real_measure_countably_subadditive -> measure_subadditive_countably

1245 measure_space.real_measure_Diff -> measure_Diff

1246 measure_space.real_measure_finite_Union -> measure_finite_Union

1247 measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton

1248 measure_space.real_measure_subadditive -> measure_subadditive

1249 measure_space.real_measure_Union -> measure_Union

1250 measure_space.real_measure_UNION -> measure_UNION

1251 measure_space.simple_function_vimage -> simple_function_comp

1252 measure_space.simple_integral_vimage -> simple_integral_distr

1253 measure_space.simple_integral_vimage -> simple_integral_distr

1254 measure_unique_Int_stable -> measure_eqI_generator_eq

1255 measure_unique_Int_stable_vimage -> measure_eqI_generator_eq

1256 pair_sigma_algebra.measurable_cut_fst -> sets_Pair1

1257 pair_sigma_algebra.measurable_cut_snd -> sets_Pair2

1258 pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1

1259 pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2

1260 pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff

1261 pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap

1262 pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'

1263 pair_sigma_algebra.sets_swap -> sets_pair_swap

1264 pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1

1265 pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2

1266 pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap

1267 pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2

1268 pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt

1269 pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times

1270 prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM

1271 prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq

1272 prob_space.measure_space_1 -> prob_space.emeasure_space_1

1273 prob_space.prob_space_vimage -> prob_space_distr

1274 prob_space.random_variable_restrict -> measurable_restrict

1275 prob_space_unique_Int_stable -> measure_eqI_prob_space

1276 product_algebraE -> prod_algebraE_all

1277 product_algebra_generator_der -> prod_algebra_eq_finite

1278 product_algebra_generator_into_space -> prod_algebra_sets_into_space

1279 product_algebraI -> sets_PiM_I_finite

1280 product_measure_exists -> product_sigma_finite.sigma_finite

1281 product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator

1282 product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb

1283 product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty

1284 product_prob_space.measurable_component -> measurable_component_singleton

1285 product_prob_space.measurable_emb -> measurable_prod_emb

1286 product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single

1287 product_prob_space.measurable_singleton_infprod -> measurable_component_singleton

1288 product_prob_space.measure_emb -> emeasure_prod_emb

1289 product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict

1290 product_sigma_algebra.product_algebra_into_space -> space_closed

1291 product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge

1292 product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton

1293 product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge

1294 sequence_space.measure_infprod -> sequence_space.measure_PiM_countable

1295 sets_product_algebra -> sets_PiM

1296 sigma_algebra.measurable_sigma -> measurable_measure_of

1297 sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint

1298 sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr

1299 sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq

1300 space_product_algebra -> space_PiM

1302 * Session HOL-TPTP: support to parse and import TPTP problems (all

1303 languages) into Isabelle/HOL.

1306 *** FOL ***

1308 * New "case_product" attribute (see HOL).

1311 *** ZF ***

1313 * Greater support for structured proofs involving induction or case

1314 analysis.

1316 * Much greater use of mathematical symbols.

1318 * Removal of many ML theorem bindings. INCOMPATIBILITY.

1321 *** ML ***

1323 * Antiquotation @{keyword "name"} produces a parser for outer syntax

1324 from a minor keyword introduced via theory header declaration.

1326 * Antiquotation @{command_spec "name"} produces the

1327 Outer_Syntax.command_spec from a major keyword introduced via theory

1328 header declaration; it can be passed to Outer_Syntax.command etc.

1330 * Local_Theory.define no longer hard-wires default theorem name

1331 "foo_def", but retains the binding as given. If that is Binding.empty

1332 / Attrib.empty_binding, the result is not registered as user-level

1333 fact. The Local_Theory.define_internal variant allows to specify a

1334 non-empty name (used for the foundation in the background theory),

1335 while omitting the fact binding in the user-context. Potential

1336 INCOMPATIBILITY for derived definitional packages: need to specify

1337 naming policy for primitive definitions more explicitly.

1339 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in

1340 conformance with similar operations in structure Term and Logic.

1342 * Antiquotation @{attributes [...]} embeds attribute source

1343 representation into the ML text, which is particularly useful with

1344 declarations like Local_Theory.note.

1346 * Structure Proof_Context follows standard naming scheme. Old

1347 ProofContext has been discontinued. INCOMPATIBILITY.

1349 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle

1350 change of semantics: update is applied to auxiliary local theory

1351 context as well.

1353 * Modernized some old-style infix operations:

1355 addeqcongs ~> Simplifier.add_eqcong

1356 deleqcongs ~> Simplifier.del_eqcong

1357 addcongs ~> Simplifier.add_cong

1358 delcongs ~> Simplifier.del_cong

1359 setmksimps ~> Simplifier.set_mksimps

1360 setmkcong ~> Simplifier.set_mkcong

1361 setmksym ~> Simplifier.set_mksym

1362 setmkeqTrue ~> Simplifier.set_mkeqTrue

1363 settermless ~> Simplifier.set_termless

1364 setsubgoaler ~> Simplifier.set_subgoaler

1365 addsplits ~> Splitter.add_split

1366 delsplits ~> Splitter.del_split

1369 *** System ***

1371 * USER_HOME settings variable points to cross-platform user home

1372 directory, which coincides with HOME on POSIX systems only. Likewise,

1373 the Isabelle path specification "~" now expands to $USER_HOME, instead

1374 of former $HOME. A different default for USER_HOME may be set

1375 explicitly in shell environment, before Isabelle settings are

1376 evaluated. Minor INCOMPATIBILITY: need to adapt Isabelle path where

1377 the generic user home was intended.

1379 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name

1380 notation, which is useful for the jEdit file browser, for example.

1382 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar

1383 (not just JRE).

1387 New in Isabelle2011-1 (October 2011)

1388 ------------------------------------

1390 *** General ***

1392 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as

1393 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.

1395 - Management of multiple theory files directly from the editor

1396 buffer store -- bypassing the file-system (no requirement to save

1397 files for checking).

1399 - Markup of formal entities within the text buffer, with semantic

1400 highlighting, tooltips and hyperlinks to jump to defining source

1401 positions.

1403 - Improved text rendering, with sub/superscripts in the source

1404 buffer (including support for copy/paste wrt. output panel, HTML

1405 theory output and other non-Isabelle text boxes).

1407 - Refined scheduling of proof checking and printing of results,

1408 based on interactive editor view. (Note: jEdit folding and

1409 narrowing allows to restrict buffer perspectives explicitly.)

1411 - Reduced CPU performance requirements, usable on machines with few

1412 cores.

1414 - Reduced memory requirements due to pruning of unused document

1415 versions (garbage collection).

1417 See also ~~/src/Tools/jEdit/README.html for further information,

1418 including some remaining limitations.

1420 * Theory loader: source files are exclusively located via the master

1421 directory of each theory node (where the .thy file itself resides).

1422 The global load path (such as src/HOL/Library) has been discontinued.

1423 Note that the path element ~~ may be used to reference theories in the

1424 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".

1425 INCOMPATIBILITY.

1427 * Theory loader: source files are identified by content via SHA1

1428 digests. Discontinued former path/modtime identification and optional

1429 ISABELLE_FILE_IDENT plugin scripts.

1431 * Parallelization of nested Isar proofs is subject to

1432 Goal.parallel_proofs_threshold (default 100). See also isabelle

1433 usedir option -Q.

1435 * Name space: former unsynchronized references are now proper

1436 configuration options, with more conventional names:

1438 long_names ~> names_long

1439 short_names ~> names_short

1440 unique_names ~> names_unique

1442 Minor INCOMPATIBILITY, need to declare options in context like this:

1444 declare [[names_unique = false]]

1446 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`. Note

1447 that the result needs to be unique, which means fact specifications

1448 may have to be refined after enriching a proof context.

1450 * Attribute "case_names" has been refined: the assumptions in each case

1451 can be named now by following the case name with [name1 name2 ...].

1453 * Isabelle/Isar reference manual has been updated and extended:

1454 - "Synopsis" provides a catalog of main Isar language concepts.

1455 - Formal references in syntax diagrams, via @{rail} antiquotation.

1456 - Updated material from classic "ref" manual, notably about

1457 "Classical Reasoner".

1460 *** HOL ***

1462 * Class bot and top require underlying partial order rather than

1463 preorder: uniqueness of bot and top is guaranteed. INCOMPATIBILITY.

1465 * Class complete_lattice: generalized a couple of lemmas from sets;

1466 generalized theorems INF_cong and SUP_cong. New type classes for

1467 complete boolean algebras and complete linear orders. Lemmas

1468 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in

1469 class complete_linorder.

1471 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,

1472 Sup_fun_def, Inf_apply, Sup_apply.

1474 Removed redundant lemmas (the right hand side gives hints how to

1475 replace them for (metis ...), or (simp only: ...) proofs):

1477 Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]

1478 Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]

1479 Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right

1480 Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right

1481 Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right

1482 Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right

1483 Inter_def ~> INF_def, image_def

1484 Union_def ~> SUP_def, image_def

1485 INT_eq ~> INF_def, and image_def

1486 UN_eq ~> SUP_def, and image_def

1487 INF_subset ~> INF_superset_mono [OF _ order_refl]

1489 More consistent and comprehensive names:

1491 INTER_eq_Inter_image ~> INF_def

1492 UNION_eq_Union_image ~> SUP_def

1493 INFI_def ~> INF_def

1494 SUPR_def ~> SUP_def

1495 INF_leI ~> INF_lower

1496 INF_leI2 ~> INF_lower2

1497 le_INFI ~> INF_greatest

1498 le_SUPI ~> SUP_upper

1499 le_SUPI2 ~> SUP_upper2

1500 SUP_leI ~> SUP_least

1501 INFI_bool_eq ~> INF_bool_eq

1502 SUPR_bool_eq ~> SUP_bool_eq

1503 INFI_apply ~> INF_apply

1504 SUPR_apply ~> SUP_apply

1505 INTER_def ~> INTER_eq

1506 UNION_def ~> UNION_eq

1508 INCOMPATIBILITY.

1510 * Renamed theory Complete_Lattice to Complete_Lattices.

1511 INCOMPATIBILITY.

1513 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,

1514 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,

1515 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,

1516 Sup_insert are now declared as [simp]. INCOMPATIBILITY.

1518 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,

1519 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,

1520 sup_inf_absob, sup_left_idem are now declared as [simp]. Minor

1521 INCOMPATIBILITY.

1523 * Added syntactic classes "inf" and "sup" for the respective

1524 constants. INCOMPATIBILITY: Changes in the argument order of the

1525 (mostly internal) locale predicates for some derived classes.

1527 * Theorem collections ball_simps and bex_simps do not contain theorems

1528 referring to UNION any longer; these have been moved to collection

1529 UN_ball_bex_simps. INCOMPATIBILITY.

1531 * Theory Archimedean_Field: floor now is defined as parameter of a

1532 separate type class floor_ceiling.

1534 * Theory Finite_Set: more coherent development of fold_set locales:

1536 locale fun_left_comm ~> locale comp_fun_commute

1537 locale fun_left_comm_idem ~> locale comp_fun_idem

1539 Both use point-free characterization; interpretation proofs may need

1540 adjustment. INCOMPATIBILITY.

1542 * Theory Limits: Type "'a net" has been renamed to "'a filter", in

1543 accordance with standard mathematical terminology. INCOMPATIBILITY.

1545 * Theory Complex_Main: The locale interpretations for the

1546 bounded_linear and bounded_bilinear locales have been removed, in

1547 order to reduce the number of duplicate lemmas. Users must use the

1548 original names for distributivity theorems, potential INCOMPATIBILITY.

1550 divide.add ~> add_divide_distrib

1551 divide.diff ~> diff_divide_distrib

1552 divide.setsum ~> setsum_divide_distrib

1553 mult.add_right ~> right_distrib

1554 mult.diff_right ~> right_diff_distrib

1555 mult_right.setsum ~> setsum_right_distrib

1556 mult_left.diff ~> left_diff_distrib

1558 * Theory Complex_Main: Several redundant theorems have been removed or

1559 replaced by more general versions. INCOMPATIBILITY.

1561 real_diff_def ~> minus_real_def

1562 real_divide_def ~> divide_real_def

1563 real_less_def ~> less_le

1564 real_abs_def ~> abs_real_def

1565 real_sgn_def ~> sgn_real_def

1566 real_mult_commute ~> mult_commute

1567 real_mult_assoc ~> mult_assoc

1568 real_mult_1 ~> mult_1_left

1569 real_add_mult_distrib ~> left_distrib

1570 real_zero_not_eq_one ~> zero_neq_one

1571 real_mult_inverse_left ~> left_inverse

1572 INVERSE_ZERO ~> inverse_zero

1573 real_le_refl ~> order_refl

1574 real_le_antisym ~> order_antisym

1575 real_le_trans ~> order_trans

1576 real_le_linear ~> linear

1577 real_le_eq_diff ~> le_iff_diff_le_0

1578 real_add_left_mono ~> add_left_mono

1579 real_mult_order ~> mult_pos_pos

1580 real_mult_less_mono2 ~> mult_strict_left_mono

1581 real_of_int_real_of_nat ~> real_of_int_of_nat_eq

1582 real_0_le_divide_iff ~> zero_le_divide_iff

1583 realpow_two_disj ~> power2_eq_iff

1584 real_squared_diff_one_factored ~> square_diff_one_factored

1585 realpow_two_diff ~> square_diff_square_factored

1586 reals_complete2 ~> complete_real

1587 real_sum_squared_expand ~> power2_sum

1588 exp_ln_eq ~> ln_unique

1589 expi_add ~> exp_add

1590 expi_zero ~> exp_zero

1591 lemma_DERIV_subst ~> DERIV_cong

1592 LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff

1593 LIMSEQ_const ~> tendsto_const

1594 LIMSEQ_norm ~> tendsto_norm

1595 LIMSEQ_add ~> tendsto_add

1596 LIMSEQ_minus ~> tendsto_minus

1597 LIMSEQ_minus_cancel ~> tendsto_minus_cancel

1598 LIMSEQ_diff ~> tendsto_diff

1599 bounded_linear.LIMSEQ ~> bounded_linear.tendsto

1600 bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto

1601 LIMSEQ_mult ~> tendsto_mult

1602 LIMSEQ_inverse ~> tendsto_inverse

1603 LIMSEQ_divide ~> tendsto_divide

1604 LIMSEQ_pow ~> tendsto_power

1605 LIMSEQ_setsum ~> tendsto_setsum

1606 LIMSEQ_setprod ~> tendsto_setprod

1607 LIMSEQ_norm_zero ~> tendsto_norm_zero_iff

1608 LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff

1609 LIMSEQ_imp_rabs ~> tendsto_rabs

1610 LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]

1611 LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]

1612 LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]

1613 LIMSEQ_Complex ~> tendsto_Complex

1614 LIM_ident ~> tendsto_ident_at

1615 LIM_const ~> tendsto_const

1616 LIM_add ~> tendsto_add

1617 LIM_add_zero ~> tendsto_add_zero

1618 LIM_minus ~> tendsto_minus

1619 LIM_diff ~> tendsto_diff

1620 LIM_norm ~> tendsto_norm

1621 LIM_norm_zero ~> tendsto_norm_zero

1622 LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel

1623 LIM_norm_zero_iff ~> tendsto_norm_zero_iff

1624 LIM_rabs ~> tendsto_rabs

1625 LIM_rabs_zero ~> tendsto_rabs_zero

1626 LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel

1627 LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff

1628 LIM_compose ~> tendsto_compose

1629 LIM_mult ~> tendsto_mult

1630 LIM_scaleR ~> tendsto_scaleR

1631 LIM_of_real ~> tendsto_of_real

1632 LIM_power ~> tendsto_power

1633 LIM_inverse ~> tendsto_inverse

1634 LIM_sgn ~> tendsto_sgn

1635 isCont_LIM_compose ~> isCont_tendsto_compose

1636 bounded_linear.LIM ~> bounded_linear.tendsto

1637 bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero

1638 bounded_bilinear.LIM ~> bounded_bilinear.tendsto

1639 bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero

1640 bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero

1641 bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero

1642 LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]

1644 * Theory Complex_Main: The definition of infinite series was

1645 generalized. Now it is defined on the type class {topological_space,

1646 comm_monoid_add}. Hence it is useable also for extended real numbers.

1648 * Theory Complex_Main: The complex exponential function "expi" is now

1649 a type-constrained abbreviation for "exp :: complex => complex"; thus

1650 several polymorphic lemmas about "exp" are now applicable to "expi".

1652 * Code generation:

1654 - Theory Library/Code_Char_ord provides native ordering of

1655 characters in the target language.

1657 - Commands code_module and code_library are legacy, use export_code

1658 instead.

1660 - Method "evaluation" is legacy, use method "eval" instead.

1662 - Legacy evaluator "SML" is deactivated by default. May be

1663 reactivated by the following theory command:

1665 setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}

1667 * Declare ext [intro] by default. Rare INCOMPATIBILITY.

1669 * New proof method "induction" that gives induction hypotheses the

1670 name "IH", thus distinguishing them from further hypotheses that come

1671 from rule induction. The latter are still called "hyps". Method

1672 "induction" is a thin wrapper around "induct" and follows the same

1673 syntax.

1675 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is

1676 still available as a legacy feature for some time.

1678 * Nitpick:

1679 - Added "need" and "total_consts" options.

1680 - Reintroduced "show_skolems" option by popular demand.

1681 - Renamed attribute: nitpick_def ~> nitpick_unfold.

1682 INCOMPATIBILITY.

1684 * Sledgehammer:

1685 - Use quasi-sound (and efficient) translations by default.

1686 - Added support for the following provers: E-ToFoF, LEO-II,

1687 Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.

1688 - Automatically preplay and minimize proofs before showing them if

1689 this can be done within reasonable time.

1690 - sledgehammer available_provers ~> sledgehammer supported_provers.

1691 INCOMPATIBILITY.

1692 - Added "preplay_timeout", "slicing", "type_enc", "sound",

1693 "max_mono_iters", and "max_new_mono_instances" options.

1694 - Removed "explicit_apply" and "full_types" options as well as "Full

1695 Types" Proof General menu item. INCOMPATIBILITY.

1697 * Metis:

1698 - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.

1699 - Obsoleted "metisFT" -- use "metis (full_types)" instead.

1700 INCOMPATIBILITY.

1702 * Command 'try':

1703 - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and

1704 "elim:" options. INCOMPATIBILITY.

1705 - Introduced 'try' that not only runs 'try_methods' but also

1706 'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.

1708 * Quickcheck:

1709 - Added "eval" option to evaluate terms for the found counterexample

1710 (currently only supported by the default (exhaustive) tester).

1711 - Added post-processing of terms to obtain readable counterexamples

1712 (currently only supported by the default (exhaustive) tester).

1713 - New counterexample generator quickcheck[narrowing] enables

1714 narrowing-based testing. Requires the Glasgow Haskell compiler

1715 with its installation location defined in the Isabelle settings

1716 environment as ISABELLE_GHC.

1717 - Removed quickcheck tester "SML" based on the SML code generator

1718 (formly in HOL/Library).

1720 * Function package: discontinued option "tailrec". INCOMPATIBILITY,

1721 use 'partial_function' instead.

1723 * Theory Library/Extended_Reals replaces now the positive extended

1724 reals found in probability theory. This file is extended by

1725 Multivariate_Analysis/Extended_Real_Limits.

1727 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,

1728 from where it must be imported explicitly if it is really required.

1729 INCOMPATIBILITY.

1731 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has

1732 been moved here. INCOMPATIBILITY.

1734 * Theory Library/Saturated provides type of numbers with saturated

1735 arithmetic.

1737 * Theory Library/Product_Lattice defines a pointwise ordering for the

1738 product type 'a * 'b, and provides instance proofs for various order

1739 and lattice type classes.

1741 * Theory Library/Countable now provides the "countable_datatype" proof

1742 method for proving "countable" class instances for datatypes.

1744 * Theory Library/Cset_Monad allows do notation for computable sets

1745 (cset) via the generic monad ad-hoc overloading facility.

1747 * Library: Theories of common data structures are split into theories

1748 for implementation, an invariant-ensuring type, and connection to an

1749 abstract type. INCOMPATIBILITY.

1751 - RBT is split into RBT and RBT_Mapping.

1752 - AssocList is split and renamed into AList and AList_Mapping.

1753 - DList is split into DList_Impl, DList, and DList_Cset.

1754 - Cset is split into Cset and List_Cset.

1756 * Theory Library/Nat_Infinity has been renamed to

1757 Library/Extended_Nat, with name changes of the following types and

1758 constants:

1760 type inat ~> type enat

1761 Fin ~> enat

1762 Infty ~> infinity (overloaded)

1763 iSuc ~> eSuc

1764 the_Fin ~> the_enat

1766 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has

1767 been renamed accordingly. INCOMPATIBILITY.

1769 * Session Multivariate_Analysis: The euclidean_space type class now

1770 fixes a constant "Basis :: 'a set" consisting of the standard

1771 orthonormal basis for the type. Users now have the option of

1772 quantifying over this set instead of using the "basis" function, e.g.

1773 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".

1775 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed

1776 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants

1777 "Cart_nth" and "Cart_lambda" have been respectively renamed to

1778 "vec_nth" and "vec_lambda"; theorems mentioning those names have

1779 changed to match. Definition theorems for overloaded constants now use

1780 the standard "foo_vec_def" naming scheme. A few other theorems have

1781 been renamed as follows (INCOMPATIBILITY):

1783 Cart_eq ~> vec_eq_iff

1784 dist_nth_le_cart ~> dist_vec_nth_le

1785 tendsto_vector ~> vec_tendstoI

1786 Cauchy_vector ~> vec_CauchyI

1788 * Session Multivariate_Analysis: Several duplicate theorems have been

1789 removed, and other theorems have been renamed or replaced with more

1790 general versions. INCOMPATIBILITY.

1792 finite_choice ~> finite_set_choice

1793 eventually_conjI ~> eventually_conj

1794 eventually_and ~> eventually_conj_iff

1795 eventually_false ~> eventually_False

1796 setsum_norm ~> norm_setsum

1797 Lim_sequentially ~> LIMSEQ_def

1798 Lim_ident_at ~> LIM_ident

1799 Lim_const ~> tendsto_const

1800 Lim_cmul ~> tendsto_scaleR [OF tendsto_const]

1801 Lim_neg ~> tendsto_minus

1802 Lim_add ~> tendsto_add

1803 Lim_sub ~> tendsto_diff

1804 Lim_mul ~> tendsto_scaleR

1805 Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]

1806 Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]

1807 Lim_linear ~> bounded_linear.tendsto

1808 Lim_component ~> tendsto_euclidean_component

1809 Lim_component_cart ~> tendsto_vec_nth

1810 Lim_inner ~> tendsto_inner [OF tendsto_const]

1811 dot_lsum ~> inner_setsum_left

1812 dot_rsum ~> inner_setsum_right

1813 continuous_cmul ~> continuous_scaleR [OF continuous_const]

1814 continuous_neg ~> continuous_minus

1815 continuous_sub ~> continuous_diff

1816 continuous_vmul ~> continuous_scaleR [OF _ continuous_const]

1817 continuous_mul ~> continuous_scaleR

1818 continuous_inv ~> continuous_inverse

1819 continuous_at_within_inv ~> continuous_at_within_inverse

1820 continuous_at_inv ~> continuous_at_inverse

1821 continuous_at_norm ~> continuous_norm [OF continuous_at_id]

1822 continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]

1823 continuous_at_component ~> continuous_component [OF continuous_at_id]

1824 continuous_on_neg ~> continuous_on_minus

1825 continuous_on_sub ~> continuous_on_diff

1826 continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]

1827 continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]

1828 continuous_on_mul ~> continuous_on_scaleR

1829 continuous_on_mul_real ~> continuous_on_mult

1830 continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]

1831 continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]

1832 continuous_on_inverse ~> continuous_on_inv

1833 uniformly_continuous_on_neg ~> uniformly_continuous_on_minus

1834 uniformly_continuous_on_sub ~> uniformly_continuous_on_diff

1835 subset_interior ~> interior_mono

1836 subset_closure ~> closure_mono

1837 closure_univ ~> closure_UNIV

1838 real_arch_lt ~> reals_Archimedean2

1839 real_arch ~> reals_Archimedean3

1840 real_abs_norm ~> abs_norm_cancel

1841 real_abs_sub_norm ~> norm_triangle_ineq3

1842 norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2

1844 * Session HOL-Probability:

1845 - Caratheodory's extension lemma is now proved for ring_of_sets.

1846 - Infinite products of probability measures are now available.

1847 - Sigma closure is independent, if the generator is independent

1848 - Use extended reals instead of positive extended

1849 reals. INCOMPATIBILITY.

1851 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.

1853 expand_fun_below ~> fun_below_iff

1854 below_fun_ext ~> fun_belowI

1855 expand_cfun_eq ~> cfun_eq_iff

1856 ext_cfun ~> cfun_eqI

1857 expand_cfun_below ~> cfun_below_iff

1858 below_cfun_ext ~> cfun_belowI

1859 monofun_fun_fun ~> fun_belowD

1860 monofun_fun_arg ~> monofunE

1861 monofun_lub_fun ~> adm_monofun [THEN admD]

1862 cont_lub_fun ~> adm_cont [THEN admD]

1863 cont2cont_Rep_CFun ~> cont2cont_APP

1864 cont_Rep_CFun_app ~> cont_APP_app

1865 cont_Rep_CFun_app_app ~> cont_APP_app_app

1866 cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]

1867 cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]

1868 contlub_cfun ~> lub_APP [symmetric]

1869 contlub_LAM ~> lub_LAM [symmetric]

1870 thelubI ~> lub_eqI

1871 UU_I ~> bottomI

1872 lift_distinct1 ~> lift.distinct(1)

1873 lift_distinct2 ~> lift.distinct(2)

1874 Def_not_UU ~> lift.distinct(2)

1875 Def_inject ~> lift.inject

1876 below_UU_iff ~> below_bottom_iff

1877 eq_UU_iff ~> eq_bottom_iff

1880 *** Document preparation ***

1882 * Antiquotation @{rail} layouts railroad syntax diagrams, see also

1883 isar-ref manual, both for description and actual application of the

1884 same.

1886 * Antiquotation @{value} evaluates the given term and presents its

1887 result.

1889 * Antiquotations: term style "isub" provides ad-hoc conversion of

1890 variables x1, y23 into subscripted form x\<^isub>1,

1891 y\<^isub>2\<^isub>3.

1893 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>

1894 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).

1896 * Localized \isabellestyle switch can be used within blocks or groups

1897 like this:

1899 \isabellestyle{it} %preferred default

1900 {\isabellestylett @{text "typewriter stuff"}}

1902 * Discontinued special treatment of hard tabulators. Implicit

1903 tab-width is now defined as 1. Potential INCOMPATIBILITY for visual

1904 layouts.

1907 *** ML ***

1909 * The inner syntax of sort/type/term/prop supports inlined YXML

1910 representations within quoted string tokens. By encoding logical

1911 entities via Term_XML (in ML or Scala) concrete syntax can be

1912 bypassed, which is particularly useful for producing bits of text

1913 under external program control.

1915 * Antiquotations for ML and document preparation are managed as theory

1916 data, which requires explicit setup.

1918 * Isabelle_Process.is_active allows tools to check if the official

1919 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop

1920 (better known as Proof General).

1922 * Structure Proof_Context follows standard naming scheme. Old

1923 ProofContext is still available for some time as legacy alias.

1925 * Structure Timing provides various operations for timing; supersedes

1926 former start_timing/end_timing etc.

1928 * Path.print is the official way to show file-system paths to users

1929 (including quotes etc.).

1931 * Inner syntax: identifiers in parse trees of generic categories

1932 "logic", "aprop", "idt" etc. carry position information (disguised as

1933 type constraints). Occasional INCOMPATIBILITY with non-compliant

1934 translations that choke on unexpected type constraints. Positions can

1935 be stripped in ML translations via Syntax.strip_positions /

1936 Syntax.strip_positions_ast, or via the syntax constant

1937 "_strip_positions" within parse trees. As last resort, positions can

1938 be disabled via the configuration option Syntax.positions, which is

1939 called "syntax_positions" in Isar attribute syntax.

1941 * Discontinued special status of various ML structures that contribute

1942 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less

1943 pervasive content, no inclusion in structure Syntax. INCOMPATIBILITY,

1944 refer directly to Ast.Constant, Lexicon.is_identifier,

1945 Syntax_Trans.mk_binder_tr etc.

1947 * Typed print translation: discontinued show_sorts argument, which is

1948 already available via context of "advanced" translation.

1950 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic

1951 goal states; body tactic needs to address all subgoals uniformly.

1953 * Slightly more special eq_list/eq_set, with shortcut involving

1954 pointer equality (assumes that eq relation is reflexive).

1956 * Classical tactics use proper Proof.context instead of historic types

1957 claset/clasimpset. Old-style declarations like addIs, addEs, addDs

1958 operate directly on Proof.context. Raw type claset retains its use as

1959 snapshot of the classical context, which can be recovered via

1960 (put_claset HOL_cs) etc. Type clasimpset has been discontinued.

1961 INCOMPATIBILITY, classical tactics and derived proof methods require

1962 proper Proof.context.

1965 *** System ***

1967 * Discontinued support for Poly/ML 5.2, which was the last version

1968 without proper multithreading and TimeLimit implementation.

1970 * Discontinued old lib/scripts/polyml-platform, which has been

1971 obsolete since Isabelle2009-2.

1973 * Various optional external tools are referenced more robustly and

1974 uniformly by explicit Isabelle settings as follows:

1976 ISABELLE_CSDP (formerly CSDP_EXE)

1977 ISABELLE_GHC (formerly EXEC_GHC or GHC_PATH)

1978 ISABELLE_OCAML (formerly EXEC_OCAML)

1979 ISABELLE_SWIPL (formerly EXEC_SWIPL)

1980 ISABELLE_YAP (formerly EXEC_YAP)

1982 Note that automated detection from the file-system or search path has

1983 been discontinued. INCOMPATIBILITY.

1985 * Scala layer provides JVM method invocation service for static

1986 methods of type (String)String, see Invoke_Scala.method in ML. For

1987 example:

1989 Invoke_Scala.method "java.lang.System.getProperty" "java.home"

1991 Together with YXML.string_of_body/parse_body and XML.Encode/Decode

1992 this allows to pass structured values between ML and Scala.

1994 * The IsabelleText fonts includes some further glyphs to support the

1995 Prover IDE. Potential INCOMPATIBILITY: users who happen to have

1996 installed a local copy (which is normally *not* required) need to

1997 delete or update it from ~~/lib/fonts/.

2001 New in Isabelle2011 (January 2011)

2002 ----------------------------------

2004 *** General ***

2006 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see

2007 src/Tools/jEdit). This also serves as IDE for Isabelle/ML, with

2008 useful tooltips and hyperlinks produced from its static analysis. The

2009 bundled component provides an executable Isabelle tool that can be run

2010 like this:

2012 Isabelle2011/bin/isabelle jedit

2014 * Significantly improved Isabelle/Isar implementation manual.

2016 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER

2017 (and thus refers to something like $HOME/.isabelle/Isabelle2011),

2018 while the default heap location within that directory lacks that extra

2019 suffix. This isolates multiple Isabelle installations from each

2020 other, avoiding problems with old settings in new versions.

2021 INCOMPATIBILITY, need to copy/upgrade old user settings manually.

2023 * Source files are always encoded as UTF-8, instead of old-fashioned

2024 ISO-Latin-1. INCOMPATIBILITY. Isabelle LaTeX documents might require

2025 the following package declarations:

2027 \usepackage[utf8]{inputenc}

2028 \usepackage{textcomp}

2030 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that

2031 a Unicode character is treated as a single symbol, not a sequence of

2032 non-ASCII bytes as before. Since Isabelle/ML string literals may

2033 contain symbols without further backslash escapes, Unicode can now be

2034 used here as well. Recall that Symbol.explode in ML provides a

2035 consistent view on symbols, while raw explode (or String.explode)

2036 merely give a byte-oriented representation.

2038 * Theory loader: source files are primarily located via the master

2039 directory of each theory node (where the .thy file itself resides).

2040 The global load path is still partially available as legacy feature.

2041 Minor INCOMPATIBILITY due to subtle change in file lookup: use

2042 explicit paths, relatively to the theory.

2044 * Special treatment of ML file names has been discontinued.

2045 Historically, optional extensions .ML or .sml were added on demand --

2046 at the cost of clarity of file dependencies. Recall that Isabelle/ML

2047 files exclusively use the .ML extension. Minor INCOMPATIBILTY.

2049 * Various options that affect pretty printing etc. are now properly

2050 handled within the context via configuration options, instead of

2051 unsynchronized references or print modes. There are both ML Config.T

2052 entities and Isar declaration attributes to access these.

2054 ML (Config.T) Isar (attribute)

2056 eta_contract eta_contract

2057 show_brackets show_brackets

2058 show_sorts show_sorts

2059 show_types show_types

2060 show_question_marks show_question_marks

2061 show_consts show_consts

2062 show_abbrevs show_abbrevs

2064 Syntax.ast_trace syntax_ast_trace

2065 Syntax.ast_stat syntax_ast_stat

2066 Syntax.ambiguity_level syntax_ambiguity_level

2068 Goal_Display.goals_limit goals_limit

2069 Goal_Display.show_main_goal show_main_goal

2071 Method.rule_trace rule_trace

2073 Thy_Output.display thy_output_display

2074 Thy_Output.quotes thy_output_quotes

2075 Thy_Output.indent thy_output_indent

2076 Thy_Output.source thy_output_source

2077 Thy_Output.break thy_output_break

2079 Note that corresponding "..._default" references in ML may only be

2080 changed globally at the ROOT session setup, but *not* within a theory.

2081 The option "show_abbrevs" supersedes the former print mode

2082 "no_abbrevs" with inverted meaning.

2084 * More systematic naming of some configuration options.

2085 INCOMPATIBILITY.

2087 trace_simp ~> simp_trace

2088 debug_simp ~> simp_debug

2090 * Support for real valued configuration options, using simplistic

2091 floating-point notation that coincides with the inner syntax for

2092 float_token.

2094 * Support for real valued preferences (with approximative PGIP type):

2095 front-ends need to accept "pgint" values in float notation.

2096 INCOMPATIBILITY.

2098 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from

2099 DejaVu Sans.

2101 * Discontinued support for Poly/ML 5.0 and 5.1 versions.

2104 *** Pure ***

2106 * Command 'type_synonym' (with single argument) replaces somewhat

2107 outdated 'types', which is still available as legacy feature for some

2108 time.

2110 * Command 'nonterminal' (with 'and' separated list of arguments)

2111 replaces somewhat outdated 'nonterminals'. INCOMPATIBILITY.

2113 * Command 'notepad' replaces former 'example_proof' for

2114 experimentation in Isar without any result. INCOMPATIBILITY.

2116 * Locale interpretation commands 'interpret' and 'sublocale' accept

2117 lists of equations to map definitions in a locale to appropriate

2118 entities in the context of the interpretation. The 'interpretation'

2119 command already provided this functionality.

2121 * Diagnostic command 'print_dependencies' prints the locale instances

2122 that would be activated if the specified expression was interpreted in

2123 the current context. Variant "print_dependencies!" assumes a context

2124 without interpretations.

2126 * Diagnostic command 'print_interps' prints interpretations in proofs

2127 in addition to interpretations in theories.

2129 * Discontinued obsolete 'global' and 'local' commands to manipulate

2130 the theory name space. Rare INCOMPATIBILITY. The ML functions

2131 Sign.root_path and Sign.local_path may be applied directly where this

2132 feature is still required for historical reasons.

2134 * Discontinued obsolete 'constdefs' command. INCOMPATIBILITY, use

2135 'definition' instead.

2137 * The "prems" fact, which refers to the accidental collection of

2138 foundational premises in the context, is now explicitly marked as

2139 legacy feature and will be discontinued soon. Consider using "assms"

2140 of the head statement or reference facts by explicit names.

2142 * Document antiquotations @{class} and @{type} print classes and type

2143 constructors.

2145 * Document antiquotation @{file} checks file/directory entries within

2146 the local file system.

2149 *** HOL ***

2151 * Coercive subtyping: functions can be declared as coercions and type

2152 inference will add them as necessary upon input of a term. Theory

2153 Complex_Main declares real :: nat => real and real :: int => real as

2154 coercions. A coercion function f is declared like this:

2156 declare [[coercion f]]

2158 To lift coercions through type constructors (e.g. from nat => real to

2159 nat list => real list), map functions can be declared, e.g.

2161 declare [[coercion_map map]]

2163 Currently coercion inference is activated only in theories including

2164 real numbers, i.e. descendants of Complex_Main. This is controlled by

2165 the configuration option "coercion_enabled", e.g. it can be enabled in

2166 other theories like this:

2168 declare [[coercion_enabled]]

2170 * Command 'partial_function' provides basic support for recursive

2171 function definitions over complete partial orders. Concrete instances

2172 are provided for i) the option type, ii) tail recursion on arbitrary

2173 types, and iii) the heap monad of Imperative_HOL. See

2174 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy

2175 for examples.

2177 * Function package: f.psimps rules are no longer implicitly declared

2178 as [simp]. INCOMPATIBILITY.

2180 * Datatype package: theorems generated for executable equality (class

2181 "eq") carry proper names and are treated as default code equations.

2183 * Inductive package: now offers command 'inductive_simps' to

2184 automatically derive instantiated and simplified equations for

2185 inductive predicates, similar to 'inductive_cases'.

2187 * Command 'enriched_type' allows to register properties of the

2188 functorial structure of types.

2190 * Improved infrastructure for term evaluation using code generator

2191 techniques, in particular static evaluation conversions.

2193 * Code generator: Scala (2.8 or higher) has been added to the target

2194 languages.

2196 * Code generator: globbing constant expressions "*" and "Theory.*"

2197 have been replaced by the more idiomatic "_" and "Theory._".

2198 INCOMPATIBILITY.

2200 * Code generator: export_code without explicit file declaration prints

2201 to standard output. INCOMPATIBILITY.

2203 * Code generator: do not print function definitions for case

2204 combinators any longer.

2206 * Code generator: simplification with rules determined with

2207 src/Tools/Code/code_simp.ML and method "code_simp".

2209 * Code generator for records: more idiomatic representation of record

2210 types. Warning: records are not covered by ancient SML code

2211 generation any longer. INCOMPATIBILITY. In cases of need, a suitable

2212 rep_datatype declaration helps to succeed then:

2214 record 'a foo = ...

2215 ...

2216 rep_datatype foo_ext ...

2218 * Records: logical foundation type for records does not carry a

2219 '_type' suffix any longer (obsolete due to authentic syntax).

2220 INCOMPATIBILITY.

2222 * Quickcheck now by default uses exhaustive testing instead of random

2223 testing. Random testing can be invoked by "quickcheck [random]",

2224 exhaustive testing by "quickcheck [exhaustive]".

2226 * Quickcheck instantiates polymorphic types with small finite

2227 datatypes by default. This enables a simple execution mechanism to

2228 handle quantifiers and function equality over the finite datatypes.

2230 * Quickcheck random generator has been renamed from "code" to

2231 "random". INCOMPATIBILITY.

2233 * Quickcheck now has a configurable time limit which is set to 30

2234 seconds by default. This can be changed by adding [timeout = n] to the

2235 quickcheck command. The time limit for Auto Quickcheck is still set

2236 independently.

2238 * Quickcheck in locales considers interpretations of that locale for

2239 counter example search.

2241 * Sledgehammer:

2242 - Added "smt" and "remote_smt" provers based on the "smt" proof

2243 method. See the Sledgehammer manual for details ("isabelle doc

2244 sledgehammer").

2245 - Renamed commands:

2246 sledgehammer atp_info ~> sledgehammer running_provers

2247 sledgehammer atp_kill ~> sledgehammer kill_provers

2248 sledgehammer available_atps ~> sledgehammer available_provers

2249 INCOMPATIBILITY.

2250 - Renamed options:

2251 sledgehammer [atps = ...] ~> sledgehammer [provers = ...]

2252 sledgehammer [atp = ...] ~> sledgehammer [prover = ...]

2253 sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]

2254 (and "ms" and "min" are no longer supported)

2255 INCOMPATIBILITY.

2257 * Nitpick:

2258 - Renamed options:

2259 nitpick [timeout = 77 s] ~> nitpick [timeout = 77]

2260 nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]

2261 INCOMPATIBILITY.

2262 - Added support for partial quotient types.

2263 - Added local versions of the "Nitpick.register_xxx" functions.

2264 - Added "whack" option.

2265 - Allow registration of quotient types as codatatypes.

2266 - Improved "merge_type_vars" option to merge more types.

2267 - Removed unsound "fast_descrs" option.

2268 - Added custom symmetry breaking for datatypes, making it possible to reach

2269 higher cardinalities.

2270 - Prevent the expansion of too large definitions.

2272 * Proof methods "metis" and "meson" now have configuration options

2273 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled

2274 to diagnose these tools. E.g.

2276 using [[metis_trace = true]]

2278 * Auto Solve: Renamed "Auto Solve Direct". The tool is now available

2279 manually as command 'solve_direct'.

2281 * The default SMT solver Z3 must be enabled explicitly (due to

2282 licensing issues) by setting the environment variable

2283 Z3_NON_COMMERCIAL in etc/settings of the component, for example. For

2284 commercial applications, the SMT solver CVC3 is provided as fall-back;

2285 changing the SMT solver is done via the configuration option

2286 "smt_solver".

2288 * Remote SMT solvers need to be referred to by the "remote_" prefix,

2289 i.e. "remote_cvc3" and "remote_z3".

2291 * Added basic SMT support for datatypes, records, and typedefs using

2292 the oracle mode (no proofs). Direct support of pairs has been dropped

2293 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT

2294 support for a similar behavior). Minor INCOMPATIBILITY.

2296 * Changed SMT configuration options:

2297 - Renamed:

2298 z3_proofs ~> smt_oracle (with inverted meaning)

2299 z3_trace_assms ~> smt_trace_used_facts

2300 INCOMPATIBILITY.

2301 - Added:

2302 smt_verbose

2303 smt_random_seed

2304 smt_datatypes

2305 smt_infer_triggers

2306 smt_monomorph_limit

2307 cvc3_options

2308 remote_cvc3_options

2309 remote_z3_options

2310 yices_options

2312 * Boogie output files (.b2i files) need to be declared in the theory

2313 header.

2315 * Simplification procedure "list_to_set_comprehension" rewrites list

2316 comprehensions applied to List.set to set comprehensions. Occasional

2317 INCOMPATIBILITY, may be deactivated like this:

2319 declare [[simproc del: list_to_set_comprehension]]

2321 * Removed old version of primrec package. INCOMPATIBILITY.

2323 * Removed simplifier congruence rule of "prod_case", as has for long

2324 been the case with "split". INCOMPATIBILITY.

2326 * String.literal is a type, but not a datatype. INCOMPATIBILITY.

2328 * Removed [split_format ... and ... and ...] version of

2329 [split_format]. Potential INCOMPATIBILITY.

2331 * Predicate "sorted" now defined inductively, with nice induction

2332 rules. INCOMPATIBILITY: former sorted.simps now named sorted_simps.

2334 * Constant "contents" renamed to "the_elem", to free the generic name

2335 contents for other uses. INCOMPATIBILITY.

2337 * Renamed class eq and constant eq (for code generation) to class

2338 equal and constant equal, plus renaming of related facts and various

2339 tuning. INCOMPATIBILITY.

2341 * Dropped type classes mult_mono and mult_mono1. INCOMPATIBILITY.

2343 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".

2344 INCOMPATIBILITY.

2346 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to

2347 avoid confusion with finite sets. INCOMPATIBILITY.

2349 * Abandoned locales equiv, congruent and congruent2 for equivalence

2350 relations. INCOMPATIBILITY: use equivI rather than equiv_intro (same

2351 for congruent(2)).

2353 * Some previously unqualified names have been qualified:

2355 types

2356 bool ~> HOL.bool

2357 nat ~> Nat.nat

2359 constants

2360 Trueprop ~> HOL.Trueprop

2361 True ~> HOL.True

2362 False ~> HOL.False

2363 op & ~> HOL.conj

2364 op | ~> HOL.disj

2365 op --> ~> HOL.implies

2366 op = ~> HOL.eq

2367 Not ~> HOL.Not

2368 The ~> HOL.The

2369 All ~> HOL.All

2370 Ex ~> HOL.Ex

2371 Ex1 ~> HOL.Ex1

2372 Let ~> HOL.Let

2373 If ~> HOL.If

2374 Ball ~> Set.Ball

2375 Bex ~> Set.Bex

2376 Suc ~> Nat.Suc

2377 Pair ~> Product_Type.Pair

2378 fst ~> Product_Type.fst

2379 snd ~> Product_Type.snd

2380 curry ~> Product_Type.curry

2381 op : ~> Set.member

2382 Collect ~> Set.Collect

2384 INCOMPATIBILITY.

2386 * More canonical naming convention for some fundamental definitions:

2388 bot_bool_eq ~> bot_bool_def

2389 top_bool_eq ~> top_bool_def

2390 inf_bool_eq ~> inf_bool_def

2391 sup_bool_eq ~> sup_bool_def

2392 bot_fun_eq ~> bot_fun_def

2393 top_fun_eq ~> top_fun_def

2394 inf_fun_eq ~> inf_fun_def

2395 sup_fun_eq ~> sup_fun_def

2397 INCOMPATIBILITY.

2399 * More stylized fact names:

2401 expand_fun_eq ~> fun_eq_iff

2402 expand_set_eq ~> set_eq_iff

2403 set_ext ~> set_eqI

2404 nat_number ~> eval_nat_numeral

2406 INCOMPATIBILITY.

2408 * Refactoring of code-generation specific operations in theory List:

2410 constants

2411 null ~> List.null

2413 facts

2414 mem_iff ~> member_def

2415 null_empty ~> null_def

2417 INCOMPATIBILITY. Note that these were not supposed to be used

2418 regularly unless for striking reasons; their main purpose was code

2419 generation.

2421 Various operations from the Haskell prelude are used for generating

2422 Haskell code.

2424 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV". Term

2425 "surj f" is now an abbreviation of "range f = UNIV". The theorems

2426 bij_def and surj_def are unchanged. INCOMPATIBILITY.

2428 * Abolished some non-alphabetic type names: "prod" and "sum" replace

2429 "*" and "+" respectively. INCOMPATIBILITY.

2431 * Name "Plus" of disjoint sum operator "<+>" is now hidden. Write

2432 "Sum_Type.Plus" instead.

2434 * Constant "split" has been merged with constant "prod_case"; names of

2435 ML functions, facts etc. involving split have been retained so far,

2436 though. INCOMPATIBILITY.

2438 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"

2439 instead. INCOMPATIBILITY.

2441 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".

2442 INCOMPATIBILITY.

2444 * Former theory Library/Enum is now part of the HOL-Main image.

2445 INCOMPATIBILITY: all constants of the Enum theory now have to be

2446 referred to by its qualified name.

2448 enum ~> Enum.enum

2449 nlists ~> Enum.nlists

2450 product ~> Enum.product

2452 * Theory Library/Monad_Syntax provides do-syntax for monad types.

2453 Syntax in Library/State_Monad has been changed to avoid ambiguities.

2454 INCOMPATIBILITY.

2456 * Theory Library/SetsAndFunctions has been split into

2457 Library/Function_Algebras and Library/Set_Algebras; canonical names

2458 for instance definitions for functions; various improvements.

2459 INCOMPATIBILITY.

2461 * Theory Library/Multiset provides stable quicksort implementation of

2462 sort_key.

2464 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.

2465 INCOMPATIBILITY.

2467 * Session Multivariate_Analysis: introduced a type class for euclidean

2468 space. Most theorems are now stated in terms of euclidean spaces

2469 instead of finite cartesian products.

2471 types

2472 real ^ 'n ~> 'a::real_vector

2473 ~> 'a::euclidean_space

2474 ~> 'a::ordered_euclidean_space

2475 (depends on your needs)

2477 constants

2478 _ $ _ ~> _ $$ _

2479 \<chi> x. _ ~> \<chi>\<chi> x. _

2480 CARD('n) ~> DIM('a)

2482 Also note that the indices are now natural numbers and not from some

2483 finite type. Finite cartesian products of euclidean spaces, products

2484 of euclidean spaces the real and complex numbers are instantiated to

2485 be euclidean_spaces. INCOMPATIBILITY.

2487 * Session Probability: introduced pextreal as positive extended real

2488 numbers. Use pextreal as value for measures. Introduce the

2489 Radon-Nikodym derivative, product spaces and Fubini's theorem for

2490 arbitrary sigma finite measures. Introduces Lebesgue measure based on

2491 the integral in Multivariate Analysis. INCOMPATIBILITY.

2493 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.

2494 INCOMPATIBILITY.

2496 * Session SPARK (with image HOL-SPARK) provides commands to load and

2497 prove verification conditions generated by the SPARK Ada program

2498 verifier. See also src/HOL/SPARK and src/HOL/SPARK/Examples.

2501 *** HOL-Algebra ***

2503 * Theorems for additive ring operations (locale abelian_monoid and

2504 descendants) are generated by interpretation from their multiplicative

2505 counterparts. Names (in particular theorem names) have the mandatory

2506 qualifier 'add'. Previous theorem names are redeclared for

2507 compatibility.

2509 * Structure "int_ring" is now an abbreviation (previously a

2510 definition). This fits more natural with advanced interpretations.

2513 *** HOLCF ***

2515 * The domain package now runs in definitional mode by default: The

2516 former command 'new_domain' is now called 'domain'. To use the domain

2517 package in its original axiomatic mode, use 'domain (unsafe)'.

2518 INCOMPATIBILITY.

2520 * The new class "domain" is now the default sort. Class "predomain"

2521 is an unpointed version of "domain". Theories can be updated by

2522 replacing sort annotations as shown below. INCOMPATIBILITY.

2524 'a::type ~> 'a::countable

2525 'a::cpo ~> 'a::predomain

2526 'a::pcpo ~> 'a::domain

2528 * The old type class "rep" has been superseded by class "domain".

2529 Accordingly, users of the definitional package must remove any

2530 "default_sort rep" declarations. INCOMPATIBILITY.

2532 * The domain package (definitional mode) now supports unpointed

2533 predomain argument types, as long as they are marked 'lazy'. (Strict

2534 arguments must be in class "domain".) For example, the following

2535 domain definition now works:

2537 domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")

2539 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class

2540 instances for types from main HOL: bool, nat, int, char, 'a + 'b,

2541 'a option, and 'a list. Additionally, it configures fixrec and the

2542 domain package to work with these types. For example:

2544 fixrec isInl :: "('a + 'b) u -> tr"

2545 where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"

2547 domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")

2549 * The "(permissive)" option of fixrec has been replaced with a

2550 per-equation "(unchecked)" option. See

2551 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.

2553 * The "bifinite" class no longer fixes a constant "approx"; the class

2554 now just asserts that such a function exists. INCOMPATIBILITY.

2556 * Former type "alg_defl" has been renamed to "defl". HOLCF no longer

2557 defines an embedding of type 'a defl into udom by default; instances

2558 of "bifinite" and "domain" classes are available in

2559 src/HOL/HOLCF/Library/Defl_Bifinite.thy.

2561 * The syntax "REP('a)" has been replaced with "DEFL('a)".

2563 * The predicate "directed" has been removed. INCOMPATIBILITY.

2565 * The type class "finite_po" has been removed. INCOMPATIBILITY.

2567 * The function "cprod_map" has been renamed to "prod_map".

2568 INCOMPATIBILITY.

2570 * The monadic bind operator on each powerdomain has new binder syntax

2571 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents

2572 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".

2574 * The infix syntax for binary union on each powerdomain has changed

2575 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set

2576 syntax. INCOMPATIBILITY.

2578 * The constant "UU" has been renamed to "bottom". The syntax "UU" is

2579 still supported as an input translation.

2581 * Renamed some theorems (the original names are also still available).

2583 expand_fun_below ~> fun_below_iff

2584 below_fun_ext ~> fun_belowI

2585 expand_cfun_eq ~> cfun_eq_iff

2586 ext_cfun ~> cfun_eqI

2587 expand_cfun_below ~> cfun_below_iff

2588 below_cfun_ext ~> cfun_belowI

2589 cont2cont_Rep_CFun ~> cont2cont_APP

2591 * The Abs and Rep functions for various types have changed names.

2592 Related theorem names have also changed to match. INCOMPATIBILITY.

2594 Rep_CFun ~> Rep_cfun

2595 Abs_CFun ~> Abs_cfun

2596 Rep_Sprod ~> Rep_sprod

2597 Abs_Sprod ~> Abs_sprod

2598 Rep_Ssum ~> Rep_ssum

2599 Abs_Ssum ~> Abs_ssum

2601 * Lemmas with names of the form *_defined_iff or *_strict_iff have

2602 been renamed to *_bottom_iff. INCOMPATIBILITY.

2604 * Various changes to bisimulation/coinduction with domain package:

2606 - Definitions of "bisim" constants no longer mention definedness.

2607 - With mutual recursion, "bisim" predicate is now curried.

2608 - With mutual recursion, each type gets a separate coind theorem.

2609 - Variable names in bisim_def and coinduct rules have changed.

2611 INCOMPATIBILITY.

2613 * Case combinators generated by the domain package for type "foo" are

2614 now named "foo_case" instead of "foo_when". INCOMPATIBILITY.

2616 * Several theorems have been renamed to more accurately reflect the

2617 names of constants and types involved. INCOMPATIBILITY.

2619 thelub_const ~> lub_const

2620 lub_const ~> is_lub_const

2621 thelubI ~> lub_eqI

2622 is_lub_lub ~> is_lubD2

2623 lubI ~> is_lub_lub

2624 unique_lub ~> is_lub_unique

2625 is_ub_lub ~> is_lub_rangeD1

2626 lub_bin_chain ~> is_lub_bin_chain

2627 lub_fun ~> is_lub_fun

2628 thelub_fun ~> lub_fun

2629 thelub_cfun ~> lub_cfun

2630 thelub_Pair ~> lub_Pair

2631 lub_cprod ~> is_lub_prod

2632 thelub_cprod ~> lub_prod

2633 minimal_cprod ~> minimal_prod

2634 inst_cprod_pcpo ~> inst_prod_pcpo

2635 UU_I ~> bottomI

2636 compact_UU ~> compact_bottom

2637 deflation_UU ~> deflation_bottom

2638 finite_deflation_UU ~> finite_deflation_bottom

2640 * Many legacy theorem names have been discontinued. INCOMPATIBILITY.

2642 sq_ord_less_eq_trans ~> below_eq_trans

2643 sq_ord_eq_less_trans ~> eq_below_trans

2644 refl_less ~> below_refl

2645 trans_less ~> below_trans

2646 antisym_less ~> below_antisym

2647 antisym_less_inverse ~> po_eq_conv [THEN iffD1]

2648 box_less ~> box_below

2649 rev_trans_less ~> rev_below_trans

2650 not_less2not_eq ~> not_below2not_eq

2651 less_UU_iff ~> below_UU_iff

2652 flat_less_iff ~> flat_below_iff

2653 adm_less ~> adm_below

2654 adm_not_less ~> adm_not_below

2655 adm_compact_not_less ~> adm_compact_not_below

2656 less_fun_def ~> below_fun_def

2657 expand_fun_less ~> fun_below_iff

2658 less_fun_ext ~> fun_belowI

2659 less_discr_def ~> below_discr_def

2660 discr_less_eq ~> discr_below_eq

2661 less_unit_def ~> below_unit_def

2662 less_cprod_def ~> below_prod_def

2663 prod_lessI ~> prod_belowI

2664 Pair_less_iff ~> Pair_below_iff

2665 fst_less_iff ~> fst_below_iff

2666 snd_less_iff ~> snd_below_iff

2667 expand_cfun_less ~> cfun_below_iff

2668 less_cfun_ext ~> cfun_belowI

2669 injection_less ~> injection_below

2670 less_up_def ~> below_up_def

2671 not_Iup_less ~> not_Iup_below

2672 Iup_less ~> Iup_below

2673 up_less ~> up_below

2674 Def_inject_less_eq ~> Def_below_Def

2675 Def_less_is_eq ~> Def_below_iff

2676 spair_less_iff ~> spair_below_iff

2677 less_sprod ~> below_sprod

2678 spair_less ~> spair_below

2679 sfst_less_iff ~> sfst_below_iff

2680 ssnd_less_iff ~> ssnd_below_iff

2681 fix_least_less ~> fix_least_below

2682 dist_less_one ~> dist_below_one

2683 less_ONE ~> below_ONE

2684 ONE_less_iff ~> ONE_below_iff

2685 less_sinlD ~> below_sinlD

2686 less_sinrD ~> below_sinrD

2689 *** FOL and ZF ***

2691 * All constant names are now qualified internally and use proper

2692 identifiers, e.g. "IFOL.eq" instead of "op =". INCOMPATIBILITY.

2695 *** ML ***

2697 * Antiquotation @{assert} inlines a function bool -> unit that raises

2698 Fail if the argument is false. Due to inlining the source position of

2699 failed assertions is included in the error output.

2701 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML

2702 text is in practice always evaluated with a stable theory checkpoint.

2703 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.

2705 * Antiquotation @{theory A} refers to theory A from the ancestry of

2706 the current context, not any accidental theory loader state as before.

2707 Potential INCOMPATIBILITY, subtle change in semantics.

2709 * Syntax.pretty_priority (default 0) configures the required priority

2710 of pretty-printed output and thus affects insertion of parentheses.

2712 * Syntax.default_root (default "any") configures the inner syntax

2713 category (nonterminal symbol) for parsing of terms.

2715 * Former exception Library.UnequalLengths now coincides with

2716 ListPair.UnequalLengths.

2718 * Renamed structure MetaSimplifier to Raw_Simplifier. Note that the

2719 main functionality is provided by structure Simplifier.

2721 * Renamed raw "explode" function to "raw_explode" to emphasize its

2722 meaning. Note that internally to Isabelle, Symbol.explode is used in

2723 almost all situations.

2725 * Discontinued obsolete function sys_error and exception SYS_ERROR.

2726 See implementation manual for further details on exceptions in

2727 Isabelle/ML.

2729 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its

2730 meaning.

2732 * Renamed structure PureThy to Pure_Thy and moved most of its

2733 operations to structure Global_Theory, to emphasize that this is

2734 rarely-used global-only stuff.

2736 * Discontinued Output.debug. Minor INCOMPATIBILITY, use plain writeln

2737 instead (or tracing for high-volume output).

2739 * Configuration option show_question_marks only affects regular pretty

2740 printing of types and terms, not raw Term.string_of_vname.

2742 * ML_Context.thm and ML_Context.thms are no longer pervasive. Rare

2743 INCOMPATIBILITY, superseded by static antiquotations @{thm} and

2744 @{thms} for most purposes.

2746 * ML structure Unsynchronized is never opened, not even in Isar

2747 interaction mode as before. Old Unsynchronized.set etc. have been

2748 discontinued -- use plain := instead. This should be *rare* anyway,

2749 since modern tools always work via official context data, notably

2750 configuration options.

2752 * Parallel and asynchronous execution requires special care concerning

2753 interrupts. Structure Exn provides some convenience functions that

2754 avoid working directly with raw Interrupt. User code must not absorb

2755 interrupts -- intermediate handling (for cleanup etc.) needs to be

2756 followed by re-raising of the original exception. Another common

2757 source of mistakes are "handle _" patterns, which make the meaning of

2758 the program subject to physical effects of the environment.

2762 New in Isabelle2009-2 (June 2010)

2763 ---------------------------------

2765 *** General ***

2767 * Authentic syntax for *all* logical entities (type classes, type

2768 constructors, term constants): provides simple and robust

2769 correspondence between formal entities and concrete syntax. Within

2770 the parse tree / AST representations, "constants" are decorated by

2771 their category (class, type, const) and spelled out explicitly with

2772 their full internal name.

2774 Substantial INCOMPATIBILITY concerning low-level syntax declarations

2775 and translations (translation rules and translation functions in ML).

2776 Some hints on upgrading:

2778 - Many existing uses of 'syntax' and 'translations' can be replaced

2779 by more modern 'type_notation', 'notation' and 'abbreviation',

2780 which are independent of this issue.

2782 - 'translations' require markup within the AST; the term syntax

2783 provides the following special forms:

2785 CONST c -- produces syntax version of constant c from context

2786 XCONST c -- literally c, checked as constant from context

2787 c -- literally c, if declared by 'syntax'

2789 Plain identifiers are treated as AST variables -- occasionally the

2790 system indicates accidental variables via the error "rhs contains

2791 extra variables".

2793 Type classes and type constructors are marked according to their

2794 concrete syntax. Some old translations rules need to be written

2795 for the "type" category, using type constructor application

2796 instead of pseudo-term application of the default category

2797 "logic".

2799 - 'parse_translation' etc. in ML may use the following

2800 antiquotations:

2802 @{class_syntax c} -- type class c within parse tree / AST

2803 @{term_syntax c} -- type constructor c within parse tree / AST

2804 @{const_syntax c} -- ML version of "CONST c" above

2805 @{syntax_const c} -- literally c (checked wrt. 'syntax' declarations)

2807 - Literal types within 'typed_print_translations', i.e. those *not*

2808 represented as pseudo-terms are represented verbatim. Use @{class

2809 c} or @{type_name c} here instead of the above syntax

2810 antiquotations.

2812 Note that old non-authentic syntax was based on unqualified base

2813 names, so all of the above "constant" names would coincide. Recall

2814 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to

2815 diagnose syntax problems.

2817 * Type constructors admit general mixfix syntax, not just infix.

2819 * Concrete syntax may be attached to local entities without a proof

2820 body, too. This works via regular mixfix annotations for 'fix',

2821 'def', 'obtain' etc. or via the explicit 'write' command, which is

2822 similar to the 'notation' command in theory specifications.

2824 * Discontinued unnamed infix syntax (legacy feature for many years) --

2825 need to specify constant name and syntax separately. Internal ML

2826 datatype constructors have been renamed from InfixName to Infix etc.

2827 Minor INCOMPATIBILITY.

2829 * Schematic theorem statements need to be explicitly markup as such,

2830 via commands 'schematic_lemma', 'schematic_theorem',

2831 'schematic_corollary'. Thus the relevance of the proof is made

2832 syntactically clear, which impacts performance in a parallel or

2833 asynchronous interactive environment. Minor INCOMPATIBILITY.

2835 * Use of cumulative prems via "!" in some proof methods has been

2836 discontinued (old legacy feature).

2838 * References 'trace_simp' and 'debug_simp' have been replaced by

2839 configuration options stored in the context. Enabling tracing (the

2840 case of debugging is similar) in proofs works via

2842 using [[trace_simp = true]]

2844 Tracing is then active for all invocations of the simplifier in

2845 subsequent goal refinement steps. Tracing may also still be enabled or

2846 disabled via the ProofGeneral settings menu.

2848 * Separate commands 'hide_class', 'hide_type', 'hide_const',

2849 'hide_fact' replace the former 'hide' KIND command. Minor

2850 INCOMPATIBILITY.

2852 * Improved parallelism of proof term normalization: usedir -p2 -q0 is

2853 more efficient than combinations with -q1 or -q2.

2856 *** Pure ***

2858 * Proofterms record type-class reasoning explicitly, using the

2859 "unconstrain" operation internally. This eliminates all sort

2860 constraints from a theorem and proof, introducing explicit

2861 OFCLASS-premises. On the proof term level, this operation is

2862 automatically applied at theorem boundaries, such that closed proofs

2863 are always free of sort constraints. INCOMPATIBILITY for tools that

2864 inspect proof terms.

2866 * Local theory specifications may depend on extra type variables that

2867 are not present in the result type -- arguments TYPE('a) :: 'a itself

2868 are added internally. For example:

2870 definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"

2872 * Predicates of locales introduced by classes carry a mandatory

2873 "class" prefix. INCOMPATIBILITY.

2875 * Vacuous class specifications observe default sort. INCOMPATIBILITY.

2877 * Old 'axclass' command has been discontinued. INCOMPATIBILITY, use

2878 'class' instead.

2880 * Command 'code_reflect' allows to incorporate generated ML code into

2881 runtime environment; replaces immature code_datatype antiquotation.

2882 INCOMPATIBILITY.

2884 * Code generator: simple concept for abstract datatypes obeying

2885 invariants.

2887 * Code generator: details of internal data cache have no impact on the

2888 user space functionality any longer.

2890 * Methods "unfold_locales" and "intro_locales" ignore non-locale

2891 subgoals. This is more appropriate for interpretations with 'where'.

2892 INCOMPATIBILITY.

2894 * Command 'example_proof' opens an empty proof body. This allows to

2895 experiment with Isar, without producing any persistent result.

2897 * Commands 'type_notation' and 'no_type_notation' declare type syntax

2898 within a local theory context, with explicit checking of the

2899 constructors involved (in contrast to the raw 'syntax' versions).

2901 * Commands 'types' and 'typedecl' now work within a local theory

2902 context -- without introducing dependencies on parameters or

2903 assumptions, which is not possible in Isabelle/Pure.

2905 * Command 'defaultsort' has been renamed to 'default_sort', it works

2906 within a local theory context. Minor INCOMPATIBILITY.

2909 *** HOL ***

2911 * Command 'typedef' now works within a local theory context -- without

2912 introducing dependencies on parameters or assumptions, which is not

2913 possible in Isabelle/Pure/HOL. Note that the logical environment may

2914 contain multiple interpretations of local typedefs (with different

2915 non-emptiness proofs), even in a global theory context.

2917 * New package for quotient types. Commands 'quotient_type' and

2918 'quotient_definition' may be used for defining types and constants by

2919 quotient constructions. An example is the type of integers created by

2920 quotienting pairs of natural numbers:

2922 fun

2923 intrel :: "(nat * nat) => (nat * nat) => bool"

2924 where

2925 "intrel (x, y) (u, v) = (x + v = u + y)"

2927 quotient_type int = "nat * nat" / intrel

2928 by (auto simp add: equivp_def expand_fun_eq)

2930 quotient_definition

2931 "0::int" is "(0::nat, 0::nat)"

2933 The method "lifting" can be used to lift of theorems from the

2934 underlying "raw" type to the quotient type. The example

2935 src/HOL/Quotient_Examples/FSet.thy includes such a quotient

2936 construction and provides a reasoning infrastructure for finite sets.

2938 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid

2939 clash with new theory Quotient in Main HOL.

2941 * Moved the SMT binding into the main HOL session, eliminating

2942 separate HOL-SMT session.

2944 * List membership infix mem operation is only an input abbreviation.

2945 INCOMPATIBILITY.

2947 * Theory Library/Word.thy has been removed. Use library Word/Word.thy

2948 for future developements; former Library/Word.thy is still present in

2949 the AFP entry RSAPPS.

2951 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no

2952 longer shadowed. INCOMPATIBILITY.

2954 * Dropped theorem duplicate comp_arith; use semiring_norm instead.

2955 INCOMPATIBILITY.

2957 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.

2958 INCOMPATIBILITY.

2960 * Dropped normalizing_semiring etc; use the facts in semiring classes

2961 instead. INCOMPATIBILITY.

2963 * Dropped several real-specific versions of lemmas about floor and

2964 ceiling; use the generic lemmas from theory "Archimedean_Field"

2965 instead. INCOMPATIBILITY.

2967 floor_number_of_eq ~> floor_number_of

2968 le_floor_eq_number_of ~> number_of_le_floor

2969 le_floor_eq_zero ~> zero_le_floor

2970 le_floor_eq_one ~> one_le_floor

2971 floor_less_eq_number_of ~> floor_less_number_of

2972 floor_less_eq_zero ~> floor_less_zero

2973 floor_less_eq_one ~> floor_less_one

2974 less_floor_eq_number_of ~> number_of_less_floor

2975 less_floor_eq_zero ~> zero_less_floor

2976 less_floor_eq_one ~> one_less_floor

2977 floor_le_eq_number_of ~> floor_le_number_of

2978 floor_le_eq_zero ~> floor_le_zero

2979 floor_le_eq_one ~> floor_le_one

2980 floor_subtract_number_of ~> floor_diff_number_of

2981 floor_subtract_one ~> floor_diff_one

2982 ceiling_number_of_eq ~> ceiling_number_of

2983 ceiling_le_eq_number_of ~> ceiling_le_number_of

2984 ceiling_le_zero_eq ~> ceiling_le_zero

2985 ceiling_le_eq_one ~> ceiling_le_one

2986 less_ceiling_eq_number_of ~> number_of_less_ceiling

2987 less_ceiling_eq_zero ~> zero_less_ceiling

2988 less_ceiling_eq_one ~> one_less_ceiling

2989 ceiling_less_eq_number_of ~> ceiling_less_number_of

2990 ceiling_less_eq_zero ~> ceiling_less_zero

2991 ceiling_less_eq_one ~> ceiling_less_one

2992 le_ceiling_eq_number_of ~> number_of_le_ceiling

2993 le_ceiling_eq_zero ~> zero_le_ceiling

2994 le_ceiling_eq_one ~> one_le_ceiling

2995 ceiling_subtract_number_of ~> ceiling_diff_number_of

2996 ceiling_subtract_one ~> ceiling_diff_one

2998 * Theory "Finite_Set": various folding_XXX locales facilitate the

2999 application of the various fold combinators on finite sets.

3001 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"

3002 provides abstract red-black tree type which is backed by "RBT_Impl" as

3003 implementation. INCOMPATIBILTY.

3005 * Theory Library/Coinductive_List has been removed -- superseded by

3006 AFP/thys/Coinductive.

3008 * Theory PReal, including the type "preal" and related operations, has

3009 been removed. INCOMPATIBILITY.

3011 * Real: new development using Cauchy Sequences.

3013 * Split off theory "Big_Operators" containing setsum, setprod,

3014 Inf_fin, Sup_fin, Min, Max from theory Finite_Set. INCOMPATIBILITY.

3016 * Theory "Rational" renamed to "Rat", for consistency with "Nat",

3017 "Int" etc. INCOMPATIBILITY.

3019 * Constant Rat.normalize needs to be qualified. INCOMPATIBILITY.

3021 * New set of rules "ac_simps" provides combined assoc / commute

3022 rewrites for all interpretations of the appropriate generic locales.

3024 * Renamed theory "OrderedGroup" to "Groups" and split theory

3025 "Ring_and_Field" into theories "Rings" and "Fields"; for more

3026 appropriate and more consistent names suitable for name prefixes

3027 within the HOL theories. INCOMPATIBILITY.

3029 * Some generic constants have been put to appropriate theories:

3030 - less_eq, less: Orderings

3031 - zero, one, plus, minus, uminus, times, abs, sgn: Groups

3032 - inverse, divide: Rings

3033 INCOMPATIBILITY.

3035 * More consistent naming of type classes involving orderings (and

3036 lattices):

3038 lower_semilattice ~> semilattice_inf

3039 upper_semilattice ~> semilattice_sup

3041 dense_linear_order ~> dense_linorder

3043 pordered_ab_group_add ~> ordered_ab_group_add

3044 pordered_ab_group_add_abs ~> ordered_ab_group_add_abs

3045 pordered_ab_semigroup_add ~> ordered_ab_semigroup_add

3046 pordered_ab_semigroup_add_imp_le ~> ordered_ab_semigroup_add_imp_le

3047 pordered_cancel_ab_semigroup_add ~> ordered_cancel_ab_semigroup_add

3048 pordered_cancel_comm_semiring ~> ordered_cancel_comm_semiring

3049 pordered_cancel_semiring ~> ordered_cancel_semiring

3050 pordered_comm_monoid_add ~> ordered_comm_monoid_add

3051 pordered_comm_ring ~> ordered_comm_ring

3052 pordered_comm_semiring ~> ordered_comm_semiring

3053 pordered_ring ~> ordered_ring

3054 pordered_ring_abs ~> ordered_ring_abs

3055 pordered_semiring ~> ordered_semiring

3057 ordered_ab_group_add ~> linordered_ab_group_add

3058 ordered_ab_semigroup_add ~> linordered_ab_semigroup_add

3059 ordered_cancel_ab_semigroup_add ~> linordered_cancel_ab_semigroup_add

3060 ordered_comm_semiring_strict ~> linordered_comm_semiring_strict

3061 ordered_field ~> linordered_field

3062 ordered_field_no_lb ~> linordered_field_no_lb

3063 ordered_field_no_ub ~> linordered_field_no_ub

3064 ordered_field_dense_linear_order ~> dense_linordered_field

3065 ordered_idom ~> linordered_idom

3066 ordered_ring ~> linordered_ring

3067 ordered_ring_le_cancel_factor ~> linordered_ring_le_cancel_factor

3068 ordered_ring_less_cancel_factor ~> linordered_ring_less_cancel_factor

3069 ordered_ring_strict ~> linordered_ring_strict

3070 ordered_semidom ~> linordered_semidom

3071 ordered_semiring ~> linordered_semiring

3072 ordered_semiring_1 ~> linordered_semiring_1

3073 ordered_semiring_1_strict ~> linordered_semiring_1_strict

3074 ordered_semiring_strict ~> linordered_semiring_strict

3076 The following slightly odd type classes have been moved to a

3077 separate theory Library/Lattice_Algebras:

3079 lordered_ab_group_add ~> lattice_ab_group_add

3080 lordered_ab_group_add_abs ~> lattice_ab_group_add_abs

3081 lordered_ab_group_add_meet ~> semilattice_inf_ab_group_add

3082 lordered_ab_group_add_join ~> semilattice_sup_ab_group_add

3083 lordered_ring ~> lattice_ring

3085 INCOMPATIBILITY.

3087 * Refined field classes:

3088 - classes division_ring_inverse_zero, field_inverse_zero,

3089 linordered_field_inverse_zero include rule inverse 0 = 0 --

3090 subsumes former division_by_zero class;

3091 - numerous lemmas have been ported from field to division_ring.

3092 INCOMPATIBILITY.

3094 * Refined algebra theorem collections:

3095 - dropped theorem group group_simps, use algebra_simps instead;

3096 - dropped theorem group ring_simps, use field_simps instead;

3097 - proper theorem collection field_simps subsumes former theorem

3098 groups field_eq_simps and field_simps;

3099 - dropped lemma eq_minus_self_iff which is a duplicate for

3100 equal_neg_zero.

3101 INCOMPATIBILITY.

3103 * Theory Finite_Set and List: some lemmas have been generalized from

3104 sets to lattices:

3106 fun_left_comm_idem_inter ~> fun_left_comm_idem_inf

3107 fun_left_comm_idem_union ~> fun_left_comm_idem_sup

3108 inter_Inter_fold_inter ~> inf_Inf_fold_inf

3109 union_Union_fold_union ~> sup_Sup_fold_sup

3110 Inter_fold_inter ~> Inf_fold_inf

3111 Union_fold_union ~> Sup_fold_sup

3112 inter_INTER_fold_inter ~> inf_INFI_fold_inf

3113 union_UNION_fold_union ~> sup_SUPR_fold_sup

3114 INTER_fold_inter ~> INFI_fold_inf

3115 UNION_fold_union ~> SUPR_fold_sup

3117 * Theory "Complete_Lattice": lemmas top_def and bot_def have been

3118 replaced by the more convenient lemmas Inf_empty and Sup_empty.

3119 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed

3120 by Inf_insert and Sup_insert. Lemmas Inf_UNIV and Sup_UNIV replace

3121 former Inf_Univ and Sup_Univ. Lemmas inf_top_right and sup_bot_right

3122 subsume inf_top and sup_bot respectively. INCOMPATIBILITY.

3124 * Reorganized theory Multiset: swapped notation of pointwise and

3125 multiset order:

3127 - pointwise ordering is instance of class order with standard syntax

3128 <= and <;

3129 - multiset ordering has syntax <=# and <#; partial order properties

3130 are provided by means of interpretation with prefix

3131 multiset_order;

3132 - less duplication, less historical organization of sections,

3133 conversion from associations lists to multisets, rudimentary code

3134 generation;

3135 - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,

3136 if needed.

3138 Renamed:

3140 multiset_eq_conv_count_eq ~> multiset_ext_iff

3141 multi_count_ext ~> multiset_ext

3142 diff_union_inverse2 ~> diff_union_cancelR

3144 INCOMPATIBILITY.

3146 * Theory Permutation: replaced local "remove" by List.remove1.

3148 * Code generation: ML and OCaml code is decorated with signatures.

3150 * Theory List: added transpose.

3152 * Library/Nat_Bijection.thy is a collection of bijective functions

3153 between nat and other types, which supersedes the older libraries

3154 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy. INCOMPATIBILITY.

3156 Constants:

3157 Nat_Int_Bij.nat2_to_nat ~> prod_encode

3158 Nat_Int_Bij.nat_to_nat2 ~> prod_decode

3159 Nat_Int_Bij.int_to_nat_bij ~> int_encode

3160 Nat_Int_Bij.nat_to_int_bij ~> int_decode

3161 Countable.pair_encode ~> prod_encode

3162 NatIso.prod2nat ~> prod_encode

3163 NatIso.nat2prod ~> prod_decode

3164 NatIso.sum2nat ~> sum_encode

3165 NatIso.nat2sum ~> sum_decode

3166 NatIso.list2nat ~> list_encode

3167 NatIso.nat2list ~> list_decode

3168 NatIso.set2nat ~> set_encode

3169 NatIso.nat2set ~> set_decode

3171 Lemmas:

3172 Nat_Int_Bij.bij_nat_to_int_bij ~> bij_int_decode

3173 Nat_Int_Bij.nat2_to_nat_inj ~> inj_prod_encode

3174 Nat_Int_Bij.nat2_to_nat_surj ~> surj_prod_encode

3175 Nat_Int_Bij.nat_to_nat2_inj ~> inj_prod_decode

3176 Nat_Int_Bij.nat_to_nat2_surj ~> surj_prod_decode

3177 Nat_Int_Bij.i2n_n2i_id ~> int_encode_inverse

3178 Nat_Int_Bij.n2i_i2n_id ~> int_decode_inverse

3179 Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode

3180 Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode

3181 Nat_Int_Bij.inj_nat_to_int_bij ~> inj_int_encode

3182 Nat_Int_Bij.inj_int_to_nat_bij ~> inj_int_decode

3183 Nat_Int_Bij.bij_nat_to_int_bij ~> bij_int_encode

3184 Nat_Int_Bij.bij_int_to_nat_bij ~> bij_int_decode

3186 * Sledgehammer:

3187 - Renamed ATP commands:

3188 atp_info ~> sledgehammer running_atps

3189 atp_kill ~> sledgehammer kill_atps

3190 atp_messages ~> sledgehammer messages

3191 atp_minimize ~> sledgehammer minimize

3192 print_atps ~> sledgehammer available_atps

3193 INCOMPATIBILITY.

3194 - Added user's manual ("isabelle doc sledgehammer").

3195 - Added option syntax and "sledgehammer_params" to customize

3196 Sledgehammer's behavior. See the manual for details.

3197 - Modified the Isar proof reconstruction code so that it produces

3198 direct proofs rather than proofs by contradiction. (This feature

3199 is still experimental.)

3200 - Made Isar proof reconstruction work for SPASS, remote ATPs, and in

3201 full-typed mode.

3202 - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.

3204 * Nitpick:

3205 - Added and implemented "binary_ints" and "bits" options.

3206 - Added "std" option and implemented support for nonstandard models.

3207 - Added and implemented "finitize" option to improve the precision

3208 of infinite datatypes based on a monotonicity analysis.

3209 - Added support for quotient types.

3210 - Added support for "specification" and "ax_specification"

3211 constructs.

3212 - Added support for local definitions (for "function" and

3213 "termination" proofs).

3214 - Added support for term postprocessors.

3215 - Optimized "Multiset.multiset" and "FinFun.finfun".

3216 - Improved efficiency of "destroy_constrs" optimization.

3217 - Fixed soundness bugs related to "destroy_constrs" optimization and

3218 record getters.

3219 - Fixed soundness bug related to higher-order constructors.

3220 - Fixed soundness bug when "full_descrs" is enabled.

3221 - Improved precision of set constructs.

3222 - Added "atoms" option.

3223 - Added cache to speed up repeated Kodkod invocations on the same

3224 problems.

3225 - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and

3226 "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and

3227 "SAT4J_Light". INCOMPATIBILITY.

3228 - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",

3229 "sharing_depth", and "show_skolems" options. INCOMPATIBILITY.

3230 - Removed "nitpick_intro" attribute. INCOMPATIBILITY.

3232 * Method "induct" now takes instantiations of the form t, where t is not

3233 a variable, as a shorthand for "x == t", where x is a fresh variable.

3234 If this is not intended, t has to be enclosed in parentheses.

3235 By default, the equalities generated by definitional instantiations

3236 are pre-simplified, which may cause parameters of inductive cases

3237 to disappear, or may even delete some of the inductive cases.

3238 Use "induct (no_simp)" instead of "induct" to restore the old

3239 behaviour. The (no_simp) option is also understood by the "cases"

3240 and "nominal_induct" methods, which now perform pre-simplification, too.

3241 INCOMPATIBILITY.

3244 *** HOLCF ***

3246 * Variable names in lemmas generated by the domain package have

3247 changed; the naming scheme is now consistent with the HOL datatype

3248 package. Some proof scripts may be affected, INCOMPATIBILITY.

3250 * The domain package no longer defines the function "foo_copy" for

3251 recursive domain "foo". The reach lemma is now stated directly in

3252 terms of "foo_take". Lemmas and proofs that mention "foo_copy" must

3253 be reformulated in terms of "foo_take", INCOMPATIBILITY.

3255 * Most definedness lemmas generated by the domain package (previously

3256 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form

3257 like "foo$x = UU <-> x = UU", which works better as a simp rule.

3258 Proofs that used definedness lemmas as intro rules may break,

3259 potential INCOMPATIBILITY.

3261 * Induction and casedist rules generated by the domain package now

3262 declare proper case_names (one called "bottom", and one named for each

3263 constructor). INCOMPATIBILITY.

3265 * For mutually-recursive domains, separate "reach" and "take_lemma"

3266 rules are generated for each domain, INCOMPATIBILITY.

3268 foo_bar.reach ~> foo.reach bar.reach

3269 foo_bar.take_lemmas ~> foo.take_lemma bar.take_lemma

3271 * Some lemmas generated by the domain package have been renamed for

3272 consistency with the datatype package, INCOMPATIBILITY.

3274 foo.ind ~> foo.induct

3275 foo.finite_ind ~> foo.finite_induct

3276 foo.coind ~> foo.coinduct

3277 foo.casedist ~> foo.exhaust

3278 foo.exhaust ~> foo.nchotomy

3280 * For consistency with other definition packages, the fixrec package

3281 now generates qualified theorem names, INCOMPATIBILITY.

3283 foo_simps ~> foo.simps

3284 foo_unfold ~> foo.unfold

3285 foo_induct ~> foo.induct

3287 * The "fixrec_simp" attribute has been removed. The "fixrec_simp"

3288 method and internal fixrec proofs now use the default simpset instead.

3289 INCOMPATIBILITY.

3291 * The "contlub" predicate has been removed. Proof scripts should use

3292 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.

3294 * The "admw" predicate has been removed, INCOMPATIBILITY.

3296 * The constants cpair, cfst, and csnd have been removed in favor of

3297 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.

3300 *** ML ***

3302 * Antiquotations for basic formal entities:

3304 @{class NAME} -- type class

3305 @{class_syntax NAME} -- syntax representation of the above

3307 @{type_name NAME} -- logical type

3308 @{type_abbrev NAME} -- type abbreviation

3309 @{nonterminal NAME} -- type of concrete syntactic category

3310 @{type_syntax NAME} -- syntax representation of any of the above

3312 @{const_name NAME} -- logical constant (INCOMPATIBILITY)

3313 @{const_abbrev NAME} -- abbreviated constant

3314 @{const_syntax NAME} -- syntax representation of any of the above

3316 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw

3317 syntax constant (cf. 'syntax' command).

3319 * Antiquotation @{make_string} inlines a function to print arbitrary

3320 values similar to the ML toplevel. The result is compiler dependent

3321 and may fall back on "?" in certain situations.

3323 * Diagnostic commands 'ML_val' and 'ML_command' may refer to

3324 antiquotations @{Isar.state} and @{Isar.goal}. This replaces impure

3325 Isar.state() and Isar.goal(), which belong to the old TTY loop and do

3326 not work with the asynchronous Isar document model.

3328 * Configuration options now admit dynamic default values, depending on

3329 the context or even global references.

3331 * SHA1.digest digests strings according to SHA-1 (see RFC 3174). It

3332 uses an efficient external library if available (for Poly/ML).

3334 * Renamed some important ML structures, while keeping the old names

3335 for some time as aliases within the structure Legacy:

3337 OuterKeyword ~> Keyword

3338 OuterLex ~> Token

3339 OuterParse ~> Parse

3340 OuterSyntax ~> Outer_Syntax

3341 PrintMode ~> Print_Mode

3342 SpecParse ~> Parse_Spec

3343 ThyInfo ~> Thy_Info

3344 ThyLoad ~> Thy_Load

3345 ThyOutput ~> Thy_Output

3346 TypeInfer ~> Type_Infer

3348 Note that "open Legacy" simplifies porting of sources, but forgetting

3349 to remove it again will complicate porting again in the future.

3351 * Most operations that refer to a global context are named

3352 accordingly, e.g. Simplifier.global_context or

3353 ProofContext.init_global. There are some situations where a global

3354 context actually works, but under normal circumstances one needs to

3355 pass the proper local context through the code!

3357 * Discontinued old TheoryDataFun with its copy/init operation -- data

3358 needs to be pure. Functor Theory_Data_PP retains the traditional

3359 Pretty.pp argument to merge, which is absent in the standard

3360 Theory_Data version.

3362 * Sorts.certify_sort and derived "cert" operations for types and terms

3363 no longer minimize sorts. Thus certification at the boundary of the

3364 inference kernel becomes invariant under addition of class relations,

3365 which is an important monotonicity principle. Sorts are now minimized

3366 in the syntax layer only, at the boundary between the end-user and the

3367 system. Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort

3368 explicitly in rare situations.

3370 * Renamed old-style Drule.standard to Drule.export_without_context, to

3371 emphasize that this is in no way a standard operation.

3372 INCOMPATIBILITY.

3374 * Subgoal.FOCUS (and variants): resulting goal state is normalized as

3375 usual for resolution. Rare INCOMPATIBILITY.

3377 * Renamed varify/unvarify operations to varify_global/unvarify_global

3378 to emphasize that these only work in a global situation (which is

3379 quite rare).

3381 * Curried take and drop in library.ML; negative length is interpreted

3382 as infinity (as in chop). Subtle INCOMPATIBILITY.

3384 * Proof terms: type substitutions on proof constants now use canonical

3385 order of type variables. INCOMPATIBILITY for tools working with proof

3386 terms.

3388 * Raw axioms/defs may no longer carry sort constraints, and raw defs

3389 may no longer carry premises. User-level specifications are

3390 transformed accordingly by Thm.add_axiom/add_def.

3393 *** System ***

3395 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;

3396 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions. Note that

3397 proof terms are enabled unconditionally in the new HOL-Proofs image.

3399 * Discontinued old ISABELLE and ISATOOL environment settings (legacy

3400 feature since Isabelle2009). Use ISABELLE_PROCESS and ISABELLE_TOOL,

3401 respectively.

3403 * Old lib/scripts/polyml-platform is superseded by the

3404 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit

3405 variant, even on a 64 bit machine. The following example setting

3406 prefers 64 bit if available:

3408 ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"

3410 * The preliminary Isabelle/jEdit application demonstrates the emerging

3411 Isabelle/Scala layer for advanced prover interaction and integration.

3412 See src/Tools/jEdit or "isabelle jedit" provided by the properly built

3413 component.

3415 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono

3416 and Bluesky TeX fonts. It provides the usual Isabelle symbols,

3417 similar to the default assignment of the document preparation system

3418 (cf. isabellesym.sty). The Isabelle/Scala class Isabelle_System

3419 provides some operations for direct access to the font without asking

3420 the user for manual installation.

3424 New in Isabelle2009-1 (December 2009)

3425 -------------------------------------

3427 *** General ***

3429 * Discontinued old form of "escaped symbols" such as \\<forall>. Only

3430 one backslash should be used, even in ML sources.

3433 *** Pure ***

3435 * Locale interpretation propagates mixins along the locale hierarchy.

3436 The currently only available mixins are the equations used to map

3437 local definitions to terms of the target domain of an interpretation.

3439 * Reactivated diagnostic command 'print_interps'. Use "print_interps

3440 loc" to print all interpretations of locale "loc" in the theory.

3441 Interpretations in proofs are not shown.

3443 * Thoroughly revised locales tutorial. New section on conditional

3444 interpretation.

3446 * On instantiation of classes, remaining undefined class parameters

3447 are formally declared. INCOMPATIBILITY.

3450 *** Document preparation ***

3452 * New generalized style concept for printing terms: @{foo (style) ...}

3453 instead of @{foo_style style ...} (old form is still retained for

3454 backward compatibility). Styles can be also applied for

3455 antiquotations prop, term_type and typeof.

3458 *** HOL ***

3460 * New proof method "smt" for a combination of first-order logic with

3461 equality, linear and nonlinear (natural/integer/real) arithmetic, and

3462 fixed-size bitvectors; there is also basic support for higher-order

3463 features (esp. lambda abstractions). It is an incomplete decision

3464 procedure based on external SMT solvers using the oracle mechanism;

3465 for the SMT solver Z3, this method is proof-producing. Certificates

3466 are provided to avoid calling the external solvers solely for

3467 re-checking proofs. Due to a remote SMT service there is no need for

3468 installing SMT solvers locally. See src/HOL/SMT.

3470 * New commands to load and prove verification conditions generated by

3471 the Boogie program verifier or derived systems (e.g. the Verifying C

3472 Compiler (VCC) or Spec#). See src/HOL/Boogie.

3474 * New counterexample generator tool 'nitpick' based on the Kodkod

3475 relational model finder. See src/HOL/Tools/Nitpick and

3476 src/HOL/Nitpick_Examples.

3478 * New commands 'code_pred' and 'values' to invoke the predicate

3479 compiler and to enumerate values of inductive predicates.

3481 * A tabled implementation of the reflexive transitive closure.

3483 * New implementation of quickcheck uses generic code generator;

3484 default generators are provided for all suitable HOL types, records

3485 and datatypes. Old quickcheck can be re-activated importing theory

3486 Library/SML_Quickcheck.

3488 * New testing tool Mirabelle for automated proof tools. Applies

3489 several tools and tactics like sledgehammer, metis, or quickcheck, to

3490 every proof step in a theory. To be used in batch mode via the

3491 "mirabelle" utility.

3493 * New proof method "sos" (sum of squares) for nonlinear real

3494 arithmetic (originally due to John Harison). It requires theory

3495 Library/Sum_Of_Squares. It is not a complete decision procedure but

3496 works well in practice on quantifier-free real arithmetic with +, -,

3497 *, ^, =, <= and <, i.e. boolean combinations of equalities and

3498 inequalities between polynomials. It makes use of external

3499 semidefinite programming solvers. Method "sos" generates a

3500 certificate that can be pasted into the proof thus avoiding the need

3501 to call an external tool every time the proof is checked. See

3502 src/HOL/Library/Sum_Of_Squares.

3504 * New method "linarith" invokes existing linear arithmetic decision

3505 procedure only.

3507 * New command 'atp_minimal' reduces result produced by Sledgehammer.

3509 * New Sledgehammer option "Full Types" in Proof General settings menu.

3510 Causes full type information to be output to the ATPs. This slows

3511 ATPs down considerably but eliminates a source of unsound "proofs"

3512 that fail later.

3514 * New method "metisFT": A version of metis that uses full type

3515 information in order to avoid failures of proof reconstruction.

3517 * New evaluator "approximate" approximates an real valued term using

3518 the same method as the approximation method.

3520 * Method "approximate" now supports arithmetic expressions as

3521 boundaries of intervals and implements interval splitting and Taylor

3522 series expansion.

3524 * ML antiquotation @{code_datatype} inserts definition of a datatype

3525 generated by the code generator; e.g. see src/HOL/Predicate.thy.

3527 * New theory SupInf of the supremum and infimum operators for sets of

3528 reals.

3530 * New theory Probability, which contains a development of measure

3531 theory, eventually leading to Lebesgue integration and probability.

3533 * Extended Multivariate Analysis to include derivation and Brouwer's

3534 fixpoint theorem.

3536 * Reorganization of number theory, INCOMPATIBILITY:

3537 - new number theory development for nat and int, in theories Divides

3538 and GCD as well as in new session Number_Theory

3539 - some constants and facts now suffixed with _nat and _int

3540 accordingly

3541 - former session NumberTheory now named Old_Number_Theory, including

3542 theories Legacy_GCD and Primes (prefer Number_Theory if possible)

3543 - moved theory Pocklington from src/HOL/Library to

3544 src/HOL/Old_Number_Theory

3546 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and

3547 lcm of finite and infinite sets. It is shown that they form a complete

3548 lattice.

3550 * Class semiring_div requires superclass no_zero_divisors and proof of

3551 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,

3552 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been

3553 generalized to class semiring_div, subsuming former theorems

3554 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and

3555 zdiv_zmult_zmult2. div_mult_mult1 is now [simp] by default.

3556 INCOMPATIBILITY.

3558 * Refinements to lattice classes and sets:

3559 - less default intro/elim rules in locale variant, more default

3560 intro/elim rules in class variant: more uniformity

3561 - lemma ge_sup_conv renamed to le_sup_iff, in accordance with

3562 le_inf_iff

3563 - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and

3564 sup_aci)

3565 - renamed ACI to inf_sup_aci

3566 - new class "boolean_algebra"

3567 - class "complete_lattice" moved to separate theory

3568 "Complete_Lattice"; corresponding constants (and abbreviations)

3569 renamed and with authentic syntax:

3570 Set.Inf ~> Complete_Lattice.Inf

3571 Set.Sup ~> Complete_Lattice.Sup

3572 Set.INFI ~> Complete_Lattice.INFI

3573 Set.SUPR ~> Complete_Lattice.SUPR

3574 Set.Inter ~> Complete_Lattice.Inter

3575 Set.Union ~> Complete_Lattice.Union

3576 Set.INTER ~> Complete_Lattice.INTER

3577 Set.UNION ~> Complete_Lattice.UNION

3578 - authentic syntax for

3579 Set.Pow

3580 Set.image

3581 - mere abbreviations:

3582 Set.empty (for bot)

3583 Set.UNIV (for top)

3584 Set.inter (for inf, formerly Set.Int)

3585 Set.union (for sup, formerly Set.Un)

3586 Complete_Lattice.Inter (for Inf)

3587 Complete_Lattice.Union (for Sup)

3588 Complete_Lattice.INTER (for INFI)

3589 Complete_Lattice.UNION (for SUPR)

3590 - object-logic definitions as far as appropriate

3592 INCOMPATIBILITY. Care is required when theorems Int_subset_iff or

3593 Un_subset_iff are explicitly deleted as default simp rules; then also

3594 their lattice counterparts le_inf_iff and le_sup_iff have to be

3595 deleted to achieve the desired effect.

3597 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp

3598 rules by default any longer; the same applies to min_max.inf_absorb1

3599 etc. INCOMPATIBILITY.

3601 * Rules sup_Int_eq and sup_Un_eq are no longer declared as

3602 pred_set_conv by default. INCOMPATIBILITY.

3604 * Power operations on relations and functions are now one dedicated

3605 constant "compow" with infix syntax "^^". Power operation on

3606 multiplicative monoids retains syntax "^" and is now defined generic

3607 in class power. INCOMPATIBILITY.

3609 * Relation composition "R O S" now has a more standard argument order:

3610 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}". INCOMPATIBILITY,

3611 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally

3612 break, since the O_assoc rule was not rewritten like this. Fix using

3613 O_assoc[symmetric]. The same applies to the curried version "R OO S".

3615 * Function "Inv" is renamed to "inv_into" and function "inv" is now an

3616 abbreviation for "inv_into UNIV". Lemmas are renamed accordingly.

3617 INCOMPATIBILITY.

3619 * Most rules produced by inductive and datatype package have mandatory

3620 prefixes. INCOMPATIBILITY.

3622 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by

3623 the attribute of the same name. Each of the theorems in the list

3624 DERIV_intros assumes composition with an additional function and

3625 matches a variable to the derivative, which has to be solved by the

3626 Simplifier. Hence (auto intro!: DERIV_intros) computes the derivative

3627 of most elementary terms. Former Maclauren.DERIV_tac and

3628 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).

3629 INCOMPATIBILITY.

3631 * Code generator attributes follow the usual underscore convention:

3632 code_unfold replaces code unfold

3633 code_post replaces code post

3634 etc.

3635 INCOMPATIBILITY.

3637 * Renamed methods:

3638 sizechange -> size_change

3639 induct_scheme -> induction_schema

3640 INCOMPATIBILITY.

3642 * Discontinued abbreviation "arbitrary" of constant "undefined".

3643 INCOMPATIBILITY, use "undefined" directly.

3645 * Renamed theorems:

3646 Suc_eq_add_numeral_1 -> Suc_eq_plus1

3647 Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left

3648 Suc_plus1 -> Suc_eq_plus1

3649 *anti_sym -> *antisym*

3650 vector_less_eq_def -> vector_le_def

3651 INCOMPATIBILITY.

3653 * Added theorem List.map_map as [simp]. Removed List.map_compose.

3654 INCOMPATIBILITY.

3656 * Removed predicate "M hassize n" (<--> card M = n & finite M).

3657 INCOMPATIBILITY.

3660 *** HOLCF ***

3662 * Theory Representable defines a class "rep" of domains that are

3663 representable (via an ep-pair) in the universal domain type "udom".

3664 Instances are provided for all type constructors defined in HOLCF.

3666 * The 'new_domain' command is a purely definitional version of the

3667 domain package, for representable domains. Syntax is identical to the

3668 old domain package. The 'new_domain' package also supports indirect

3669 recursion using previously-defined type constructors. See

3670 src/HOLCF/ex/New_Domain.thy for examples.

3672 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant

3673 on the left-hand side of an equation, and then performs

3674 simplification. Rewriting is done using rules declared with the

3675 "fixrec_simp" attribute. The "fixrec_simp" method is intended as a

3676 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.

3678 * The pattern-match compiler in 'fixrec' can now handle constructors

3679 with HOL function types. Pattern-match combinators for the Pair

3680 constructor are pre-configured.

3682 * The 'fixrec' package now produces better fixed-point induction rules

3683 for mutually-recursive definitions: Induction rules have conclusions

3684 of the form "P foo bar" instead of "P <foo, bar>".

3686 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has

3687 been renamed to "below". The name "below" now replaces "less" in many

3688 theorem names. (Legacy theorem names using "less" are still supported

3689 as well.)

3691 * The 'fixrec' package now supports "bottom patterns". Bottom

3692 patterns can be used to generate strictness rules, or to make

3693 functions more strict (much like the bang-patterns supported by the

3694 Glasgow Haskell Compiler). See src/HOLCF/ex/Fixrec_ex.thy for

3695 examples.

3698 *** ML ***

3700 * Support for Poly/ML 5.3.0, with improved reporting of compiler

3701 errors and run-time exceptions, including detailed source positions.

3703 * Structure Name_Space (formerly NameSpace) now manages uniquely

3704 identified entries, with some additional information such as source

3705 position, logical grouping etc.

3707 * Theory and context data is now introduced by the simplified and

3708 modernized functors Theory_Data, Proof_Data, Generic_Data. Data needs

3709 to be pure, but the old TheoryDataFun for mutable data (with explicit

3710 copy operation) is still available for some time.

3712 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)

3713 provides a high-level programming interface to synchronized state

3714 variables with atomic update. This works via pure function

3715 application within a critical section -- its runtime should be as

3716 short as possible; beware of deadlocks if critical code is nested,

3717 either directly or indirectly via other synchronized variables!

3719 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)

3720 wraps raw ML references, explicitly indicating their non-thread-safe

3721 behaviour. The Isar toplevel keeps this structure open, to

3722 accommodate Proof General as well as quick and dirty interactive

3723 experiments with references.

3725 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for

3726 parallel tactical reasoning.

3728 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS

3729 are similar to SUBPROOF, but are slightly more flexible: only the

3730 specified parts of the subgoal are imported into the context, and the

3731 body tactic may introduce new subgoals and schematic variables.

3733 * Old tactical METAHYPS, which does not observe the proof context, has

3734 been renamed to Old_Goals.METAHYPS and awaits deletion. Use SUBPROOF

3735 or Subgoal.FOCUS etc.

3737 * Renamed functor TableFun to Table, and GraphFun to Graph. (Since

3738 functors have their own ML name space there is no point to mark them

3739 separately.) Minor INCOMPATIBILITY.

3741 * Renamed NamedThmsFun to Named_Thms. INCOMPATIBILITY.

3743 * Renamed several structures FooBar to Foo_Bar. Occasional,

3744 INCOMPATIBILITY.

3746 * Operations of structure Skip_Proof no longer require quick_and_dirty

3747 mode, which avoids critical setmp.

3749 * Eliminated old Attrib.add_attributes, Method.add_methods and related

3750 combinators for "args". INCOMPATIBILITY, need to use simplified

3751 Attrib/Method.setup introduced in Isabelle2009.

3753 * Proper context for simpset_of, claset_of, clasimpset_of. May fall

3754 back on global_simpset_of, global_claset_of, global_clasimpset_of as

3755 last resort. INCOMPATIBILITY.

3757 * Display.pretty_thm now requires a proper context (cf. former

3758 ProofContext.pretty_thm). May fall back on Display.pretty_thm_global

3759 or even Display.pretty_thm_without_context as last resort.

3760 INCOMPATIBILITY.

3762 * Discontinued Display.pretty_ctyp/cterm etc. INCOMPATIBILITY, use

3763 Syntax.pretty_typ/term directly, preferably with proper context

3764 instead of global theory.

3767 *** System ***

3769 * Further fine tuning of parallel proof checking, scales up to 8 cores

3770 (max. speedup factor 5.0). See also Goal.parallel_proofs in ML and

3771 usedir option -q.

3773 * Support for additional "Isabelle components" via etc/components, see

3774 also the system manual.

3776 * The isabelle makeall tool now operates on all components with

3777 IsaMakefile, not just hardwired "logics".

3779 * Removed "compress" option from isabelle-process and isabelle usedir;

3780 this is always enabled.

3782 * Discontinued support for Poly/ML 4.x versions.

3784 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'

3785 on a given logic image. This requires the lighttpd webserver and is

3786 currently supported on Linux only.

3790 New in Isabelle2009 (April 2009)

3791 --------------------------------

3793 *** General ***

3795 * Simplified main Isabelle executables, with less surprises on

3796 case-insensitive file-systems (such as Mac OS).

3798 - The main Isabelle tool wrapper is now called "isabelle" instead of

3799 "isatool."

3801 - The former "isabelle" alias for "isabelle-process" has been

3802 removed (should rarely occur to regular users).

3804 - The former "isabelle-interface" and its alias "Isabelle" have been

3805 removed (interfaces are now regular Isabelle tools).

3807 Within scripts and make files, the Isabelle environment variables

3808 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,

3809 respectively. (The latter are still available as legacy feature.)

3811 The old isabelle-interface wrapper could react in confusing ways if

3812 the interface was uninstalled or changed otherwise. Individual

3813 interface tool configuration is now more explicit, see also the

3814 Isabelle system manual. In particular, Proof General is now available

3815 via "isabelle emacs".

3817 INCOMPATIBILITY, need to adapt derivative scripts. Users may need to

3818 purge installed copies of Isabelle executables and re-run "isabelle

3819 install -p ...", or use symlinks.

3821 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the

3822 old ~/isabelle, which was slightly non-standard and apt to cause

3823 surprises on case-insensitive file-systems (such as Mac OS).

3825 INCOMPATIBILITY, need to move existing ~/isabelle/etc,

3826 ~/isabelle/heaps, ~/isabelle/browser_info to the new place. Special

3827 care is required when using older releases of Isabelle. Note that

3828 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any

3829 Isabelle distribution, in order to use the new ~/.isabelle uniformly.

3831 * Proofs of fully specified statements are run in parallel on

3832 multi-core systems. A speedup factor of 2.5 to 3.2 can be expected on

3833 a regular 4-core machine, if the initial heap space is made reasonably

3834 large (cf. Poly/ML option -H). (Requires Poly/ML 5.2.1 or later.)

3836 * The main reference manuals ("isar-ref", "implementation", and

3837 "system") have been updated and extended. Formally checked references

3838 as hyperlinks are now available uniformly.

3841 *** Pure ***

3843 * Complete re-implementation of locales. INCOMPATIBILITY in several

3844 respects. The most important changes are listed below. See the

3845 Tutorial on Locales ("locales" manual) for details.

3847 - In locale expressions, instantiation replaces renaming. Parameters

3848 must be declared in a for clause. To aid compatibility with previous

3849 parameter inheritance, in locale declarations, parameters that are not

3850 'touched' (instantiation position "_" or omitted) are implicitly added

3851 with their syntax at the beginning of the for clause.

3853 - Syntax from abbreviations and definitions in locales is available in

3854 locale expressions and context elements. The latter is particularly

3855 useful in locale declarations.

3857 - More flexible mechanisms to qualify names generated by locale

3858 expressions. Qualifiers (prefixes) may be specified in locale

3859 expressions, and can be marked as mandatory (syntax: "name!:") or

3860 optional (syntax "name?:"). The default depends for plain "name:"

3861 depends on the situation where a locale expression is used: in

3862 commands 'locale' and 'sublocale' prefixes are optional, in

3863 'interpretation' and 'interpret' prefixes are mandatory. The old

3864 implicit qualifiers derived from the parameter names of a locale are

3865 no longer generated.

3867 - Command "sublocale l < e" replaces "interpretation l < e". The

3868 instantiation clause in "interpretation" and "interpret" (square

3869 brackets) is no longer available. Use locale expressions.

3871 - When converting proof scripts, mandatory qualifiers in

3872 'interpretation' and 'interpret' should be retained by default, even

3873 if this is an INCOMPATIBILITY compared to former behavior. In the

3874 worst case, use the "name?:" form for non-mandatory ones. Qualifiers

3875 in locale expressions range over a single locale instance only.

3877 - Dropped locale element "includes". This is a major INCOMPATIBILITY.

3878 In existing theorem specifications replace the includes element by the

3879 respective context elements of the included locale, omitting those

3880 that are already present in the theorem specification. Multiple

3881 assume elements of a locale should be replaced by a single one

3882 involving the locale predicate. In the proof body, declarations (most

3883 notably theorems) may be regained by interpreting the respective

3884 locales in the proof context as required (command "interpret").

3886 If using "includes" in replacement of a target solely because the

3887 parameter types in the theorem are not as general as in the target,

3888 consider declaring a new locale with additional type constraints on

3889 the parameters (context element "constrains").

3891 - Discontinued "locale (open)". INCOMPATIBILITY.

3893 - Locale interpretation commands no longer attempt to simplify goal.

3894 INCOMPATIBILITY: in rare situations the generated goal differs. Use

3895 methods intro_locales and unfold_locales to clarify.

3897 - Locale interpretation commands no longer accept interpretation

3898 attributes. INCOMPATIBILITY.

3900 * Class declaration: so-called "base sort" must not be given in import

3901 list any longer, but is inferred from the specification. Particularly

3902 in HOL, write

3904 class foo = ...

3906 instead of

3908 class foo = type + ...

3910 * Class target: global versions of theorems stemming do not carry a

3911 parameter prefix any longer. INCOMPATIBILITY.

3913 * Class 'instance' command no longer accepts attached definitions.

3914 INCOMPATIBILITY, use proper 'instantiation' target instead.

3916 * Recovered hiding of consts, which was accidentally broken in

3917 Isabelle2007. Potential INCOMPATIBILITY, ``hide const c'' really

3918 makes c inaccessible; consider using ``hide (open) const c'' instead.

3920 * Slightly more coherent Pure syntax, with updated documentation in

3921 isar-ref manual. Removed locales meta_term_syntax and

3922 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,

3923 INCOMPATIBILITY in rare situations. Note that &&& should not be used

3924 directly in regular applications.

3926 * There is a new syntactic category "float_const" for signed decimal

3927 fractions (e.g. 123.45 or -123.45).

3929 * Removed exotic 'token_translation' command. INCOMPATIBILITY, use ML

3930 interface with 'setup' command instead.

3932 * Command 'local_setup' is similar to 'setup', but operates on a local

3933 theory context.

3935 * The 'axiomatization' command now only works within a global theory

3936 context. INCOMPATIBILITY.

3938 * Goal-directed proof now enforces strict proof irrelevance wrt. sort

3939 hypotheses. Sorts required in the course of reasoning need to be

3940 covered by the constraints in the initial statement, completed by the

3941 type instance information of the background theory. Non-trivial sort

3942 hypotheses, which rarely occur in practice, may be specified via

3943 vacuous propositions of the form SORT_CONSTRAINT('a::c). For example:

3945 lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...

3947 The result contains an implicit sort hypotheses as before --

3948 SORT_CONSTRAINT premises are eliminated as part of the canonical rule

3949 normalization.

3951 * Generalized Isar history, with support for linear undo, direct state

3952 addressing etc.

3954 * Changed defaults for unify configuration options:

3956 unify_trace_bound = 50 (formerly 25)

3957 unify_search_bound = 60 (formerly 30)

3959 * Different bookkeeping for code equations (INCOMPATIBILITY):

3961 a) On theory merge, the last set of code equations for a particular

3962 constant is taken (in accordance with the policy applied by other

3963 parts of the code generator framework).

3965 b) Code equations stemming from explicit declarations (e.g. code

3966 attribute) gain priority over default code equations stemming

3967 from definition, primrec, fun etc.

3969 * Keyword 'code_exception' now named 'code_abort'. INCOMPATIBILITY.

3971 * Unified theorem tables for both code generators. Thus [code

3972 func] has disappeared and only [code] remains. INCOMPATIBILITY.

3974 * Command 'find_consts' searches for constants based on type and name

3975 patterns, e.g.

3977 find_consts "_ => bool"

3979 By default, matching is against subtypes, but it may be restricted to

3980 the whole type. Searching by name is possible. Multiple queries are

3981 conjunctive and queries may be negated by prefixing them with a

3982 hyphen:

3984 find_consts strict: "_ => bool" name: "Int" -"int => int"

3986 * New 'find_theorems' criterion "solves" matches theorems that

3987 directly solve the current goal (modulo higher-order unification).

3989 * Auto solve feature for main theorem statements: whenever a new goal

3990 is stated, "find_theorems solves" is called; any theorems that could

3991 solve the lemma directly are listed as part of the goal state.

3992 Cf. associated options in Proof General Isabelle settings menu,

3993 enabled by default, with reasonable timeout for pathological cases of

3994 higher-order unification.

3997 *** Document preparation ***

3999 * Antiquotation @{lemma} now imitates a regular terminal proof,

4000 demanding keyword 'by' and supporting the full method expression

4001 syntax just like the Isar command 'by'.

4004 *** HOL ***

4006 * Integrated main parts of former image HOL-Complex with HOL. Entry

4007 points Main and Complex_Main remain as before.

4009 * Logic image HOL-Plain provides a minimal HOL with the most important

4010 tools available (inductive, datatype, primrec, ...). This facilitates

4011 experimentation and tool development. Note that user applications

4012 (and library theories) should never refer to anything below theory

4013 Main, as before.

4015 * Logic image HOL-Main stops at theory Main, and thus facilitates

4016 experimentation due to shorter build times.

4018 * Logic image HOL-NSA contains theories of nonstandard analysis which

4019 were previously part of former HOL-Complex. Entry point Hyperreal

4020 remains valid, but theories formerly using Complex_Main should now use

4021 new entry point Hypercomplex.

4023 * Generic ATP manager for Sledgehammer, based on ML threads instead of

4024 Posix processes. Avoids potentially expensive forking of the ML

4025 process. New thread-based implementation also works on non-Unix

4026 platforms (Cygwin). Provers are no longer hardwired, but defined

4027 within the theory via plain ML wrapper functions. Basic Sledgehammer

4028 commands are covered in the isar-ref manual.

4030 * Wrapper scripts for remote SystemOnTPTP service allows to use

4031 sledgehammer without local ATP installation (Vampire etc.). Other

4032 provers may be included via suitable ML wrappers, see also

4033 src/HOL/ATP_Linkup.thy.

4035 * ATP selection (E/Vampire/Spass) is now via Proof General's settings

4036 menu.

4038 * The metis method no longer fails because the theorem is too trivial

4039 (contains the empty clause).

4041 * The metis method now fails in the usual manner, rather than raising

4042 an exception, if it determines that it cannot prove the theorem.

4044 * Method "coherent" implements a prover for coherent logic (see also

4045 src/Tools/coherent.ML).

4047 * Constants "undefined" and "default" replace "arbitrary". Usually

4048 "undefined" is the right choice to replace "arbitrary", though

4049 logically there is no difference. INCOMPATIBILITY.

4051 * Command "value" now integrates different evaluation mechanisms. The

4052 result of the first successful evaluation mechanism is printed. In

4053 square brackets a particular named evaluation mechanisms may be

4054 specified (currently, [SML], [code] or [nbe]). See further

4055 src/HOL/ex/Eval_Examples.thy.

4057 * Normalization by evaluation now allows non-leftlinear equations.

4058 Declare with attribute [code nbe].

4060 * Methods "case_tac" and "induct_tac" now refer to the very same rules

4061 as the structured Isar versions "cases" and "induct", cf. the

4062 corresponding "cases" and "induct" attributes. Mutual induction rules

4063 are now presented as a list of individual projections

4064 (e.g. foo_bar.inducts for types foo and bar); the old format with

4065 explicit HOL conjunction is no longer supported. INCOMPATIBILITY, in

4066 rare situations a different rule is selected --- notably nested tuple

4067 elimination instead of former prod.exhaust: use explicit (case_tac t

4068 rule: prod.exhaust) here.

4070 * Attributes "cases", "induct", "coinduct" support "del" option.

4072 * Removed fact "case_split_thm", which duplicates "case_split".

4074 * The option datatype has been moved to a new theory Option. Renamed

4075 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.

4077 * New predicate "strict_mono" classifies strict functions on partial

4078 orders. With strict functions on linear orders, reasoning about

4079 (in)equalities is facilitated by theorems "strict_mono_eq",

4080 "strict_mono_less_eq" and "strict_mono_less".

4082 * Some set operations are now proper qualified constants with

4083 authentic syntax. INCOMPATIBILITY:

4085 op Int ~> Set.Int

4086 op Un ~> Set.Un

4087 INTER ~> Set.INTER

4088 UNION ~> Set.UNION

4089 Inter ~> Set.Inter

4090 Union ~> Set.Union

4091 {} ~> Set.empty

4092 UNIV ~> Set.UNIV

4094 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in

4095 theory Set.

4097 * Auxiliary class "itself" has disappeared -- classes without any

4098 parameter are treated as expected by the 'class' command.

4100 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.

4102 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,

4103 Dense_Linear_Order) are now in directory HOL/Decision_Procs.

4105 * Theory src/HOL/Decision_Procs/Approximation provides the new proof

4106 method "approximation". It proves formulas on real values by using

4107 interval arithmetic. In the formulas are also the transcendental

4108 functions sin, cos, tan, atan, ln, exp and the constant pi are

4109 allowed. For examples see

4110 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.

4112 * Theory "Reflection" now resides in HOL/Library.

4114 * Entry point to Word library now simply named "Word".

4115 INCOMPATIBILITY.

4117 * Made source layout more coherent with logical distribution

4118 structure:

4120 src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy

4121 src/HOL/Library/Code_Message.thy ~> src/HOL/

4122 src/HOL/Library/GCD.thy ~> src/HOL/

4123 src/HOL/Library/Order_Relation.thy ~> src/HOL/

4124 src/HOL/Library/Parity.thy ~> src/HOL/

4125 src/HOL/Library/Univ_Poly.thy ~> src/HOL/

4126 src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/

4127 src/HOL/Real/Lubs.thy ~> src/HOL/

4128 src/HOL/Real/PReal.thy ~> src/HOL/

4129 src/HOL/Real/Rational.thy ~> src/HOL/

4130 src/HOL/Real/RComplete.thy ~> src/HOL/

4131 src/HOL/Real/RealDef.thy ~> src/HOL/

4132 src/HOL/Real/RealPow.thy ~> src/HOL/

4133 src/HOL/Real/Real.thy ~> src/HOL/

4134 src/HOL/Complex/Complex_Main.thy ~> src/HOL/

4135 src/HOL/Complex/Complex.thy ~> src/HOL/

4136 src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/

4137 src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/

4138 src/HOL/Hyperreal/Deriv.thy ~> src/HOL/

4139 src/HOL/Hyperreal/Fact.thy ~> src/HOL/

4140 src/HOL/Hyperreal/Integration.thy ~> src/HOL/

4141 src/HOL/Hyperreal/Lim.thy ~> src/HOL/

4142 src/HOL/Hyperreal/Ln.thy ~> src/HOL/

4143 src/HOL/Hyperreal/Log.thy ~> src/HOL/

4144 src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/

4145 src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/

4146 src/HOL/Hyperreal/Series.thy ~> src/HOL/

4147 src/HOL/Hyperreal/SEQ.thy ~> src/HOL/

4148 src/HOL/Hyperreal/Taylor.thy ~> src/HOL/

4149 src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/

4150 src/HOL/Real/Float ~> src/HOL/Library/

4151 src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach

4152 src/HOL/Real/RealVector.thy ~> src/HOL/

4154 src/HOL/arith_data.ML ~> src/HOL/Tools

4155 src/HOL/hologic.ML ~> src/HOL/Tools

4156 src/HOL/simpdata.ML ~> src/HOL/Tools

4157 src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML

4158 src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools

4159 src/HOL/nat_simprocs.ML ~> src/HOL/Tools

4160 src/HOL/Real/float_arith.ML ~> src/HOL/Tools

4161 src/HOL/Real/float_syntax.ML ~> src/HOL/Tools

4162 src/HOL/Real/rat_arith.ML ~> src/HOL/Tools

4163 src/HOL/Real/real_arith.ML ~> src/HOL/Tools

4165 src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL

4166 src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL

4167 src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL

4168 src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL

4169 src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL

4170 src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL

4172 * If methods "eval" and "evaluation" encounter a structured proof

4173 state with !!/==>, only the conclusion is evaluated to True (if

4174 possible), avoiding strange error messages.

4176 * Method "sizechange" automates termination proofs using (a

4177 modification of) the size-change principle. Requires SAT solver. See

4178 src/HOL/ex/Termination.thy for examples.

4180 * Simplifier: simproc for let expressions now unfolds if bound

4181 variable occurs at most once in let expression body. INCOMPATIBILITY.

4183 * Method "arith": Linear arithmetic now ignores all inequalities when

4184 fast_arith_neq_limit is exceeded, instead of giving up entirely.

4186 * New attribute "arith" for facts that should always be used

4187 automatically by arithmetic. It is intended to be used locally in

4188 proofs, e.g.

4190 assumes [arith]: "x > 0"

4192 Global usage is discouraged because of possible performance impact.

4194 * New classes "top" and "bot" with corresponding operations "top" and

4195 "bot" in theory Orderings; instantiation of class "complete_lattice"

4196 requires instantiation of classes "top" and "bot". INCOMPATIBILITY.

4198 * Changed definition lemma "less_fun_def" in order to provide an

4199 instance for preorders on functions; use lemma "less_le" instead.

4200 INCOMPATIBILITY.

4202 * Theory Orderings: class "wellorder" moved here, with explicit

4203 induction rule "less_induct" as assumption. For instantiation of

4204 "wellorder" by means of predicate "wf", use rule wf_wellorderI.

4205 INCOMPATIBILITY.

4207 * Theory Orderings: added class "preorder" as superclass of "order".

4208 INCOMPATIBILITY: Instantiation proofs for order, linorder

4209 etc. slightly changed. Some theorems named order_class.* now named

4210 preorder_class.*.

4212 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,

4213 "diag" to "Id_on".

4215 * Theory Finite_Set: added a new fold combinator of type

4217 ('a => 'b => 'b) => 'b => 'a set => 'b

4219 Occasionally this is more convenient than the old fold combinator

4220 which is now defined in terms of the new one and renamed to

4221 fold_image.

4223 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"

4224 and "ring_simps" have been replaced by "algebra_simps" (which can be

4225 extended with further lemmas!). At the moment both still exist but

4226 the former will disappear at some point.

4228 * Theory Power: Lemma power_Suc is now declared as a simp rule in

4229 class recpower. Type-specific simp rules for various recpower types

4230 have been removed. INCOMPATIBILITY, rename old lemmas as follows:

4232 rat_power_0 -> power_0

4233 rat_power_Suc -> power_Suc

4234 realpow_0 -> power_0

4235 realpow_Suc -> power_Suc

4236 complexpow_0 -> power_0

4237 complexpow_Suc -> power_Suc

4238 power_poly_0 -> power_0

4239 power_poly_Suc -> power_Suc

4241 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been

4242 moved to separate class dvd in Ring_and_Field; a couple of lemmas on

4243 dvd has been generalized to class comm_semiring_1. Likewise a bunch

4244 of lemmas from Divides has been generalized from nat to class

4245 semiring_div. INCOMPATIBILITY. This involves the following theorem

4246 renames resulting from duplicate elimination:

4248 dvd_def_mod ~> dvd_eq_mod_eq_0

4249 zero_dvd_iff ~> dvd_0_left_iff

4250 dvd_0 ~> dvd_0_right

4251 DIVISION_BY_ZERO_DIV ~> div_by_0

4252 DIVISION_BY_ZERO_MOD ~> mod_by_0

4253 mult_div ~> div_mult_self2_is_id

4254 mult_mod ~> mod_mult_self2_is_0

4256 * Theory IntDiv: removed many lemmas that are instances of class-based

4257 generalizations (from Divides and Ring_and_Field). INCOMPATIBILITY,

4258 rename old lemmas as follows:

4260 dvd_diff -> nat_dvd_diff

4261 dvd_zminus_iff -> dvd_minus_iff

4262 mod_add1_eq -> mod_add_eq

4263 mod_mult1_eq -> mod_mult_right_eq

4264 mod_mult1_eq' -> mod_mult_left_eq

4265 mod_mult_distrib_mod -> mod_mult_eq

4266 nat_mod_add_left_eq -> mod_add_left_eq

4267 nat_mod_add_right_eq -> mod_add_right_eq

4268 nat_mod_div_trivial -> mod_div_trivial

4269 nat_mod_mod_trivial -> mod_mod_trivial

4270 zdiv_zadd_self1 -> div_add_self1

4271 zdiv_zadd_self2 -> div_add_self2

4272 zdiv_zmult_self1 -> div_mult_self2_is_id

4273 zdiv_zmult_self2 -> div_mult_self1_is_id

4274 zdvd_triv_left -> dvd_triv_left

4275 zdvd_triv_right -> dvd_triv_right

4276 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left

4277 zmod_eq0_zdvd_iff -> dvd_eq_mod_eq_0[symmetric]

4278 zmod_zadd_left_eq -> mod_add_left_eq

4279 zmod_zadd_right_eq -> mod_add_right_eq

4280 zmod_zadd_self1 -> mod_add_self1

4281 zmod_zadd_self2 -> mod_add_self2

4282 zmod_zadd1_eq -> mod_add_eq

4283 zmod_zdiff1_eq -> mod_diff_eq

4284 zmod_zdvd_zmod -> mod_mod_cancel

4285 zmod_zmod_cancel -> mod_mod_cancel

4286 zmod_zmult_self1 -> mod_mult_self2_is_0

4287 zmod_zmult_self2 -> mod_mult_self1_is_0

4288 zmod_1 -> mod_by_1

4289 zdiv_1 -> div_by_1

4290 zdvd_abs1 -> abs_dvd_iff

4291 zdvd_abs2 -> dvd_abs_iff

4292 zdvd_refl -> dvd_refl

4293 zdvd_trans -> dvd_trans

4294 zdvd_zadd -> dvd_add

4295 zdvd_zdiff -> dvd_diff

4296 zdvd_zminus_iff -> dvd_minus_iff

4297 zdvd_zminus2_iff -> minus_dvd_iff

4298 zdvd_zmultD -> dvd_mult_right

4299 zdvd_zmultD2 -> dvd_mult_left

4300 zdvd_zmult_mono -> mult_dvd_mono

4301 zdvd_0_right -> dvd_0_right

4302 zdvd_0_left -> dvd_0_left_iff

4303 zdvd_1_left -> one_dvd

4304 zminus_dvd_iff -> minus_dvd_iff

4306 * Theory Rational: 'Fract k 0' now equals '0'. INCOMPATIBILITY.

4308 * The real numbers offer decimal input syntax: 12.34 is translated

4309 into 1234/10^2. This translation is not reversed upon output.

4311 * Theory Library/Polynomial defines an abstract type 'a poly of

4312 univariate polynomials with coefficients of type 'a. In addition to

4313 the standard ring operations, it also supports div and mod. Code

4314 generation is also supported, using list-style constructors.

4316 * Theory Library/Inner_Product defines a class of real_inner for real

4317 inner product spaces, with an overloaded operation inner :: 'a => 'a

4318 => real. Class real_inner is a subclass of real_normed_vector from

4319 theory RealVector.

4321 * Theory Library/Product_Vector provides instances for the product

4322 type 'a * 'b of several classes from RealVector and Inner_Product.

4323 Definitions of addition, subtraction, scalar multiplication, norms,

4324 and inner products are included.

4326 * Theory Library/Bit defines the field "bit" of integers modulo 2. In

4327 addition to the field operations, numerals and case syntax are also

4328 supported.

4330 * Theory Library/Diagonalize provides constructive version of Cantor's

4331 first diagonalization argument.

4333 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,

4334 zlcm (for int); carried together from various gcd/lcm developements in

4335 the HOL Distribution. Constants zgcd and zlcm replace former igcd and

4336 ilcm; corresponding theorems renamed accordingly. INCOMPATIBILITY,

4337 may recover tupled syntax as follows:

4339 hide (open) const gcd

4340 abbreviation gcd where

4341 "gcd == (%(a, b). GCD.gcd a b)"

4342 notation (output)

4343 GCD.gcd ("gcd '(_, _')")

4345 The same works for lcm, zgcd, zlcm.

4347 * Theory Library/Nat_Infinity: added addition, numeral syntax and more

4348 instantiations for algebraic structures. Removed some duplicate

4349 theorems. Changes in simp rules. INCOMPATIBILITY.

4351 * ML antiquotation @{code} takes a constant as argument and generates

4352 corresponding code in background and inserts name of the corresponding

4353 resulting ML value/function/datatype constructor binding in place.

4354 All occurrences of @{code} with a single ML block are generated

4355 simultaneously. Provides a generic and safe interface for

4356 instrumentalizing code generation. See

4357 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.

4358 In future you ought to refrain from ad-hoc compiling generated SML

4359 code on the ML toplevel. Note that (for technical reasons) @{code}

4360 cannot refer to constants for which user-defined serializations are

4361 set. Refer to the corresponding ML counterpart directly in that

4362 cases.

4364 * Command 'rep_datatype': instead of theorem names the command now

4365 takes a list of terms denoting the constructors of the type to be

4366 represented as datatype. The characteristic theorems have to be

4367 proven. INCOMPATIBILITY. Also observe that the following theorems

4368 have disappeared in favour of existing ones:

4370 unit_induct ~> unit.induct

4371 prod_induct ~> prod.induct

4372 sum_induct ~> sum.induct

4373 Suc_Suc_eq ~> nat.inject

4374 Suc_not_Zero Zero_not_Suc ~> nat.distinct

4377 *** HOL-Algebra ***

4379 * New locales for orders and lattices where the equivalence relation

4380 is not restricted to equality. INCOMPATIBILITY: all order and lattice

4381 locales use a record structure with field eq for the equivalence.

4383 * New theory of factorial domains.

4385 * Units_l_inv and Units_r_inv are now simp rules by default.

4386 INCOMPATIBILITY. Simplifier proof that require deletion of l_inv

4387 and/or r_inv will now also require deletion of these lemmas.

4389 * Renamed the following theorems, INCOMPATIBILITY:

4391 UpperD ~> Upper_memD

4392 LowerD ~> Lower_memD

4393 least_carrier ~> least_closed

4394 greatest_carrier ~> greatest_closed

4395 greatest_Lower_above ~> greatest_Lower_below

4396 one_zero ~> carrier_one_zero

4397 one_not_zero ~> carrier_one_not_zero (collision with assumption)

4400 *** HOL-Nominal ***

4402 * Nominal datatypes can now contain type-variables.

4404 * Commands 'nominal_inductive' and 'equivariance' work with local

4405 theory targets.

4407 * Nominal primrec can now works with local theory targets and its

4408 specification syntax now conforms to the general format as seen in

4409 'inductive' etc.

4411 * Method "perm_simp" honours the standard simplifier attributes

4412 (no_asm), (no_asm_use) etc.

4414 * The new predicate #* is defined like freshness, except that on the

4415 left hand side can be a set or list of atoms.

4417 * Experimental command 'nominal_inductive2' derives strong induction

4418 principles for inductive definitions. In contrast to

4419 'nominal_inductive', which can only deal with a fixed number of

4420 binders, it can deal with arbitrary expressions standing for sets of

4421 atoms to be avoided. The only inductive definition we have at the

4422 moment that needs this generalisation is the typing rule for Lets in

4423 the algorithm W:

4425 Gamma |- t1 : T1 (x,close Gamma T1)::Gamma |- t2 : T2 x#Gamma

4426 -----------------------------------------------------------------

4427 Gamma |- Let x be t1 in t2 : T2

4429 In this rule one wants to avoid all the binders that are introduced by

4430 "close Gamma T1". We are looking for other examples where this

4431 feature might be useful. Please let us know.

4434 *** HOLCF ***

4436 * Reimplemented the simplification procedure for proving continuity

4437 subgoals. The new simproc is extensible; users can declare additional

4438 continuity introduction rules with the attribute [cont2cont].

4440 * The continuity simproc now uses a different introduction rule for

4441 solving continuity subgoals on terms with lambda abstractions. In

4442 some rare cases the new simproc may fail to solve subgoals that the

4443 old one could solve, and "simp add: cont2cont_LAM" may be necessary.

4444 Potential INCOMPATIBILITY.

4446 * Command 'fixrec': specification syntax now conforms to the general

4447 format as seen in 'inductive' etc. See src/HOLCF/ex/Fixrec_ex.thy for

4448 examples. INCOMPATIBILITY.

4451 *** ZF ***

4453 * Proof of Zorn's Lemma for partial orders.

4456 *** ML ***

4458 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for

4459 Poly/ML 5.2.1 or later. Important note: the TimeLimit facility

4460 depends on multithreading, so timouts will not work before Poly/ML

4461 5.2.1!

4463 * High-level support for concurrent ML programming, see

4464 src/Pure/Cuncurrent. The data-oriented model of "future values" is

4465 particularly convenient to organize independent functional

4466 computations. The concept of "synchronized variables" provides a

4467 higher-order interface for components with shared state, avoiding the

4468 delicate details of mutexes and condition variables. (Requires

4469 Poly/ML 5.2.1 or later.)

4471 * ML bindings produced via Isar commands are stored within the Isar

4472 context (theory or proof). Consequently, commands like 'use' and 'ML'

4473 become thread-safe and work with undo as expected (concerning

4474 top-level bindings, not side-effects on global references).

4475 INCOMPATIBILITY, need to provide proper Isar context when invoking the

4476 compiler at runtime; really global bindings need to be given outside a

4477 theory. (Requires Poly/ML 5.2 or later.)

4479 * Command 'ML_prf' is analogous to 'ML' but works within a proof

4480 context. Top-level ML bindings are stored within the proof context in

4481 a purely sequential fashion, disregarding the nested proof structure.

4482 ML bindings introduced by 'ML_prf' are discarded at the end of the

4483 proof. (Requires Poly/ML 5.2 or later.)

4485 * Simplified ML attribute and method setup, cf. functions Attrib.setup

4486 and Method.setup, as well as Isar commands 'attribute_setup' and

4487 'method_setup'. INCOMPATIBILITY for 'method_setup', need to simplify

4488 existing code accordingly, or use plain 'setup' together with old

4489 Method.add_method.

4491 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm

4492 to 'a -> thm, while results are always tagged with an authentic oracle

4493 name. The Isar command 'oracle' is now polymorphic, no argument type

4494 is specified. INCOMPATIBILITY, need to simplify existing oracle code

4495 accordingly. Note that extra performance may be gained by producing

4496 the cterm carefully, avoiding slow Thm.cterm_of.

4498 * Simplified interface for defining document antiquotations via

4499 ThyOutput.antiquotation, ThyOutput.output, and optionally

4500 ThyOutput.maybe_pretty_source. INCOMPATIBILITY, need to simplify user

4501 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common

4502 examples.

4504 * More systematic treatment of long names, abstract name bindings, and

4505 name space operations. Basic operations on qualified names have been

4506 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,

4507 Long_Name.append. Old type bstring has been mostly replaced by

4508 abstract type binding (see structure Binding), which supports precise

4509 qualification by packages and local theory targets, as well as proper

4510 tracking of source positions. INCOMPATIBILITY, need to wrap old

4511 bstring values into Binding.name, or better pass through abstract

4512 bindings everywhere. See further src/Pure/General/long_name.ML,

4513 src/Pure/General/binding.ML and src/Pure/General/name_space.ML

4515 * Result facts (from PureThy.note_thms, ProofContext.note_thms,

4516 LocalTheory.note etc.) now refer to the *full* internal name, not the

4517 bstring as before. INCOMPATIBILITY, not detected by ML type-checking!

4519 * Disposed old type and term read functions (Sign.read_def_typ,

4520 Sign.read_typ, Sign.read_def_terms, Sign.read_term,

4521 Thm.read_def_cterms, Thm.read_cterm etc.). INCOMPATIBILITY, should

4522 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,

4523 Syntax.read_term_global etc.; see also OldGoals.read_term as last

4524 resort for legacy applications.

4526 * Disposed old declarations, tactics, tactic combinators that refer to

4527 the simpset or claset of an implicit theory (such as Addsimps,

4528 Simp_tac, SIMPSET). INCOMPATIBILITY, should use @{simpset} etc. in

4529 embedded ML text, or local_simpset_of with a proper context passed as

4530 explicit runtime argument.

4532 * Rules and tactics that read instantiations (read_instantiate,

4533 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof

4534 context, which is required for parsing and type-checking. Moreover,

4535 the variables are specified as plain indexnames, not string encodings

4536 thereof. INCOMPATIBILITY.

4538 * Generic Toplevel.add_hook interface allows to analyze the result of

4539 transactions. E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML

4540 for theorem dependency output of transactions resulting in a new

4541 theory state.

4543 * ML antiquotations: block-structured compilation context indicated by

4544 \<lbrace> ... \<rbrace>; additional antiquotation forms:

4546 @{binding name} - basic name binding

4547 @{let ?pat = term} - term abbreviation (HO matching)

4548 @{note name = fact} - fact abbreviation

4549 @{thm fact} - singleton fact (with attributes)

4550 @{thms fact} - general fact (with attributes)

4551 @{lemma prop by method} - singleton goal

4552 @{lemma prop by meth1 meth2} - singleton goal

4553 @{lemma prop1 ... propN by method} - general goal

4554 @{lemma prop1 ... propN by meth1 meth2} - general goal

4555 @{lemma (open) ...} - open derivation

4558 *** System ***

4560 * The Isabelle "emacs" tool provides a specific interface to invoke

4561 Proof General / Emacs, with more explicit failure if that is not

4562 installed (the old isabelle-interface script silently falls back on

4563 isabelle-process). The PROOFGENERAL_HOME setting determines the

4564 installation location of the Proof General distribution.

4566 * Isabelle/lib/classes/Pure.jar provides basic support to integrate

4567 the Isabelle process into a JVM/Scala application. See

4568 Isabelle/lib/jedit/plugin for a minimal example. (The obsolete Java

4569 process wrapper has been discontinued.)

4571 * Added homegrown Isabelle font with unicode layout, see lib/fonts.

4573 * Various status messages (with exact source position information) are

4574 emitted, if proper markup print mode is enabled. This allows

4575 user-interface components to provide detailed feedback on internal

4576 prover operations.

4580 New in Isabelle2008 (June 2008)

4581 -------------------------------

4583 *** General ***

4585 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized

4586 and updated, with formally checked references as hyperlinks.

4588 * Theory loader: use_thy (and similar operations) no longer set the

4589 implicit ML context, which was occasionally hard to predict and in

4590 conflict with concurrency. INCOMPATIBILITY, use ML within Isar which

4591 provides a proper context already.

4593 * Theory loader: old-style ML proof scripts being *attached* to a thy

4594 file are no longer supported. INCOMPATIBILITY, regular 'uses' and

4595 'use' within a theory file will do the job.

4597 * Name space merge now observes canonical order, i.e. the second space

4598 is inserted into the first one, while existing entries in the first

4599 space take precedence. INCOMPATIBILITY in rare situations, may try to

4600 swap theory imports.

4602 * Syntax: symbol \<chi> is now considered a letter. Potential

4603 INCOMPATIBILITY in identifier syntax etc.

4605 * Outer syntax: string tokens no longer admit escaped white space,

4606 which was an accidental (undocumented) feature. INCOMPATIBILITY, use

4607 white space without escapes.

4609 * Outer syntax: string tokens may contain arbitrary character codes

4610 specified via 3 decimal digits (as in SML). E.g. "foo\095bar" for

4611 "foo_bar".

4614 *** Pure ***

4616 * Context-dependent token translations. Default setup reverts locally

4617 fixed variables, and adds hilite markup for undeclared frees.

4619 * Unused theorems can be found using the new command 'unused_thms'.

4620 There are three ways of invoking it:

4622 (1) unused_thms

4623 Only finds unused theorems in the current theory.

4625 (2) unused_thms thy_1 ... thy_n -

4626 Finds unused theorems in the current theory and all of its ancestors,

4627 excluding the theories thy_1 ... thy_n and all of their ancestors.

4629 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m

4630 Finds unused theorems in the theories thy'_1 ... thy'_m and all of

4631 their ancestors, excluding the theories thy_1 ... thy_n and all of

4632 their ancestors.

4634 In order to increase the readability of the list produced by

4635 unused_thms, theorems that have been created by a particular instance

4636 of a theory command such as 'inductive' or 'function' are considered

4637 to belong to the same "group", meaning that if at least one theorem in

4638 this group is used, the other theorems in the same group are no longer

4639 reported as unused. Moreover, if all theorems in the group are

4640 unused, only one theorem in the group is displayed.

4642 Note that proof objects have to be switched on in order for

4643 unused_thms to work properly (i.e. !proofs must be >= 1, which is

4644 usually the case when using Proof General with the default settings).

4646 * Authentic naming of facts disallows ad-hoc overwriting of previous

4647 theorems within the same name space. INCOMPATIBILITY, need to remove

4648 duplicate fact bindings, or even accidental fact duplications. Note

4649 that tools may maintain dynamically scoped facts systematically, using

4650 PureThy.add_thms_dynamic.

4652 * Command 'hide' now allows to hide from "fact" name space as well.

4654 * Eliminated destructive theorem database, simpset, claset, and

4655 clasimpset. Potential INCOMPATIBILITY, really need to observe linear

4656 update of theories within ML code.

4658 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.

4659 INCOMPATIBILITY, object-logics depending on former Pure require

4660 additional setup PureThy.old_appl_syntax_setup; object-logics

4661 depending on former CPure need to refer to Pure.

4663 * Commands 'use' and 'ML' are now purely functional, operating on

4664 theory/local_theory. Removed former 'ML_setup' (on theory), use 'ML'

4665 instead. Added 'ML_val' as mere diagnostic replacement for 'ML'.

4666 INCOMPATIBILITY.

4668 * Command 'setup': discontinued implicit version with ML reference.

4670 * Instantiation target allows for simultaneous specification of class

4671 instance operations together with an instantiation proof.

4672 Type-checking phase allows to refer to class operations uniformly.

4673 See src/HOL/Complex/Complex.thy for an Isar example and

4674 src/HOL/Library/Eval.thy for an ML example.

4676 * Indexing of literal facts: be more serious about including only

4677 facts from the visible specification/proof context, but not the

4678 background context (locale etc.). Affects `prop` notation and method

4679 "fact". INCOMPATIBILITY: need to name facts explicitly in rare

4680 situations.

4682 * Method "cases", "induct", "coinduct": removed obsolete/undocumented

4683 "(open)" option, which used to expose internal bound variables to the

4684 proof text.

4686 * Isar statements: removed obsolete case "rule_context".

4687 INCOMPATIBILITY, better use explicit fixes/assumes.

4689 * Locale proofs: default proof step now includes 'unfold_locales';

4690 hence 'proof' without argument may be used to unfold locale

4691 predicates.

4694 *** Document preparation ***

4696 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for

4697 both pdf and dvi (hyperlinks usually work in xdvi as well); removed

4698 obsolete thumbpdf setup (contemporary PDF viewers do this on the

4699 spot); renamed link color from "darkblue" to "linkcolor" (default

4700 value unchanged, can be redefined via \definecolor); no longer sets

4701 "a4paper" option (unnecessary or even intrusive).

4703 * Antiquotation @{lemma A method} proves proposition A by the given

4704 method (either a method name or a method name plus (optional) method

4705 arguments in parentheses) and prints A just like @{prop A}.

4708 *** HOL ***

4710 * New primrec package. Specification syntax conforms in style to

4711 definition/function/.... No separate induction rule is provided. The

4712 "primrec" command distinguishes old-style and new-style specifications

4713 by syntax. The former primrec package is now named OldPrimrecPackage.

4714 When adjusting theories, beware: constants stemming from new-style

4715 primrec specifications have authentic syntax.

4717 * Metis prover is now an order of magnitude faster, and also works

4718 with multithreading.

4720 * Metis: the maximum number of clauses that can be produced from a

4721 theorem is now given by the attribute max_clauses. Theorems that

4722 exceed this number are ignored, with a warning printed.

4724 * Sledgehammer no longer produces structured proofs by default. To

4725 enable, declare [[sledgehammer_full = true]]. Attributes

4726 reconstruction_modulus, reconstruction_sorts renamed

4727 sledgehammer_modulus, sledgehammer_sorts. INCOMPATIBILITY.

4729 * Method "induct_scheme" derives user-specified induction rules

4730 from well-founded induction and completeness of patterns. This factors

4731 out some operations that are done internally by the function package

4732 and makes them available separately. See

4733 src/HOL/ex/Induction_Scheme.thy for examples.

4735 * More flexible generation of measure functions for termination

4736 proofs: Measure functions can be declared by proving a rule of the

4737 form "is_measure f" and giving it the [measure_function] attribute.

4738 The "is_measure" predicate is logically meaningless (always true), and

4739 just guides the heuristic. To find suitable measure functions, the

4740 termination prover sets up the goal "is_measure ?f" of the appropriate

4741 type and generates all solutions by prolog-style backwards proof using

4742 the declared rules.

4744 This setup also deals with rules like

4746 "is_measure f ==> is_measure (list_size f)"

4748 which accommodates nested datatypes that recurse through lists.

4749 Similar rules are predeclared for products and option types.

4751 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"

4753 INCOMPATIBILITIES:

4755 - Definitions of overloaded constants on sets have to be replaced by

4756 definitions on => and bool.

4758 - Some definitions of overloaded operators on sets can now be proved

4759 using the definitions of the operators on => and bool. Therefore,

4760 the following theorems have been renamed:

4762 subset_def -> subset_eq

4763 psubset_def -> psubset_eq

4764 set_diff_def -> set_diff_eq

4765 Compl_def -> Compl_eq

4766 Sup_set_def -> Sup_set_eq

4767 Inf_set_def -> Inf_set_eq

4768 sup_set_def -> sup_set_eq

4769 inf_set_def -> inf_set_eq

4771 - Due to the incompleteness of the HO unification algorithm, some

4772 rules such as subst may require manual instantiation, if some of

4773 the unknowns in the rule is a set.

4775 - Higher order unification and forward proofs:

4776 The proof pattern

4778 have "P (S::'a set)" <...>

4779 then have "EX S. P S" ..

4781 no longer works (due to the incompleteness of the HO unification

4782 algorithm) and must be replaced by the pattern

4784 have "EX S. P S"

4785 proof

4786 show "P S" <...>

4787 qed

4789 - Calculational reasoning with subst (or similar rules):

4790 The proof pattern

4792 have "P (S::'a set)" <...>

4793 also have "S = T" <...>

4794 finally have "P T" .

4796 no longer works (for similar reasons as the previous example) and

4797 must be replaced by something like

4799 have "P (S::'a set)" <...>

4800 moreover have "S = T" <...>

4801 ultimately have "P T" by simp

4803 - Tactics or packages written in ML code:

4804 Code performing pattern matching on types via

4806 Type ("set", [T]) => ...

4808 must be rewritten. Moreover, functions like strip_type or

4809 binder_types no longer return the right value when applied to a

4810 type of the form

4812 T1 => ... => Tn => U => bool

4814 rather than

4816 T1 => ... => Tn => U set

4818 * Merged theories Wellfounded_Recursion, Accessible_Part and

4819 Wellfounded_Relations to theory Wellfounded.

4821 * Explicit class "eq" for executable equality. INCOMPATIBILITY.

4823 * Class finite no longer treats UNIV as class parameter. Use class

4824 enum from theory Library/Enum instead to achieve a similar effect.

4825 INCOMPATIBILITY.

4827 * Theory List: rule list_induct2 now has explicitly named cases "Nil"

4828 and "Cons". INCOMPATIBILITY.

4830 * HOL (and FOL): renamed variables in rules imp_elim and swap.

4831 Potential INCOMPATIBILITY.

4833 * Theory Product_Type: duplicated lemmas split_Pair_apply and

4834 injective_fst_snd removed, use split_eta and prod_eqI instead.

4835 Renamed upd_fst to apfst and upd_snd to apsnd. INCOMPATIBILITY.

4837 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of

4838 the same name in theory Orderings:

4840 less_trans

4841 less_linear

4842 le_imp_less_or_eq

4843 le_less_trans

4844 less_le_trans

4845 less_not_sym

4846 less_asym

4848 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to

4849 less_irrefl_nat. Potential INCOMPATIBILITY due to more general types

4850 and different variable names.

4852 * Library/Option_ord.thy: Canonical order on option type.

4854 * Library/RBT.thy: Red-black trees, an efficient implementation of

4855 finite maps.

4857 * Library/Countable.thy: Type class for countable types.

4859 * Theory Int: The representation of numerals has changed. The infix

4860 operator BIT and the bit datatype with constructors B0 and B1 have

4861 disappeared. INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in

4862 place of "x BIT bit.B0" and "y BIT bit.B1", respectively. Theorems

4863 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"

4864 accordingly.

4866 * Theory Nat: definition of <= and < on natural numbers no longer

4867 depend on well-founded relations. INCOMPATIBILITY. Definitions

4868 le_def and less_def have disappeared. Consider lemmas not_less

4869 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.

4871 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin

4872 (whose purpose mainly is for various fold_set functionals) have been

4873 abandoned in favor of the existing algebraic classes

4874 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,

4875 lower_semilattice (resp. upper_semilattice) and linorder.

4876 INCOMPATIBILITY.

4878 * Theory Transitive_Closure: induct and cases rules now declare proper

4879 case_names ("base" and "step"). INCOMPATIBILITY.

4881 * Theorem Inductive.lfp_ordinal_induct generalized to complete

4882 lattices. The form set-specific version is available as

4883 Inductive.lfp_ordinal_induct_set.

4885 * Renamed theorems "power.simps" to "power_int.simps".

4886 INCOMPATIBILITY.

4888 * Class semiring_div provides basic abstract properties of semirings

4889 with division and modulo operations. Subsumes former class dvd_mod.

4891 * Merged theories IntDef, Numeral and IntArith into unified theory

4892 Int. INCOMPATIBILITY.

4894 * Theory Library/Code_Index: type "index" now represents natural

4895 numbers rather than integers. INCOMPATIBILITY.

4897 * New class "uminus" with operation "uminus" (split of from class

4898 "minus" which now only has operation "minus", binary).

4899 INCOMPATIBILITY.

4901 * Constants "card", "internal_split", "option_map" now with authentic

4902 syntax. INCOMPATIBILITY.

4904 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,

4905 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,

4906 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,

4907 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,

4908 Sup_set_def, le_def, less_def, option_map_def now with object

4909 equality. INCOMPATIBILITY.

4911 * Records. Removed K_record, and replaced it by pure lambda term

4912 %x. c. The simplifier setup is now more robust against eta expansion.

4913 INCOMPATIBILITY: in cases explicitly referring to K_record.

4915 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.

4917 * Library/ListVector: new theory of arithmetic vector operations.

4919 * Library/Order_Relation: new theory of various orderings as sets of

4920 pairs. Defines preorders, partial orders, linear orders and

4921 well-orders on sets and on types.

4924 *** ZF ***

4926 * Renamed some theories to allow to loading both ZF and HOL in the

4927 same session:

4929 Datatype -> Datatype_ZF

4930 Inductive -> Inductive_ZF

4931 Int -> Int_ZF

4932 IntDiv -> IntDiv_ZF

4933 Nat -> Nat_ZF

4934 List -> List_ZF

4935 Main -> Main_ZF

4937 INCOMPATIBILITY: ZF theories that import individual theories below

4938 Main might need to be adapted. Regular theory Main is still

4939 available, as trivial extension of Main_ZF.

4942 *** ML ***

4944 * ML within Isar: antiquotation @{const name} or @{const

4945 name(typargs)} produces statically-checked Const term.

4947 * Functor NamedThmsFun: data is available to the user as dynamic fact

4948 (of the same name). Removed obsolete print command.

4950 * Removed obsolete "use_legacy_bindings" function.

4952 * The ``print mode'' is now a thread-local value derived from a global

4953 template (the former print_mode reference), thus access becomes

4954 non-critical. The global print_mode reference is for session

4955 management only; user-code should use print_mode_value,

4956 print_mode_active, PrintMode.setmp etc. INCOMPATIBILITY.

4958 * Functions system/system_out provide a robust way to invoke external

4959 shell commands, with propagation of interrupts (requires Poly/ML

4960 5.2.1). Do not use OS.Process.system etc. from the basis library!

4963 *** System ***

4965 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---

4966 in accordance with Proof General 3.7, which prefers GNU emacs.

4968 * isatool tty runs Isabelle process with plain tty interaction;

4969 optional line editor may be specified via ISABELLE_LINE_EDITOR

4970 setting, the default settings attempt to locate "ledit" and "rlwrap".

4972 * isatool browser now works with Cygwin as well, using general

4973 "javapath" function defined in Isabelle process environment.

4975 * YXML notation provides a simple and efficient alternative to

4976 standard XML transfer syntax. See src/Pure/General/yxml.ML and

4977 isatool yxml as described in the Isabelle system manual.

4979 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)

4980 provides general wrapper for managing an Isabelle process in a robust

4981 fashion, with ``cooked'' output from stdin/stderr.

4983 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),

4984 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).

4986 * Removed obsolete THIS_IS_ISABELLE_BUILD feature. NB: the documented

4987 way of changing the user's settings is via

4988 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash

4989 script.

4991 * Multithreading.max_threads := 0 refers to the number of actual CPU

4992 cores of the underlying machine, which is a good starting point for

4993 optimal performance tuning. The corresponding usedir option -M allows

4994 "max" as an alias for "0". WARNING: does not work on certain versions

4995 of Mac OS (with Poly/ML 5.1).

4997 * isabelle-process: non-ML sessions are run with "nice", to reduce the

4998 adverse effect of Isabelle flooding interactive front-ends (notably

4999 ProofGeneral / XEmacs).

5003 New in Isabelle2007 (November 2007)

5004 -----------------------------------

5006 *** General ***

5008 * More uniform information about legacy features, notably a

5009 warning/error of "Legacy feature: ...", depending on the state of the

5010 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:

5011 legacy features will disappear eventually.

5013 * Theory syntax: the header format ``theory A = B + C:'' has been

5014 discontinued in favour of ``theory A imports B C begin''. Use isatool

5015 fixheaders to convert existing theory files. INCOMPATIBILITY.

5017 * Theory syntax: the old non-Isar theory file format has been

5018 discontinued altogether. Note that ML proof scripts may still be used

5019 with Isar theories; migration is usually quite simple with the ML

5020 function use_legacy_bindings. INCOMPATIBILITY.

5022 * Theory syntax: some popular names (e.g. 'class', 'declaration',

5023 'fun', 'help', 'if') are now keywords. INCOMPATIBILITY, use double

5024 quotes.

5026 * Theory loader: be more serious about observing the static theory

5027 header specifications (including optional directories), but not the

5028 accidental file locations of previously successful loads. The strict

5029 update policy of former update_thy is now already performed by

5030 use_thy, so the former has been removed; use_thys updates several

5031 theories simultaneously, just as 'imports' within a theory header

5032 specification, but without merging the results. Potential

5033 INCOMPATIBILITY: may need to refine theory headers and commands

5034 ROOT.ML which depend on load order.

5036 * Theory loader: optional support for content-based file

5037 identification, instead of the traditional scheme of full physical

5038 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting

5039 (cf. the system manual). The new scheme allows to work with

5040 non-finished theories in persistent session images, such that source

5041 files may be moved later on without requiring reloads.

5043 * Theory loader: old-style ML proof scripts being *attached* to a thy

5044 file (with the same base name as the theory) are considered a legacy

5045 feature, which will disappear eventually. Even now, the theory loader

5046 no longer maintains dependencies on such files.

5048 * Syntax: the scope for resolving ambiguities via type-inference is

5049 now limited to individual terms, instead of whole simultaneous

5050 specifications as before. This greatly reduces the complexity of the

5051 syntax module and improves flexibility by separating parsing and

5052 type-checking. INCOMPATIBILITY: additional type-constraints (explicit

5053 'fixes' etc.) are required in rare situations.

5055 * Syntax: constants introduced by new-style packages ('definition',

5056 'abbreviation' etc.) are passed through the syntax module in

5057 ``authentic mode''. This means that associated mixfix annotations

5058 really stick to such constants, independently of potential name space

5059 ambiguities introduced later on. INCOMPATIBILITY: constants in parse

5060 trees are represented slightly differently, may need to adapt syntax

5061 translations accordingly. Use CONST marker in 'translations' and

5062 @{const_syntax} antiquotation in 'parse_translation' etc.

5064 * Legacy goal package: reduced interface to the bare minimum required

5065 to keep existing proof scripts running. Most other user-level

5066 functions are now part of the OldGoals structure, which is *not* open

5067 by default (consider isatool expandshort before open OldGoals).

5068 Removed top_sg, prin, printyp, pprint_term/typ altogether, because

5069 these tend to cause confusion about the actual goal (!) context being

5070 used here, which is not necessarily the same as the_context().

5072 * Command 'find_theorems': supports "*" wild-card in "name:"

5073 criterion; "with_dups" option. Certain ProofGeneral versions might

5074 support a specific search form (see ProofGeneral/CHANGES).

5076 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1

5077 by default, which means that "prems" (and also "fixed variables") are

5078 suppressed from proof state output. Note that the ProofGeneral

5079 settings mechanism allows to change and save options persistently, but

5080 older versions of Isabelle will fail to start up if a negative prems

5081 limit is imposed.

5083 * Local theory targets may be specified by non-nested blocks of

5084 ``context/locale/class ... begin'' followed by ``end''. The body may

5085 contain definitions, theorems etc., including any derived mechanism

5086 that has been implemented on top of these primitives. This concept

5087 generalizes the existing ``theorem (in ...)'' towards more versatility

5088 and scalability.

5090 * Proof General interface: proper undo of final 'end' command;

5091 discontinued Isabelle/classic mode (ML proof scripts).

5094 *** Document preparation ***

5096 * Added antiquotation @{theory name} which prints the given name,

5097 after checking that it refers to a valid ancestor theory in the

5098 current context.

5100 * Added antiquotations @{ML_type text} and @{ML_struct text} which

5101 check the given source text as ML type/structure, printing verbatim.

5103 * Added antiquotation @{abbrev "c args"} which prints the abbreviation

5104 "c args == rhs" given in the current context. (Any number of

5105 arguments may be given on the LHS.)

5108 *** Pure ***

5110 * The 'class' package offers a combination of axclass and locale to

5111 achieve Haskell-like type classes in Isabelle. Definitions and

5112 theorems within a class context produce both relative results (with

5113 implicit parameters according to the locale context), and polymorphic

5114 constants with qualified polymorphism (according to the class

5115 context). Within the body context of a 'class' target, a separate

5116 syntax layer ("user space type system") takes care of converting

5117 between global polymorphic consts and internal locale representation.

5118 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).

5119 "isatool doc classes" provides a tutorial.

5121 * Generic code generator framework allows to generate executable

5122 code for ML and Haskell (including Isabelle classes). A short usage

5123 sketch:

5125 internal compilation:

5126 export_code <list of constants (term syntax)> in SML

5127 writing SML code to a file:

5128 export_code <list of constants (term syntax)> in SML <filename>

5129 writing OCaml code to a file:

5130 export_code <list of constants (term syntax)> in OCaml <filename>

5131 writing Haskell code to a bunch of files:

5132 export_code <list of constants (term syntax)> in Haskell <filename>

5134 evaluating closed propositions to True/False using code generation:

5135 method ``eval''

5137 Reasonable default setup of framework in HOL.

5139 Theorem attributs for selecting and transforming function equations theorems:

5141 [code fun]: select a theorem as function equation for a specific constant

5142 [code fun del]: deselect a theorem as function equation for a specific constant

5143 [code inline]: select an equation theorem for unfolding (inlining) in place

5144 [code inline del]: deselect an equation theorem for unfolding (inlining) in place

5146 User-defined serializations (target in {SML, OCaml, Haskell}):

5148 code_const <and-list of constants (term syntax)>

5149 {(target) <and-list of const target syntax>}+

5151 code_type <and-list of type constructors>

5152 {(target) <and-list of type target syntax>}+

5154 code_instance <and-list of instances>

5155 {(target)}+

5156 where instance ::= <type constructor> :: <class>

5158 code_class <and_list of classes>

5159 {(target) <and-list of class target syntax>}+

5160 where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?

5162 code_instance and code_class only are effective to target Haskell.

5164 For example usage see src/HOL/ex/Codegenerator.thy and

5165 src/HOL/ex/Codegenerator_Pretty.thy. A separate tutorial on code

5166 generation from Isabelle/HOL theories is available via "isatool doc

5167 codegen".

5169 * Code generator: consts in 'consts_code' Isar commands are now

5170 referred to by usual term syntax (including optional type

5171 annotations).

5173 * Command 'no_translations' removes translation rules from theory

5174 syntax.

5176 * Overloaded definitions are now actually checked for acyclic

5177 dependencies. The overloading scheme is slightly more general than

5178 that of Haskell98, although Isabelle does not demand an exact

5179 correspondence to type class and instance declarations.

5180 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more

5181 exotic versions of overloading -- at the discretion of the user!

5183 Polymorphic constants are represented via type arguments, i.e. the

5184 instantiation that matches an instance against the most general

5185 declaration given in the signature. For example, with the declaration

5186 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented

5187 as c(nat). Overloading is essentially simultaneous structural

5188 recursion over such type arguments. Incomplete specification patterns

5189 impose global constraints on all occurrences, e.g. c('a * 'a) on the

5190 LHS means that more general c('a * 'b) will be disallowed on any RHS.

5191 Command 'print_theory' outputs the normalized system of recursive

5192 equations, see section "definitions".

5194 * Configuration options are maintained within the theory or proof

5195 context (with name and type bool/int/string), providing a very simple

5196 interface to a poor-man's version of general context data. Tools may

5197 declare options in ML (e.g. using Attrib.config_int) and then refer to

5198 these values using Config.get etc. Users may change options via an

5199 associated attribute of the same name. This form of context

5200 declaration works particularly well with commands 'declare' or

5201 'using', for example ``declare [[foo = 42]]''. Thus it has become

5202 very easy to avoid global references, which would not observe Isar

5203 toplevel undo/redo and fail to work with multithreading.

5205 Various global ML references of Pure and HOL have been turned into

5206 configuration options:

5208 Unify.search_bound unify_search_bound

5209 Unify.trace_bound unify_trace_bound

5210 Unify.trace_simp unify_trace_simp

5211 Unify.trace_types unify_trace_types

5212 Simplifier.simp_depth_limit simp_depth_limit

5213 Blast.depth_limit blast_depth_limit

5214 DatatypeProp.dtK datatype_distinctness_limit

5215 fast_arith_neq_limit fast_arith_neq_limit

5216 fast_arith_split_limit fast_arith_split_limit

5218 * Named collections of theorems may be easily installed as context

5219 data using the functor NamedThmsFun (see also

5220 src/Pure/Tools/named_thms.ML). The user may add or delete facts via

5221 attributes; there is also a toplevel print command. This facility is

5222 just a common case of general context data, which is the preferred way

5223 for anything more complex than just a list of facts in canonical

5224 order.

5226 * Isar: command 'declaration' augments a local theory by generic

5227 declaration functions written in ML. This enables arbitrary content

5228 being added to the context, depending on a morphism that tells the

5229 difference of the original declaration context wrt. the application

5230 context encountered later on.

5232 * Isar: proper interfaces for simplification procedures. Command

5233 'simproc_setup' declares named simprocs (with match patterns, and body

5234 text in ML). Attribute "simproc" adds/deletes simprocs in the current

5235 context. ML antiquotation @{simproc name} retrieves named simprocs.

5237 * Isar: an extra pair of brackets around attribute declarations

5238 abbreviates a theorem reference involving an internal dummy fact,

5239 which will be ignored later --- only the effect of the attribute on

5240 the background context will persist. This form of in-place

5241 declarations is particularly useful with commands like 'declare' and

5242 'using', for example ``have A using [[simproc a]] by simp''.

5244 * Isar: method "assumption" (and implicit closing of subproofs) now

5245 takes simple non-atomic goal assumptions into account: after applying

5246 an assumption as a rule the resulting subgoals are solved by atomic

5247 assumption steps. This is particularly useful to finish 'obtain'

5248 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",

5249 without referring to the original premise "!!x. P x ==> thesis" in the

5250 Isar proof context. POTENTIAL INCOMPATIBILITY: method "assumption" is

5251 more permissive.

5253 * Isar: implicit use of prems from the Isar proof context is

5254 considered a legacy feature. Common applications like ``have A .''

5255 may be replaced by ``have A by fact'' or ``note `A`''. In general,

5256 referencing facts explicitly here improves readability and

5257 maintainability of proof texts.

5259 * Isar: improper proof element 'guess' is like 'obtain', but derives

5260 the obtained context from the course of reasoning! For example:

5262 assume "EX x y. A x & B y" -- "any previous fact"

5263 then guess x and y by clarify

5265 This technique is potentially adventurous, depending on the facts and

5266 proof tools being involved here.

5268 * Isar: known facts from the proof context may be specified as literal

5269 propositions, using ASCII back-quote syntax. This works wherever

5270 named facts used to be allowed so far, in proof commands, proof

5271 methods, attributes etc. Literal facts are retrieved from the context

5272 according to unification of type and term parameters. For example,

5273 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known

5274 theorems in the current context, then these are valid literal facts:

5275 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.

5277 There is also a proof method "fact" which does the same composition

5278 for explicit goal states, e.g. the following proof texts coincide with

5279 certain special cases of literal facts:

5281 have "A" by fact == note `A`

5282 have "A ==> B" by fact == note `A ==> B`

5283 have "!!x. P x ==> Q x" by fact == note `!!x. P x ==> Q x`

5284 have "P a ==> Q a" by fact == note `P a ==> Q a`

5286 * Isar: ":" (colon) is no longer a symbolic identifier character in

5287 outer syntax. Thus symbolic identifiers may be used without

5288 additional white space in declarations like this: ``assume *: A''.

5290 * Isar: 'print_facts' prints all local facts of the current context,

5291 both named and unnamed ones.

5293 * Isar: 'def' now admits simultaneous definitions, e.g.:

5295 def x == "t" and y == "u"

5297 * Isar: added command 'unfolding', which is structurally similar to

5298 'using', but affects both the goal state and facts by unfolding given

5299 rewrite rules. Thus many occurrences of the 'unfold' method or

5300 'unfolded' attribute may be replaced by first-class proof text.

5302 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',

5303 and command 'unfolding' now all support object-level equalities

5304 (potentially conditional). The underlying notion of rewrite rule is

5305 analogous to the 'rule_format' attribute, but *not* that of the

5306 Simplifier (which is usually more generous).

5308 * Isar: the new attribute [rotated n] (default n = 1) rotates the

5309 premises of a theorem by n. Useful in conjunction with drule.

5311 * Isar: the goal restriction operator [N] (default N = 1) evaluates a

5312 method expression within a sandbox consisting of the first N

5313 sub-goals, which need to exist. For example, ``simp_all [3]''

5314 simplifies the first three sub-goals, while (rule foo, simp_all)[]

5315 simplifies all new goals that emerge from applying rule foo to the

5316 originally first one.

5318 * Isar: schematic goals are no longer restricted to higher-order

5319 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as

5320 expected.

5322 * Isar: the conclusion of a long theorem statement is now either

5323 'shows' (a simultaneous conjunction, as before), or 'obtains'

5324 (essentially a disjunction of cases with local parameters and

5325 assumptions). The latter allows to express general elimination rules

5326 adequately; in this notation common elimination rules look like this:

5328 lemma exE: -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"

5329 assumes "EX x. P x"

5330 obtains x where "P x"

5332 lemma conjE: -- "A & B ==> (A ==> B ==> thesis) ==> thesis"

5333 assumes "A & B"

5334 obtains A and B

5336 lemma disjE: -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"

5337 assumes "A | B"

5338 obtains

5339 A

5340 | B

5342 The subsequent classical rules even refer to the formal "thesis"

5343 explicitly:

5345 lemma classical: -- "(~ thesis ==> thesis) ==> thesis"

5346 obtains "~ thesis"

5348 lemma Peirce's_Law: -- "((thesis ==> something) ==> thesis) ==> thesis"

5349 obtains "thesis ==> something"

5351 The actual proof of an 'obtains' statement is analogous to that of the

5352 Isar proof element 'obtain', only that there may be several cases.

5353 Optional case names may be specified in parentheses; these will be

5354 available both in the present proof and as annotations in the

5355 resulting rule, for later use with the 'cases' method (cf. attribute

5356 case_names).

5358 * Isar: the assumptions of a long theorem statement are available as

5359 "assms" fact in the proof context. This is more appropriate than the

5360 (historical) "prems", which refers to all assumptions of the current

5361 context, including those from the target locale, proof body etc.

5363 * Isar: 'print_statement' prints theorems from the current theory or

5364 proof context in long statement form, according to the syntax of a

5365 top-level lemma.

5367 * Isar: 'obtain' takes an optional case name for the local context

5368 introduction rule (default "that").

5370 * Isar: removed obsolete 'concl is' patterns. INCOMPATIBILITY, use

5371 explicit (is "_ ==> ?foo") in the rare cases where this still happens

5372 to occur.

5374 * Pure: syntax "CONST name" produces a fully internalized constant

5375 according to the current context. This is particularly useful for

5376 syntax translations that should refer to internal constant

5377 representations independently of name spaces.

5379 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"

5380 instead of "FOO ". This allows multiple binder declarations to coexist

5381 in the same context. INCOMPATIBILITY.

5383 * Isar/locales: 'notation' provides a robust interface to the 'syntax'

5384 primitive that also works in a locale context (both for constants and

5385 fixed variables). Type declaration and internal syntactic representation

5386 of given constants retrieved from the context. Likewise, the

5387 'no_notation' command allows to remove given syntax annotations from the

5388 current context.

5390 * Isar/locales: new derived specification elements 'axiomatization',

5391 'definition', 'abbreviation', which support type-inference, admit

5392 object-level specifications (equality, equivalence). See also the

5393 isar-ref manual. Examples:

5395 axiomatization

5396 eq (infix "===" 50) where

5397 eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"

5399 definition "f x y = x + y + 1"

5400 definition g where "g x = f x x"

5402 abbreviation

5403 neq (infix "=!=" 50) where

5404 "x =!= y == ~ (x === y)"

5406 These specifications may be also used in a locale context. Then the

5407 constants being introduced depend on certain fixed parameters, and the

5408 constant name is qualified by the locale base name. An internal

5409 abbreviation takes care for convenient input and output, making the

5410 parameters implicit and using the original short name. See also

5411 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic

5412 entities from a monomorphic theory.

5414 Presently, abbreviations are only available 'in' a target locale, but

5415 not inherited by general import expressions. Also note that

5416 'abbreviation' may be used as a type-safe replacement for 'syntax' +

5417 'translations' in common applications. The "no_abbrevs" print mode

5418 prevents folding of abbreviations in term output.

5420 Concrete syntax is attached to specified constants in internal form,

5421 independently of name spaces. The parse tree representation is

5422 slightly different -- use 'notation' instead of raw 'syntax', and

5423 'translations' with explicit "CONST" markup to accommodate this.

5425 * Pure/Isar: unified syntax for new-style specification mechanisms

5426 (e.g. 'definition', 'abbreviation', or 'inductive' in HOL) admits

5427 full type inference and dummy patterns ("_"). For example:

5429 definition "K x _ = x"

5431 inductive conj for A B

5432 where "A ==> B ==> conj A B"

5434 * Pure: command 'print_abbrevs' prints all constant abbreviations of

5435 the current context. Print mode "no_abbrevs" prevents inversion of

5436 abbreviations on output.

5438 * Isar/locales: improved parameter handling: use of locales "var" and

5439 "struct" no longer necessary; - parameter renamings are no longer

5440 required to be injective. For example, this allows to define

5441 endomorphisms as locale endom = homom mult mult h.

5443 * Isar/locales: changed the way locales with predicates are defined.

5444 Instead of accumulating the specification, the imported expression is

5445 now an interpretation. INCOMPATIBILITY: different normal form of

5446 locale expressions. In particular, in interpretations of locales with

5447 predicates, goals repesenting already interpreted fragments are not

5448 removed automatically. Use methods `intro_locales' and

5449 `unfold_locales'; see below.

5451 * Isar/locales: new methods `intro_locales' and `unfold_locales'

5452 provide backward reasoning on locales predicates. The methods are

5453 aware of interpretations and discharge corresponding goals.

5454 `intro_locales' is less aggressive then `unfold_locales' and does not

5455 unfold predicates to assumptions.

5457 * Isar/locales: the order in which locale fragments are accumulated

5458 has changed. This enables to override declarations from fragments due

5459 to interpretations -- for example, unwanted simp rules.

5461 * Isar/locales: interpretation in theories and proof contexts has been

5462 extended. One may now specify (and prove) equations, which are

5463 unfolded in interpreted theorems. This is useful for replacing

5464 defined concepts (constants depending on locale parameters) by

5465 concepts already existing in the target context. Example:

5467 interpretation partial_order ["op <= :: [int, int] => bool"]

5468 where "partial_order.less (op <=) (x::int) y = (x < y)"

5470 Typically, the constant `partial_order.less' is created by a

5471 definition specification element in the context of locale

5472 partial_order.

5474 * Method "induct": improved internal context management to support

5475 local fixes and defines on-the-fly. Thus explicit meta-level

5476 connectives !! and ==> are rarely required anymore in inductive goals

5477 (using object-logic connectives for this purpose has been long

5478 obsolete anyway). Common proof patterns are explained in

5479 src/HOL/Induct/Common_Patterns.thy, see also

5480 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic

5481 examples.

5483 * Method "induct": improved handling of simultaneous goals. Instead of

5484 introducing object-level conjunction, the statement is now split into

5485 several conclusions, while the corresponding symbolic cases are nested

5486 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,

5487 see src/HOL/Induct/Common_Patterns.thy, for example.

5489 * Method "induct": mutual induction rules are now specified as a list

5490 of rule sharing the same induction cases. HOL packages usually provide

5491 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive

5492 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify

5493 mutual induction rules differently, i.e. like this:

5495 (induct rule: foo_bar.inducts)

5496 (induct set: foo bar)

5497 (induct pred: foo bar)

5498 (induct type: foo bar)

5500 The ML function ProjectRule.projections turns old-style rules into the

5501 new format.

5503 * Method "coinduct": dual of induction, see

5504 src/HOL/Library/Coinductive_List.thy for various examples.

5506 * Method "cases", "induct", "coinduct": the ``(open)'' option is

5507 considered a legacy feature.

5509 * Attribute "symmetric" produces result with standardized schematic

5510 variables (index 0). Potential INCOMPATIBILITY.

5512 * Simplifier: by default the simplifier trace only shows top level

5513 rewrites now. That is, trace_simp_depth_limit is set to 1 by

5514 default. Thus there is less danger of being flooded by the trace. The

5515 trace indicates where parts have been suppressed.

5517 * Provers/classical: removed obsolete classical version of elim_format

5518 attribute; classical elim/dest rules are now treated uniformly when

5519 manipulating the claset.

5521 * Provers/classical: stricter checks to ensure that supplied intro,

5522 dest and elim rules are well-formed; dest and elim rules must have at

5523 least one premise.

5525 * Provers/classical: attributes dest/elim/intro take an optional

5526 weight argument for the rule (just as the Pure versions). Weights are

5527 ignored by automated tools, but determine the search order of single

5528 rule steps.

5530 * Syntax: input syntax now supports dummy variable binding "%_. b",

5531 where the body does not mention the bound variable. Note that dummy

5532 patterns implicitly depend on their context of bounds, which makes

5533 "{_. _}" match any set comprehension as expected. Potential

5534 INCOMPATIBILITY -- parse translations need to cope with syntactic

5535 constant "_idtdummy" in the binding position.

5537 * Syntax: removed obsolete syntactic constant "_K" and its associated

5538 parse translation. INCOMPATIBILITY -- use dummy abstraction instead,

5539 for example "A -> B" => "Pi A (%_. B)".

5541 * Pure: 'class_deps' command visualizes the subclass relation, using

5542 the graph browser tool.

5544 * Pure: 'print_theory' now suppresses certain internal declarations by

5545 default; use '!' option for full details.

5548 *** HOL ***

5550 * Method "metis" proves goals by applying the Metis general-purpose

5551 resolution prover (see also http://gilith.com/software/metis/).

5552 Examples are in the directory MetisExamples. WARNING: the

5553 Isabelle/HOL-Metis integration does not yet work properly with

5554 multi-threading.

5556 * Command 'sledgehammer' invokes external automatic theorem provers as

5557 background processes. It generates calls to the "metis" method if

5558 successful. These can be pasted into the proof. Users do not have to

5559 wait for the automatic provers to return. WARNING: does not really

5560 work with multi-threading.

5562 * New "auto_quickcheck" feature tests outermost goal statements for

5563 potential counter-examples. Controlled by ML references

5564 auto_quickcheck (default true) and auto_quickcheck_time_limit (default

5565 5000 milliseconds). Fails silently if statements is outside of

5566 executable fragment, or any other codgenerator problem occurs.

5568 * New constant "undefined" with axiom "undefined x = undefined".

5570 * Added class "HOL.eq", allowing for code generation with polymorphic

5571 equality.

5573 * Some renaming of class constants due to canonical name prefixing in

5574 the new 'class' package:

5576 HOL.abs ~> HOL.abs_class.abs

5577 HOL.divide ~> HOL.divide_class.divide

5578 0 ~> HOL.zero_class.zero

5579 1 ~> HOL.one_class.one

5580 op + ~> HOL.plus_class.plus

5581 op - ~> HOL.minus_class.minus

5582 uminus ~> HOL.minus_class.uminus

5583 op * ~> HOL.times_class.times

5584 op < ~> HOL.ord_class.less

5585 op <= > HOL.ord_class.less_eq

5586 Nat.power ~> Power.power_class.power

5587 Nat.size ~> Nat.size_class.size

5588 Numeral.number_of ~> Numeral.number_class.number_of

5589 FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf

5590 FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup

5591 Orderings.min ~> Orderings.ord_class.min

5592 Orderings.max ~> Orderings.ord_class.max

5593 Divides.op div ~> Divides.div_class.div

5594 Divides.op mod ~> Divides.div_class.mod

5595 Divides.op dvd ~> Divides.div_class.dvd

5597 INCOMPATIBILITY. Adaptions may be required in the following cases:

5599 a) User-defined constants using any of the names "plus", "minus",

5600 "times", "less" or "less_eq". The standard syntax translations for

5601 "+", "-" and "*" may go wrong. INCOMPATIBILITY: use more specific

5602 names.

5604 b) Variables named "plus", "minus", "times", "less", "less_eq"

5605 INCOMPATIBILITY: use more specific names.

5607 c) Permutative equations (e.g. "a + b = b + a")

5608 Since the change of names also changes the order of terms, permutative

5609 rewrite rules may get applied in a different order. Experience shows

5610 that this is rarely the case (only two adaptions in the whole Isabelle

5611 distribution). INCOMPATIBILITY: rewrite proofs

5613 d) ML code directly refering to constant names

5614 This in general only affects hand-written proof tactics, simprocs and

5615 so on. INCOMPATIBILITY: grep your sourcecode and replace names.

5616 Consider using @{const_name} antiquotation.

5618 * New class "default" with associated constant "default".

5620 * Function "sgn" is now overloaded and available on int, real, complex

5621 (and other numeric types), using class "sgn". Two possible defs of

5622 sgn are given as equational assumptions in the classes sgn_if and

5623 sgn_div_norm; ordered_idom now also inherits from sgn_if.

5624 INCOMPATIBILITY.

5626 * Locale "partial_order" now unified with class "order" (cf. theory

5627 Orderings), added parameter "less". INCOMPATIBILITY.

5629 * Renamings in classes "order" and "linorder": facts "refl", "trans" and

5630 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid

5631 clashes with HOL "refl" and "trans". INCOMPATIBILITY.

5633 * Classes "order" and "linorder": potential INCOMPATIBILITY due to

5634 changed order of proof goals in instance proofs.

5636 * The transitivity reasoner for partial and linear orders is set up

5637 for classes "order" and "linorder". Instances of the reasoner are available

5638 in all contexts importing or interpreting the corresponding locales.

5639 Method "order" invokes the reasoner separately; the reasoner

5640 is also integrated with the Simplifier as a solver. Diagnostic

5641 command 'print_orders' shows the available instances of the reasoner

5642 in the current context.

5644 * Localized monotonicity predicate in theory "Orderings"; integrated

5645 lemmas max_of_mono and min_of_mono with this predicate.

5646 INCOMPATIBILITY.

5648 * Formulation of theorem "dense" changed slightly due to integration

5649 with new class dense_linear_order.

5651 * Uniform lattice theory development in HOL.

5653 constants "meet" and "join" now named "inf" and "sup"

5654 constant "Meet" now named "Inf"

5656 classes "meet_semilorder" and "join_semilorder" now named

5657 "lower_semilattice" and "upper_semilattice"

5658 class "lorder" now named "lattice"

5659 class "comp_lat" now named "complete_lattice"

5661 Instantiation of lattice classes allows explicit definitions

5662 for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).

5664 INCOMPATIBILITY. Theorem renames:

5666 meet_left_le ~> inf_le1

5667 meet_right_le ~> inf_le2

5668 join_left_le ~> sup_ge1

5669 join_right_le ~> sup_ge2

5670 meet_join_le ~> inf_sup_ord

5671 le_meetI ~> le_infI

5672 join_leI ~> le_supI

5673 le_meet ~> le_inf_iff

5674 le_join ~> ge_sup_conv

5675 meet_idempotent ~> inf_idem

5676 join_idempotent ~> sup_idem

5677 meet_comm ~> inf_commute

5678 join_comm ~> sup_commute

5679 meet_leI1 ~> le_infI1

5680 meet_leI2 ~> le_infI2

5681 le_joinI1 ~> le_supI1

5682 le_joinI2 ~> le_supI2

5683 meet_assoc ~> inf_assoc

5684 join_assoc ~> sup_assoc

5685 meet_left_comm ~> inf_left_commute

5686 meet_left_idempotent ~> inf_left_idem

5687 join_left_comm ~> sup_left_commute

5688 join_left_idempotent ~> sup_left_idem

5689 meet_aci ~> inf_aci

5690 join_aci ~> sup_aci

5691 le_def_meet ~> le_iff_inf

5692 le_def_join ~> le_iff_sup

5693 join_absorp2 ~> sup_absorb2

5694 join_absorp1 ~> sup_absorb1

5695 meet_absorp1 ~> inf_absorb1

5696 meet_absorp2 ~> inf_absorb2

5697 meet_join_absorp ~> inf_sup_absorb

5698 join_meet_absorp ~> sup_inf_absorb

5699 distrib_join_le ~> distrib_sup_le

5700 distrib_meet_le ~> distrib_inf_le

5702 add_meet_distrib_left ~> add_inf_distrib_left

5703 add_join_distrib_left ~> add_sup_distrib_left

5704 is_join_neg_meet ~> is_join_neg_inf

5705 is_meet_neg_join ~> is_meet_neg_sup

5706 add_meet_distrib_right ~> add_inf_distrib_right

5707 add_join_distrib_right ~> add_sup_distrib_right

5708 add_meet_join_distribs ~> add_sup_inf_distribs

5709 join_eq_neg_meet ~> sup_eq_neg_inf

5710 meet_eq_neg_join ~> inf_eq_neg_sup

5711 add_eq_meet_join ~> add_eq_inf_sup

5712 meet_0_imp_0 ~> inf_0_imp_0

5713 join_0_imp_0 ~> sup_0_imp_0

5714 meet_0_eq_0 ~> inf_0_eq_0

5715 join_0_eq_0 ~> sup_0_eq_0

5716 neg_meet_eq_join ~> neg_inf_eq_sup

5717 neg_join_eq_meet ~> neg_sup_eq_inf

5718 join_eq_if ~> sup_eq_if

5720 mono_meet ~> mono_inf

5721 mono_join ~> mono_sup

5722 meet_bool_eq ~> inf_bool_eq

5723 join_bool_eq ~> sup_bool_eq

5724 meet_fun_eq ~> inf_fun_eq

5725 join_fun_eq ~> sup_fun_eq

5726 meet_set_eq ~> inf_set_eq

5727 join_set_eq ~> sup_set_eq

5728 meet1_iff ~> inf1_iff

5729 meet2_iff ~> inf2_iff

5730 meet1I ~> inf1I

5731 meet2I ~> inf2I

5732 meet1D1 ~> inf1D1

5733 meet2D1 ~> inf2D1

5734 meet1D2 ~> inf1D2

5735 meet2D2 ~> inf2D2

5736 meet1E ~> inf1E

5737 meet2E ~> inf2E

5738 join1_iff ~> sup1_iff

5739 join2_iff ~> sup2_iff

5740 join1I1 ~> sup1I1

5741 join2I1 ~> sup2I1

5742 join1I1 ~> sup1I1

5743 join2I2 ~> sup1I2

5744 join1CI ~> sup1CI

5745 join2CI ~> sup2CI

5746 join1E ~> sup1E

5747 join2E ~> sup2E

5749 is_meet_Meet ~> is_meet_Inf

5750 Meet_bool_def ~> Inf_bool_def

5751 Meet_fun_def ~> Inf_fun_def

5752 Meet_greatest ~> Inf_greatest

5753 Meet_lower ~> Inf_lower

5754 Meet_set_def ~> Inf_set_def

5756 Sup_def ~> Sup_Inf

5757 Sup_bool_eq ~> Sup_bool_def

5758 Sup_fun_eq ~> Sup_fun_def

5759 Sup_set_eq ~> Sup_set_def

5761 listsp_meetI ~> listsp_infI

5762 listsp_meet_eq ~> listsp_inf_eq

5764 meet_min ~> inf_min

5765 join_max ~> sup_max

5767 * Added syntactic class "size"; overloaded constant "size" now has

5768 type "'a::size ==> bool"

5770 * Internal reorganisation of `size' of datatypes: size theorems

5771 "foo.size" are no longer subsumed by "foo.simps" (but are still

5772 simplification rules by default!); theorems "prod.size" now named

5773 "*.size".

5775 * Class "div" now inherits from class "times" rather than "type".

5776 INCOMPATIBILITY.

5778 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,

5779 Linorder etc. have disappeared; operations defined in terms of

5780 fold_set now are named Inf_fin, Sup_fin. INCOMPATIBILITY.

5782 * HOL/Nat: neq0_conv no longer declared as iff. INCOMPATIBILITY.

5784 * HOL-Word: New extensive library and type for generic, fixed size

5785 machine words, with arithmetic, bit-wise, shifting and rotating

5786 operations, reflection into int, nat, and bool lists, automation for

5787 linear arithmetic (by automatic reflection into nat or int), including

5788 lemmas on overflow and monotonicity. Instantiated to all appropriate

5789 arithmetic type classes, supporting automatic simplification of

5790 numerals on all operations.

5792 * Library/Boolean_Algebra: locales for abstract boolean algebras.

5794 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).

5796 * Code generator library theories:

5797 - Code_Integer represents HOL integers by big integer literals in target

5798 languages.

5799 - Code_Char represents HOL characters by character literals in target

5800 languages.

5801 - Code_Char_chr like Code_Char, but also offers treatment of character

5802 codes; includes Code_Integer.

5803 - Executable_Set allows to generate code for finite sets using lists.

5804 - Executable_Rat implements rational numbers as triples (sign, enumerator,

5805 denominator).

5806 - Executable_Real implements a subset of real numbers, namly those

5807 representable by rational numbers.

5808 - Efficient_Nat implements natural numbers by integers, which in general will

5809 result in higher efficency; pattern matching with 0/Suc is eliminated;

5810 includes Code_Integer.

5811 - Code_Index provides an additional datatype index which is mapped to

5812 target-language built-in integers.

5813 - Code_Message provides an additional datatype message_string which is isomorphic to

5814 strings; messages are mapped to target-language strings.

5816 * New package for inductive predicates

5818 An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via

5820 inductive

5821 p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"

5822 for z_1 :: U_1 and ... and z_n :: U_m

5823 where

5824 rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"

5825 | ...

5827 with full support for type-inference, rather than

5829 consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"

5831 abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"

5832 where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"

5834 inductive "s z_1 ... z_m"

5835 intros

5836 rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"

5837 ...

5839 For backward compatibility, there is a wrapper allowing inductive

5840 sets to be defined with the new package via

5842 inductive_set

5843 s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"

5844 for z_1 :: U_1 and ... and z_n :: U_m

5845 where

5846 rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"

5847 | ...

5849 or

5851 inductive_set

5852 s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"

5853 and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"

5854 for z_1 :: U_1 and ... and z_n :: U_m

5855 where

5856 "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"

5857 | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"

5858 | ...

5860 if the additional syntax "p ..." is required.

5862 Numerous examples can be found in the subdirectories src/HOL/Auth,

5863 src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.

5865 INCOMPATIBILITIES:

5867 - Since declaration and definition of inductive sets or predicates

5868 is no longer separated, abbreviations involving the newly

5869 introduced sets or predicates must be specified together with the

5870 introduction rules after the 'where' keyword (see above), rather

5871 than before the actual inductive definition.

5873 - The variables in induction and elimination rules are now

5874 quantified in the order of their occurrence in the introduction

5875 rules, rather than in alphabetical order. Since this may break

5876 some proofs, these proofs either have to be repaired, e.g. by

5877 reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'

5878 statements of the form

5880 case (rule_i a_i_1 ... a_i_{k_i})

5882 or the old order of quantification has to be restored by explicitly adding

5883 meta-level quantifiers in the introduction rules, i.e.

5885 | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"

5887 - The format of the elimination rules is now

5889 p z_1 ... z_m x_1 ... x_n ==>

5890 (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)

5891 ==> ... ==> P

5893 for predicates and

5895 (x_1, ..., x_n) : s z_1 ... z_m ==>

5896 (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)

5897 ==> ... ==> P

5899 for sets rather than

5901 x : s z_1 ... z_m ==>

5902 (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)

5903 ==> ... ==> P

5905 This may require terms in goals to be expanded to n-tuples

5906 (e.g. using case_tac or simplification with the split_paired_all

5907 rule) before the above elimination rule is applicable.

5909 - The elimination or case analysis rules for (mutually) inductive

5910 sets or predicates are now called "p_1.cases" ... "p_k.cases". The

5911 list of rules "p_1_..._p_k.elims" is no longer available.

5913 * New package "function"/"fun" for general recursive functions,

5914 supporting mutual and nested recursion, definitions in local contexts,

5915 more general pattern matching and partiality. See HOL/ex/Fundefs.thy

5916 for small examples, and the separate tutorial on the function

5917 package. The old recdef "package" is still available as before, but

5918 users are encouraged to use the new package.

5920 * Method "lexicographic_order" automatically synthesizes termination

5921 relations as lexicographic combinations of size measures.

5923 * Case-expressions allow arbitrary constructor-patterns (including

5924 "_") and take their order into account, like in functional

5925 programming. Internally, this is translated into nested

5926 case-expressions; missing cases are added and mapped to the predefined

5927 constant "undefined". In complicated cases printing may no longer show

5928 the original input but the internal form. Lambda-abstractions allow

5929 the same form of pattern matching: "% pat1 => e1 | ..." is an

5930 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new

5931 variable.

5933 * IntDef: The constant "int :: nat => int" has been removed; now "int"

5934 is an abbreviation for "of_nat :: nat => int". The simplification

5935 rules for "of_nat" have been changed to work like "int" did

5936 previously. Potential INCOMPATIBILITY:

5937 - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"

5938 - of_nat_diff and of_nat_mult are no longer default simp rules

5940 * Method "algebra" solves polynomial equations over (semi)rings using

5941 Groebner bases. The (semi)ring structure is defined by locales and the

5942 tool setup depends on that generic context. Installing the method for

5943 a specific type involves instantiating the locale and possibly adding

5944 declarations for computation on the coefficients. The method is

5945 already instantiated for natural numbers and for the axiomatic class

5946 of idoms with numerals. See also the paper by Chaieb and Wenzel at

5947 CALCULEMUS 2007 for the general principles underlying this

5948 architecture of context-aware proof-tools.

5950 * Method "ferrack" implements quantifier elimination over

5951 special-purpose dense linear orders using locales (analogous to

5952 "algebra"). The method is already installed for class

5953 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,

5954 rat, etc.

5956 * Former constant "List.op @" now named "List.append". Use ML

5957 antiquotations @{const_name List.append} or @{term " ... @ ... "} to

5958 circumvent possible incompatibilities when working on ML level.

5960 * primrec: missing cases mapped to "undefined" instead of "arbitrary".

5962 * New function listsum :: 'a list => 'a for arbitrary monoids.

5963 Special syntax: "SUM x <- xs. f x" (and latex variants)

5965 * New syntax for Haskell-like list comprehension (input only), eg.

5966 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.

5968 * The special syntax for function "filter" has changed from [x :

5969 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list

5970 comprehension syntax, and for uniformity. INCOMPATIBILITY.

5972 * [a..b] is now defined for arbitrary linear orders. It used to be

5973 defined on nat only, as an abbreviation for [a..<Suc b]

5974 INCOMPATIBILITY.

5976 * Renamed lemma "set_take_whileD" to "set_takeWhileD".

5978 * New functions "sorted" and "sort" in src/HOL/List.thy.

5980 * New lemma collection field_simps (an extension of ring_simps) for

5981 manipulating (in)equations involving division. Multiplies with all

5982 denominators that can be proved to be non-zero (in equations) or

5983 positive/negative (in inequations).

5985 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib

5986 have been improved and renamed to ring_simps, group_simps and

5987 ring_distribs. Removed lemmas field_xyz in theory Ring_and_Field

5988 because they were subsumed by lemmas xyz. INCOMPATIBILITY.

5990 * Theory Library/Commutative_Ring: switched from recdef to function

5991 package; constants add, mul, pow now curried. Infix syntax for

5992 algebraic operations.

5994 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.

5995 INCOMPATIBILITY.

5997 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.

5998 INCOMPATIBILITY.

6000 * HOL/records: generalised field-update to take a function on the

6001 field rather than the new value: r(|A := x|) is translated to A_update

6002 (K x) r The K-combinator that is internally used is called K_record.

6003 INCOMPATIBILITY: Usage of the plain update functions has to be

6004 adapted.

6006 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0

6007 * x = 0, which are required for a semiring. Richer structures do not

6008 inherit from semiring_0 anymore, because this property is a theorem

6009 there, not an axiom. INCOMPATIBILITY: In instances of semiring_0,

6010 there is more to prove, but this is mostly trivial.

6012 * Class "recpower" is generalized to arbitrary monoids, not just

6013 commutative semirings. INCOMPATIBILITY: may need to incorporate

6014 commutativity or semiring properties additionally.

6016 * Constant "List.list_all2" in List.thy now uses authentic syntax.

6017 INCOMPATIBILITY: translations containing list_all2 may go wrong,

6018 better use 'abbreviation'.

6020 * Renamed constant "List.op mem" to "List.member". INCOMPATIBILITY.

6022 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has

6023 been abandoned in favour of plain 'int'. INCOMPATIBILITY --

6024 significant changes for setting up numeral syntax for types:

6025 - New constants Numeral.pred and Numeral.succ instead

6026 of former Numeral.bin_pred and Numeral.bin_succ.

6027 - Use integer operations instead of bin_add, bin_mult and so on.

6028 - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.

6029 - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.

6031 See src/HOL/Integ/IntArith.thy for an example setup.

6033 * Command 'normal_form' computes the normal form of a term that may

6034 contain free variables. For example ``normal_form "rev [a, b, c]"''

6035 produces ``[b, c, a]'' (without proof). This command is suitable for

6036 heavy-duty computations because the functions are compiled to ML

6037 first. Correspondingly, a method "normalization" is provided. See

6038 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.

6040 * Alternative iff syntax "A <-> B" for equality on bool (with priority

6041 25 like -->); output depends on the "iff" print_mode, the default is

6042 "A = B" (with priority 50).

6044 * Relations less (<) and less_eq (<=) are also available on type bool.

6045 Modified syntax to disallow nesting without explicit parentheses,

6046 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z". Potential

6047 INCOMPATIBILITY.

6049 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).

6051 * Relation composition operator "op O" now has precedence 75 and binds

6052 stronger than union and intersection. INCOMPATIBILITY.

6054 * The old set interval syntax "{m..n(}" (and relatives) has been

6055 removed. Use "{m..<n}" (and relatives) instead.

6057 * In the context of the assumption "~(s = t)" the Simplifier rewrites

6058 "t = s" to False (by simproc "neq"). INCOMPATIBILITY, consider using

6059 ``declare [[simproc del: neq]]''.

6061 * Simplifier: "m dvd n" where m and n are numbers is evaluated to

6062 True/False.

6064 * Theorem Cons_eq_map_conv no longer declared as "simp".

6066 * Theorem setsum_mult renamed to setsum_right_distrib.

6068 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the

6069 ``rule'' method.

6071 * Reimplemented methods "sat" and "satx", with several improvements:

6072 goals no longer need to be stated as "<prems> ==> False", equivalences

6073 (i.e. "=" on type bool) are handled, variable names of the form

6074 "lit_<n>" are no longer reserved, significant speedup.

6076 * Methods "sat" and "satx" can now replay MiniSat proof traces.

6077 zChaff is still supported as well.

6079 * 'inductive' and 'datatype': provide projections of mutual rules,

6080 bundled as foo_bar.inducts;

6082 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to

6083 Library.

6085 * Library: moved theory Accessible_Part to main HOL.

6087 * Library: added theory Coinductive_List of potentially infinite lists

6088 as greatest fixed-point.

6090 * Library: added theory AssocList which implements (finite) maps as

6091 association lists.

6093 * Method "evaluation" solves goals (i.e. a boolean expression)

6094 efficiently by compiling it to ML. The goal is "proved" (via an

6095 oracle) if it evaluates to True.

6097 * Linear arithmetic now splits certain operators (e.g. min, max, abs)

6098 also when invoked by the simplifier. This results in the Simplifier

6099 being more powerful on arithmetic goals. INCOMPATIBILITY.

6100 Configuration option fast_arith_split_limit=0 recovers the old

6101 behavior.

6103 * Support for hex (0x20) and binary (0b1001) numerals.

6105 * New method: reify eqs (t), where eqs are equations for an

6106 interpretation I :: 'a list => 'b => 'c and t::'c is an optional

6107 parameter, computes a term s::'b and a list xs::'a list and proves the

6108 theorem I xs s = t. This is also known as reification or quoting. The

6109 resulting theorem is applied to the subgoal to substitute t with I xs

6110 s. If t is omitted, the subgoal itself is reified.

6112 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)

6113 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,

6114 where f is supposed to be a computable function (in the sense of code

6115 generattion). The method uses reify to compute s and xs as above then

6116 applies corr_thm and uses normalization by evaluation to "prove" f s =

6117 r and finally gets the theorem t = r, which is again applied to the

6118 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.

6120 * Reflection: Automatic reification now handels binding, an example is

6121 available in src/HOL/ex/ReflectionEx.thy

6123 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a

6124 command 'statespace' that is similar to 'record', but introduces an

6125 abstract specification based on the locale infrastructure instead of

6126 HOL types. This leads to extra flexibility in composing state spaces,

6127 in particular multiple inheritance and renaming of components.

6130 *** HOL-Complex ***

6132 * Hyperreal: Functions root and sqrt are now defined on negative real

6133 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.

6134 Nonnegativity side conditions have been removed from many lemmas, so

6135 that more subgoals may now be solved by simplification; potential

6136 INCOMPATIBILITY.

6138 * Real: new type classes formalize real normed vector spaces and

6139 algebras, using new overloaded constants scaleR :: real => 'a => 'a

6140 and norm :: 'a => real.

6142 * Real: constant of_real :: real => 'a::real_algebra_1 injects from

6143 reals into other types. The overloaded constant Reals :: 'a set is now

6144 defined as range of_real; potential INCOMPATIBILITY.

6146 * Real: proper support for ML code generation, including 'quickcheck'.

6147 Reals are implemented as arbitrary precision rationals.

6149 * Hyperreal: Several constants that previously worked only for the

6150 reals have been generalized, so they now work over arbitrary vector

6151 spaces. Type annotations may need to be added in some cases; potential

6152 INCOMPATIBILITY.

6154 Infinitesimal :: ('a::real_normed_vector) star set

6155 HFinite :: ('a::real_normed_vector) star set

6156 HInfinite :: ('a::real_normed_vector) star set

6157 approx :: ('a::real_normed_vector) star => 'a star => bool

6158 monad :: ('a::real_normed_vector) star => 'a star set

6159 galaxy :: ('a::real_normed_vector) star => 'a star set

6160 (NS)LIMSEQ :: [nat => 'a::real_normed_vector, 'a] => bool

6161 (NS)convergent :: (nat => 'a::real_normed_vector) => bool

6162 (NS)Bseq :: (nat => 'a::real_normed_vector) => bool

6163 (NS)Cauchy :: (nat => 'a::real_normed_vector) => bool

6164 (NS)LIM :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool

6165 is(NS)Cont :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool

6166 deriv :: ['a::real_normed_field => 'a, 'a, 'a] => bool

6167 sgn :: 'a::real_normed_vector => 'a

6168 exp :: 'a::{recpower,real_normed_field,banach} => 'a

6170 * Complex: Some complex-specific constants are now abbreviations for

6171 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =

6172 hnorm. Other constants have been entirely removed in favor of the

6173 polymorphic versions (INCOMPATIBILITY):

6175 approx <-- capprox

6176 HFinite <-- CFinite

6177 HInfinite <-- CInfinite

6178 Infinitesimal <-- CInfinitesimal

6179 monad <-- cmonad

6180 galaxy <-- cgalaxy

6181 (NS)LIM <-- (NS)CLIM, (NS)CRLIM

6182 is(NS)Cont <-- is(NS)Contc, is(NS)contCR

6183 (ns)deriv <-- (ns)cderiv

6186 *** HOL-Algebra ***

6188 * Formalisation of ideals and the quotient construction over rings.

6190 * Order and lattice theory no longer based on records.

6191 INCOMPATIBILITY.

6193 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->

6194 greatest_closed. INCOMPATIBILITY.

6196 * Method algebra is now set up via an attribute. For examples see

6197 Ring.thy. INCOMPATIBILITY: the method is now weaker on combinations

6198 of algebraic structures.

6200 * Renamed theory CRing to Ring.

6203 *** HOL-Nominal ***

6205 * Substantial, yet incomplete support for nominal datatypes (binding

6206 structures) based on HOL-Nominal logic. See src/HOL/Nominal and

6207 src/HOL/Nominal/Examples. Prospective users should consult

6208 http://isabelle.in.tum.de/nominal/

6211 *** ML ***

6213 * ML basics: just one true type int, which coincides with IntInf.int

6214 (even on SML/NJ).

6216 * ML within Isar: antiquotations allow to embed statically-checked

6217 formal entities in the source, referring to the context available at

6218 compile-time. For example:

6220 ML {* @{sort "{zero,one}"} *}

6221 ML {* @{typ "'a => 'b"} *}

6222 ML {* @{term "%x. x"} *}

6223 ML {* @{prop "x == y"} *}

6224 ML {* @{ctyp "'a => 'b"} *}

6225 ML {* @{cterm "%x. x"} *}

6226 ML {* @{cprop "x == y"} *}

6227 ML {* @{thm asm_rl} *}

6228 ML {* @{thms asm_rl} *}

6229 ML {* @{type_name c} *}

6230 ML {* @{type_syntax c} *}

6231 ML {* @{const_name c} *}

6232 ML {* @{const_syntax c} *}

6233 ML {* @{context} *}

6234 ML {* @{theory} *}

6235 ML {* @{theory Pure} *}

6236 ML {* @{theory_ref} *}

6237 ML {* @{theory_ref Pure} *}

6238 ML {* @{simpset} *}

6239 ML {* @{claset} *}

6240 ML {* @{clasimpset} *}

6242 The same works for sources being ``used'' within an Isar context.