NEWS
author wenzelm
Sun Aug 12 19:00:58 2007 +0200 (2007-08-12)
changeset 24234 4714e04fb8e9
parent 24213 71c57c5099d6
child 24238 ae70f95e31de
permissions -rw-r--r--
* Syntax: scope for resolving ambiguities via type-inference is now limited to individual terms;
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * More uniform information about legacy features, notably a
    10 warning/error of "Legacy feature: ...", depending on the state of the
    11 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
    12 legacy features will disappear eventually.
    13 
    14 * Theory syntax: the header format ``theory A = B + C:'' has been
    15 discontinued in favour of ``theory A imports B C begin''.  Use isatool
    16 fixheaders to convert existing theory files.  INCOMPATIBILITY.
    17 
    18 * Theory syntax: the old non-Isar theory file format has been
    19 discontinued altogether.  Note that ML proof scripts may still be used
    20 with Isar theories; migration is usually quite simple with the ML
    21 function use_legacy_bindings.  INCOMPATIBILITY.
    22 
    23 * Theory syntax: some popular names (e.g. 'class', 'declaration',
    24 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
    25 quotes.
    26 
    27 * Theory loader: be more serious about observing the static theory
    28 header specifications (including optional directories), but not the
    29 accidental file locations of previously successful loads.  The strict
    30 update policy of former update_thy is now already performed by
    31 use_thy, so the former has been removed; use_thys updates several
    32 theories simultaneously, just as 'imports' within a theory header
    33 specification, but without merging the results.  Potential
    34 INCOMPATIBILITY: may need to refine theory headers and commands
    35 ROOT.ML which depend on load order.
    36 
    37 * Theory loader: optional support for content-based file
    38 identification, instead of the traditional scheme of full physical
    39 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
    40 (cf. the system manual).  The new scheme allows to work with
    41 non-finished theories in persistent session images, such that source
    42 files may be moved later on without requiring reloads.
    43 
    44 * Theory loader: old-style ML proof scripts being *attached* to a thy
    45 file (with the same base name as the theory) are considered a legacy
    46 feature, which will disappear eventually. Even now, the theory loader no
    47 longer maintains dependencies on such files.
    48 
    49 * Syntax: the scope for resolving ambiguities via type-inference is now
    50 limited to individual terms, instead of whole simultaneous
    51 specifications as before. This greatly reduces the complexity of the
    52 syntax module and improves flexibility by separating parsing and
    53 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
    54 'fixes' etc.) are required in rare situations.
    55 
    56 * Legacy goal package: reduced interface to the bare minimum required
    57 to keep existing proof scripts running.  Most other user-level
    58 functions are now part of the OldGoals structure, which is *not* open
    59 by default (consider isatool expandshort before open OldGoals).
    60 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
    61 these tend to cause confusion about the actual goal (!) context being
    62 used here, which is not necessarily the same as the_context().
    63 
    64 * Command 'find_theorems': supports "*" wild-card in "name:"
    65 criterion; "with_dups" option.  Certain ProofGeneral versions might
    66 support a specific search form (see ProofGeneral/CHANGES).
    67 
    68 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
    69 by default, which means that "prems" (and also "fixed variables") are
    70 suppressed from proof state output.  Note that the ProofGeneral
    71 settings mechanism allows to change and save options persistently, but
    72 older versions of Isabelle will fail to start up if a negative prems
    73 limit is imposed.
    74 
    75 * Local theory targets may be specified by non-nested blocks of
    76 ``context/locale/class ... begin'' followed by ``end''.  The body may
    77 contain definitions, theorems etc., including any derived mechanism
    78 that has been implemented on top of these primitives.  This concept
    79 generalizes the existing ``theorem (in ...)'' towards more versatility
    80 and scalability.
    81 
    82 * Proof General interface: proper undo of final 'end' command;
    83 discontinued Isabelle/classic mode (ML proof scripts).
    84 
    85 
    86 *** Document preparation ***
    87 
    88 * Added antiquotation @{theory name} which prints the given name,
    89 after checking that it refers to a valid ancestor theory in the
    90 current context.
    91 
    92 * Added antiquotations @{ML_type text} and @{ML_struct text} which
    93 check the given source text as ML type/structure, printing verbatim.
    94 
    95 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
    96 "c args == rhs" given in the current context.  (Any number of
    97 arguments may be given on the LHS.)
    98 
    99 
   100 
   101 *** Pure ***
   102 
   103 * code generator: consts in 'consts_code' Isar commands are now referred
   104   to by usual term syntax (including optional type annotations).
   105 
   106 * code generator: 
   107   - Isar 'definition's, 'constdef's and primitive instance definitions are added
   108     explicitly to the table of defining equations
   109   - primitive definitions are not used as defining equations by default any longer
   110   - defining equations are now definitly restricted to meta "==" and object
   111         equality "="
   112   - HOL theories have been adopted accordingly
   113 
   114 * class_package.ML offers a combination of axclasses and locales to
   115 achieve Haskell-like type classes in Isabelle.  See
   116 HOL/ex/Classpackage.thy for examples.
   117 
   118 * Yet another code generator framework allows to generate executable
   119 code for ML and Haskell (including "class"es).  A short usage sketch:
   120 
   121     internal compilation:
   122         code_gen <list of constants (term syntax)> in SML
   123     writing SML code to a file:
   124         code_gen <list of constants (term syntax)> in SML <filename>
   125     writing OCaml code to a file:
   126         code_gen <list of constants (term syntax)> in OCaml <filename>
   127     writing Haskell code to a bunch of files:
   128         code_gen <list of constants (term syntax)> in Haskell <filename>
   129 
   130 Reasonable default setup of framework in HOL/Main.
   131 
   132 Theorem attributs for selecting and transforming function equations theorems:
   133 
   134     [code fun]:        select a theorem as function equation for a specific constant
   135     [code fun del]:    deselect a theorem as function equation for a specific constant
   136     [code inline]:     select an equation theorem for unfolding (inlining) in place
   137     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
   138 
   139 User-defined serializations (target in {SML, OCaml, Haskell}):
   140 
   141     code_const <and-list of constants (term syntax)>
   142       {(target) <and-list of const target syntax>}+
   143 
   144     code_type <and-list of type constructors>
   145       {(target) <and-list of type target syntax>}+
   146 
   147     code_instance <and-list of instances>
   148       {(target)}+
   149         where instance ::= <type constructor> :: <class>
   150 
   151     code_class <and_list of classes>
   152       {(target) <and-list of class target syntax>}+
   153         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
   154 
   155 code_instance and code_class only apply to target Haskell.
   156 
   157 See HOL theories and HOL/ex/Codegenerator*.thy for usage examples.
   158 Doc/Isar/Advanced/Codegen/ provides a tutorial.
   159 
   160 * Command 'no_translations' removes translation rules from theory
   161 syntax.
   162 
   163 * Overloaded definitions are now actually checked for acyclic
   164 dependencies.  The overloading scheme is slightly more general than
   165 that of Haskell98, although Isabelle does not demand an exact
   166 correspondence to type class and instance declarations.
   167 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
   168 exotic versions of overloading -- at the discretion of the user!
   169 
   170 Polymorphic constants are represented via type arguments, i.e. the
   171 instantiation that matches an instance against the most general
   172 declaration given in the signature.  For example, with the declaration
   173 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
   174 as c(nat).  Overloading is essentially simultaneous structural
   175 recursion over such type arguments.  Incomplete specification patterns
   176 impose global constraints on all occurrences, e.g. c('a * 'a) on the
   177 LHS means that more general c('a * 'b) will be disallowed on any RHS.
   178 Command 'print_theory' outputs the normalized system of recursive
   179 equations, see section "definitions".
   180 
   181 * Configuration options are maintained within the theory or proof
   182 context (with name and type bool/int/string), providing a very simple
   183 interface to a poor-man's version of general context data.  Tools may
   184 declare options in ML (e.g. using Attrib.config_int) and then refer to
   185 these values using Config.get etc.  Users may change options via an
   186 associated attribute of the same name.  This form of context
   187 declaration works particularly well with commands 'declare' or
   188 'using', for example ``declare [[foo = 42]]''.  Thus it has become
   189 very easy to avoid global references, which would not observe Isar
   190 toplevel undo/redo and fail to work with multithreading.
   191 
   192 Various global ML references of Pure and HOL have been turned into
   193 configuration options:
   194 
   195   Unify.search_bound		unify_search_bound
   196   Unify.trace_bound		unify_trace_bound
   197   Unify.trace_simp		unify_trace_simp
   198   Unify.trace_types		unify_trace_types
   199   Simplifier.simp_depth_limit	simp_depth_limit
   200   Blast.depth_limit		blast_depth_limit
   201   DatatypeProp.dtK		datatype_distinctness_limit
   202   fast_arith_neq_limit  	fast_arith_neq_limit
   203   fast_arith_split_limit	fast_arith_split_limit
   204 
   205 * Named collections of theorems may be easily installed as context
   206 data using the functor NamedThmsFun (see
   207 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
   208 attributes; there is also a toplevel print command.  This facility is
   209 just a common case of general context data, which is the preferred way
   210 for anything more complex than just a list of facts in canonical
   211 order.
   212 
   213 * Isar: command 'declaration' augments a local theory by generic
   214 declaration functions written in ML.  This enables arbitrary content
   215 being added to the context, depending on a morphism that tells the
   216 difference of the original declaration context wrt. the application
   217 context encountered later on.
   218 
   219 * Isar: proper interfaces for simplification procedures.  Command
   220 'simproc_setup' declares named simprocs (with match patterns, and body
   221 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
   222 context.  ML antiquotation @{simproc name} retrieves named simprocs.
   223 
   224 * Isar: an extra pair of brackets around attribute declarations
   225 abbreviates a theorem reference involving an internal dummy fact,
   226 which will be ignored later --- only the effect of the attribute on
   227 the background context will persist.  This form of in-place
   228 declarations is particularly useful with commands like 'declare' and
   229 'using', for example ``have A using [[simproc a]] by simp''.
   230 
   231 * Isar: method "assumption" (and implicit closing of subproofs) now
   232 takes simple non-atomic goal assumptions into account: after applying
   233 an assumption as a rule the resulting subgoals are solved by atomic
   234 assumption steps.  This is particularly useful to finish 'obtain'
   235 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
   236 without referring to the original premise "!!x. P x ==> thesis" in the
   237 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
   238 more permissive.
   239 
   240 * Isar: implicit use of prems from the Isar proof context is
   241 considered a legacy feature.  Common applications like ``have A .''
   242 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
   243 referencing facts explicitly here improves readability and
   244 maintainability of proof texts.
   245 
   246 * Isar: improper proof element 'guess' is like 'obtain', but derives
   247 the obtained context from the course of reasoning!  For example:
   248 
   249   assume "EX x y. A x & B y"   -- "any previous fact"
   250   then guess x and y by clarify
   251 
   252 This technique is potentially adventurous, depending on the facts and
   253 proof tools being involved here.
   254 
   255 * Isar: known facts from the proof context may be specified as literal
   256 propositions, using ASCII back-quote syntax.  This works wherever
   257 named facts used to be allowed so far, in proof commands, proof
   258 methods, attributes etc.  Literal facts are retrieved from the context
   259 according to unification of type and term parameters.  For example,
   260 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
   261 theorems in the current context, then these are valid literal facts:
   262 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
   263 
   264 There is also a proof method "fact" which does the same composition
   265 for explicit goal states, e.g. the following proof texts coincide with
   266 certain special cases of literal facts:
   267 
   268   have "A" by fact                 ==  note `A`
   269   have "A ==> B" by fact           ==  note `A ==> B`
   270   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
   271   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
   272 
   273 * Isar: ":" (colon) is no longer a symbolic identifier character in
   274 outer syntax.  Thus symbolic identifiers may be used without
   275 additional white space in declarations like this: ``assume *: A''.
   276 
   277 * Isar: 'print_facts' prints all local facts of the current context,
   278 both named and unnamed ones.
   279 
   280 * Isar: 'def' now admits simultaneous definitions, e.g.:
   281 
   282   def x == "t" and y == "u"
   283 
   284 * Isar: added command 'unfolding', which is structurally similar to
   285 'using', but affects both the goal state and facts by unfolding given
   286 rewrite rules.  Thus many occurrences of the 'unfold' method or
   287 'unfolded' attribute may be replaced by first-class proof text.
   288 
   289 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
   290 and command 'unfolding' now all support object-level equalities
   291 (potentially conditional).  The underlying notion of rewrite rule is
   292 analogous to the 'rule_format' attribute, but *not* that of the
   293 Simplifier (which is usually more generous).
   294 
   295 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
   296 method expression within a sandbox consisting of the first N
   297 sub-goals, which need to exist.  For example, ``simp_all [3]''
   298 simplifies the first three sub-goals, while (rule foo, simp_all)[]
   299 simplifies all new goals that emerge from applying rule foo to the
   300 originally first one.
   301 
   302 * Isar: schematic goals are no longer restricted to higher-order
   303 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
   304 expected.
   305 
   306 * Isar: the conclusion of a long theorem statement is now either
   307 'shows' (a simultaneous conjunction, as before), or 'obtains'
   308 (essentially a disjunction of cases with local parameters and
   309 assumptions).  The latter allows to express general elimination rules
   310 adequately; in this notation common elimination rules look like this:
   311 
   312   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
   313     assumes "EX x. P x"
   314     obtains x where "P x"
   315 
   316   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
   317     assumes "A & B"
   318     obtains A and B
   319 
   320   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
   321     assumes "A | B"
   322     obtains
   323       A
   324     | B
   325 
   326 The subsequent classical rules even refer to the formal "thesis"
   327 explicitly:
   328 
   329   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
   330     obtains "~ thesis"
   331 
   332   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
   333     obtains "thesis ==> something"
   334 
   335 The actual proof of an 'obtains' statement is analogous to that of the
   336 Isar proof element 'obtain', only that there may be several cases.
   337 Optional case names may be specified in parentheses; these will be
   338 available both in the present proof and as annotations in the
   339 resulting rule, for later use with the 'cases' method (cf. attribute
   340 case_names).
   341 
   342 * Isar: the assumptions of a long theorem statement are available as
   343 "assms" fact in the proof context.  This is more appropriate than the
   344 (historical) "prems", which refers to all assumptions of the current
   345 context, including those from the target locale, proof body etc.
   346 
   347 * Isar: 'print_statement' prints theorems from the current theory or
   348 proof context in long statement form, according to the syntax of a
   349 top-level lemma.
   350 
   351 * Isar: 'obtain' takes an optional case name for the local context
   352 introduction rule (default "that").
   353 
   354 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
   355 explicit (is "_ ==> ?foo") in the rare cases where this still happens
   356 to occur.
   357 
   358 * Pure: syntax "CONST name" produces a fully internalized constant
   359 according to the current context.  This is particularly useful for
   360 syntax translations that should refer to internal constant
   361 representations independently of name spaces.
   362 
   363 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
   364 instead of "FOO ". This allows multiple binder declarations to coexist
   365 in the same context.  INCOMPATIBILITY.
   366 
   367 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
   368 primitive that also works in a locale context (both for constants and
   369 fixed variables).  Type declaration and internal syntactic
   370 representation of given constants retrieved from the context.
   371 
   372 * Isar/locales: new derived specification elements 'axiomatization',
   373 'definition', 'abbreviation', which support type-inference, admit
   374 object-level specifications (equality, equivalence).  See also the
   375 isar-ref manual.  Examples:
   376 
   377   axiomatization
   378     eq  (infix "===" 50) where
   379     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
   380 
   381   definition "f x y = x + y + 1"
   382   definition g where "g x = f x x"
   383 
   384   abbreviation
   385     neq  (infix "=!=" 50) where
   386     "x =!= y == ~ (x === y)"
   387 
   388 These specifications may be also used in a locale context.  Then the
   389 constants being introduced depend on certain fixed parameters, and the
   390 constant name is qualified by the locale base name.  An internal
   391 abbreviation takes care for convenient input and output, making the
   392 parameters implicit and using the original short name.  See also
   393 HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
   394 entities from a monomorphic theory.
   395 
   396 Presently, abbreviations are only available 'in' a target locale, but
   397 not inherited by general import expressions.  Also note that
   398 'abbreviation' may be used as a type-safe replacement for 'syntax' +
   399 'translations' in common applications.
   400 
   401 Concrete syntax is attached to specified constants in internal form,
   402 independently of name spaces.  The parse tree representation is
   403 slightly different -- use 'notation' instead of raw 'syntax', and
   404 'translations' with explicit "CONST" markup to accommodate this.
   405 
   406 * Pure: command 'print_abbrevs' prints all constant abbreviations of
   407 the current context.  Print mode "no_abbrevs" prevents inversion of
   408 abbreviations on output.
   409 
   410 * Isar/locales: improved parameter handling:
   411 - use of locales "var" and "struct" no longer necessary;
   412 - parameter renamings are no longer required to be injective.
   413   This enables, for example, to define a locale for endomorphisms thus:
   414   locale endom = homom mult mult h.
   415 
   416 * Isar/locales: changed the way locales with predicates are defined.
   417 Instead of accumulating the specification, the imported expression is
   418 now an interpretation.  INCOMPATIBILITY: different normal form of
   419 locale expressions.  In particular, in interpretations of locales with
   420 predicates, goals repesenting already interpreted fragments are not
   421 removed automatically.  Use methods `intro_locales' and
   422 `unfold_locales'; see below.
   423 
   424 * Isar/locales: new methods `intro_locales' and `unfold_locales'
   425 provide backward reasoning on locales predicates.  The methods are
   426 aware of interpretations and discharge corresponding goals.
   427 `intro_locales' is less aggressive then `unfold_locales' and does not
   428 unfold predicates to assumptions.
   429 
   430 * Isar/locales: the order in which locale fragments are accumulated
   431 has changed.  This enables to override declarations from fragments due
   432 to interpretations -- for example, unwanted simp rules.
   433 
   434 * Isar/locales: interpretation in theories and proof contexts has been
   435 extended.  One may now specify (and prove) equations, which are
   436 unfolded in interpreted theorems.  This is useful for replacing
   437 defined concepts (constants depending on locale parameters) by
   438 concepts already existing in the target context.  Example:
   439 
   440   interpretation partial_order ["op <= :: [int, int] => bool"]
   441     where "partial_order.less (op <=) (x::int) y = (x < y)"
   442 
   443 Typically, the constant `partial_order.less' is created by a definition
   444 specification element in the context of locale partial_order.
   445 
   446 * Provers/induct: improved internal context management to support
   447 local fixes and defines on-the-fly.  Thus explicit meta-level
   448 connectives !! and ==> are rarely required anymore in inductive goals
   449 (using object-logic connectives for this purpose has been long
   450 obsolete anyway).  The subsequent proof patterns illustrate advanced
   451 techniques of natural induction; general datatypes and inductive sets
   452 work analogously (see also src/HOL/Lambda for realistic examples).
   453 
   454 (1) This is how to ``strengthen'' an inductive goal wrt. certain
   455 parameters:
   456 
   457   lemma
   458     fixes n :: nat and x :: 'a
   459     assumes a: "A n x"
   460     shows "P n x"
   461     using a                     -- {* make induct insert fact a *}
   462   proof (induct n arbitrary: x) -- {* generalize goal to "!!x. A n x ==> P n x" *}
   463     case 0
   464     show ?case sorry
   465   next
   466     case (Suc n)
   467     note `!!x. A n x ==> P n x` -- {* induction hypothesis, according to induction rule *}
   468     note `A (Suc n) x`          -- {* induction premise, stemming from fact a *}
   469     show ?case sorry
   470   qed
   471 
   472 (2) This is how to perform induction over ``expressions of a certain
   473 form'', using a locally defined inductive parameter n == "a x"
   474 together with strengthening (the latter is usually required to get
   475 sufficiently flexible induction hypotheses):
   476 
   477   lemma
   478     fixes a :: "'a => nat"
   479     assumes a: "A (a x)"
   480     shows "P (a x)"
   481     using a
   482   proof (induct n == "a x" arbitrary: x)
   483     ...
   484 
   485 See also HOL/Isar_examples/Puzzle.thy for an application of the this
   486 particular technique.
   487 
   488 (3) This is how to perform existential reasoning ('obtains' or
   489 'obtain') by induction, while avoiding explicit object-logic
   490 encodings:
   491 
   492   lemma
   493     fixes n :: nat
   494     obtains x :: 'a where "P n x" and "Q n x"
   495   proof (induct n arbitrary: thesis)
   496     case 0
   497     obtain x where "P 0 x" and "Q 0 x" sorry
   498     then show thesis by (rule 0)
   499   next
   500     case (Suc n)
   501     obtain x where "P n x" and "Q n x" by (rule Suc.hyps)
   502     obtain x where "P (Suc n) x" and "Q (Suc n) x" sorry
   503     then show thesis by (rule Suc.prems)
   504   qed
   505 
   506 Here the 'arbitrary: thesis' specification essentially modifies the
   507 scope of the formal thesis parameter, in order to the get the whole
   508 existence statement through the induction as expected.
   509 
   510 * Provers/induct: mutual induction rules are now specified as a list
   511 of rule sharing the same induction cases.  HOL packages usually
   512 provide foo_bar.inducts for mutually defined items foo and bar
   513 (e.g. inductive sets or datatypes).  INCOMPATIBILITY, users need to
   514 specify mutual induction rules differently, i.e. like this:
   515 
   516   (induct rule: foo_bar.inducts)
   517   (induct set: foo bar)
   518   (induct type: foo bar)
   519 
   520 The ML function ProjectRule.projections turns old-style rules into the
   521 new format.
   522 
   523 * Provers/induct: improved handling of simultaneous goals.  Instead of
   524 introducing object-level conjunction, the statement is now split into
   525 several conclusions, while the corresponding symbolic cases are
   526 nested accordingly.  INCOMPATIBILITY, proofs need to be structured
   527 explicitly.  For example:
   528 
   529   lemma
   530     fixes n :: nat
   531     shows "P n" and "Q n"
   532   proof (induct n)
   533     case 0 case 1
   534     show "P 0" sorry
   535   next
   536     case 0 case 2
   537     show "Q 0" sorry
   538   next
   539     case (Suc n) case 1
   540     note `P n` and `Q n`
   541     show "P (Suc n)" sorry
   542   next
   543     case (Suc n) case 2
   544     note `P n` and `Q n`
   545     show "Q (Suc n)" sorry
   546   qed
   547 
   548 The split into subcases may be deferred as follows -- this is
   549 particularly relevant for goal statements with local premises.
   550 
   551   lemma
   552     fixes n :: nat
   553     shows "A n ==> P n" and "B n ==> Q n"
   554   proof (induct n)
   555     case 0
   556     {
   557       case 1
   558       note `A 0`
   559       show "P 0" sorry
   560     next
   561       case 2
   562       note `B 0`
   563       show "Q 0" sorry
   564     }
   565   next
   566     case (Suc n)
   567     note `A n ==> P n` and `B n ==> Q n`
   568     {
   569       case 1
   570       note `A (Suc n)`
   571       show "P (Suc n)" sorry
   572     next
   573       case 2
   574       note `B (Suc n)`
   575       show "Q (Suc n)" sorry
   576     }
   577   qed
   578 
   579 If simultaneous goals are to be used with mutual rules, the statement
   580 needs to be structured carefully as a two-level conjunction, using
   581 lists of propositions separated by 'and':
   582 
   583   lemma
   584     shows "a : A ==> P1 a"
   585           "a : A ==> P2 a"
   586       and "b : B ==> Q1 b"
   587           "b : B ==> Q2 b"
   588           "b : B ==> Q3 b"
   589   proof (induct set: A B)
   590 
   591 * Provers/induct: support coinduction as well.  See
   592 src/HOL/Library/Coinductive_List.thy for various examples.
   593 
   594 * Attribute "symmetric" produces result with standardized schematic
   595 variables (index 0).  Potential INCOMPATIBILITY.
   596 
   597 * Simplifier: by default the simplifier trace only shows top level
   598 rewrites now. That is, trace_simp_depth_limit is set to 1 by
   599 default. Thus there is less danger of being flooded by the trace. The
   600 trace indicates where parts have been suppressed.
   601   
   602 * Provers/classical: removed obsolete classical version of elim_format
   603 attribute; classical elim/dest rules are now treated uniformly when
   604 manipulating the claset.
   605 
   606 * Provers/classical: stricter checks to ensure that supplied intro,
   607 dest and elim rules are well-formed; dest and elim rules must have at
   608 least one premise.
   609 
   610 * Provers/classical: attributes dest/elim/intro take an optional
   611 weight argument for the rule (just as the Pure versions).  Weights are
   612 ignored by automated tools, but determine the search order of single
   613 rule steps.
   614 
   615 * Syntax: input syntax now supports dummy variable binding "%_. b",
   616 where the body does not mention the bound variable.  Note that dummy
   617 patterns implicitly depend on their context of bounds, which makes
   618 "{_. _}" match any set comprehension as expected.  Potential
   619 INCOMPATIBILITY -- parse translations need to cope with syntactic
   620 constant "_idtdummy" in the binding position.
   621 
   622 * Syntax: removed obsolete syntactic constant "_K" and its associated
   623 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
   624 for example "A -> B" => "Pi A (%_. B)".
   625 
   626 * Pure: 'class_deps' command visualizes the subclass relation, using
   627 the graph browser tool.
   628 
   629 * Pure: 'print_theory' now suppresses entities with internal name
   630 (trailing "_") by default; use '!' option for full details.
   631 
   632 
   633 *** HOL ***
   634 
   635 * Code generator library theories:
   636   * Pretty_Int represents HOL integers by big integer literals in target
   637     languages.
   638   * Pretty_Char represents HOL characters by character literals in target
   639     languages.
   640   * Pretty_Char_chr like Pretty_Char, but also offers treatment of character
   641     codes; includes Pretty_Int.
   642   * Executable_Set allows to generate code for finite sets using lists.
   643   * Executable_Rat implements rational numbers as triples (sign, enumerator,
   644     denominator).
   645   * Executable_Real implements a subset of real numbers, namly those
   646     representable by rational numbers.
   647   * Efficient_Nat implements natural numbers by integers, which in general will
   648     result in higher efficency; pattern matching with 0/Suc is eliminated;
   649     includes Pretty_Int.
   650   * ML_String provides an additional datatype ml_string; in the HOL default
   651     setup, strings in HOL are mapped to lists of HOL characters in SML; values
   652     of type ml_string are mapped to strings in SML.
   653   * ML_Int provides an additional datatype ml_int which is mapped to to SML
   654     built-in integers.
   655 
   656 * New package for inductive predicates
   657 
   658   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
   659 
   660     inductive
   661       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
   662       for z_1 :: U_1 and ... and z_n :: U_m
   663     where
   664       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
   665     | ...
   666 
   667   rather than
   668 
   669     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
   670 
   671     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
   672     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
   673 
   674     inductive "s z_1 ... z_m"
   675     intros
   676       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
   677       ...
   678 
   679   For backward compatibility, there is a wrapper allowing inductive
   680   sets to be defined with the new package via
   681 
   682     inductive_set
   683       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
   684       for z_1 :: U_1 and ... and z_n :: U_m
   685     where
   686       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
   687     | ...
   688 
   689   or
   690 
   691     inductive_set
   692       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
   693       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
   694       for z_1 :: U_1 and ... and z_n :: U_m
   695     where
   696       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
   697     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
   698     | ...
   699 
   700   if the additional syntax "p ..." is required.
   701 
   702   Many examples can be found in the subdirectories Auth, Bali, Induct,
   703   or MicroJava.
   704 
   705   INCOMPATIBILITIES:
   706 
   707   - Since declaration and definition of inductive sets or predicates
   708     is no longer separated, abbreviations involving the newly introduced
   709     sets or predicates must be specified together with the introduction
   710     rules after the "where" keyword (see example above), rather than before
   711     the actual inductive definition.
   712 
   713   - The variables in induction and elimination rules are now quantified
   714     in the order of their occurrence in the introduction rules, rather than
   715     in alphabetical order. Since this may break some proofs, these proofs
   716     either have to be repaired, e.g. by reordering the variables
   717     a_i_1 ... a_i_{k_i} in Isar "case" statements of the form
   718 
   719       case (rule_i a_i_1 ... a_i_{k_i})
   720 
   721     or the old order of quantification has to be restored by explicitly adding
   722     meta-level quantifiers in the introduction rules, i.e.
   723 
   724       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
   725 
   726   - The format of the elimination rules is now
   727 
   728       p z_1 ... z_m x_1 ... x_n ==>
   729         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
   730         ==> ... ==> P
   731 
   732     for predicates and
   733 
   734       (x_1, ..., x_n) : s z_1 ... z_m ==>
   735         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
   736         ==> ... ==> P
   737 
   738     for sets rather than
   739 
   740       x : s z_1 ... z_m ==>
   741         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
   742         ==> ... ==> P
   743 
   744     This may require terms in goals to be expanded to n-tuples (e.g. using case_tac
   745     or simplification with the split_paired_all rule) before the above elimination
   746     rule is applicable.
   747 
   748   - The elimination or case analysis rules for (mutually) inductive sets or
   749     predicates are now called "p_1.cases" ... "p_k.cases". The list of rules
   750     "p_1_..._p_k.elims" is no longer available.
   751 
   752 * Method "metis" proves goals by applying the Metis general-purpose
   753 resolution prover.  Examples are in the directory MetisExamples.  See
   754 also http://gilith.com/software/metis/
   755   
   756 * Command 'sledgehammer' invokes external automatic theorem provers as
   757 background processes.  It generates calls to the "metis" method if
   758 successful. These can be pasted into the proof.  Users do not have to
   759 wait for the automatic provers to return.
   760 
   761 * Case-expressions allow arbitrary constructor-patterns (including "_") and
   762   take their order into account, like in functional programming.
   763   Internally, this is translated into nested case-expressions; missing cases
   764   are added and mapped to the predefined constant "undefined". In complicated
   765   cases printing may no longer show the original input but the internal
   766   form. Lambda-abstractions allow the same form of pattern matching:
   767   "% pat1 => e1 | ..." is an abbreviation for
   768   "%x. case x of pat1 => e1 | ..." where x is a new variable.
   769 
   770 * IntDef: The constant "int :: nat => int" has been removed; now "int"
   771   is an abbreviation for "of_nat :: nat => int". The simplification rules
   772   for "of_nat" have been changed to work like "int" did previously.
   773   (potential INCOMPATIBILITY)
   774   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
   775   - of_nat_diff and of_nat_mult are no longer default simp rules
   776 
   777 * Method "algebra" solves polynomial equations over (semi)rings using
   778   Groebner bases. The (semi)ring structure is defined by locales and
   779   the tool setup depends on that generic context. Installing the
   780   method for a specific type involves instantiating the locale and
   781   possibly adding declarations for computation on the coefficients.
   782   The method is already instantiated for natural numbers and for the
   783   axiomatic class of idoms with numerals.  See also the paper by
   784   Chaieb and Wenzel at CALCULEMUS 2007 for the general principles
   785   underlying this architecture of context-aware proof-tools.
   786 
   787 * constant "List.op @" now named "List.append".  Use ML antiquotations
   788 @{const_name List.append} or @{term " ... @ ... "} to circumvent
   789 possible incompatibilities when working on ML level.
   790 
   791 * Constant renames due to introduction of canonical name prefixing for
   792   class package:
   793 
   794     HOL.abs ~> HOL.minus_class.abs
   795     HOL.divide ~> HOL.divide_class.divide
   796     Nat.power ~> Nat.power_class.power
   797     Nat.size ~> Nat.size_class.size
   798     Numeral.number_of ~> Numeral.number_class.number_of
   799     FixedPoint.Inf ~> FixedPoint.complete_lattice_class.Inf
   800 
   801 * Rudimentary class target mechanism involves constant renames:
   802 
   803     Orderings.min ~> Orderings.ord_class.min
   804     Orderings.max ~> Orderings.ord_class.max
   805     FixedPoint.Sup ~> FixedPoint.complete_lattice_class.Sup
   806 
   807 * primrec: missing cases mapped to "undefined" instead of "arbitrary"
   808 
   809 * new constant "undefined" with axiom "undefined x = undefined"
   810 
   811 * new class "default" with associated constant "default"
   812 
   813 * new function listsum :: 'a list => 'a for arbitrary monoids.
   814   Special syntax: "SUM x <- xs. f x" (and latex variants)
   815 
   816 * new (input only) syntax for Haskell-like list comprehension, eg
   817   [(x,y). x <- xs, y <- ys, x ~= y]
   818   For details see List.thy.
   819 
   820 * The special syntax for function "filter" has changed from [x : xs. P] to
   821   [x <- xs. P] to avoid an ambiguity caused by list comprehension syntax,
   822   and for uniformity. INCOMPATIBILITY
   823 
   824 * Lemma "set_take_whileD" renamed to "set_takeWhileD"
   825 
   826 * New lemma collection field_simps (an extension of ring_simps)
   827   for manipulating (in)equations involving division. Multiplies
   828   with all denominators that can be proved to be non-zero (in equations)
   829   or positive/negative (in inequations).
   830 
   831 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
   832   have been improved and renamed to ring_simps, group_simps and ring_distribs.
   833   Removed lemmas field_xyz in Ring_and_Field
   834   because they were subsumed by lemmas xyz.
   835 INCOMPATIBILITY.
   836 
   837 * Library/Pretty_Int.thy: maps HOL numerals on target language integer literals
   838   when generating code.
   839 
   840 * Library/Pretty_Char.thy: maps HOL characters on target language character literals
   841   when generating code.
   842 
   843 * Library/Commutative_Ring.thy: switched from recdef to function package;
   844   constants add, mul, pow now curried.  Infix syntax for algebraic operations.
   845 
   846 * Some steps towards more uniform lattice theory development in HOL.
   847 
   848     constants "meet" and "join" now named "inf" and "sup"
   849     constant "Meet" now named "Inf"
   850 
   851     classes "meet_semilorder" and "join_semilorder" now named
   852       "lower_semilattice" and "upper_semilattice"
   853     class "lorder" now named "lattice"
   854     class "comp_lat" now named "complete_lattice"
   855 
   856     Instantiation of lattice classes allows explicit definitions
   857     for "inf" and "sup" operations.
   858 
   859   INCOMPATIBILITY.  Theorem renames:
   860 
   861     meet_left_le            ~> inf_le1
   862     meet_right_le           ~> inf_le2
   863     join_left_le            ~> sup_ge1
   864     join_right_le           ~> sup_ge2
   865     meet_join_le            ~> inf_sup_ord
   866     le_meetI                ~> le_infI
   867     join_leI                ~> le_supI
   868     le_meet                 ~> le_inf_iff
   869     le_join                 ~> ge_sup_conv
   870     meet_idempotent         ~> inf_idem
   871     join_idempotent         ~> sup_idem
   872     meet_comm               ~> inf_commute
   873     join_comm               ~> sup_commute
   874     meet_leI1               ~> le_infI1
   875     meet_leI2               ~> le_infI2
   876     le_joinI1               ~> le_supI1
   877     le_joinI2               ~> le_supI2
   878     meet_assoc              ~> inf_assoc
   879     join_assoc              ~> sup_assoc
   880     meet_left_comm          ~> inf_left_commute
   881     meet_left_idempotent    ~> inf_left_idem
   882     join_left_comm          ~> sup_left_commute
   883     join_left_idempotent    ~> sup_left_idem
   884     meet_aci                ~> inf_aci
   885     join_aci                ~> sup_aci
   886     le_def_meet             ~> le_iff_inf
   887     le_def_join             ~> le_iff_sup
   888     join_absorp2            ~> sup_absorb2
   889     join_absorp1            ~> sup_absorb1
   890     meet_absorp1            ~> inf_absorb1
   891     meet_absorp2            ~> inf_absorb2
   892     meet_join_absorp        ~> inf_sup_absorb
   893     join_meet_absorp        ~> sup_inf_absorb
   894     distrib_join_le         ~> distrib_sup_le
   895     distrib_meet_le         ~> distrib_inf_le
   896 
   897     add_meet_distrib_left   ~> add_inf_distrib_left
   898     add_join_distrib_left   ~> add_sup_distrib_left
   899     is_join_neg_meet        ~> is_join_neg_inf
   900     is_meet_neg_join        ~> is_meet_neg_sup
   901     add_meet_distrib_right  ~> add_inf_distrib_right
   902     add_join_distrib_right  ~> add_sup_distrib_right
   903     add_meet_join_distribs  ~> add_sup_inf_distribs
   904     join_eq_neg_meet        ~> sup_eq_neg_inf
   905     meet_eq_neg_join        ~> inf_eq_neg_sup
   906     add_eq_meet_join        ~> add_eq_inf_sup
   907     meet_0_imp_0            ~> inf_0_imp_0
   908     join_0_imp_0            ~> sup_0_imp_0
   909     meet_0_eq_0             ~> inf_0_eq_0
   910     join_0_eq_0             ~> sup_0_eq_0
   911     neg_meet_eq_join        ~> neg_inf_eq_sup
   912     neg_join_eq_meet        ~> neg_sup_eq_inf
   913     join_eq_if              ~> sup_eq_if
   914 
   915     mono_meet               ~> mono_inf
   916     mono_join               ~> mono_sup
   917     meet_bool_eq            ~> inf_bool_eq
   918     join_bool_eq            ~> sup_bool_eq
   919     meet_fun_eq             ~> inf_fun_eq
   920     join_fun_eq             ~> sup_fun_eq
   921     meet_set_eq             ~> inf_set_eq
   922     join_set_eq             ~> sup_set_eq
   923     meet1_iff               ~> inf1_iff
   924     meet2_iff               ~> inf2_iff
   925     meet1I                  ~> inf1I
   926     meet2I                  ~> inf2I
   927     meet1D1                 ~> inf1D1
   928     meet2D1                 ~> inf2D1
   929     meet1D2                 ~> inf1D2
   930     meet2D2                 ~> inf2D2
   931     meet1E                  ~> inf1E
   932     meet2E                  ~> inf2E
   933     join1_iff               ~> sup1_iff
   934     join2_iff               ~> sup2_iff
   935     join1I1                 ~> sup1I1
   936     join2I1                 ~> sup2I1
   937     join1I1                 ~> sup1I1
   938     join2I2                 ~> sup1I2
   939     join1CI                 ~> sup1CI
   940     join2CI                 ~> sup2CI
   941     join1E                  ~> sup1E
   942     join2E                  ~> sup2E
   943 
   944     is_meet_Meet            ~> is_meet_Inf
   945     Meet_bool_def           ~> Inf_bool_def
   946     Meet_fun_def            ~> Inf_fun_def
   947     Meet_greatest           ~> Inf_greatest
   948     Meet_lower              ~> Inf_lower
   949     Meet_set_def            ~> Inf_set_def
   950 
   951     listsp_meetI            ~> listsp_infI
   952     listsp_meet_eq          ~> listsp_inf_eq
   953 
   954     meet_min                ~> inf_min
   955     join_max                ~> sup_max
   956 
   957 * Classes "order" and "linorder": facts "refl", "trans" and
   958 "cases" renamed ro "order_refl", "order_trans" and "linorder_cases", to
   959 avoid clashes with HOL "refl" and "trans". INCOMPATIBILITY.
   960 
   961 * Classes "order" and "linorder": 
   962 potential INCOMPATIBILITY: order of proof goals in order/linorder instance
   963 proofs changed.
   964 
   965 * Dropped lemma duplicate def_imp_eq in favor of meta_eq_to_obj_eq.
   966 INCOMPATIBILITY.
   967 
   968 * Dropped lemma duplicate if_def2 in favor of if_bool_eq_conj.
   969 INCOMPATIBILITY.
   970 
   971 * Added syntactic class "size"; overloaded constant "size" now has
   972 type "'a::size ==> bool"
   973 
   974 * Renamed constants "Divides.op div", "Divides.op mod" and "Divides.op
   975 dvd" to "Divides.div_class.div", "Divides.div_class.mod" and "Divides.dvd". INCOMPATIBILITY.
   976 
   977 * Added method "lexicographic_order" automatically synthesizes
   978 termination relations as lexicographic combinations of size measures
   979 -- 'function' package.
   980 
   981 * HOL/records: generalised field-update to take a function on the
   982 field rather than the new value: r(|A := x|) is translated to A_update
   983 (K x) r The K-combinator that is internally used is called K_record.
   984 INCOMPATIBILITY: Usage of the plain update functions has to be
   985 adapted.
   986  
   987 * axclass "semiring_0" now contains annihilation axioms x * 0 = 0 and
   988 0 * x = 0, which are required for a semiring.  Richer structures do
   989 not inherit from semiring_0 anymore, because this property is a
   990 theorem there, not an axiom.  INCOMPATIBILITY: In instances of
   991 semiring_0, there is more to prove, but this is mostly trivial.
   992 
   993 * axclass "recpower" was generalized to arbitrary monoids, not just
   994 commutative semirings.  INCOMPATIBILITY: If you use recpower and need
   995 commutativity or a semiring property, add the corresponding classes.
   996 
   997 * Unified locale partial_order with class definition (cf. theory
   998 Orderings), added parameter ``less''.  INCOMPATIBILITY.
   999 
  1000 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  1001 INCOMPATIBILITY: translations containing list_all2 may go wrong.  On
  1002 Isar level, use abbreviations instead.
  1003 
  1004 * Renamed constant "List.op mem" to "List.memberl" INCOMPATIBILITY:
  1005 rarely occuring name references (e.g. ``List.op mem.simps'') require
  1006 renaming (e.g. ``List.memberl.simps'').
  1007 
  1008 * Renamed constants "0" to "HOL.zero_class.zero" and "1" to "HOL.one_class.one".
  1009 INCOMPATIBILITY.
  1010 
  1011 * Added class "HOL.eq", allowing for code generation with polymorphic equality.
  1012 
  1013 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  1014 been abandoned in favour of plain 'int'. INCOMPATIBILITY --
  1015 significant changes for setting up numeral syntax for types:
  1016 
  1017   - new constants Numeral.pred and Numeral.succ instead
  1018       of former Numeral.bin_pred and Numeral.bin_succ.
  1019   - Use integer operations instead of bin_add, bin_mult and so on.
  1020   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  1021   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  1022 
  1023 See HOL/Integ/IntArith.thy for an example setup.
  1024 
  1025 * New top level command 'normal_form' computes the normal form of a
  1026 term that may contain free variables. For example ``normal_form
  1027 "rev[a,b,c]"'' produces ``[b,c,a]'' (without proof).  This command is
  1028 suitable for heavy-duty computations because the functions are
  1029 compiled to ML first.
  1030 
  1031 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  1032 25 like -->); output depends on the "iff" print_mode, the default is
  1033 "A = B" (with priority 50).
  1034 
  1035 * Renamed constants in HOL.thy and Orderings.thy:
  1036     op +   ~> HOL.plus_class.plus
  1037     op -   ~> HOL.minus_class.minus
  1038     uminus ~> HOL.minus_class.uminus
  1039     abs    ~> HOL.abs_class.abs
  1040     op *   ~> HOL.times_class.times
  1041     op <   ~> HOL.ord_class.less
  1042     op <=  ~> HOL.ord_class.less_eq
  1043 
  1044 Adaptions may be required in the following cases:
  1045 
  1046 a) User-defined constants using any of the names "plus", "minus", "times",
  1047 "less" or "less_eq". The standard syntax translations for "+", "-" and "*"
  1048 may go wrong.
  1049 INCOMPATIBILITY: use more specific names.
  1050 
  1051 b) Variables named "plus", "minus", "times", "less", "less_eq"
  1052 INCOMPATIBILITY: use more specific names.
  1053 
  1054 c) Permutative equations (e.g. "a + b = b + a")
  1055 Since the change of names also changes the order of terms, permutative
  1056 rewrite rules may get applied in a different order. Experience shows that
  1057 this is rarely the case (only two adaptions in the whole Isabelle
  1058 distribution).
  1059 INCOMPATIBILITY: rewrite proofs
  1060 
  1061 d) ML code directly refering to constant names
  1062 This in general only affects hand-written proof tactics, simprocs and so on.
  1063 INCOMPATIBILITY: grep your sourcecode and replace names.  Consider use
  1064 of const_name ML antiquotations.
  1065 
  1066 * Relations less (<) and less_eq (<=) are also available on type bool.
  1067 Modified syntax to disallow nesting without explicit parentheses,
  1068 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".
  1069 
  1070 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  1071 
  1072 * Relation composition operator "op O" now has precedence 75 and binds
  1073 stronger than union and intersection. INCOMPATIBILITY.
  1074 
  1075 * The old set interval syntax "{m..n(}" (and relatives) has been
  1076 removed.  Use "{m..<n}" (and relatives) instead.
  1077 
  1078 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  1079 "t = s" to False (by simproc "neq_simproc").  For backward
  1080 compatibility this can be disabled by ML "reset use_neq_simproc".
  1081 
  1082 * "m dvd n" where m and n are numbers is evaluated to True/False by
  1083 simp.
  1084 
  1085 * Theorem Cons_eq_map_conv no longer declared as ``simp''.
  1086 
  1087 * Theorem setsum_mult renamed to setsum_right_distrib.
  1088 
  1089 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  1090 ``rule'' method.
  1091 
  1092 * Reimplemented methods ``sat'' and ``satx'', with several
  1093 improvements: goals no longer need to be stated as "<prems> ==>
  1094 False", equivalences (i.e. "=" on type bool) are handled, variable
  1095 names of the form "lit_<n>" are no longer reserved, significant
  1096 speedup.
  1097 
  1098 * Methods ``sat'' and ``satx'' can now replay MiniSat proof traces.
  1099 zChaff is still supported as well.
  1100 
  1101 * 'inductive' and 'datatype': provide projections of mutual rules,
  1102 bundled as foo_bar.inducts;
  1103 
  1104 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  1105 Library.
  1106 
  1107 * Library: moved theory Accessible_Part to main HOL.
  1108 
  1109 * Library: added theory Coinductive_List of potentially infinite lists
  1110 as greatest fixed-point.
  1111 
  1112 * Library: added theory AssocList which implements (finite) maps as
  1113 association lists.
  1114 
  1115 * Added proof method ``evaluation'' for efficiently solving a goal
  1116 (i.e. a boolean expression) by compiling it to ML. The goal is
  1117 "proved" (via an oracle) if it evaluates to True.
  1118 
  1119 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  1120 also when invoked by the simplifier.  This results in the simplifier
  1121 being more powerful on arithmetic goals.  INCOMPATIBILITY.  Set
  1122 fast_arith_split_limit to 0 to obtain the old behavior.
  1123 
  1124 * Support for hex (0x20) and binary (0b1001) numerals.
  1125 
  1126 * New method: reify eqs (t), where eqs are equations for an
  1127 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  1128 parameter, computes a term s::'b and a list xs::'a list and proves the
  1129 theorem I xs s = t. This is also known as reification or quoting. The
  1130 resulting theorem is applied to the subgoal to substitute t with I xs
  1131 s.  If t is omitted, the subgoal itself is reified.
  1132 
  1133 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  1134 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  1135 where f is supposed to be a computable function (in the sense of code
  1136 generattion). The method uses reify to compute s and xs as above then
  1137 applies corr_thm and uses normalization by evaluation to "prove" f s =
  1138 r and finally gets the theorem t = r, which is again applied to the
  1139 subgoal. An Example is available in HOL/ex/ReflectionEx.thy.
  1140 
  1141 * Reflection: Automatic reification now handels binding, an example
  1142 is available in HOL/ex/ReflectionEx.thy
  1143 
  1144 
  1145 *** HOL-Algebra ***
  1146 
  1147 * Formalisation of ideals and the quotient construction over rings.
  1148 
  1149 * Order and lattice theory no longer based on records.
  1150 INCOMPATIBILITY.
  1151 
  1152 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  1153 greatest_closed.  INCOMPATIBILITY.
  1154 
  1155 * Method algebra is now set up via an attribute.  For examples see
  1156 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  1157 of algebraic structures.
  1158 
  1159 * Renamed theory CRing to Ring.
  1160 
  1161 
  1162 *** HOL-Complex ***
  1163 
  1164 * Theory Real: new method ferrack implements quantifier elimination
  1165 for linear arithmetic over the reals. The quantifier elimination
  1166 feature is used only for decision, for compatibility with arith. This
  1167 means a goal is either solved or left unchanged, no simplification.
  1168 
  1169 * Hyperreal: Functions root and sqrt are now defined on negative real
  1170 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  1171 Nonnegativity side conditions have been removed from many lemmas, so
  1172 that more subgoals may now be solved by simplification; potential
  1173 INCOMPATIBILITY.
  1174 
  1175 * Real: New axiomatic classes formalize real normed vector spaces and
  1176 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  1177 and norm :: 'a => real.
  1178 
  1179 * Real: New constant of_real :: real => 'a::real_algebra_1 injects
  1180 from reals into other types. The overloaded constant Reals :: 'a set
  1181 is now defined as range of_real; potential INCOMPATIBILITY.
  1182 
  1183 * Real: ML code generation is supported now and hence also quickcheck.
  1184 Reals are implemented as arbitrary precision rationals.
  1185 
  1186 * Hyperreal: Several constants that previously worked only for the
  1187 reals have been generalized, so they now work over arbitrary vector
  1188 spaces. Type annotations may need to be added in some cases; potential
  1189 INCOMPATIBILITY.
  1190 
  1191   Infinitesimal  :: ('a::real_normed_vector) star set
  1192   HFinite        :: ('a::real_normed_vector) star set
  1193   HInfinite      :: ('a::real_normed_vector) star set
  1194   approx         :: ('a::real_normed_vector) star => 'a star => bool
  1195   monad          :: ('a::real_normed_vector) star => 'a star set
  1196   galaxy         :: ('a::real_normed_vector) star => 'a star set
  1197   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  1198   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  1199   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  1200   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  1201   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  1202   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  1203   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  1204   sgn            :: 'a::real_normed_vector => 'a
  1205   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  1206 
  1207 * Complex: Some complex-specific constants are now abbreviations for
  1208 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  1209 hnorm.  Other constants have been entirely removed in favor of the
  1210 polymorphic versions (INCOMPATIBILITY):
  1211 
  1212   approx        <-- capprox
  1213   HFinite       <-- CFinite
  1214   HInfinite     <-- CInfinite
  1215   Infinitesimal <-- CInfinitesimal
  1216   monad         <-- cmonad
  1217   galaxy        <-- cgalaxy
  1218   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  1219   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  1220   (ns)deriv     <-- (ns)cderiv
  1221 
  1222 
  1223 *** ML ***
  1224 
  1225 * Generic arithmetic modules: Tools/integer.ML, Tools/rat.ML, Tools/float.ML
  1226 
  1227 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  1228 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  1229 init only required for impure data.  INCOMPATIBILITY: empty really
  1230 need to be empty (no dependencies on theory content!)
  1231 
  1232 * ML within Isar: antiquotations allow to embed statically-checked
  1233 formal entities in the source, referring to the context available at
  1234 compile-time.  For example:
  1235 
  1236 ML {* @{typ "'a => 'b"} *}
  1237 ML {* @{term "%x. x"} *}
  1238 ML {* @{prop "x == y"} *}
  1239 ML {* @{ctyp "'a => 'b"} *}
  1240 ML {* @{cterm "%x. x"} *}
  1241 ML {* @{cprop "x == y"} *}
  1242 ML {* @{thm asm_rl} *}
  1243 ML {* @{thms asm_rl} *}
  1244 ML {* @{const_name c} *}
  1245 ML {* @{const_syntax c} *}
  1246 ML {* @{context} *}
  1247 ML {* @{theory} *}
  1248 ML {* @{theory Pure} *}
  1249 ML {* @{simpset} *}
  1250 ML {* @{claset} *}
  1251 ML {* @{clasimpset} *}
  1252 
  1253 The same works for sources being ``used'' within an Isar context.
  1254 
  1255 * ML in Isar: improved error reporting; extra verbosity with
  1256 Toplevel.debug enabled.
  1257 
  1258 * Pure/library:
  1259 
  1260   val burrow: ('a list -> 'b list) -> 'a list list -> 'b list list
  1261   val fold_burrow: ('a list -> 'c -> 'b list * 'd) -> 'a list list -> 'c -> 'b list list * 'd
  1262 
  1263 The semantics of "burrow" is: "take a function with *simulatanously*
  1264 transforms a list of value, and apply it *simulatanously* to a list of
  1265 list of values of the appropriate type". Compare this with "map" which
  1266 would *not* apply its argument function simulatanously but in
  1267 sequence; "fold_burrow" has an additional context.
  1268 
  1269 * Pure/library: functions map2 and fold2 with curried syntax for
  1270 simultanous mapping and folding:
  1271 
  1272     val map2: ('a -> 'b -> 'c) -> 'a list -> 'b list -> 'c list
  1273     val fold2: ('a -> 'b -> 'c -> 'c) -> 'a list -> 'b list -> 'c -> 'c
  1274 
  1275 * Pure/library: indexed lists - some functions in the Isabelle library
  1276 treating lists over 'a as finite mappings from [0...n] to 'a have been
  1277 given more convenient names and signatures reminiscent of similar
  1278 functions for alists, tables, etc:
  1279 
  1280   val nth: 'a list -> int -> 'a 
  1281   val nth_map: int -> ('a -> 'a) -> 'a list -> 'a list
  1282   val fold_index: (int * 'a -> 'b -> 'b) -> 'a list -> 'b -> 'b
  1283 
  1284 Note that fold_index starts counting at index 0, not 1 like foldln
  1285 used to.
  1286 
  1287 * Pure/library: added general ``divide_and_conquer'' combinator on
  1288 lists.
  1289 
  1290 * Pure/General/table.ML: the join operations now works via exceptions
  1291 DUP/SAME instead of type option.  This is simpler in simple cases, and
  1292 admits slightly more efficient complex applications.
  1293 
  1294 * Pure: datatype Context.generic joins theory/Proof.context and
  1295 provides some facilities for code that works in either kind of
  1296 context, notably GenericDataFun for uniform theory and proof data.
  1297 
  1298 * Pure: 'advanced' translation functions (parse_translation etc.) now
  1299 use Context.generic instead of just theory.
  1300 
  1301 * Pure: simplified internal attribute type, which is now always
  1302 Context.generic * thm -> Context.generic * thm.  Global (theory)
  1303 vs. local (Proof.context) attributes have been discontinued, while
  1304 minimizing code duplication.  Thm.rule_attribute and
  1305 Thm.declaration_attribute build canonical attributes; see also
  1306 structure Context for further operations on Context.generic, notably
  1307 GenericDataFun.  INCOMPATIBILITY, need to adapt attribute type
  1308 declarations and definitions.
  1309 
  1310 * Pure/kernel: consts certification ignores sort constraints given in
  1311 signature declarations.  (This information is not relevant to the
  1312 logic, but only for type inference.)  IMPORTANT INTERNAL CHANGE,
  1313 potential INCOMPATIBILITY.
  1314 
  1315 * Pure: axiomatic type classes are now purely definitional, with
  1316 explicit proofs of class axioms and super class relations performed
  1317 internally.  See Pure/axclass.ML for the main internal interfaces --
  1318 notably AxClass.define_class supercedes AxClass.add_axclass, and
  1319 AxClass.axiomatize_class/classrel/arity supercede
  1320 Sign.add_classes/classrel/arities.
  1321 
  1322 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  1323 global/local versions on theory vs. Proof.context have been
  1324 discontinued; Attrib.syntax and Method.syntax have been adapted
  1325 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  1326 attributes, methods, etc.
  1327 
  1328 * Pure: several functions of signature "... -> theory -> theory * ..."
  1329 have been reoriented to "... -> theory -> ... * theory" in order to
  1330 allow natural usage in combination with the ||>, ||>>, |-> and
  1331 fold_map combinators.
  1332 
  1333 * Pure: official theorem names (closed derivations) and additional
  1334 comments (tags) are now strictly separate.  Name hints -- which are
  1335 maintained as tags -- may be attached any time without affecting the
  1336 derivation.
  1337 
  1338 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  1339 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  1340 achieve the old behaviour.
  1341 
  1342 * Pure: the "Goal" constant is now called "prop", supporting a
  1343 slightly more general idea of ``protecting'' meta-level rule
  1344 statements.
  1345 
  1346 * Pure: Logic.(un)varify only works in a global context, which is now
  1347 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  1348 Logic.legacy_(un)varify as temporary workaround.
  1349 
  1350 * Pure: structure Name provides scalable operations for generating
  1351 internal variable names, notably Name.variants etc.  This replaces
  1352 some popular functions from term.ML:
  1353 
  1354   Term.variant		->  Name.variant
  1355   Term.variantlist	->  Name.variant_list  (*canonical argument order*)
  1356   Term.invent_names	->  Name.invent_list
  1357 
  1358 Note that low-level renaming rarely occurs in new code -- operations
  1359 from structure Variable are used instead (see below).
  1360 
  1361 * Pure: structure Variable provides fundamental operations for proper
  1362 treatment of fixed/schematic variables in a context.  For example,
  1363 Variable.import introduces fixes for schematics of given facts and
  1364 Variable.export reverses the effect (up to renaming) -- this replaces
  1365 various freeze_thaw operations.
  1366 
  1367 * Pure: structure Goal provides simple interfaces for
  1368 init/conclude/finish and tactical prove operations (replacing former
  1369 Tactic.prove).  Goal.prove is the canonical way to prove results
  1370 within a given context; Goal.prove_global is a degraded version for
  1371 theory level goals, including a global Drule.standard.  Note that
  1372 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  1373 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  1374 in a locale context).
  1375 
  1376 * Isar: simplified treatment of user-level errors, using exception
  1377 ERROR of string uniformly.  Function error now merely raises ERROR,
  1378 without any side effect on output channels.  The Isar toplevel takes
  1379 care of proper display of ERROR exceptions.  ML code may use plain
  1380 handle/can/try; cat_error may be used to concatenate errors like this:
  1381 
  1382   ... handle ERROR msg => cat_error msg "..."
  1383 
  1384 Toplevel ML code (run directly or through the Isar toplevel) may be
  1385 embedded into the Isar toplevel with exception display/debug like
  1386 this:
  1387 
  1388   Isar.toplevel (fn () => ...)
  1389 
  1390 INCOMPATIBILITY, removed special transform_error facilities, removed
  1391 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  1392 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  1393 -- use plain ERROR instead.
  1394 
  1395 * Isar: theory setup now has type (theory -> theory), instead of a
  1396 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  1397 
  1398 * Isar: installed ML toplevel pretty printer for type Proof.context,
  1399 subject to ProofContext.debug/verbose flags.
  1400 
  1401 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  1402 theory before entering a proof state.  Transactions now always see a
  1403 quasi-functional intermediate checkpoint, both in interactive and
  1404 batch mode.
  1405 
  1406 * Simplifier: the simpset of a running simplification process now
  1407 contains a proof context (cf. Simplifier.the_context), which is the
  1408 very context that the initial simpset has been retrieved from (by
  1409 simpset_of/local_simpset_of).  Consequently, all plug-in components
  1410 (solver, looper etc.) may depend on arbitrary proof data.
  1411 
  1412 * Simplifier.inherit_context inherits the proof context (plus the
  1413 local bounds) of the current simplification process; any simproc
  1414 etc. that calls the Simplifier recursively should do this!  Removed
  1415 former Simplifier.inherit_bounds, which is already included here --
  1416 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  1417 specify an explicit context using Simplifier.context/theory_context.
  1418 
  1419 * Simplifier/Classical Reasoner: more abstract interfaces
  1420 change_simpset/claset for modifying the simpset/claset reference of a
  1421 theory; raw versions simpset/claset_ref etc. have been discontinued --
  1422 INCOMPATIBILITY.
  1423 
  1424 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  1425 "Trueprop" etc.
  1426 
  1427 
  1428 *** System ***
  1429 
  1430 * settings: ML_IDENTIFIER -- which is appended to user specific heap
  1431 locations -- now includes the Isabelle version identifier as well.
  1432 This simplifies use of multiple Isabelle installations.
  1433 
  1434 * isabelle-process: option -S (secure mode) disables some critical
  1435 operations, notably runtime compilation and evaluation of ML source
  1436 code.
  1437 
  1438 * Experimental support for multithreading, using Poly/ML 5.1 (internal
  1439 version from CVS). The theory loader exploits parallelism when
  1440 processing independent theories, following the header specifications.
  1441 The maximum number of worker threads is specified via usedir option -M
  1442 or the "max-threads" setting in Proof General.  User-code needs to
  1443 observe certain guidelines for thread-safe programming, see appendix A
  1444 in the Isar Implementation manual.
  1445 
  1446 
  1447 New in Isabelle2005 (October 2005)
  1448 ----------------------------------
  1449 
  1450 *** General ***
  1451 
  1452 * Theory headers: the new header syntax for Isar theories is
  1453 
  1454   theory <name>
  1455   imports <theory1> ... <theoryN>
  1456   uses <file1> ... <fileM>
  1457   begin
  1458 
  1459 where the 'uses' part is optional.  The previous syntax
  1460 
  1461   theory <name> = <theory1> + ... + <theoryN>:
  1462 
  1463 will disappear in the next release.  Use isatool fixheaders to convert
  1464 existing theory files.  Note that there is no change in ancient
  1465 non-Isar theories now, but these will disappear soon.
  1466 
  1467 * Theory loader: parent theories can now also be referred to via
  1468 relative and absolute paths.
  1469 
  1470 * Command 'find_theorems' searches for a list of criteria instead of a
  1471 list of constants. Known criteria are: intro, elim, dest, name:string,
  1472 simp:term, and any term. Criteria can be preceded by '-' to select
  1473 theorems that do not match. Intro, elim, dest select theorems that
  1474 match the current goal, name:s selects theorems whose fully qualified
  1475 name contain s, and simp:term selects all simplification rules whose
  1476 lhs match term.  Any other term is interpreted as pattern and selects
  1477 all theorems matching the pattern. Available in ProofGeneral under
  1478 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  1479 
  1480   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  1481 
  1482 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  1483 matching the current goal as introduction rule and not having "HOL."
  1484 in their name (i.e. not being defined in theory HOL).
  1485 
  1486 * Command 'thms_containing' has been discontinued in favour of
  1487 'find_theorems'; INCOMPATIBILITY.
  1488 
  1489 * Communication with Proof General is now 8bit clean, which means that
  1490 Unicode text in UTF-8 encoding may be used within theory texts (both
  1491 formal and informal parts).  Cf. option -U of the Isabelle Proof
  1492 General interface.  Here are some simple examples (cf. src/HOL/ex):
  1493 
  1494   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  1495   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  1496 
  1497 * Improved efficiency of the Simplifier and, to a lesser degree, the
  1498 Classical Reasoner.  Typical big applications run around 2 times
  1499 faster.
  1500 
  1501 
  1502 *** Document preparation ***
  1503 
  1504 * Commands 'display_drafts' and 'print_drafts' perform simple output
  1505 of raw sources.  Only those symbols that do not require additional
  1506 LaTeX packages (depending on comments in isabellesym.sty) are
  1507 displayed properly, everything else is left verbatim.  isatool display
  1508 and isatool print are used as front ends (these are subject to the
  1509 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  1510 
  1511 * Command tags control specific markup of certain regions of text,
  1512 notably folding and hiding.  Predefined tags include "theory" (for
  1513 theory begin and end), "proof" for proof commands, and "ML" for
  1514 commands involving ML code; the additional tags "visible" and
  1515 "invisible" are unused by default.  Users may give explicit tag
  1516 specifications in the text, e.g. ''by %invisible (auto)''.  The
  1517 interpretation of tags is determined by the LaTeX job during document
  1518 preparation: see option -V of isatool usedir, or options -n and -t of
  1519 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  1520 \isadroptag.
  1521 
  1522 Several document versions may be produced at the same time via isatool
  1523 usedir (the generated index.html will link all of them).  Typical
  1524 specifications include ''-V document=theory,proof,ML'' to present
  1525 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  1526 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  1527 these parts without any formal replacement text.  The Isabelle site
  1528 default settings produce ''document'' and ''outline'' versions as
  1529 specified above.
  1530 
  1531 * Several new antiquotations:
  1532 
  1533   @{term_type term} prints a term with its type annotated;
  1534 
  1535   @{typeof term} prints the type of a term;
  1536 
  1537   @{const const} is the same as @{term const}, but checks that the
  1538   argument is a known logical constant;
  1539 
  1540   @{term_style style term} and @{thm_style style thm} print a term or
  1541   theorem applying a "style" to it
  1542 
  1543   @{ML text}
  1544 
  1545 Predefined styles are 'lhs' and 'rhs' printing the lhs/rhs of
  1546 definitions, equations, inequations etc., 'concl' printing only the
  1547 conclusion of a meta-logical statement theorem, and 'prem1' .. 'prem19'
  1548 to print the specified premise.  TermStyle.add_style provides an ML
  1549 interface for introducing further styles.  See also the "LaTeX Sugar"
  1550 document practical applications.  The ML antiquotation prints
  1551 type-checked ML expressions verbatim.
  1552 
  1553 * Markup commands 'chapter', 'section', 'subsection', 'subsubsection',
  1554 and 'text' support optional locale specification '(in loc)', which
  1555 specifies the default context for interpreting antiquotations.  For
  1556 example: 'text (in lattice) {* @{thm inf_assoc}*}'.
  1557 
  1558 * Option 'locale=NAME' of antiquotations specifies an alternative
  1559 context interpreting the subsequent argument.  For example: @{thm
  1560 [locale=lattice] inf_assoc}.
  1561 
  1562 * Proper output of proof terms (@{prf ...} and @{full_prf ...}) within
  1563 a proof context.
  1564 
  1565 * Proper output of antiquotations for theory commands involving a
  1566 proof context (such as 'locale' or 'theorem (in loc) ...').
  1567 
  1568 * Delimiters of outer tokens (string etc.) now produce separate LaTeX
  1569 macros (\isachardoublequoteopen, isachardoublequoteclose etc.).
  1570 
  1571 * isatool usedir: new option -C (default true) controls whether option
  1572 -D should include a copy of the original document directory; -C false
  1573 prevents unwanted effects such as copying of administrative CVS data.
  1574 
  1575 
  1576 *** Pure ***
  1577 
  1578 * Considerably improved version of 'constdefs' command.  Now performs
  1579 automatic type-inference of declared constants; additional support for
  1580 local structure declarations (cf. locales and HOL records), see also
  1581 isar-ref manual.  Potential INCOMPATIBILITY: need to observe strictly
  1582 sequential dependencies of definitions within a single 'constdefs'
  1583 section; moreover, the declared name needs to be an identifier.  If
  1584 all fails, consider to fall back on 'consts' and 'defs' separately.
  1585 
  1586 * Improved indexed syntax and implicit structures.  First of all,
  1587 indexed syntax provides a notational device for subscripted
  1588 application, using the new syntax \<^bsub>term\<^esub> for arbitrary
  1589 expressions.  Secondly, in a local context with structure
  1590 declarations, number indexes \<^sub>n or the empty index (default
  1591 number 1) refer to a certain fixed variable implicitly; option
  1592 show_structs controls printing of implicit structures.  Typical
  1593 applications of these concepts involve record types and locales.
  1594 
  1595 * New command 'no_syntax' removes grammar declarations (and
  1596 translations) resulting from the given syntax specification, which is
  1597 interpreted in the same manner as for the 'syntax' command.
  1598 
  1599 * 'Advanced' translation functions (parse_translation etc.) may depend
  1600 on the signature of the theory context being presently used for
  1601 parsing/printing, see also isar-ref manual.
  1602 
  1603 * Improved 'oracle' command provides a type-safe interface to turn an
  1604 ML expression of type theory -> T -> term into a primitive rule of
  1605 type theory -> T -> thm (i.e. the functionality of Thm.invoke_oracle
  1606 is already included here); see also FOL/ex/IffExample.thy;
  1607 INCOMPATIBILITY.
  1608 
  1609 * axclass: name space prefix for class "c" is now "c_class" (was "c"
  1610 before); "cI" is no longer bound, use "c.intro" instead.
  1611 INCOMPATIBILITY.  This change avoids clashes of fact bindings for
  1612 axclasses vs. locales.
  1613 
  1614 * Improved internal renaming of symbolic identifiers -- attach primes
  1615 instead of base 26 numbers.
  1616 
  1617 * New flag show_question_marks controls printing of leading question
  1618 marks in schematic variable names.
  1619 
  1620 * In schematic variable names, *any* symbol following \<^isub> or
  1621 \<^isup> is now treated as part of the base name.  For example, the
  1622 following works without printing of awkward ".0" indexes:
  1623 
  1624   lemma "x\<^isub>1 = x\<^isub>2 ==> x\<^isub>2 = x\<^isub>1"
  1625     by simp
  1626 
  1627 * Inner syntax includes (*(*nested*) comments*).
  1628 
  1629 * Pretty printer now supports unbreakable blocks, specified in mixfix
  1630 annotations as "(00...)".
  1631 
  1632 * Clear separation of logical types and nonterminals, where the latter
  1633 may only occur in 'syntax' specifications or type abbreviations.
  1634 Before that distinction was only partially implemented via type class
  1635 "logic" vs. "{}".  Potential INCOMPATIBILITY in rare cases of improper
  1636 use of 'types'/'consts' instead of 'nonterminals'/'syntax'.  Some very
  1637 exotic syntax specifications may require further adaption
  1638 (e.g. Cube/Cube.thy).
  1639 
  1640 * Removed obsolete type class "logic", use the top sort {} instead.
  1641 Note that non-logical types should be declared as 'nonterminals'
  1642 rather than 'types'.  INCOMPATIBILITY for new object-logic
  1643 specifications.
  1644 
  1645 * Attributes 'induct' and 'cases': type or set names may now be
  1646 locally fixed variables as well.
  1647 
  1648 * Simplifier: can now control the depth to which conditional rewriting
  1649 is traced via the PG menu Isabelle -> Settings -> Trace Simp Depth
  1650 Limit.
  1651 
  1652 * Simplifier: simplification procedures may now take the current
  1653 simpset into account (cf. Simplifier.simproc(_i) / mk_simproc
  1654 interface), which is very useful for calling the Simplifier
  1655 recursively.  Minor INCOMPATIBILITY: the 'prems' argument of simprocs
  1656 is gone -- use prems_of_ss on the simpset instead.  Moreover, the
  1657 low-level mk_simproc no longer applies Logic.varify internally, to
  1658 allow for use in a context of fixed variables.
  1659 
  1660 * thin_tac now works even if the assumption being deleted contains !!
  1661 or ==>.  More generally, erule now works even if the major premise of
  1662 the elimination rule contains !! or ==>.
  1663 
  1664 * Method 'rules' has been renamed to 'iprover'. INCOMPATIBILITY.
  1665 
  1666 * Reorganized bootstrapping of the Pure theories; CPure is now derived
  1667 from Pure, which contains all common declarations already.  Both
  1668 theories are defined via plain Isabelle/Isar .thy files.
  1669 INCOMPATIBILITY: elements of CPure (such as the CPure.intro /
  1670 CPure.elim / CPure.dest attributes) now appear in the Pure name space;
  1671 use isatool fixcpure to adapt your theory and ML sources.
  1672 
  1673 * New syntax 'name(i-j, i-, i, ...)' for referring to specific
  1674 selections of theorems in named facts via index ranges.
  1675 
  1676 * 'print_theorems': in theory mode, really print the difference
  1677 wrt. the last state (works for interactive theory development only),
  1678 in proof mode print all local facts (cf. 'print_facts');
  1679 
  1680 * 'hide': option '(open)' hides only base names.
  1681 
  1682 * More efficient treatment of intermediate checkpoints in interactive
  1683 theory development.
  1684 
  1685 * Code generator is now invoked via code_module (incremental code
  1686 generation) and code_library (modular code generation, ML structures
  1687 for each theory).  INCOMPATIBILITY: new keywords 'file' and 'contains'
  1688 must be quoted when used as identifiers.
  1689 
  1690 * New 'value' command for reading, evaluating and printing terms using
  1691 the code generator.  INCOMPATIBILITY: command keyword 'value' must be
  1692 quoted when used as identifier.
  1693 
  1694 
  1695 *** Locales ***
  1696 
  1697 * New commands for the interpretation of locale expressions in
  1698 theories (1), locales (2) and proof contexts (3).  These generate
  1699 proof obligations from the expression specification.  After the
  1700 obligations have been discharged, theorems of the expression are added
  1701 to the theory, target locale or proof context.  The synopsis of the
  1702 commands is a follows:
  1703 
  1704   (1) interpretation expr inst
  1705   (2) interpretation target < expr
  1706   (3) interpret expr inst
  1707 
  1708 Interpretation in theories and proof contexts require a parameter
  1709 instantiation of terms from the current context.  This is applied to
  1710 specifications and theorems of the interpreted expression.
  1711 Interpretation in locales only permits parameter renaming through the
  1712 locale expression.  Interpretation is smart in that interpretations
  1713 that are active already do not occur in proof obligations, neither are
  1714 instantiated theorems stored in duplicate.  Use 'print_interps' to
  1715 inspect active interpretations of a particular locale.  For details,
  1716 see the Isar Reference manual.  Examples can be found in
  1717 HOL/Finite_Set.thy and HOL/Algebra/UnivPoly.thy.
  1718 
  1719 INCOMPATIBILITY: former 'instantiate' has been withdrawn, use
  1720 'interpret' instead.
  1721 
  1722 * New context element 'constrains' for adding type constraints to
  1723 parameters.
  1724 
  1725 * Context expressions: renaming of parameters with syntax
  1726 redeclaration.
  1727 
  1728 * Locale declaration: 'includes' disallowed.
  1729 
  1730 * Proper static binding of attribute syntax -- i.e. types / terms /
  1731 facts mentioned as arguments are always those of the locale definition
  1732 context, independently of the context of later invocations.  Moreover,
  1733 locale operations (renaming and type / term instantiation) are applied
  1734 to attribute arguments as expected.
  1735 
  1736 INCOMPATIBILITY of the ML interface: always pass Attrib.src instead of
  1737 actual attributes; rare situations may require Attrib.attribute to
  1738 embed those attributes into Attrib.src that lack concrete syntax.
  1739 Attribute implementations need to cooperate properly with the static
  1740 binding mechanism.  Basic parsers Args.XXX_typ/term/prop and
  1741 Attrib.XXX_thm etc. already do the right thing without further
  1742 intervention.  Only unusual applications -- such as "where" or "of"
  1743 (cf. src/Pure/Isar/attrib.ML), which process arguments depending both
  1744 on the context and the facts involved -- may have to assign parsed
  1745 values to argument tokens explicitly.
  1746 
  1747 * Changed parameter management in theorem generation for long goal
  1748 statements with 'includes'.  INCOMPATIBILITY: produces a different
  1749 theorem statement in rare situations.
  1750 
  1751 * Locale inspection command 'print_locale' omits notes elements.  Use
  1752 'print_locale!' to have them included in the output.
  1753 
  1754 
  1755 *** Provers ***
  1756 
  1757 * Provers/hypsubst.ML: improved version of the subst method, for
  1758 single-step rewriting: it now works in bound variable contexts. New is
  1759 'subst (asm)', for rewriting an assumption.  INCOMPATIBILITY: may
  1760 rewrite a different subterm than the original subst method, which is
  1761 still available as 'simplesubst'.
  1762 
  1763 * Provers/quasi.ML: new transitivity reasoners for transitivity only
  1764 and quasi orders.
  1765 
  1766 * Provers/trancl.ML: new transitivity reasoner for transitive and
  1767 reflexive-transitive closure of relations.
  1768 
  1769 * Provers/blast.ML: new reference depth_limit to make blast's depth
  1770 limit (previously hard-coded with a value of 20) user-definable.
  1771 
  1772 * Provers/simplifier.ML has been moved to Pure, where Simplifier.setup
  1773 is peformed already.  Object-logics merely need to finish their
  1774 initial simpset configuration as before.  INCOMPATIBILITY.
  1775 
  1776 
  1777 *** HOL ***
  1778 
  1779 * Symbolic syntax of Hilbert Choice Operator is now as follows:
  1780 
  1781   syntax (epsilon)
  1782     "_Eps" :: "[pttrn, bool] => 'a"    ("(3\<some>_./ _)" [0, 10] 10)
  1783 
  1784 The symbol \<some> is displayed as the alternative epsilon of LaTeX
  1785 and x-symbol; use option '-m epsilon' to get it actually printed.
  1786 Moreover, the mathematically important symbolic identifier \<epsilon>
  1787 becomes available as variable, constant etc.  INCOMPATIBILITY,
  1788 
  1789 * "x > y" abbreviates "y < x" and "x >= y" abbreviates "y <= x".
  1790 Similarly for all quantifiers: "ALL x > y" etc.  The x-symbol for >=
  1791 is \<ge>. New transitivity rules have been added to HOL/Orderings.thy to
  1792 support corresponding Isar calculations.
  1793 
  1794 * "{x:A. P}" abbreviates "{x. x:A & P}", and similarly for "\<in>"
  1795 instead of ":".
  1796 
  1797 * theory SetInterval: changed the syntax for open intervals:
  1798 
  1799   Old       New
  1800   {..n(}    {..<n}
  1801   {)n..}    {n<..}
  1802   {m..n(}   {m..<n}
  1803   {)m..n}   {m<..n}
  1804   {)m..n(}  {m<..<n}
  1805 
  1806 The old syntax is still supported but will disappear in the next
  1807 release.  For conversion use the following Emacs search and replace
  1808 patterns (these are not perfect but work quite well):
  1809 
  1810   {)\([^\.]*\)\.\.  ->  {\1<\.\.}
  1811   \.\.\([^(}]*\)(}  ->  \.\.<\1}
  1812 
  1813 * Theory Commutative_Ring (in Library): method comm_ring for proving
  1814 equalities in commutative rings; method 'algebra' provides a generic
  1815 interface.
  1816 
  1817 * Theory Finite_Set: changed the syntax for 'setsum', summation over
  1818 finite sets: "setsum (%x. e) A", which used to be "\<Sum>x:A. e", is
  1819 now either "SUM x:A. e" or "\<Sum>x \<in> A. e". The bound variable can
  1820 be a tuple pattern.
  1821 
  1822 Some new syntax forms are available:
  1823 
  1824   "\<Sum>x | P. e"      for     "setsum (%x. e) {x. P}"
  1825   "\<Sum>x = a..b. e"   for     "setsum (%x. e) {a..b}"
  1826   "\<Sum>x = a..<b. e"  for     "setsum (%x. e) {a..<b}"
  1827   "\<Sum>x < k. e"      for     "setsum (%x. e) {..<k}"
  1828 
  1829 The latter form "\<Sum>x < k. e" used to be based on a separate
  1830 function "Summation", which has been discontinued.
  1831 
  1832 * theory Finite_Set: in structured induction proofs, the insert case
  1833 is now 'case (insert x F)' instead of the old counterintuitive 'case
  1834 (insert F x)'.
  1835 
  1836 * The 'refute' command has been extended to support a much larger
  1837 fragment of HOL, including axiomatic type classes, constdefs and
  1838 typedefs, inductive datatypes and recursion.
  1839 
  1840 * New tactics 'sat' and 'satx' to prove propositional tautologies.
  1841 Requires zChaff with proof generation to be installed.  See
  1842 HOL/ex/SAT_Examples.thy for examples.
  1843 
  1844 * Datatype induction via method 'induct' now preserves the name of the
  1845 induction variable. For example, when proving P(xs::'a list) by
  1846 induction on xs, the induction step is now P(xs) ==> P(a#xs) rather
  1847 than P(list) ==> P(a#list) as previously.  Potential INCOMPATIBILITY
  1848 in unstructured proof scripts.
  1849 
  1850 * Reworked implementation of records.  Improved scalability for
  1851 records with many fields, avoiding performance problems for type
  1852 inference. Records are no longer composed of nested field types, but
  1853 of nested extension types. Therefore the record type only grows linear
  1854 in the number of extensions and not in the number of fields.  The
  1855 top-level (users) view on records is preserved.  Potential
  1856 INCOMPATIBILITY only in strange cases, where the theory depends on the
  1857 old record representation. The type generated for a record is called
  1858 <record_name>_ext_type.
  1859 
  1860 Flag record_quick_and_dirty_sensitive can be enabled to skip the
  1861 proofs triggered by a record definition or a simproc (if
  1862 quick_and_dirty is enabled).  Definitions of large records can take
  1863 quite long.
  1864 
  1865 New simproc record_upd_simproc for simplification of multiple record
  1866 updates enabled by default.  Moreover, trivial updates are also
  1867 removed: r(|x := x r|) = r.  INCOMPATIBILITY: old proofs break
  1868 occasionally, since simplification is more powerful by default.
  1869 
  1870 * typedef: proper support for polymorphic sets, which contain extra
  1871 type-variables in the term.
  1872 
  1873 * Simplifier: automatically reasons about transitivity chains
  1874 involving "trancl" (r^+) and "rtrancl" (r^*) by setting up tactics
  1875 provided by Provers/trancl.ML as additional solvers.  INCOMPATIBILITY:
  1876 old proofs break occasionally as simplification may now solve more
  1877 goals than previously.
  1878 
  1879 * Simplifier: converts x <= y into x = y if assumption y <= x is
  1880 present.  Works for all partial orders (class "order"), in particular
  1881 numbers and sets.  For linear orders (e.g. numbers) it treats ~ x < y
  1882 just like y <= x.
  1883 
  1884 * Simplifier: new simproc for "let x = a in f x".  If a is a free or
  1885 bound variable or a constant then the let is unfolded.  Otherwise
  1886 first a is simplified to b, and then f b is simplified to g. If
  1887 possible we abstract b from g arriving at "let x = b in h x",
  1888 otherwise we unfold the let and arrive at g.  The simproc can be
  1889 enabled/disabled by the reference use_let_simproc.  Potential
  1890 INCOMPATIBILITY since simplification is more powerful by default.
  1891 
  1892 * Classical reasoning: the meson method now accepts theorems as arguments.
  1893 
  1894 * Prover support: pre-release of the Isabelle-ATP linkup, which runs background
  1895 jobs to provide advice on the provability of subgoals.
  1896 
  1897 * Theory OrderedGroup and Ring_and_Field: various additions and
  1898 improvements to faciliate calculations involving equalities and
  1899 inequalities.
  1900 
  1901 The following theorems have been eliminated or modified
  1902 (INCOMPATIBILITY):
  1903 
  1904   abs_eq             now named abs_of_nonneg
  1905   abs_of_ge_0        now named abs_of_nonneg
  1906   abs_minus_eq       now named abs_of_nonpos
  1907   imp_abs_id         now named abs_of_nonneg
  1908   imp_abs_neg_id     now named abs_of_nonpos
  1909   mult_pos           now named mult_pos_pos
  1910   mult_pos_le        now named mult_nonneg_nonneg
  1911   mult_pos_neg_le    now named mult_nonneg_nonpos
  1912   mult_pos_neg2_le   now named mult_nonneg_nonpos2
  1913   mult_neg           now named mult_neg_neg
  1914   mult_neg_le        now named mult_nonpos_nonpos
  1915 
  1916 * The following lemmas in Ring_and_Field have been added to the simplifier:
  1917      
  1918      zero_le_square
  1919      not_square_less_zero 
  1920 
  1921   The following lemmas have been deleted from Real/RealPow:
  1922   
  1923      realpow_zero_zero
  1924      realpow_two
  1925      realpow_less
  1926      zero_le_power
  1927      realpow_two_le
  1928      abs_realpow_two
  1929      realpow_two_abs     
  1930 
  1931 * Theory Parity: added rules for simplifying exponents.
  1932 
  1933 * Theory List:
  1934 
  1935 The following theorems have been eliminated or modified
  1936 (INCOMPATIBILITY):
  1937 
  1938   list_all_Nil       now named list_all.simps(1)
  1939   list_all_Cons      now named list_all.simps(2)
  1940   list_all_conv      now named list_all_iff
  1941   set_mem_eq         now named mem_iff
  1942 
  1943 * Theories SetsAndFunctions and BigO (see HOL/Library) support
  1944 asymptotic "big O" calculations.  See the notes in BigO.thy.
  1945 
  1946 
  1947 *** HOL-Complex ***
  1948 
  1949 * Theory RealDef: better support for embedding natural numbers and
  1950 integers in the reals.
  1951 
  1952 The following theorems have been eliminated or modified
  1953 (INCOMPATIBILITY):
  1954 
  1955   exp_ge_add_one_self  now requires no hypotheses
  1956   real_of_int_add      reversed direction of equality (use [symmetric])
  1957   real_of_int_minus    reversed direction of equality (use [symmetric])
  1958   real_of_int_diff     reversed direction of equality (use [symmetric])
  1959   real_of_int_mult     reversed direction of equality (use [symmetric])
  1960 
  1961 * Theory RComplete: expanded support for floor and ceiling functions.
  1962 
  1963 * Theory Ln is new, with properties of the natural logarithm
  1964 
  1965 * Hyperreal: There is a new type constructor "star" for making
  1966 nonstandard types.  The old type names are now type synonyms:
  1967 
  1968   hypreal = real star
  1969   hypnat = nat star
  1970   hcomplex = complex star
  1971 
  1972 * Hyperreal: Many groups of similarly-defined constants have been
  1973 replaced by polymorphic versions (INCOMPATIBILITY):
  1974 
  1975   star_of <-- hypreal_of_real, hypnat_of_nat, hcomplex_of_complex
  1976 
  1977   starset      <-- starsetNat, starsetC
  1978   *s*          <-- *sNat*, *sc*
  1979   starset_n    <-- starsetNat_n, starsetC_n
  1980   *sn*         <-- *sNatn*, *scn*
  1981   InternalSets <-- InternalNatSets, InternalCSets
  1982 
  1983   starfun      <-- starfun{Nat,Nat2,C,RC,CR}
  1984   *f*          <-- *fNat*, *fNat2*, *fc*, *fRc*, *fcR*
  1985   starfun_n    <-- starfun{Nat,Nat2,C,RC,CR}_n
  1986   *fn*         <-- *fNatn*, *fNat2n*, *fcn*, *fRcn*, *fcRn*
  1987   InternalFuns <-- InternalNatFuns, InternalNatFuns2, Internal{C,RC,CR}Funs
  1988 
  1989 * Hyperreal: Many type-specific theorems have been removed in favor of
  1990 theorems specific to various axiomatic type classes (INCOMPATIBILITY):
  1991 
  1992   add_commute <-- {hypreal,hypnat,hcomplex}_add_commute
  1993   add_assoc   <-- {hypreal,hypnat,hcomplex}_add_assocs
  1994   OrderedGroup.add_0 <-- {hypreal,hypnat,hcomplex}_add_zero_left
  1995   OrderedGroup.add_0_right <-- {hypreal,hcomplex}_add_zero_right
  1996   right_minus <-- hypreal_add_minus
  1997   left_minus <-- {hypreal,hcomplex}_add_minus_left
  1998   mult_commute <-- {hypreal,hypnat,hcomplex}_mult_commute
  1999   mult_assoc <-- {hypreal,hypnat,hcomplex}_mult_assoc
  2000   mult_1_left <-- {hypreal,hypnat}_mult_1, hcomplex_mult_one_left
  2001   mult_1_right <-- hcomplex_mult_one_right
  2002   mult_zero_left <-- hcomplex_mult_zero_left
  2003   left_distrib <-- {hypreal,hypnat,hcomplex}_add_mult_distrib
  2004   right_distrib <-- hypnat_add_mult_distrib2
  2005   zero_neq_one <-- {hypreal,hypnat,hcomplex}_zero_not_eq_one
  2006   right_inverse <-- hypreal_mult_inverse
  2007   left_inverse <-- hypreal_mult_inverse_left, hcomplex_mult_inv_left
  2008   order_refl <-- {hypreal,hypnat}_le_refl
  2009   order_trans <-- {hypreal,hypnat}_le_trans
  2010   order_antisym <-- {hypreal,hypnat}_le_anti_sym
  2011   order_less_le <-- {hypreal,hypnat}_less_le
  2012   linorder_linear <-- {hypreal,hypnat}_le_linear
  2013   add_left_mono <-- {hypreal,hypnat}_add_left_mono
  2014   mult_strict_left_mono <-- {hypreal,hypnat}_mult_less_mono2
  2015   add_nonneg_nonneg <-- hypreal_le_add_order
  2016 
  2017 * Hyperreal: Separate theorems having to do with type-specific
  2018 versions of constants have been merged into theorems that apply to the
  2019 new polymorphic constants (INCOMPATIBILITY):
  2020 
  2021   STAR_UNIV_set <-- {STAR_real,NatStar_real,STARC_complex}_set
  2022   STAR_empty_set <-- {STAR,NatStar,STARC}_empty_set
  2023   STAR_Un <-- {STAR,NatStar,STARC}_Un
  2024   STAR_Int <-- {STAR,NatStar,STARC}_Int
  2025   STAR_Compl <-- {STAR,NatStar,STARC}_Compl
  2026   STAR_subset <-- {STAR,NatStar,STARC}_subset
  2027   STAR_mem <-- {STAR,NatStar,STARC}_mem
  2028   STAR_mem_Compl <-- {STAR,STARC}_mem_Compl
  2029   STAR_diff <-- {STAR,STARC}_diff
  2030   STAR_star_of_image_subset <-- {STAR_hypreal_of_real, NatStar_hypreal_of_real,
  2031     STARC_hcomplex_of_complex}_image_subset
  2032   starset_n_Un <-- starset{Nat,C}_n_Un
  2033   starset_n_Int <-- starset{Nat,C}_n_Int
  2034   starset_n_Compl <-- starset{Nat,C}_n_Compl
  2035   starset_n_diff <-- starset{Nat,C}_n_diff
  2036   InternalSets_Un <-- Internal{Nat,C}Sets_Un
  2037   InternalSets_Int <-- Internal{Nat,C}Sets_Int
  2038   InternalSets_Compl <-- Internal{Nat,C}Sets_Compl
  2039   InternalSets_diff <-- Internal{Nat,C}Sets_diff
  2040   InternalSets_UNIV_diff <-- Internal{Nat,C}Sets_UNIV_diff
  2041   InternalSets_starset_n <-- Internal{Nat,C}Sets_starset{Nat,C}_n
  2042   starset_starset_n_eq <-- starset{Nat,C}_starset{Nat,C}_n_eq
  2043   starset_n_starset <-- starset{Nat,C}_n_starset{Nat,C}
  2044   starfun_n_starfun <-- starfun{Nat,Nat2,C,RC,CR}_n_starfun{Nat,Nat2,C,RC,CR}
  2045   starfun <-- starfun{Nat,Nat2,C,RC,CR}
  2046   starfun_mult <-- starfun{Nat,Nat2,C,RC,CR}_mult
  2047   starfun_add <-- starfun{Nat,Nat2,C,RC,CR}_add
  2048   starfun_minus <-- starfun{Nat,Nat2,C,RC,CR}_minus
  2049   starfun_diff <-- starfun{C,RC,CR}_diff
  2050   starfun_o <-- starfun{NatNat2,Nat2,_stafunNat,C,C_starfunRC,_starfunCR}_o
  2051   starfun_o2 <-- starfun{NatNat2,_stafunNat,C,C_starfunRC,_starfunCR}_o2
  2052   starfun_const_fun <-- starfun{Nat,Nat2,C,RC,CR}_const_fun
  2053   starfun_inverse <-- starfun{Nat,C,RC,CR}_inverse
  2054   starfun_eq <-- starfun{Nat,Nat2,C,RC,CR}_eq
  2055   starfun_eq_iff <-- starfun{C,RC,CR}_eq_iff
  2056   starfun_Id <-- starfunC_Id
  2057   starfun_approx <-- starfun{Nat,CR}_approx
  2058   starfun_capprox <-- starfun{C,RC}_capprox
  2059   starfun_abs <-- starfunNat_rabs
  2060   starfun_lambda_cancel <-- starfun{C,CR,RC}_lambda_cancel
  2061   starfun_lambda_cancel2 <-- starfun{C,CR,RC}_lambda_cancel2
  2062   starfun_mult_HFinite_approx <-- starfunCR_mult_HFinite_capprox
  2063   starfun_mult_CFinite_capprox <-- starfun{C,RC}_mult_CFinite_capprox
  2064   starfun_add_capprox <-- starfun{C,RC}_add_capprox
  2065   starfun_add_approx <-- starfunCR_add_approx
  2066   starfun_inverse_inverse <-- starfunC_inverse_inverse
  2067   starfun_divide <-- starfun{C,CR,RC}_divide
  2068   starfun_n <-- starfun{Nat,C}_n
  2069   starfun_n_mult <-- starfun{Nat,C}_n_mult
  2070   starfun_n_add <-- starfun{Nat,C}_n_add
  2071   starfun_n_add_minus <-- starfunNat_n_add_minus
  2072   starfun_n_const_fun <-- starfun{Nat,C}_n_const_fun
  2073   starfun_n_minus <-- starfun{Nat,C}_n_minus
  2074   starfun_n_eq <-- starfun{Nat,C}_n_eq
  2075 
  2076   star_n_add <-- {hypreal,hypnat,hcomplex}_add
  2077   star_n_minus <-- {hypreal,hcomplex}_minus
  2078   star_n_diff <-- {hypreal,hcomplex}_diff
  2079   star_n_mult <-- {hypreal,hcomplex}_mult
  2080   star_n_inverse <-- {hypreal,hcomplex}_inverse
  2081   star_n_le <-- {hypreal,hypnat}_le
  2082   star_n_less <-- {hypreal,hypnat}_less
  2083   star_n_zero_num <-- {hypreal,hypnat,hcomplex}_zero_num
  2084   star_n_one_num <-- {hypreal,hypnat,hcomplex}_one_num
  2085   star_n_abs <-- hypreal_hrabs
  2086   star_n_divide <-- hcomplex_divide
  2087 
  2088   star_of_add <-- {hypreal_of_real,hypnat_of_nat,hcomplex_of_complex}_add
  2089   star_of_minus <-- {hypreal_of_real,hcomplex_of_complex}_minus
  2090   star_of_diff <-- hypreal_of_real_diff
  2091   star_of_mult <-- {hypreal_of_real,hypnat_of_nat,hcomplex_of_complex}_mult
  2092   star_of_one <-- {hypreal_of_real,hcomplex_of_complex}_one
  2093   star_of_zero <-- {hypreal_of_real,hypnat_of_nat,hcomplex_of_complex}_zero
  2094   star_of_le <-- {hypreal_of_real,hypnat_of_nat}_le_iff
  2095   star_of_less <-- {hypreal_of_real,hypnat_of_nat}_less_iff
  2096   star_of_eq <-- {hypreal_of_real,hypnat_of_nat,hcomplex_of_complex}_eq_iff
  2097   star_of_inverse <-- {hypreal_of_real,hcomplex_of_complex}_inverse
  2098   star_of_divide <-- {hypreal_of_real,hcomplex_of_complex}_divide
  2099   star_of_of_nat <-- {hypreal_of_real,hcomplex_of_complex}_of_nat
  2100   star_of_of_int <-- {hypreal_of_real,hcomplex_of_complex}_of_int
  2101   star_of_number_of <-- {hypreal,hcomplex}_number_of
  2102   star_of_number_less <-- number_of_less_hypreal_of_real_iff
  2103   star_of_number_le <-- number_of_le_hypreal_of_real_iff
  2104   star_of_eq_number <-- hypreal_of_real_eq_number_of_iff
  2105   star_of_less_number <-- hypreal_of_real_less_number_of_iff
  2106   star_of_le_number <-- hypreal_of_real_le_number_of_iff
  2107   star_of_power <-- hypreal_of_real_power
  2108   star_of_eq_0 <-- hcomplex_of_complex_zero_iff
  2109 
  2110 * Hyperreal: new method "transfer" that implements the transfer
  2111 principle of nonstandard analysis. With a subgoal that mentions
  2112 nonstandard types like "'a star", the command "apply transfer"
  2113 replaces it with an equivalent one that mentions only standard types.
  2114 To be successful, all free variables must have standard types; non-
  2115 standard variables must have explicit universal quantifiers.
  2116 
  2117 * Hyperreal: A theory of Taylor series.
  2118 
  2119 
  2120 *** HOLCF ***
  2121 
  2122 * Discontinued special version of 'constdefs' (which used to support
  2123 continuous functions) in favor of the general Pure one with full
  2124 type-inference.
  2125 
  2126 * New simplification procedure for solving continuity conditions; it
  2127 is much faster on terms with many nested lambda abstractions (cubic
  2128 instead of exponential time).
  2129 
  2130 * New syntax for domain package: selector names are now optional.
  2131 Parentheses should be omitted unless argument is lazy, for example:
  2132 
  2133   domain 'a stream = cons "'a" (lazy "'a stream")
  2134 
  2135 * New command 'fixrec' for defining recursive functions with pattern
  2136 matching; defining multiple functions with mutual recursion is also
  2137 supported.  Patterns may include the constants cpair, spair, up, sinl,
  2138 sinr, or any data constructor defined by the domain package. The given
  2139 equations are proven as rewrite rules. See HOLCF/ex/Fixrec_ex.thy for
  2140 syntax and examples.
  2141 
  2142 * New commands 'cpodef' and 'pcpodef' for defining predicate subtypes
  2143 of cpo and pcpo types. Syntax is exactly like the 'typedef' command,
  2144 but the proof obligation additionally includes an admissibility
  2145 requirement. The packages generate instances of class cpo or pcpo,
  2146 with continuity and strictness theorems for Rep and Abs.
  2147 
  2148 * HOLCF: Many theorems have been renamed according to a more standard naming
  2149 scheme (INCOMPATIBILITY):
  2150 
  2151   foo_inject:  "foo$x = foo$y ==> x = y"
  2152   foo_eq:      "(foo$x = foo$y) = (x = y)"
  2153   foo_less:    "(foo$x << foo$y) = (x << y)"
  2154   foo_strict:  "foo$UU = UU"
  2155   foo_defined: "... ==> foo$x ~= UU"
  2156   foo_defined_iff: "(foo$x = UU) = (x = UU)"
  2157 
  2158 
  2159 *** ZF ***
  2160 
  2161 * ZF/ex: theories Group and Ring provide examples in abstract algebra,
  2162 including the First Isomorphism Theorem (on quotienting by the kernel
  2163 of a homomorphism).
  2164 
  2165 * ZF/Simplifier: install second copy of type solver that actually
  2166 makes use of TC rules declared to Isar proof contexts (or locales);
  2167 the old version is still required for ML proof scripts.
  2168 
  2169 
  2170 *** Cube ***
  2171 
  2172 * Converted to Isar theory format; use locales instead of axiomatic
  2173 theories.
  2174 
  2175 
  2176 *** ML ***
  2177 
  2178 * Pure/library.ML: added ##>, ##>>, #>> -- higher-order counterparts
  2179 for ||>, ||>>, |>>,
  2180 
  2181 * Pure/library.ML no longer defines its own option datatype, but uses
  2182 that of the SML basis, which has constructors NONE and SOME instead of
  2183 None and Some, as well as exception Option.Option instead of OPTION.
  2184 The functions the, if_none, is_some, is_none have been adapted
  2185 accordingly, while Option.map replaces apsome.
  2186 
  2187 * Pure/library.ML: the exception LIST has been given up in favour of
  2188 the standard exceptions Empty and Subscript, as well as
  2189 Library.UnequalLengths.  Function like Library.hd and Library.tl are
  2190 superceded by the standard hd and tl functions etc.
  2191 
  2192 A number of basic list functions are no longer exported to the ML
  2193 toplevel, as they are variants of predefined functions.  The following
  2194 suggests how one can translate existing code:
  2195 
  2196     rev_append xs ys = List.revAppend (xs, ys)
  2197     nth_elem (i, xs) = List.nth (xs, i)
  2198     last_elem xs = List.last xs
  2199     flat xss = List.concat xss
  2200     seq fs = List.app fs
  2201     partition P xs = List.partition P xs
  2202     mapfilter f xs = List.mapPartial f xs
  2203 
  2204 * Pure/library.ML: several combinators for linear functional
  2205 transformations, notably reverse application and composition:
  2206 
  2207   x |> f                f #> g
  2208   (x, y) |-> f          f #-> g
  2209 
  2210 * Pure/library.ML: introduced/changed precedence of infix operators:
  2211 
  2212   infix 1 |> |-> ||> ||>> |>> |>>> #> #->;
  2213   infix 2 ?;
  2214   infix 3 o oo ooo oooo;
  2215   infix 4 ~~ upto downto;
  2216 
  2217 Maybe INCOMPATIBILITY when any of those is used in conjunction with other
  2218 infix operators.
  2219 
  2220 * Pure/library.ML: natural list combinators fold, fold_rev, and
  2221 fold_map support linear functional transformations and nesting.  For
  2222 example:
  2223 
  2224   fold f [x1, ..., xN] y =
  2225     y |> f x1 |> ... |> f xN
  2226 
  2227   (fold o fold) f [xs1, ..., xsN] y =
  2228     y |> fold f xs1 |> ... |> fold f xsN
  2229 
  2230   fold f [x1, ..., xN] =
  2231     f x1 #> ... #> f xN
  2232 
  2233   (fold o fold) f [xs1, ..., xsN] =
  2234     fold f xs1 #> ... #> fold f xsN
  2235 
  2236 * Pure/library.ML: the following selectors on type 'a option are
  2237 available:
  2238 
  2239   the:               'a option -> 'a  (*partial*)
  2240   these:             'a option -> 'a  where 'a = 'b list
  2241   the_default: 'a -> 'a option -> 'a
  2242   the_list:          'a option -> 'a list
  2243 
  2244 * Pure/General: structure AList (cf. Pure/General/alist.ML) provides
  2245 basic operations for association lists, following natural argument
  2246 order; moreover the explicit equality predicate passed here avoids
  2247 potentially expensive polymorphic runtime equality checks.
  2248 The old functions may be expressed as follows:
  2249 
  2250   assoc = uncurry (AList.lookup (op =))
  2251   assocs = these oo AList.lookup (op =)
  2252   overwrite = uncurry (AList.update (op =)) o swap
  2253 
  2254 * Pure/General: structure AList (cf. Pure/General/alist.ML) provides
  2255 
  2256   val make: ('a -> 'b) -> 'a list -> ('a * 'b) list
  2257   val find: ('a * 'b -> bool) -> ('c * 'b) list -> 'a -> 'c list
  2258 
  2259 replacing make_keylist and keyfilter (occassionally used)
  2260 Naive rewrites:
  2261 
  2262   make_keylist = AList.make
  2263   keyfilter = AList.find (op =)
  2264 
  2265 * eq_fst and eq_snd now take explicit equality parameter, thus
  2266   avoiding eqtypes. Naive rewrites:
  2267 
  2268     eq_fst = eq_fst (op =)
  2269     eq_snd = eq_snd (op =)
  2270 
  2271 * Removed deprecated apl and apr (rarely used).
  2272   Naive rewrites:
  2273 
  2274     apl (n, op) =>>= curry op n
  2275     apr (op, m) =>>= fn n => op (n, m)
  2276 
  2277 * Pure/General: structure OrdList (cf. Pure/General/ord_list.ML)
  2278 provides a reasonably efficient light-weight implementation of sets as
  2279 lists.
  2280 
  2281 * Pure/General: generic tables (cf. Pure/General/table.ML) provide a
  2282 few new operations; existing lookup and update are now curried to
  2283 follow natural argument order (for use with fold etc.);
  2284 INCOMPATIBILITY, use (uncurry Symtab.lookup) etc. as last resort.
  2285 
  2286 * Pure/General: output via the Isabelle channels of
  2287 writeln/warning/error etc. is now passed through Output.output, with a
  2288 hook for arbitrary transformations depending on the print_mode
  2289 (cf. Output.add_mode -- the first active mode that provides a output
  2290 function wins).  Already formatted output may be embedded into further
  2291 text via Output.raw; the result of Pretty.string_of/str_of and derived
  2292 functions (string_of_term/cterm/thm etc.) is already marked raw to
  2293 accommodate easy composition of diagnostic messages etc.  Programmers
  2294 rarely need to care about Output.output or Output.raw at all, with
  2295 some notable exceptions: Output.output is required when bypassing the
  2296 standard channels (writeln etc.), or in token translations to produce
  2297 properly formatted results; Output.raw is required when capturing
  2298 already output material that will eventually be presented to the user
  2299 a second time.  For the default print mode, both Output.output and
  2300 Output.raw have no effect.
  2301 
  2302 * Pure/General: Output.time_accumulator NAME creates an operator ('a
  2303 -> 'b) -> 'a -> 'b to measure runtime and count invocations; the
  2304 cumulative results are displayed at the end of a batch session.
  2305 
  2306 * Pure/General: File.sysify_path and File.quote_sysify path have been
  2307 replaced by File.platform_path and File.shell_path (with appropriate
  2308 hooks).  This provides a clean interface for unusual systems where the
  2309 internal and external process view of file names are different.
  2310 
  2311 * Pure: more efficient orders for basic syntactic entities: added
  2312 fast_string_ord, fast_indexname_ord, fast_term_ord; changed sort_ord
  2313 and typ_ord to use fast_string_ord and fast_indexname_ord (term_ord is
  2314 NOT affected); structures Symtab, Vartab, Typtab, Termtab use the fast
  2315 orders now -- potential INCOMPATIBILITY for code that depends on a
  2316 particular order for Symtab.keys, Symtab.dest, etc. (consider using
  2317 Library.sort_strings on result).
  2318 
  2319 * Pure/term.ML: combinators fold_atyps, fold_aterms, fold_term_types,
  2320 fold_types traverse types/terms from left to right, observing natural
  2321 argument order.  Supercedes previous foldl_XXX versions, add_frees,
  2322 add_vars etc. have been adapted as well: INCOMPATIBILITY.
  2323 
  2324 * Pure: name spaces have been refined, with significant changes of the
  2325 internal interfaces -- INCOMPATIBILITY.  Renamed cond_extern(_table)
  2326 to extern(_table).  The plain name entry path is superceded by a
  2327 general 'naming' context, which also includes the 'policy' to produce
  2328 a fully qualified name and external accesses of a fully qualified
  2329 name; NameSpace.extend is superceded by context dependent
  2330 Sign.declare_name.  Several theory and proof context operations modify
  2331 the naming context.  Especially note Theory.restore_naming and
  2332 ProofContext.restore_naming to get back to a sane state; note that
  2333 Theory.add_path is no longer sufficient to recover from
  2334 Theory.absolute_path in particular.
  2335 
  2336 * Pure: new flags short_names (default false) and unique_names
  2337 (default true) for controlling output of qualified names.  If
  2338 short_names is set, names are printed unqualified.  If unique_names is
  2339 reset, the name prefix is reduced to the minimum required to achieve
  2340 the original result when interning again, even if there is an overlap
  2341 with earlier declarations.
  2342 
  2343 * Pure/TheoryDataFun: change of the argument structure; 'prep_ext' is
  2344 now 'extend', and 'merge' gets an additional Pretty.pp argument
  2345 (useful for printing error messages).  INCOMPATIBILITY.
  2346 
  2347 * Pure: major reorganization of the theory context.  Type Sign.sg and
  2348 Theory.theory are now identified, referring to the universal
  2349 Context.theory (see Pure/context.ML).  Actual signature and theory
  2350 content is managed as theory data.  The old code and interfaces were
  2351 spread over many files and structures; the new arrangement introduces
  2352 considerable INCOMPATIBILITY to gain more clarity:
  2353 
  2354   Context -- theory management operations (name, identity, inclusion,
  2355     parents, ancestors, merge, etc.), plus generic theory data;
  2356 
  2357   Sign -- logical signature and syntax operations (declaring consts,
  2358     types, etc.), plus certify/read for common entities;
  2359 
  2360   Theory -- logical theory operations (stating axioms, definitions,
  2361     oracles), plus a copy of logical signature operations (consts,
  2362     types, etc.); also a few basic management operations (Theory.copy,
  2363     Theory.merge, etc.)
  2364 
  2365 The most basic sign_of operations (Theory.sign_of, Thm.sign_of_thm
  2366 etc.) as well as the sign field in Thm.rep_thm etc. have been retained
  2367 for convenience -- they merely return the theory.
  2368 
  2369 * Pure: type Type.tsig is superceded by theory in most interfaces.
  2370 
  2371 * Pure: the Isar proof context type is already defined early in Pure
  2372 as Context.proof (note that ProofContext.context and Proof.context are
  2373 aliases, where the latter is the preferred name).  This enables other
  2374 Isabelle components to refer to that type even before Isar is present.
  2375 
  2376 * Pure/sign/theory: discontinued named name spaces (i.e. classK,
  2377 typeK, constK, axiomK, oracleK), but provide explicit operations for
  2378 any of these kinds.  For example, Sign.intern typeK is now
  2379 Sign.intern_type, Theory.hide_space Sign.typeK is now
  2380 Theory.hide_types.  Also note that former
  2381 Theory.hide_classes/types/consts are now
  2382 Theory.hide_classes_i/types_i/consts_i, while the non '_i' versions
  2383 internalize their arguments!  INCOMPATIBILITY.
  2384 
  2385 * Pure: get_thm interface (of PureThy and ProofContext) expects
  2386 datatype thmref (with constructors Name and NameSelection) instead of
  2387 plain string -- INCOMPATIBILITY;
  2388 
  2389 * Pure: cases produced by proof methods specify options, where NONE
  2390 means to remove case bindings -- INCOMPATIBILITY in
  2391 (RAW_)METHOD_CASES.
  2392 
  2393 * Pure: the following operations retrieve axioms or theorems from a
  2394 theory node or theory hierarchy, respectively:
  2395 
  2396   Theory.axioms_of: theory -> (string * term) list
  2397   Theory.all_axioms_of: theory -> (string * term) list
  2398   PureThy.thms_of: theory -> (string * thm) list
  2399   PureThy.all_thms_of: theory -> (string * thm) list
  2400 
  2401 * Pure: print_tac now outputs the goal through the trace channel.
  2402 
  2403 * Isar toplevel: improved diagnostics, mostly for Poly/ML only.
  2404 Reference Toplevel.debug (default false) controls detailed printing
  2405 and tracing of low-level exceptions; Toplevel.profiling (default 0)
  2406 controls execution profiling -- set to 1 for time and 2 for space
  2407 (both increase the runtime).
  2408 
  2409 * Isar session: The initial use of ROOT.ML is now always timed,
  2410 i.e. the log will show the actual process times, in contrast to the
  2411 elapsed wall-clock time that the outer shell wrapper produces.
  2412 
  2413 * Simplifier: improved handling of bound variables (nameless
  2414 representation, avoid allocating new strings).  Simprocs that invoke
  2415 the Simplifier recursively should use Simplifier.inherit_bounds to
  2416 avoid local name clashes.  Failure to do so produces warnings
  2417 "Simplifier: renamed bound variable ..."; set Simplifier.debug_bounds
  2418 for further details.
  2419 
  2420 * ML functions legacy_bindings and use_legacy_bindings produce ML fact
  2421 bindings for all theorems stored within a given theory; this may help
  2422 in porting non-Isar theories to Isar ones, while keeping ML proof
  2423 scripts for the time being.
  2424 
  2425 * ML operator HTML.with_charset specifies the charset begin used for
  2426 generated HTML files.  For example:
  2427 
  2428   HTML.with_charset "utf-8" use_thy "Hebrew";
  2429   HTML.with_charset "utf-8" use_thy "Chinese";
  2430 
  2431 
  2432 *** System ***
  2433 
  2434 * Allow symlinks to all proper Isabelle executables (Isabelle,
  2435 isabelle, isatool etc.).
  2436 
  2437 * ISABELLE_DOC_FORMAT setting specifies preferred document format (for
  2438 isatool doc, isatool mkdir, display_drafts etc.).
  2439 
  2440 * isatool usedir: option -f allows specification of the ML file to be
  2441 used by Isabelle; default is ROOT.ML.
  2442 
  2443 * New isatool version outputs the version identifier of the Isabelle
  2444 distribution being used.
  2445 
  2446 * HOL: new isatool dimacs2hol converts files in DIMACS CNF format
  2447 (containing Boolean satisfiability problems) into Isabelle/HOL
  2448 theories.
  2449 
  2450 
  2451 
  2452 New in Isabelle2004 (April 2004)
  2453 --------------------------------
  2454 
  2455 *** General ***
  2456 
  2457 * Provers/order.ML:  new efficient reasoner for partial and linear orders.
  2458   Replaces linorder.ML.
  2459 
  2460 * Pure: Greek letters (except small lambda, \<lambda>), as well as Gothic
  2461   (\<aa>...\<zz>\<AA>...\<ZZ>), calligraphic (\<A>...\<Z>), and Euler
  2462   (\<a>...\<z>), are now considered normal letters, and can therefore
  2463   be used anywhere where an ASCII letter (a...zA...Z) has until
  2464   now. COMPATIBILITY: This obviously changes the parsing of some
  2465   terms, especially where a symbol has been used as a binder, say
  2466   '\<Pi>x. ...', which is now a type error since \<Pi>x will be parsed
  2467   as an identifier.  Fix it by inserting a space around former
  2468   symbols.  Call 'isatool fixgreek' to try to fix parsing errors in
  2469   existing theory and ML files.
  2470 
  2471 * Pure: Macintosh and Windows line-breaks are now allowed in theory files.
  2472 
  2473 * Pure: single letter sub/superscripts (\<^isub> and \<^isup>) are now
  2474   allowed in identifiers. Similar to Greek letters \<^isub> is now considered
  2475   a normal (but invisible) letter. For multiple letter subscripts repeat
  2476   \<^isub> like this: x\<^isub>1\<^isub>2.
  2477 
  2478 * Pure: There are now sub-/superscripts that can span more than one
  2479   character. Text between \<^bsub> and \<^esub> is set in subscript in
  2480   ProofGeneral and LaTeX, text between \<^bsup> and \<^esup> in
  2481   superscript. The new control characters are not identifier parts.
  2482 
  2483 * Pure: Control-symbols of the form \<^raw:...> will literally print the
  2484   content of "..." to the latex file instead of \isacntrl... . The "..."
  2485   may consist of any printable characters excluding the end bracket >.
  2486 
  2487 * Pure: Using new Isar command "finalconsts" (or the ML functions
  2488   Theory.add_finals or Theory.add_finals_i) it is now possible to
  2489   declare constants "final", which prevents their being given a definition
  2490   later.  It is useful for constants whose behaviour is fixed axiomatically
  2491   rather than definitionally, such as the meta-logic connectives.
  2492 
  2493 * Pure: 'instance' now handles general arities with general sorts
  2494   (i.e. intersections of classes),
  2495 
  2496 * Presentation: generated HTML now uses a CSS style sheet to make layout
  2497   (somewhat) independent of content. It is copied from lib/html/isabelle.css.
  2498   It can be changed to alter the colors/layout of generated pages.
  2499 
  2500 
  2501 *** Isar ***
  2502 
  2503 * Tactic emulation methods rule_tac, erule_tac, drule_tac, frule_tac,
  2504   cut_tac, subgoal_tac and thin_tac:
  2505   - Now understand static (Isar) contexts.  As a consequence, users of Isar
  2506     locales are no longer forced to write Isar proof scripts.
  2507     For details see Isar Reference Manual, paragraph 4.3.2: Further tactic
  2508     emulations.
  2509   - INCOMPATIBILITY: names of variables to be instantiated may no
  2510     longer be enclosed in quotes.  Instead, precede variable name with `?'.
  2511     This is consistent with the instantiation attribute "where".
  2512 
  2513 * Attributes "where" and "of":
  2514   - Now take type variables of instantiated theorem into account when reading
  2515     the instantiation string.  This fixes a bug that caused instantiated
  2516     theorems to have too special types in some circumstances.
  2517   - "where" permits explicit instantiations of type variables.
  2518 
  2519 * Calculation commands "moreover" and "also" no longer interfere with
  2520   current facts ("this"), admitting arbitrary combinations with "then"
  2521   and derived forms.
  2522 
  2523 * Locales:
  2524   - Goal statements involving the context element "includes" no longer
  2525     generate theorems with internal delta predicates (those ending on
  2526     "_axioms") in the premise.
  2527     Resolve particular premise with <locale>.intro to obtain old form.
  2528   - Fixed bug in type inference ("unify_frozen") that prevented mix of target
  2529     specification and "includes" elements in goal statement.
  2530   - Rule sets <locale>.intro and <locale>.axioms no longer declared as
  2531     [intro?] and [elim?] (respectively) by default.
  2532   - Experimental command for instantiation of locales in proof contexts:
  2533         instantiate <label>[<attrs>]: <loc>
  2534     Instantiates locale <loc> and adds all its theorems to the current context
  2535     taking into account their attributes.  Label and attrs are optional
  2536     modifiers, like in theorem declarations.  If present, names of
  2537     instantiated theorems are qualified with <label>, and the attributes
  2538     <attrs> are applied after any attributes these theorems might have already.
  2539       If the locale has assumptions, a chained fact of the form
  2540     "<loc> t1 ... tn" is expected from which instantiations of the parameters
  2541     are derived.  The command does not support old-style locales declared
  2542     with "locale (open)".
  2543       A few (very simple) examples can be found in FOL/ex/LocaleInst.thy.
  2544 
  2545 * HOL: Tactic emulation methods induct_tac and case_tac understand static
  2546   (Isar) contexts.
  2547 
  2548 
  2549 *** HOL ***
  2550 
  2551 * Proof import: new image HOL4 contains the imported library from
  2552   the HOL4 system with about 2500 theorems. It is imported by
  2553   replaying proof terms produced by HOL4 in Isabelle. The HOL4 image
  2554   can be used like any other Isabelle image.  See
  2555   HOL/Import/HOL/README for more information.
  2556 
  2557 * Simplifier:
  2558   - Much improved handling of linear and partial orders.
  2559     Reasoners for linear and partial orders are set up for type classes
  2560     "linorder" and "order" respectively, and are added to the default simpset
  2561     as solvers.  This means that the simplifier can build transitivity chains
  2562     to solve goals from the assumptions.
  2563   - INCOMPATIBILITY: old proofs break occasionally.  Typically, applications
  2564     of blast or auto after simplification become unnecessary because the goal
  2565     is solved by simplification already.
  2566 
  2567 * Numerics: new theory Ring_and_Field contains over 250 basic numerical laws,
  2568     all proved in axiomatic type classes for semirings, rings and fields.
  2569 
  2570 * Numerics:
  2571   - Numeric types (nat, int, and in HOL-Complex rat, real, complex, etc.) are
  2572     now formalized using the Ring_and_Field theory mentioned above.
  2573   - INCOMPATIBILITY: simplification and arithmetic behaves somewhat differently
  2574     than before, because now they are set up once in a generic manner.
  2575   - INCOMPATIBILITY: many type-specific arithmetic laws have gone.
  2576     Look for the general versions in Ring_and_Field (and Power if they concern
  2577     exponentiation).
  2578 
  2579 * Type "rat" of the rational numbers is now available in HOL-Complex.
  2580 
  2581 * Records:
  2582   - Record types are now by default printed with their type abbreviation
  2583     instead of the list of all field types. This can be configured via
  2584     the reference "print_record_type_abbr".
  2585   - Simproc "record_upd_simproc" for simplification of multiple updates added
  2586     (not enabled by default).
  2587   - Simproc "record_ex_sel_eq_simproc" to simplify EX x. sel r = x resp.
  2588     EX x. x = sel r to True (not enabled by default).
  2589   - Tactic "record_split_simp_tac" to split and simplify records added.
  2590 
  2591 * 'specification' command added, allowing for definition by
  2592   specification.  There is also an 'ax_specification' command that
  2593   introduces the new constants axiomatically.
  2594 
  2595 * arith(_tac) is now able to generate counterexamples for reals as well.
  2596 
  2597 * HOL-Algebra: new locale "ring" for non-commutative rings.
  2598 
  2599 * HOL-ex: InductiveInvariant_examples illustrates advanced recursive function
  2600   definitions, thanks to Sava Krsti\'{c} and John Matthews.
  2601 
  2602 * HOL-Matrix: a first theory for matrices in HOL with an application of
  2603   matrix theory to linear programming.
  2604 
  2605 * Unions and Intersections:
  2606   The latex output syntax of UN and INT has been changed
  2607   from "\Union x \in A. B" to "\Union_{x \in A} B"
  2608   i.e. the index formulae has become a subscript.
  2609   Similarly for "\Union x. B", and for \Inter instead of \Union.
  2610 
  2611 * Unions and Intersections over Intervals:
  2612   There is new short syntax "UN i<=n. A" for "UN i:{0..n}. A". There is
  2613   also an x-symbol version with subscripts "\<Union>\<^bsub>i <= n\<^esub>. A"
  2614   like in normal math, and corresponding versions for < and for intersection.
  2615 
  2616 * HOL/List: Ordering "lexico" is renamed "lenlex" and the standard
  2617   lexicographic dictonary ordering has been added as "lexord".
  2618 
  2619 * ML: the legacy theory structures Int and List have been removed. They had
  2620   conflicted with ML Basis Library structures having the same names.
  2621 
  2622 * 'refute' command added to search for (finite) countermodels.  Only works
  2623   for a fragment of HOL.  The installation of an external SAT solver is
  2624   highly recommended.  See "HOL/Refute.thy" for details.
  2625 
  2626 * 'quickcheck' command: Allows to find counterexamples by evaluating
  2627   formulae under an assignment of free variables to random values.
  2628   In contrast to 'refute', it can deal with inductive datatypes,
  2629   but cannot handle quantifiers. See "HOL/ex/Quickcheck_Examples.thy"
  2630   for examples.
  2631 
  2632 
  2633 *** HOLCF ***
  2634 
  2635 * Streams now come with concatenation and are part of the HOLCF image
  2636 
  2637 
  2638 
  2639 New in Isabelle2003 (May 2003)
  2640 ------------------------------
  2641 
  2642 *** General ***
  2643 
  2644 * Provers/simplifier:
  2645 
  2646   - Completely reimplemented method simp (ML: Asm_full_simp_tac):
  2647     Assumptions are now subject to complete mutual simplification,
  2648     not just from left to right. The simplifier now preserves
  2649     the order of assumptions.
  2650 
  2651     Potential INCOMPATIBILITY:
  2652 
  2653     -- simp sometimes diverges where the old version did
  2654        not, e.g. invoking simp on the goal
  2655 
  2656         [| P (f x); y = x; f x = f y |] ==> Q
  2657 
  2658        now gives rise to the infinite reduction sequence
  2659 
  2660         P(f x) --(f x = f y)--> P(f y) --(y = x)--> P(f x) --(f x = f y)--> ...
  2661 
  2662        Using "simp (asm_lr)" (ML: Asm_lr_simp_tac) instead often solves this
  2663        kind of problem.
  2664 
  2665     -- Tactics combining classical reasoner and simplification (such as auto)
  2666        are also affected by this change, because many of them rely on
  2667        simp. They may sometimes diverge as well or yield a different numbers
  2668        of subgoals. Try to use e.g. force, fastsimp, or safe instead of auto
  2669        in case of problems. Sometimes subsequent calls to the classical
  2670        reasoner will fail because a preceeding call to the simplifier too
  2671        eagerly simplified the goal, e.g. deleted redundant premises.
  2672 
  2673   - The simplifier trace now shows the names of the applied rewrite rules
  2674 
  2675   - You can limit the number of recursive invocations of the simplifier
  2676     during conditional rewriting (where the simplifie tries to solve the
  2677     conditions before applying the rewrite rule):
  2678     ML "simp_depth_limit := n"
  2679     where n is an integer. Thus you can force termination where previously
  2680     the simplifier would diverge.
  2681 
  2682   - Accepts free variables as head terms in congruence rules.  Useful in Isar.
  2683 
  2684   - No longer aborts on failed congruence proof.  Instead, the
  2685     congruence is ignored.
  2686 
  2687 * Pure: New generic framework for extracting programs from constructive
  2688   proofs. See HOL/Extraction.thy for an example instantiation, as well
  2689   as HOL/Extraction for some case studies.
  2690 
  2691 * Pure: The main goal of the proof state is no longer shown by default, only
  2692 the subgoals. This behaviour is controlled by a new flag.
  2693    PG menu: Isabelle/Isar -> Settings -> Show Main Goal
  2694 (ML: Proof.show_main_goal).
  2695 
  2696 * Pure: You can find all matching introduction rules for subgoal 1, i.e. all
  2697 rules whose conclusion matches subgoal 1:
  2698       PG menu: Isabelle/Isar -> Show me -> matching rules
  2699 The rules are ordered by how closely they match the subgoal.
  2700 In particular, rules that solve a subgoal outright are displayed first
  2701 (or rather last, the way they are printed).
  2702 (ML: ProofGeneral.print_intros())
  2703 
  2704 * Pure: New flag trace_unify_fail causes unification to print
  2705 diagnostic information (PG: in trace buffer) when it fails. This is
  2706 useful for figuring out why single step proofs like rule, erule or
  2707 assumption failed.
  2708 
  2709 * Pure: Locale specifications now produce predicate definitions
  2710 according to the body of text (covering assumptions modulo local
  2711 definitions); predicate "loc_axioms" covers newly introduced text,
  2712 while "loc" is cumulative wrt. all included locale expressions; the
  2713 latter view is presented only on export into the global theory
  2714 context; potential INCOMPATIBILITY, use "(open)" option to fall back
  2715 on the old view without predicates;
  2716 
  2717 * Pure: predefined locales "var" and "struct" are useful for sharing
  2718 parameters (as in CASL, for example); just specify something like
  2719 ``var x + var y + struct M'' as import;
  2720 
  2721 * Pure: improved thms_containing: proper indexing of facts instead of
  2722 raw theorems; check validity of results wrt. current name space;
  2723 include local facts of proof configuration (also covers active
  2724 locales), cover fixed variables in index; may use "_" in term
  2725 specification; an optional limit for the number of printed facts may
  2726 be given (the default is 40);
  2727 
  2728 * Pure: disallow duplicate fact bindings within new-style theory files
  2729 (batch-mode only);
  2730 
  2731 * Provers: improved induct method: assumptions introduced by case
  2732 "foo" are split into "foo.hyps" (from the rule) and "foo.prems" (from
  2733 the goal statement); "foo" still refers to all facts collectively;
  2734 
  2735 * Provers: the function blast.overloaded has been removed: all constants
  2736 are regarded as potentially overloaded, which improves robustness in exchange
  2737 for slight decrease in efficiency;
  2738 
  2739 * Provers/linorder: New generic prover for transitivity reasoning over
  2740 linear orders.  Note: this prover is not efficient!
  2741 
  2742 * Isar: preview of problems to finish 'show' now produce an error
  2743 rather than just a warning (in interactive mode);
  2744 
  2745 
  2746 *** HOL ***
  2747 
  2748 * arith(_tac)
  2749 
  2750  - Produces a counter example if it cannot prove a goal.
  2751    Note that the counter example may be spurious if the goal is not a formula
  2752    of quantifier-free linear arithmetic.
  2753    In ProofGeneral the counter example appears in the trace buffer.
  2754 
  2755  - Knows about div k and mod k where k is a numeral of type nat or int.
  2756 
  2757  - Calls full Presburger arithmetic (by Amine Chaieb) if quantifier-free
  2758    linear arithmetic fails. This takes account of quantifiers and divisibility.
  2759    Presburger arithmetic can also be called explicitly via presburger(_tac).
  2760 
  2761 * simp's arithmetic capabilities have been enhanced a bit: it now
  2762 takes ~= in premises into account (by performing a case split);
  2763 
  2764 * simp reduces "m*(n div m) + n mod m" to n, even if the two summands
  2765 are distributed over a sum of terms;
  2766 
  2767 * New tactic "trans_tac" and method "trans" instantiate
  2768 Provers/linorder.ML for axclasses "order" and "linorder" (predicates
  2769 "<=", "<" and "=").
  2770 
  2771 * function INCOMPATIBILITIES: Pi-sets have been redefined and moved from main
  2772 HOL to Library/FuncSet; constant "Fun.op o" is now called "Fun.comp";
  2773 
  2774 * 'typedef' command has new option "open" to suppress the set
  2775 definition;
  2776 
  2777 * functions Min and Max on finite sets have been introduced (theory
  2778 Finite_Set);
  2779 
  2780 * attribute [symmetric] now works for relations as well; it turns
  2781 (x,y) : R^-1 into (y,x) : R, and vice versa;
  2782 
  2783 * induct over a !!-quantified statement (say !!x1..xn):
  2784   each "case" automatically performs "fix x1 .. xn" with exactly those names.
  2785 
  2786 * Map: `empty' is no longer a constant but a syntactic abbreviation for
  2787 %x. None. Warning: empty_def now refers to the previously hidden definition
  2788 of the empty set.
  2789 
  2790 * Algebra: formalization of classical algebra.  Intended as base for
  2791 any algebraic development in Isabelle.  Currently covers group theory
  2792 (up to Sylow's theorem) and ring theory (Universal Property of
  2793 Univariate Polynomials).  Contributions welcome;
  2794 
  2795 * GroupTheory: deleted, since its material has been moved to Algebra;
  2796 
  2797 * Complex: new directory of the complex numbers with numeric constants,
  2798 nonstandard complex numbers, and some complex analysis, standard and
  2799 nonstandard (Jacques Fleuriot);
  2800 
  2801 * HOL-Complex: new image for analysis, replacing HOL-Real and HOL-Hyperreal;
  2802 
  2803 * Hyperreal: introduced Gauge integration and hyperreal logarithms (Jacques
  2804 Fleuriot);
  2805 
  2806 * Real/HahnBanach: updated and adapted to locales;
  2807 
  2808 * NumberTheory: added Gauss's law of quadratic reciprocity (by Avigad,
  2809 Gray and Kramer);
  2810 
  2811 * UNITY: added the Meier-Sanders theory of progress sets;
  2812 
  2813 * MicroJava: bytecode verifier and lightweight bytecode verifier
  2814 as abstract algorithms, instantiated to the JVM;
  2815 
  2816 * Bali: Java source language formalization. Type system, operational
  2817 semantics, axiomatic semantics. Supported language features:
  2818 classes, interfaces, objects,virtual methods, static methods,
  2819 static/instance fields, arrays, access modifiers, definite
  2820 assignment, exceptions.
  2821 
  2822 
  2823 *** ZF ***
  2824 
  2825 * ZF/Constructible: consistency proof for AC (Gdel's constructible
  2826 universe, etc.);
  2827 
  2828 * Main ZF: virtually all theories converted to new-style format;
  2829 
  2830 
  2831 *** ML ***
  2832 
  2833 * Pure: Tactic.prove provides sane interface for internal proofs;
  2834 omits the infamous "standard" operation, so this is more appropriate
  2835 than prove_goalw_cterm in many situations (e.g. in simprocs);
  2836 
  2837 * Pure: improved error reporting of simprocs;
  2838 
  2839 * Provers: Simplifier.simproc(_i) provides sane interface for setting
  2840 up simprocs;
  2841 
  2842 
  2843 *** Document preparation ***
  2844 
  2845 * uses \par instead of \\ for line breaks in theory text. This may
  2846 shift some page breaks in large documents. To get the old behaviour
  2847 use \renewcommand{\isanewline}{\mbox{}\\\mbox{}} in root.tex.
  2848 
  2849 * minimized dependencies of isabelle.sty and isabellesym.sty on
  2850 other packages
  2851 
  2852 * \<euro> now needs package babel/greek instead of marvosym (which
  2853 broke \Rightarrow)
  2854 
  2855 * normal size for \<zero>...\<nine> (uses \mathbf instead of
  2856 textcomp package)
  2857 
  2858 
  2859 
  2860 New in Isabelle2002 (March 2002)
  2861 --------------------------------
  2862 
  2863 *** Document preparation ***
  2864 
  2865 * greatly simplified document preparation setup, including more
  2866 graceful interpretation of isatool usedir -i/-d/-D options, and more
  2867 instructive isatool mkdir; users should basically be able to get
  2868 started with "isatool mkdir HOL Test && isatool make"; alternatively,
  2869 users may run a separate document processing stage manually like this:
  2870 "isatool usedir -D output HOL Test && isatool document Test/output";
  2871 
  2872 * theory dependency graph may now be incorporated into documents;
  2873 isatool usedir -g true will produce session_graph.eps/.pdf for use
  2874 with \includegraphics of LaTeX;
  2875 
  2876 * proper spacing of consecutive markup elements, especially text
  2877 blocks after section headings;
  2878 
  2879 * support bold style (for single symbols only), input syntax is like
  2880 this: "\<^bold>\<alpha>" or "\<^bold>A";
  2881 
  2882 * \<bullet> is now output as bold \cdot by default, which looks much
  2883 better in printed text;
  2884 
  2885 * added default LaTeX bindings for \<tturnstile> and \<TTurnstile>;
  2886 note that these symbols are currently unavailable in Proof General /
  2887 X-Symbol; new symbols \<zero>, \<one>, ..., \<nine>, and \<euro>;
  2888 
  2889 * isatool latex no longer depends on changed TEXINPUTS, instead
  2890 isatool document copies the Isabelle style files to the target
  2891 location;
  2892 
  2893 
  2894 *** Isar ***
  2895 
  2896 * Pure/Provers: improved proof by cases and induction;
  2897   - 'case' command admits impromptu naming of parameters (such as
  2898     "case (Suc n)");
  2899   - 'induct' method divinates rule instantiation from the inductive
  2900     claim; no longer requires excessive ?P bindings for proper
  2901     instantiation of cases;
  2902   - 'induct' method properly enumerates all possibilities of set/type
  2903     rules; as a consequence facts may be also passed through *type*
  2904     rules without further ado;
  2905   - 'induct' method now derives symbolic cases from the *rulified*
  2906     rule (before it used to rulify cases stemming from the internal
  2907     atomized version); this means that the context of a non-atomic
  2908     statement becomes is included in the hypothesis, avoiding the
  2909     slightly cumbersome show "PROP ?case" form;
  2910   - 'induct' may now use elim-style induction rules without chaining
  2911     facts, using ``missing'' premises from the goal state; this allows
  2912     rules stemming from inductive sets to be applied in unstructured
  2913     scripts, while still benefitting from proper handling of non-atomic
  2914     statements; NB: major inductive premises need to be put first, all
  2915     the rest of the goal is passed through the induction;
  2916   - 'induct' proper support for mutual induction involving non-atomic
  2917     rule statements (uses the new concept of simultaneous goals, see
  2918     below);
  2919   - append all possible rule selections, but only use the first
  2920     success (no backtracking);
  2921   - removed obsolete "(simplified)" and "(stripped)" options of methods;
  2922   - undeclared rule case names default to numbers 1, 2, 3, ...;
  2923   - added 'print_induct_rules' (covered by help item in recent Proof
  2924     General versions);
  2925   - moved induct/cases attributes to Pure, methods to Provers;
  2926   - generic method setup instantiated for FOL and HOL;
  2927 
  2928 * Pure: support multiple simultaneous goal statements, for example
  2929 "have a: A and b: B" (same for 'theorem' etc.); being a pure
  2930 meta-level mechanism, this acts as if several individual goals had
  2931 been stated separately; in particular common proof methods need to be
  2932 repeated in order to cover all claims; note that a single elimination
  2933 step is *not* sufficient to establish the two conjunctions, so this
  2934 fails:
  2935 
  2936   assume "A & B" then have A and B ..   (*".." fails*)
  2937 
  2938 better use "obtain" in situations as above; alternative refer to
  2939 multi-step methods like 'auto', 'simp_all', 'blast+' etc.;
  2940 
  2941 * Pure: proper integration with ``locales''; unlike the original
  2942 version by Florian Kammller, Isar locales package high-level proof
  2943 contexts rather than raw logical ones (e.g. we admit to include
  2944 attributes everywhere); operations on locales include merge and
  2945 rename; support for implicit arguments (``structures''); simultaneous
  2946 type-inference over imports and text; see also HOL/ex/Locales.thy for
  2947 some examples;
  2948 
  2949 * Pure: the following commands have been ``localized'', supporting a
  2950 target locale specification "(in name)": 'lemma', 'theorem',
  2951 'corollary', 'lemmas', 'theorems', 'declare'; the results will be
  2952 stored both within the locale and at the theory level (exported and
  2953 qualified by the locale name);
  2954 
  2955 * Pure: theory goals may now be specified in ``long'' form, with
  2956 ad-hoc contexts consisting of arbitrary locale elements. for example
  2957 ``lemma foo: fixes x assumes "A x" shows "B x"'' (local syntax and
  2958 definitions may be given, too); the result is a meta-level rule with
  2959 the context elements being discharged in the obvious way;
  2960 
  2961 * Pure: new proof command 'using' allows to augment currently used
  2962 facts after a goal statement ('using' is syntactically analogous to
  2963 'apply', but acts on the goal's facts only); this allows chained facts
  2964 to be separated into parts given before and after a claim, as in
  2965 ``from a and b have C using d and e <proof>'';
  2966 
  2967 * Pure: renamed "antecedent" case to "rule_context";
  2968 
  2969 * Pure: new 'judgment' command records explicit information about the
  2970 object-logic embedding (used by several tools internally); no longer
  2971 use hard-wired "Trueprop";
  2972 
  2973 * Pure: added 'corollary' command;
  2974 
  2975 * Pure: fixed 'token_translation' command;
  2976 
  2977 * Pure: removed obsolete 'exported' attribute;
  2978 
  2979 * Pure: dummy pattern "_" in is/let is now automatically lifted over
  2980 bound variables: "ALL x. P x --> Q x" (is "ALL x. _ --> ?C x")
  2981 supersedes more cumbersome ... (is "ALL x. _ x --> ?C x");
  2982 
  2983 * Pure: method 'atomize' presents local goal premises as object-level
  2984 statements (atomic meta-level propositions); setup controlled via
  2985 rewrite rules declarations of 'atomize' attribute; example
  2986 application: 'induct' method with proper rule statements in improper
  2987 proof *scripts*;
  2988 
  2989 * Pure: emulation of instantiation tactics (rule_tac, cut_tac, etc.)
  2990 now consider the syntactic context of assumptions, giving a better
  2991 chance to get type-inference of the arguments right (this is
  2992 especially important for locales);
  2993 
  2994 * Pure: "sorry" no longer requires quick_and_dirty in interactive
  2995 mode;
  2996 
  2997 * Pure/obtain: the formal conclusion "thesis", being marked as
  2998 ``internal'', may no longer be reference directly in the text;
  2999 potential INCOMPATIBILITY, may need to use "?thesis" in rare
  3000 situations;
  3001 
  3002 * Pure: generic 'sym' attribute which declares a rule both as pure
  3003 'elim?' and for the 'symmetric' operation;
  3004 
  3005 * Pure: marginal comments ``--'' may now occur just anywhere in the
  3006 text; the fixed correlation with particular command syntax has been
  3007 discontinued;
  3008 
  3009 * Pure: new method 'rules' is particularly well-suited for proof
  3010 search in intuitionistic logic; a bit slower than 'blast' or 'fast',
  3011 but often produces more compact proof terms with less detours;
  3012 
  3013 * Pure/Provers/classical: simplified integration with pure rule
  3014 attributes and methods; the classical "intro?/elim?/dest?"
  3015 declarations coincide with the pure ones; the "rule" method no longer
  3016 includes classically swapped intros; "intro" and "elim" methods no
  3017 longer pick rules from the context; also got rid of ML declarations
  3018 AddXIs/AddXEs/AddXDs; all of this has some potential for
  3019 INCOMPATIBILITY;
  3020 
  3021 * Provers/classical: attribute 'swapped' produces classical inversions
  3022 of introduction rules;
  3023 
  3024 * Provers/simplifier: 'simplified' attribute may refer to explicit
  3025 rules instead of full simplifier context; 'iff' attribute handles
  3026 conditional rules;
  3027 
  3028 * HOL: 'typedef' now allows alternative names for Rep/Abs morphisms;
  3029 
  3030 * HOL: 'recdef' now fails on unfinished automated proofs, use
  3031 "(permissive)" option to recover old behavior;
  3032 
  3033 * HOL: 'inductive' no longer features separate (collective) attributes
  3034 for 'intros' (was found too confusing);
  3035 
  3036 * HOL: properly declared induction rules less_induct and
  3037 wf_induct_rule;
  3038 
  3039 
  3040 *** HOL ***
  3041 
  3042 * HOL: moved over to sane numeral syntax; the new policy is as
  3043 follows:
  3044 
  3045   - 0 and 1 are polymorphic constants, which are defined on any
  3046   numeric type (nat, int, real etc.);
  3047 
  3048   - 2, 3, 4, ... and -1, -2, -3, ... are polymorphic numerals, based
  3049   binary representation internally;
  3050 
  3051   - type nat has special constructor Suc, and generally prefers Suc 0
  3052   over 1::nat and Suc (Suc 0) over 2::nat;
  3053 
  3054 This change may cause significant problems of INCOMPATIBILITY; here
  3055 are some hints on converting existing sources:
  3056 
  3057   - due to the new "num" token, "-0" and "-1" etc. are now atomic
  3058   entities, so expressions involving "-" (unary or binary minus) need
  3059   to be spaced properly;
  3060 
  3061   - existing occurrences of "1" may need to be constraint "1::nat" or
  3062   even replaced by Suc 0; similar for old "2";
  3063 
  3064   - replace "#nnn" by "nnn", and "#-nnn" by "-nnn";
  3065 
  3066   - remove all special provisions on numerals in proofs;
  3067 
  3068 * HOL: simp rules nat_number expand numerals on nat to Suc/0
  3069 representation (depends on bin_arith_simps in the default context);
  3070 
  3071 * HOL: symbolic syntax for x^2 (numeral 2);
  3072 
  3073 * HOL: the class of all HOL types is now called "type" rather than
  3074 "term"; INCOMPATIBILITY, need to adapt references to this type class
  3075 in axclass/classes, instance/arities, and (usually rare) occurrences
  3076 in typings (of consts etc.); internally the class is called
  3077 "HOL.type", ML programs should refer to HOLogic.typeS;
  3078 
  3079 * HOL/record package improvements:
  3080   - new derived operations "fields" to build a partial record section,
  3081     "extend" to promote a fixed record to a record scheme, and
  3082     "truncate" for the reverse; cf. theorems "xxx.defs", which are *not*
  3083     declared as simp by default;
  3084   - shared operations ("more", "fields", etc.) now need to be always
  3085     qualified) --- potential INCOMPATIBILITY;
  3086   - removed "make_scheme" operations (use "make" with "extend") --
  3087     INCOMPATIBILITY;
  3088   - removed "more" class (simply use "term") -- INCOMPATIBILITY;
  3089   - provides cases/induct rules for use with corresponding Isar
  3090     methods (for concrete records, record schemes, concrete more
  3091     parts, and schematic more parts -- in that order);
  3092   - internal definitions directly based on a light-weight abstract
  3093     theory of product types over typedef rather than datatype;
  3094 
  3095 * HOL: generic code generator for generating executable ML code from
  3096 specifications; specific support for HOL constructs such as inductive
  3097 datatypes and sets, as well as recursive functions; can be invoked
  3098 via 'generate_code' theory section;
  3099 
  3100 * HOL: canonical cases/induct rules for n-tuples (n = 3..7);
  3101 
  3102 * HOL: consolidated and renamed several theories.  In particular:
  3103         Ord.thy has been absorbed into HOL.thy
  3104         String.thy has been absorbed into List.thy
  3105 
  3106 * HOL: concrete setsum syntax "\<Sum>i:A. b" == "setsum (%i. b) A"
  3107 (beware of argument permutation!);
  3108 
  3109 * HOL: linorder_less_split superseded by linorder_cases;
  3110 
  3111 * HOL/List: "nodups" renamed to "distinct";
  3112 
  3113 * HOL: added "The" definite description operator; move Hilbert's "Eps"
  3114 to peripheral theory "Hilbert_Choice"; some INCOMPATIBILITIES:
  3115   - Ex_def has changed, now need to use some_eq_ex
  3116 
  3117 * HOL: made split_all_tac safe; EXISTING PROOFS MAY FAIL OR LOOP, so
  3118 in this (rare) case use:
  3119 
  3120   delSWrapper "split_all_tac"
  3121   addSbefore ("unsafe_split_all_tac", unsafe_split_all_tac)
  3122 
  3123 * HOL: added safe wrapper "split_conv_tac" to claset; EXISTING PROOFS
  3124 MAY FAIL;
  3125 
  3126 * HOL: introduced f^n = f o ... o f; warning: due to the limits of
  3127 Isabelle's type classes, ^ on functions and relations has too general
  3128 a domain, namely ('a * 'b) set and 'a => 'b; this means that it may be
  3129 necessary to attach explicit type constraints;
  3130 
  3131 * HOL/Relation: the prefix name of the infix "O" has been changed from
  3132 "comp" to "rel_comp"; INCOMPATIBILITY: a few theorems have been
  3133 renamed accordingly (eg "compI" -> "rel_compI").
  3134 
  3135 * HOL: syntax translations now work properly with numerals and records
  3136 expressions;
  3137 
  3138 * HOL: bounded abstraction now uses syntax "%" / "\<lambda>" instead
  3139 of "lam" -- INCOMPATIBILITY;
  3140 
  3141 * HOL: got rid of some global declarations (potential INCOMPATIBILITY
  3142 for ML tools): const "()" renamed "Product_Type.Unity", type "unit"
  3143 renamed "Product_Type.unit";
  3144 
  3145 * HOL: renamed rtrancl_into_rtrancl2 to converse_rtrancl_into_rtrancl
  3146 
  3147 * HOL: removed obsolete theorem "optionE" (use "option.exhaust", or
  3148 the "cases" method);
  3149 
  3150 * HOL/GroupTheory: group theory examples including Sylow's theorem (by
  3151 Florian Kammller);
  3152 
  3153 * HOL/IMP: updated and converted to new-style theory format; several
  3154 parts turned into readable document, with proper Isar proof texts and
  3155 some explanations (by Gerwin Klein);
  3156 
  3157 * HOL-Real: added Complex_Numbers (by Gertrud Bauer);
  3158 
  3159 * HOL-Hyperreal is now a logic image;
  3160 
  3161 
  3162 *** HOLCF ***
  3163 
  3164 * Isar: consts/constdefs supports mixfix syntax for continuous
  3165 operations;
  3166 
  3167 * Isar: domain package adapted to new-style theory format, e.g. see
  3168 HOLCF/ex/Dnat.thy;
  3169 
  3170 * theory Lift: proper use of rep_datatype lift instead of ML hacks --
  3171 potential INCOMPATIBILITY; now use plain induct_tac instead of former
  3172 lift.induct_tac, always use UU instead of Undef;
  3173 
  3174 * HOLCF/IMP: updated and converted to new-style theory;
  3175 
  3176 
  3177 *** ZF ***
  3178 
  3179 * Isar: proper integration of logic-specific tools and packages,
  3180 including theory commands '(co)inductive', '(co)datatype',
  3181 'rep_datatype', 'inductive_cases', as well as methods 'ind_cases',
  3182 'induct_tac', 'case_tac', and 'typecheck' (with attribute 'TC');
  3183 
  3184 * theory Main no longer includes AC; for the Axiom of Choice, base
  3185 your theory on Main_ZFC;
  3186 
  3187 * the integer library now covers quotients and remainders, with many
  3188 laws relating division to addition, multiplication, etc.;
  3189 
  3190 * ZF/UNITY: Chandy and Misra's UNITY is now available in ZF, giving a
  3191 typeless version of the formalism;
  3192 
  3193 * ZF/AC, Coind, IMP, Resid: updated and converted to new-style theory
  3194 format;
  3195 
  3196 * ZF/Induct: new directory for examples of inductive definitions,
  3197 including theory Multiset for multiset orderings; converted to
  3198 new-style theory format;
  3199 
  3200 * ZF: many new theorems about lists, ordinals, etc.;
  3201 
  3202 
  3203 *** General ***
  3204 
  3205 * Pure/kernel: meta-level proof terms (by Stefan Berghofer); reference
  3206 variable proof controls level of detail: 0 = no proofs (only oracle
  3207 dependencies), 1 = lemma dependencies, 2 = compact proof terms; see
  3208 also ref manual for further ML interfaces;
  3209 
  3210 * Pure/axclass: removed obsolete ML interface
  3211 goal_subclass/goal_arity;
  3212 
  3213 * Pure/syntax: new token syntax "num" for plain numerals (without "#"
  3214 of "xnum"); potential INCOMPATIBILITY, since -0, -1 etc. are now
  3215 separate tokens, so expressions involving minus need to be spaced
  3216 properly;
  3217 
  3218 * Pure/syntax: support non-oriented infixes, using keyword "infix"
  3219 rather than "infixl" or "infixr";
  3220 
  3221 * Pure/syntax: concrete syntax for dummy type variables admits genuine
  3222 sort constraint specifications in type inference; e.g. "x::_::foo"
  3223 ensures that the type of "x" is of sort "foo" (but not necessarily a
  3224 type variable);
  3225 
  3226 * Pure/syntax: print modes "type_brackets" and "no_type_brackets"
  3227 control output of nested => (types); the default behavior is
  3228 "type_brackets";
  3229 
  3230 * Pure/syntax: builtin parse translation for "_constify" turns valued
  3231 tokens into AST constants;
  3232 
  3233 * Pure/syntax: prefer later declarations of translations and print
  3234 translation functions; potential INCOMPATIBILITY: need to reverse
  3235 multiple declarations for same syntax element constant;
  3236 
  3237 * Pure/show_hyps reset by default (in accordance to existing Isar
  3238 practice);
  3239 
  3240 * Provers/classical: renamed addaltern to addafter, addSaltern to
  3241 addSafter;
  3242 
  3243 * Provers/clasimp: ``iff'' declarations now handle conditional rules
  3244 as well;
  3245 
  3246 * system: tested support for MacOS X; should be able to get Isabelle +
  3247 Proof General to work in a plain Terminal after installing Poly/ML
  3248 (e.g. from the Isabelle distribution area) and GNU bash alone
  3249 (e.g. from http://www.apple.com); full X11, XEmacs and X-Symbol
  3250 support requires further installations, e.g. from
  3251 http://fink.sourceforge.net/);
  3252 
  3253 * system: support Poly/ML 4.1.1 (able to manage larger heaps);
  3254 
  3255 * system: reduced base memory usage by Poly/ML (approx. 20 MB instead
  3256 of 40 MB), cf. ML_OPTIONS;
  3257 
  3258 * system: Proof General keywords specification is now part of the
  3259 Isabelle distribution (see etc/isar-keywords.el);
  3260 
  3261 * system: support for persistent Proof General sessions (refrain from
  3262 outdating all loaded theories on startup); user may create writable
  3263 logic images like this: ``isabelle -q HOL Test'';
  3264 
  3265 * system: smart selection of Isabelle process versus Isabelle
  3266 interface, accommodates case-insensitive file systems (e.g. HFS+); may
  3267 run both "isabelle" and "Isabelle" even if file names are badly
  3268 damaged (executable inspects the case of the first letter of its own
  3269 name); added separate "isabelle-process" and "isabelle-interface";
  3270 
  3271 * system: refrain from any attempt at filtering input streams; no
  3272 longer support ``8bit'' encoding of old isabelle font, instead proper
  3273 iso-latin characters may now be used; the related isatools
  3274 "symbolinput" and "nonascii" have disappeared as well;
  3275 
  3276 * system: removed old "xterm" interface (the print modes "xterm" and
  3277 "xterm_color" are still available for direct use in a suitable
  3278 terminal);
  3279 
  3280 
  3281 
  3282 New in Isabelle99-2 (February 2001)
  3283 -----------------------------------
  3284 
  3285 *** Overview of INCOMPATIBILITIES ***
  3286 
  3287 * HOL: please note that theories in the Library and elsewhere often use the
  3288 new-style (Isar) format; to refer to their theorems in an ML script you must
  3289 bind them to ML identifers by e.g.      val thm_name = thm "thm_name";
  3290 
  3291 * HOL: inductive package no longer splits induction rule aggressively,
  3292 but only as far as specified by the introductions given; the old
  3293 format may be recovered via ML function complete_split_rule or attribute
  3294 'split_rule (complete)';
  3295 
  3296 * HOL: induct renamed to lfp_induct, lfp_Tarski to lfp_unfold,
  3297 gfp_Tarski to gfp_unfold;
  3298 
  3299 * HOL: contrapos, contrapos2 renamed to contrapos_nn, contrapos_pp;
  3300 
  3301 * HOL: infix "dvd" now has priority 50 rather than 70 (because it is a
  3302 relation); infix "^^" has been renamed "``"; infix "``" has been
  3303 renamed "`"; "univalent" has been renamed "single_valued";
  3304 
  3305 * HOL/Real: "rinv" and "hrinv" replaced by overloaded "inverse"
  3306 operation;
  3307 
  3308 * HOLCF: infix "`" has been renamed "$"; the symbol syntax is \<cdot>;
  3309 
  3310 * Isar: 'obtain' no longer declares "that" fact as simp/intro;
  3311 
  3312 * Isar/HOL: method 'induct' now handles non-atomic goals; as a
  3313 consequence, it is no longer monotonic wrt. the local goal context
  3314 (which is now passed through the inductive cases);
  3315 
  3316 * Document preparation: renamed standard symbols \<ll> to \<lless> and
  3317 \<gg> to \<ggreater>;
  3318 
  3319 
  3320 *** Document preparation ***
  3321 
  3322 * \isabellestyle{NAME} selects version of Isabelle output (currently
  3323 available: are "it" for near math-mode best-style output, "sl" for
  3324 slanted text style, and "tt" for plain type-writer; if no
  3325 \isabellestyle command is given, output is according to slanted
  3326 type-writer);
  3327 
  3328 * support sub/super scripts (for single symbols only), input syntax is
  3329 like this: "A\<^sup>*" or "A\<^sup>\<star>";
  3330 
  3331 * some more standard symbols; see Appendix A of the system manual for
  3332 the complete list of symbols defined in isabellesym.sty;
  3333 
  3334 * improved isabelle style files; more abstract symbol implementation
  3335 (should now use \isamath{...} and \isatext{...} in custom symbol
  3336 definitions);
  3337 
  3338 * antiquotation @{goals} and @{subgoals} for output of *dynamic* goals
  3339 state; Note that presentation of goal states does not conform to
  3340 actual human-readable proof documents.  Please do not include goal
  3341 states into document output unless you really know what you are doing!
  3342 
  3343 * proper indentation of antiquoted output with proportional LaTeX
  3344 fonts;
  3345 
  3346 * no_document ML operator temporarily disables LaTeX document
  3347 generation;
  3348 
  3349 * isatool unsymbolize tunes sources for plain ASCII communication;
  3350 
  3351 
  3352 *** Isar ***
  3353 
  3354 * Pure: Isar now suffers initial goal statements to contain unbound
  3355 schematic variables (this does not conform to actual readable proof
  3356 documents, due to unpredictable outcome and non-compositional proof
  3357 checking); users who know what they are doing may use schematic goals
  3358 for Prolog-style synthesis of proven results;
  3359 
  3360 * Pure: assumption method (an implicit finishing) now handles actual
  3361 rules as well;
  3362 
  3363 * Pure: improved 'obtain' --- moved to Pure, insert "that" into
  3364 initial goal, declare "that" only as Pure intro (only for single
  3365 steps); the "that" rule assumption may now be involved in implicit
  3366 finishing, thus ".." becomes a feasible for trivial obtains;
  3367 
  3368 * Pure: default proof step now includes 'intro_classes'; thus trivial
  3369 instance proofs may be performed by "..";
  3370 
  3371 * Pure: ?thesis / ?this / "..." now work for pure meta-level
  3372 statements as well;
  3373 
  3374 * Pure: more robust selection of calculational rules;
  3375 
  3376 * Pure: the builtin notion of 'finished' goal now includes the ==-refl
  3377 rule (as well as the assumption rule);
  3378 
  3379 * Pure: 'thm_deps' command visualizes dependencies of theorems and
  3380 lemmas, using the graph browser tool;
  3381 
  3382 * Pure: predict failure of "show" in interactive mode;
  3383 
  3384 * Pure: 'thms_containing' now takes actual terms as arguments;
  3385 
  3386 * HOL: improved method 'induct' --- now handles non-atomic goals
  3387 (potential INCOMPATIBILITY); tuned error handling;
  3388 
  3389 * HOL: cases and induct rules now provide explicit hints about the
  3390 number of facts to be consumed (0 for "type" and 1 for "set" rules);
  3391 any remaining facts are inserted into the goal verbatim;
  3392 
  3393 * HOL: local contexts (aka cases) may now contain term bindings as
  3394 well; the 'cases' and 'induct' methods new provide a ?case binding for
  3395 the result to be shown in each case;
  3396 
  3397 * HOL: added 'recdef_tc' command;
  3398 
  3399 * isatool convert assists in eliminating legacy ML scripts;
  3400 
  3401 
  3402 *** HOL ***
  3403 
  3404 * HOL/Library: a collection of generic theories to be used together
  3405 with main HOL; the theory loader path already includes this directory
  3406 by default; the following existing theories have been moved here:
  3407 HOL/Induct/Multiset, HOL/Induct/Acc (as Accessible_Part), HOL/While
  3408 (as While_Combinator), HOL/Lex/Prefix (as List_Prefix);
  3409 
  3410 * HOL/Unix: "Some aspects of Unix file-system security", a typical
  3411 modelling and verification task performed in Isabelle/HOL +
  3412 Isabelle/Isar + Isabelle document preparation (by Markus Wenzel).
  3413 
  3414 * HOL/Algebra: special summation operator SUM no longer exists, it has
  3415 been replaced by setsum; infix 'assoc' now has priority 50 (like
  3416 'dvd'); axiom 'one_not_zero' has been moved from axclass 'ring' to
  3417 'domain', this makes the theory consistent with mathematical
  3418 literature;
  3419 
  3420 * HOL basics: added overloaded operations "inverse" and "divide"
  3421 (infix "/"), syntax for generic "abs" operation, generic summation
  3422 operator \<Sum>;
  3423 
  3424 * HOL/typedef: simplified package, provide more useful rules (see also
  3425 HOL/subset.thy);
  3426 
  3427 * HOL/datatype: induction rule for arbitrarily branching datatypes is
  3428 now expressed as a proper nested rule (old-style tactic scripts may
  3429 require atomize_strip_tac to cope with non-atomic premises);
  3430 
  3431 * HOL: renamed theory "Prod" to "Product_Type", renamed "split" rule
  3432 to "split_conv" (old name still available for compatibility);
  3433 
  3434 * HOL: improved concrete syntax for strings (e.g. allows translation
  3435 rules with string literals);
  3436 
  3437 * HOL-Real-Hyperreal: this extends HOL-Real with the hyperreals
  3438  and Fleuriot's mechanization of analysis, including the transcendental
  3439  functions for the reals;
  3440 
  3441 * HOL/Real, HOL/Hyperreal: improved arithmetic simplification;
  3442 
  3443 
  3444 *** CTT ***
  3445 
  3446 * CTT: x-symbol support for Pi, Sigma, -->, : (membership); note that
  3447 "lam" is displayed as TWO lambda-symbols
  3448 
  3449 * CTT: theory Main now available, containing everything (that is, Bool
  3450 and Arith);
  3451 
  3452 
  3453 *** General ***
  3454 
  3455 * Pure: the Simplifier has been implemented properly as a derived rule
  3456 outside of the actual kernel (at last!); the overall performance
  3457 penalty in practical applications is about 50%, while reliability of
  3458 the Isabelle inference kernel has been greatly improved;
  3459 
  3460 * print modes "brackets" and "no_brackets" control output of nested =>
  3461 (types) and ==> (props); the default behaviour is "brackets";
  3462 
  3463 * Provers: fast_tac (and friends) now handle actual object-logic rules
  3464 as assumptions as well;
  3465 
  3466 * system: support Poly/ML 4.0;
  3467 
  3468 * system: isatool install handles KDE version 1 or 2;
  3469 
  3470 
  3471 
  3472 New in Isabelle99-1 (October 2000)
  3473 ----------------------------------
  3474 
  3475 *** Overview of INCOMPATIBILITIES ***
  3476 
  3477 * HOL: simplification of natural numbers is much changed; to partly
  3478 recover the old behaviour (e.g. to prevent n+n rewriting to #2*n)
  3479 issue the following ML commands:
  3480 
  3481   Delsimprocs Nat_Numeral_Simprocs.cancel_numerals;
  3482   Delsimprocs [Nat_Numeral_Simprocs.combine_numerals];
  3483 
  3484 * HOL: simplification no longer dives into case-expressions; this is
  3485 controlled by "t.weak_case_cong" for each datatype t;
  3486 
  3487 * HOL: nat_less_induct renamed to less_induct;
  3488 
  3489 * HOL: systematic renaming of the SOME (Eps) rules, may use isatool
  3490 fixsome to patch .thy and .ML sources automatically;
  3491 
  3492   select_equality  -> some_equality
  3493   select_eq_Ex     -> some_eq_ex
  3494   selectI2EX       -> someI2_ex
  3495   selectI2         -> someI2
  3496   selectI          -> someI
  3497   select1_equality -> some1_equality
  3498   Eps_sym_eq       -> some_sym_eq_trivial
  3499   Eps_eq           -> some_eq_trivial
  3500 
  3501 * HOL: exhaust_tac on datatypes superceded by new generic case_tac;
  3502 
  3503 * HOL: removed obsolete theorem binding expand_if (refer to split_if
  3504 instead);
  3505 
  3506 * HOL: the recursion equations generated by 'recdef' are now called
  3507 f.simps instead of f.rules;
  3508 
  3509 * HOL: qed_spec_mp now also handles bounded ALL as well;
  3510 
  3511 * HOL: 0 is now overloaded, so the type constraint ":: nat" may
  3512 sometimes be needed;
  3513 
  3514 * HOL: the constant for "f``x" is now "image" rather than "op ``";
  3515 
  3516 * HOL: the constant for "f-``x" is now "vimage" rather than "op -``";
  3517 
  3518 * HOL: the disjoint sum is now "<+>" instead of "Plus"; the cartesian
  3519 product is now "<*>" instead of "Times"; the lexicographic product is
  3520 now "<*lex*>" instead of "**";
  3521 
  3522 * HOL: theory Sexp is now in HOL/Induct examples (it used to be part
  3523 of main HOL, but was unused); better use HOL's datatype package;
  3524 
  3525 * HOL: removed "symbols" syntax for constant "override" of theory Map;
  3526 the old syntax may be recovered as follows:
  3527 
  3528   syntax (symbols)
  3529     override  :: "('a ~=> 'b) => ('a ~=> 'b) => ('a ~=> 'b)"
  3530       (infixl "\\<oplus>" 100)
  3531 
  3532 * HOL/Real: "rabs" replaced by overloaded "abs" function;
  3533 
  3534 * HOL/ML: even fewer consts are declared as global (see theories Ord,
  3535 Lfp, Gfp, WF); this only affects ML packages that refer to const names
  3536 internally;
  3537 
  3538 * HOL and ZF: syntax for quotienting wrt an equivalence relation
  3539 changed from A/r to A//r;
  3540 
  3541 * ZF: new treatment of arithmetic (nat & int) may break some old
  3542 proofs;
  3543 
  3544 * Isar: renamed some attributes (RS -> THEN, simplify -> simplified,
  3545 rulify -> rule_format, elimify -> elim_format, ...);
  3546 
  3547 * Isar/Provers: intro/elim/dest attributes changed; renamed
  3548 intro/intro!/intro!! flags to intro!/intro/intro? (in most cases, one
  3549 should have to change intro!! to intro? only); replaced "delrule" by
  3550 "rule del";
  3551 
  3552 * Isar/HOL: renamed "intrs" to "intros" in inductive definitions;
  3553 
  3554 * Provers: strengthened force_tac by using new first_best_tac;
  3555 
  3556 * LaTeX document preparation: several changes of isabelle.sty (see
  3557 lib/texinputs);
  3558 
  3559 
  3560 *** Document preparation ***
  3561 
  3562 * formal comments (text blocks etc.) in new-style theories may now
  3563 contain antiquotations of thm/prop/term/typ/text to be presented
  3564 according to latex print mode; concrete syntax is like this:
  3565 @{term[show_types] "f(x) = a + x"};
  3566 
  3567 * isatool mkdir provides easy setup of Isabelle session directories,
  3568 including proper document sources;
  3569 
  3570 * generated LaTeX sources are now deleted after successful run
  3571 (isatool document -c); may retain a copy somewhere else via -D option
  3572 of isatool usedir;
  3573 
  3574 * isatool usedir -D now lets isatool latex -o sty update the Isabelle
  3575 style files, achieving self-contained LaTeX sources and simplifying
  3576 LaTeX debugging;
  3577 
  3578 * old-style theories now produce (crude) LaTeX output as well;
  3579 
  3580 * browser info session directories are now self-contained (may be put
  3581 on WWW server seperately); improved graphs of nested sessions; removed
  3582 graph for 'all sessions';
  3583 
  3584 * several improvements in isabelle style files; \isabellestyle{it}
  3585 produces fake math mode output; \isamarkupheader is now \section by
  3586 default; see lib/texinputs/isabelle.sty etc.;
  3587 
  3588 
  3589 *** Isar ***
  3590 
  3591 * Isar/Pure: local results and corresponding term bindings are now
  3592 subject to Hindley-Milner polymorphism (similar to ML); this
  3593 accommodates incremental type-inference very nicely;
  3594 
  3595 * Isar/Pure: new derived language element 'obtain' supports
  3596 generalized existence reasoning;
  3597 
  3598 * Isar/Pure: new calculational elements 'moreover' and 'ultimately'
  3599 support accumulation of results, without applying any rules yet;
  3600 useful to collect intermediate results without explicit name
  3601 references, and for use with transitivity rules with more than 2
  3602 premises;
  3603 
  3604 * Isar/Pure: scalable support for case-analysis type proofs: new
  3605 'case' language element refers to local contexts symbolically, as
  3606 produced by certain proof methods; internally, case names are attached
  3607 to theorems as "tags";
  3608 
  3609 * Isar/Pure: theory command 'hide' removes declarations from
  3610 class/type/const name spaces;
  3611 
  3612 * Isar/Pure: theory command 'defs' supports option "(overloaded)" to
  3613 indicate potential overloading;
  3614 
  3615 * Isar/Pure: changed syntax of local blocks from {{ }} to { };
  3616 
  3617 * Isar/Pure: syntax of sorts made 'inner', i.e. have to write
  3618 "{a,b,c}" instead of {a,b,c};
  3619 
  3620 * Isar/Pure now provides its own version of intro/elim/dest
  3621 attributes; useful for building new logics, but beware of confusion
  3622 with the version in Provers/classical;
  3623 
  3624 * Isar/Pure: the local context of (non-atomic) goals is provided via
  3625 case name 'antecedent';
  3626 
  3627 * Isar/Pure: removed obsolete 'transfer' attribute (transfer of thms
  3628 to the current context is now done automatically);
  3629 
  3630 * Isar/Pure: theory command 'method_setup' provides a simple interface
  3631 for definining proof methods in ML;
  3632 
  3633 * Isar/Provers: intro/elim/dest attributes changed; renamed
  3634 intro/intro!/intro!! flags to intro!/intro/intro? (INCOMPATIBILITY, in
  3635 most cases, one should have to change intro!! to intro? only);
  3636 replaced "delrule" by "rule del";
  3637 
  3638 * Isar/Provers: new 'hypsubst' method, plain 'subst' method and
  3639 'symmetric' attribute (the latter supercedes [RS sym]);
  3640 
  3641 * Isar/Provers: splitter support (via 'split' attribute and 'simp'
  3642 method modifier); 'simp' method: 'only:' modifier removes loopers as
  3643 well (including splits);
  3644 
  3645 * Isar/Provers: Simplifier and Classical methods now support all kind
  3646 of modifiers used in the past, including 'cong', 'iff', etc.
  3647 
  3648 * Isar/Provers: added 'fastsimp' and 'clarsimp' methods (combination
  3649 of Simplifier and Classical reasoner);
  3650 
  3651 * Isar/HOL: new proof method 'cases' and improved version of 'induct'
  3652 now support named cases; major packages (inductive, datatype, primrec,
  3653 recdef) support case names and properly name parameters;
  3654 
  3655 * Isar/HOL: new transitivity rules for substitution in inequalities --
  3656 monotonicity conditions are extracted to be proven at end of
  3657 calculations;
  3658 
  3659 * Isar/HOL: removed 'case_split' thm binding, should use 'cases' proof
  3660 method anyway;
  3661 
  3662 * Isar/HOL: removed old expand_if = split_if; theorems if_splits =
  3663 split_if split_if_asm; datatype package provides theorems foo.splits =
  3664 foo.split foo.split_asm for each datatype;
  3665 
  3666 * Isar/HOL: tuned inductive package, rename "intrs" to "intros"
  3667 (potential INCOMPATIBILITY), emulation of mk_cases feature for proof
  3668 scripts: new 'inductive_cases' command and 'ind_cases' method; (Note:
  3669 use "(cases (simplified))" method in proper proof texts);
  3670 
  3671 * Isar/HOL: added global 'arith_split' attribute for 'arith' method;
  3672 
  3673 * Isar: names of theorems etc. may be natural numbers as well;
  3674 
  3675 * Isar: 'pr' command: optional arguments for goals_limit and
  3676 ProofContext.prems_limit; no longer prints theory contexts, but only
  3677 proof states;
  3678 
  3679 * Isar: diagnostic commands 'pr', 'thm', 'prop', 'term', 'typ' admit
  3680 additional print modes to be specified; e.g. "pr(latex)" will print
  3681 proof state according to the Isabelle LaTeX style;
  3682 
  3683 * Isar: improved support for emulating tactic scripts, including proof
  3684 methods 'rule_tac' etc., 'cut_tac', 'thin_tac', 'subgoal_tac',
  3685 'rename_tac', 'rotate_tac', 'tactic', and 'case_tac' / 'induct_tac'
  3686 (for HOL datatypes);
  3687 
  3688 * Isar: simplified (more robust) goal selection of proof methods: 1st
  3689 goal, all goals, or explicit goal specifier (tactic emulation); thus
  3690 'proof method scripts' have to be in depth-first order;
  3691 
  3692 * Isar: tuned 'let' syntax: replaced 'as' keyword by 'and';
  3693 
  3694 * Isar: removed 'help' command, which hasn't been too helpful anyway;
  3695 should instead use individual commands for printing items
  3696 (print_commands, print_methods etc.);
  3697 
  3698 * Isar: added 'nothing' --- the empty list of theorems;
  3699 
  3700 
  3701 *** HOL ***
  3702 
  3703 * HOL/MicroJava: formalization of a fragment of Java, together with a
  3704 corresponding virtual machine and a specification of its bytecode
  3705 verifier and a lightweight bytecode verifier, including proofs of
  3706 type-safety; by Gerwin Klein, Tobias Nipkow, David von Oheimb, and
  3707 Cornelia Pusch (see also the homepage of project Bali at
  3708 http://isabelle.in.tum.de/Bali/);
  3709 
  3710 * HOL/Algebra: new theory of rings and univariate polynomials, by
  3711 Clemens Ballarin;
  3712 
  3713 * HOL/NumberTheory: fundamental Theorem of Arithmetic, Chinese
  3714 Remainder Theorem, Fermat/Euler Theorem, Wilson's Theorem, by Thomas M
  3715 Rasmussen;
  3716 
  3717 * HOL/Lattice: fundamental concepts of lattice theory and order
  3718 structures, including duals, properties of bounds versus algebraic
  3719 laws, lattice operations versus set-theoretic ones, the Knaster-Tarski
  3720 Theorem for complete lattices etc.; may also serve as a demonstration
  3721 for abstract algebraic reasoning using axiomatic type classes, and
  3722 mathematics-style proof in Isabelle/Isar; by Markus Wenzel;
  3723 
  3724 * HOL/Prolog: a (bare-bones) implementation of Lambda-Prolog, by David
  3725 von Oheimb;
  3726 
  3727 * HOL/IMPP: extension of IMP with local variables and mutually
  3728 recursive procedures, by David von Oheimb;
  3729 
  3730 * HOL/Lambda: converted into new-style theory and document;
  3731 
  3732 * HOL/ex/Multiquote: example of multiple nested quotations and
  3733 anti-quotations -- basically a generalized version of de-Bruijn
  3734 representation; very useful in avoiding lifting of operations;
  3735 
  3736 * HOL/record: added general record equality rule to simpset; fixed
  3737 select-update simplification procedure to handle extended records as
  3738 well; admit "r" as field name;
  3739 
  3740 * HOL: 0 is now overloaded over the new sort "zero", allowing its use with
  3741 other numeric types and also as the identity of groups, rings, etc.;
  3742 
  3743 * HOL: new axclass plus_ac0 for addition with the AC-laws and 0 as identity.
  3744 Types nat and int belong to this axclass;
  3745 
  3746 * HOL: greatly improved simplification involving numerals of type nat, int, real:
  3747    (i + #8 + j) = Suc k simplifies to  #7 + (i + j) = k
  3748    i*j + k + j*#3*i     simplifies to  #4*(i*j) + k
  3749   two terms #m*u and #n*u are replaced by #(m+n)*u
  3750     (where #m, #n and u can implicitly be 1; this is simproc combine_numerals)
  3751   and the term/formula #m*u+x ~~ #n*u+y simplifies simplifies to #(m-n)+x ~~ y
  3752     or x ~~ #(n-m)+y, where ~~ is one of = < <= or - (simproc cancel_numerals);
  3753 
  3754 * HOL: meson_tac is available (previously in ex/meson.ML); it is a
  3755 powerful prover for predicate logic but knows nothing of clasets; see
  3756 ex/mesontest.ML and ex/mesontest2.ML for example applications;
  3757 
  3758 * HOL: new version of "case_tac" subsumes both boolean case split and
  3759 "exhaust_tac" on datatypes; INCOMPATIBILITY: exhaust_tac no longer
  3760 exists, may define val exhaust_tac = case_tac for ad-hoc portability;
  3761 
  3762 * HOL: simplification no longer dives into case-expressions: only the
  3763 selector expression is simplified, but not the remaining arms; to
  3764 enable full simplification of case-expressions for datatype t, you may
  3765 remove t.weak_case_cong from the simpset, either globally (Delcongs
  3766 [thm"t.weak_case_cong"];) or locally (delcongs [...]).
  3767 
  3768 * HOL/recdef: the recursion equations generated by 'recdef' for
  3769 function 'f' are now called f.simps instead of f.rules; if all
  3770 termination conditions are proved automatically, these simplification
  3771 rules are added to the simpset, as in primrec; rules may be named
  3772 individually as well, resulting in a separate list of theorems for
  3773 each equation;
  3774 
  3775 * HOL/While is a new theory that provides a while-combinator. It
  3776 permits the definition of tail-recursive functions without the
  3777 provision of a termination measure. The latter is necessary once the
  3778 invariant proof rule for while is applied.
  3779 
  3780 * HOL: new (overloaded) notation for the set of elements below/above
  3781 some element: {..u}, {..u(}, {l..}, {)l..}. See theory SetInterval.
  3782 
  3783 * HOL: theorems impI, allI, ballI bound as "strip";
  3784 
  3785 * HOL: new tactic induct_thm_tac: thm -> string -> int -> tactic
  3786 induct_tac th "x1 ... xn" expects th to have a conclusion of the form
  3787 P v1 ... vn and abbreviates res_inst_tac [("v1","x1"),...,("vn","xn")] th;
  3788 
  3789 * HOL/Real: "rabs" replaced by overloaded "abs" function;
  3790 
  3791 * HOL: theory Sexp now in HOL/Induct examples (it used to be part of
  3792 main HOL, but was unused);
  3793 
  3794 * HOL: fewer consts declared as global (e.g. have to refer to
  3795 "Lfp.lfp" instead of "lfp" internally; affects ML packages only);
  3796 
  3797 * HOL: tuned AST representation of nested pairs, avoiding bogus output
  3798 in case of overlap with user translations (e.g. judgements over
  3799 tuples); (note that the underlying logical represenation is still
  3800 bogus);
  3801 
  3802 
  3803 *** ZF ***
  3804 
  3805 * ZF: simplification automatically cancels common terms in arithmetic
  3806 expressions over nat and int;
  3807 
  3808 * ZF: new treatment of nat to minimize type-checking: all operators
  3809 coerce their operands to a natural number using the function natify,
  3810 making the algebraic laws unconditional;
  3811 
  3812 * ZF: as above, for int: operators coerce their operands to an integer
  3813 using the function intify;
  3814 
  3815 * ZF: the integer library now contains many of the usual laws for the
  3816 orderings, including $<=, and monotonicity laws for $+ and $*;
  3817 
  3818 * ZF: new example ZF/ex/NatSum to demonstrate integer arithmetic
  3819 simplification;
  3820 
  3821 * FOL and ZF: AddIffs now available, giving theorems of the form P<->Q
  3822 to the simplifier and classical reasoner simultaneously;
  3823 
  3824 
  3825 *** General ***
  3826 
  3827 * Provers: blast_tac now handles actual object-logic rules as
  3828 assumptions; note that auto_tac uses blast_tac internally as well;
  3829 
  3830 * Provers: new functions rulify/rulify_no_asm: thm -> thm for turning
  3831 outer -->/All/Ball into ==>/!!; qed_spec_mp now uses rulify_no_asm;
  3832 
  3833 * Provers: delrules now handles destruct rules as well (no longer need
  3834 explicit make_elim);
  3835 
  3836 * Provers: Blast_tac now warns of and ignores "weak elimination rules" e.g.
  3837   [| inj ?f;          ?f ?x = ?f ?y; ?x = ?y ==> ?W |] ==> ?W
  3838 use instead the strong form,
  3839   [| inj ?f; ~ ?W ==> ?f ?x = ?f ?y; ?x = ?y ==> ?W |] ==> ?W
  3840 in HOL, FOL and ZF the function cla_make_elim will create such rules
  3841 from destruct-rules;
  3842 
  3843 * Provers: Simplifier.easy_setup provides a fast path to basic
  3844 Simplifier setup for new object-logics;
  3845 
  3846 * Pure: AST translation rules no longer require constant head on LHS;
  3847 
  3848 * Pure: improved name spaces: ambiguous output is qualified; support
  3849 for hiding of names;
  3850 
  3851 * system: smart setup of canonical ML_HOME, ISABELLE_INTERFACE, and
  3852 XSYMBOL_HOME; no longer need to do manual configuration in most
  3853 situations;
  3854 
  3855 * system: compression of ML heaps images may now be controlled via -c
  3856 option of isabelle and isatool usedir (currently only observed by
  3857 Poly/ML);
  3858 
  3859 * system: isatool installfonts may handle X-Symbol fonts as well (very
  3860 useful for remote X11);
  3861 
  3862 * system: provide TAGS file for Isabelle sources;
  3863 
  3864 * ML: infix 'OF' is a version of 'MRS' with more appropriate argument
  3865 order;
  3866 
  3867 * ML: renamed flags Syntax.trace_norm_ast to Syntax.trace_ast; global
  3868 timing flag supersedes proof_timing and Toplevel.trace;
  3869 
  3870 * ML: new combinators |>> and |>>> for incremental transformations
  3871 with secondary results (e.g. certain theory extensions):
  3872 
  3873 * ML: PureThy.add_defs gets additional argument to indicate potential
  3874 overloading (usually false);
  3875 
  3876 * ML: PureThy.add_thms/add_axioms/add_defs now return theorems as
  3877 results;
  3878 
  3879 
  3880 
  3881 New in Isabelle99 (October 1999)
  3882 --------------------------------
  3883 
  3884 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  3885 
  3886 * HOL: The THEN and ELSE parts of conditional expressions (if P then x else y)
  3887 are no longer simplified.  (This allows the simplifier to unfold recursive
  3888 functional programs.)  To restore the old behaviour, declare
  3889 
  3890     Delcongs [if_weak_cong];
  3891 
  3892 * HOL: Removed the obsolete syntax "Compl A"; use -A for set
  3893 complement;
  3894 
  3895 * HOL: the predicate "inj" is now defined by translation to "inj_on";
  3896 
  3897 * HOL/datatype: mutual_induct_tac no longer exists --
  3898   use induct_tac "x_1 ... x_n" instead of mutual_induct_tac ["x_1", ..., "x_n"]
  3899 
  3900 * HOL/typedef: fixed type inference for representing set; type
  3901 arguments now have to occur explicitly on the rhs as type constraints;
  3902 
  3903 * ZF: The con_defs part of an inductive definition may no longer refer
  3904 to constants declared in the same theory;
  3905 
  3906 * HOL, ZF: the function mk_cases, generated by the inductive
  3907 definition package, has lost an argument.  To simplify its result, it
  3908 uses the default simpset instead of a supplied list of theorems.
  3909 
  3910 * HOL/List: the constructors of type list are now Nil and Cons;
  3911 
  3912 * Simplifier: the type of the infix ML functions
  3913         setSSolver addSSolver setSolver addSolver
  3914 is now  simpset * solver -> simpset  where `solver' is a new abstract type
  3915 for packaging solvers. A solver is created via
  3916         mk_solver: string -> (thm list -> int -> tactic) -> solver
  3917 where the string argument is only a comment.
  3918 
  3919 
  3920 *** Proof tools ***
  3921 
  3922 * Provers/Arith/fast_lin_arith.ML contains a functor for creating a
  3923 decision procedure for linear arithmetic. Currently it is used for
  3924 types `nat', `int', and `real' in HOL (see below); it can, should and
  3925 will be instantiated for other types and logics as well.
  3926 
  3927 * The simplifier now accepts rewrite rules with flexible heads, eg
  3928      hom ?f ==> ?f(?x+?y) = ?f ?x + ?f ?y
  3929   They are applied like any rule with a non-pattern lhs, i.e. by first-order
  3930   matching.
  3931 
  3932 
  3933 *** General ***
  3934 
  3935 * New Isabelle/Isar subsystem provides an alternative to traditional
  3936 tactical theorem proving; together with the ProofGeneral/isar user
  3937 interface it offers an interactive environment for developing human
  3938 readable proof documents (Isar == Intelligible semi-automated
  3939 reasoning); for further information see isatool doc isar-ref,
  3940 src/HOL/Isar_examples and http://isabelle.in.tum.de/Isar/
  3941 
  3942 * improved and simplified presentation of theories: better HTML markup
  3943 (including colors), graph views in several sizes; isatool usedir now
  3944 provides a proper interface for user theories (via -P option); actual
  3945 document preparation based on (PDF)LaTeX is available as well (for
  3946 new-style theories only); see isatool doc system for more information;
  3947 
  3948 * native support for Proof General, both for classic Isabelle and
  3949 Isabelle/Isar;
  3950 
  3951 * ML function thm_deps visualizes dependencies of theorems and lemmas,
  3952 using the graph browser tool;
  3953 
  3954 * Isabelle manuals now also available as PDF;
  3955 
  3956 * theory loader rewritten from scratch (may not be fully
  3957 bug-compatible); old loadpath variable has been replaced by show_path,
  3958 add_path, del_path, reset_path functions; new operations such as
  3959 update_thy, touch_thy, remove_thy, use/update_thy_only (see also
  3960 isatool doc ref);
  3961 
  3962 * improved isatool install: option -k creates KDE application icon,
  3963 option -p DIR installs standalone binaries;
  3964 
  3965 * added ML_PLATFORM setting (useful for cross-platform installations);
  3966 more robust handling of platform specific ML images for SML/NJ;
  3967 
  3968 * the settings environment is now statically scoped, i.e. it is never
  3969 created again in sub-processes invoked from isabelle, isatool, or
  3970 Isabelle;
  3971 
  3972 * path element specification '~~' refers to '$ISABELLE_HOME';
  3973 
  3974 * in locales, the "assumes" and "defines" parts may be omitted if
  3975 empty;
  3976 
  3977 * new print_mode "xsymbols" for extended symbol support (e.g. genuine
  3978 long arrows);
  3979 
  3980 * new print_mode "HTML";
  3981 
  3982 * new flag show_tags controls display of tags of theorems (which are
  3983 basically just comments that may be attached by some tools);
  3984 
  3985 * Isamode 2.6 requires patch to accomodate change of Isabelle font
  3986 mode and goal output format:
  3987 
  3988 diff -r Isamode-2.6/elisp/isa-load.el Isamode/elisp/isa-load.el
  3989 244c244
  3990 <       (list (isa-getenv "ISABELLE") "-msymbols" logic-name)
  3991 ---
  3992 >       (list (isa-getenv "ISABELLE") "-misabelle_font" "-msymbols" logic-name)
  3993 diff -r Isabelle-2.6/elisp/isa-proofstate.el Isamode/elisp/isa-proofstate.el
  3994 181c181
  3995 < (defconst proofstate-proofstart-regexp "^Level [0-9]+$"
  3996 ---
  3997 > (defconst proofstate-proofstart-regexp "^Level [0-9]+"
  3998 
  3999 * function bind_thms stores lists of theorems (cf. bind_thm);
  4000 
  4001 * new shorthand tactics ftac, eatac, datac, fatac;
  4002 
  4003 * qed (and friends) now accept "" as result name; in that case the
  4004 theorem is not stored, but proper checks and presentation of the
  4005 result still apply;
  4006 
  4007 * theorem database now also indexes constants "Trueprop", "all",
  4008 "==>", "=="; thus thms_containing, findI etc. may retrieve more rules;
  4009 
  4010 
  4011 *** HOL ***
  4012 
  4013 ** HOL arithmetic **
  4014 
  4015 * There are now decision procedures for linear arithmetic over nat and
  4016 int:
  4017 
  4018 1. arith_tac copes with arbitrary formulae involving `=', `<', `<=',
  4019 `+', `-', `Suc', `min', `max' and numerical constants; other subterms
  4020 are treated as atomic; subformulae not involving type `nat' or `int'
  4021 are ignored; quantified subformulae are ignored unless they are
  4022 positive universal or negative existential. The tactic has to be
  4023 invoked by hand and can be a little bit slow. In particular, the
  4024 running time is exponential in the number of occurrences of `min' and
  4025 `max', and `-' on `nat'.
  4026 
  4027 2. fast_arith_tac is a cut-down version of arith_tac: it only takes
  4028 (negated) (in)equalities among the premises and the conclusion into
  4029 account (i.e. no compound formulae) and does not know about `min' and
  4030 `max', and `-' on `nat'. It is fast and is used automatically by the
  4031 simplifier.
  4032 
  4033 NB: At the moment, these decision procedures do not cope with mixed
  4034 nat/int formulae where the two parts interact, such as `m < n ==>
  4035 int(m) < int(n)'.
  4036 
  4037 * HOL/Numeral provides a generic theory of numerals (encoded
  4038 efficiently as bit strings); setup for types nat/int/real is in place;
  4039 INCOMPATIBILITY: since numeral syntax is now polymorphic, rather than
  4040 int, existing theories and proof scripts may require a few additional
  4041 type constraints;
  4042 
  4043 * integer division and remainder can now be performed on constant
  4044 arguments;
  4045 
  4046 * many properties of integer multiplication, division and remainder
  4047 are now available;
  4048 
  4049 * An interface to the Stanford Validity Checker (SVC) is available through the
  4050 tactic svc_tac.  Propositional tautologies and theorems of linear arithmetic
  4051 are proved automatically.  SVC must be installed separately, and its results
  4052 must be TAKEN ON TRUST (Isabelle does not check the proofs, but tags any
  4053 invocation of the underlying oracle).  For SVC see
  4054   http://verify.stanford.edu/SVC
  4055 
  4056 * IsaMakefile: the HOL-Real target now builds an actual image;
  4057 
  4058 
  4059 ** HOL misc **
  4060 
  4061 * HOL/Real/HahnBanach: the Hahn-Banach theorem for real vector spaces
  4062 (in Isabelle/Isar) -- by Gertrud Bauer;
  4063 
  4064 * HOL/BCV: generic model of bytecode verification, i.e. data-flow
  4065 analysis for assembly languages with subtypes;
  4066 
  4067 * HOL/TLA (Lamport's Temporal Logic of Actions): major reorganization
  4068 -- avoids syntactic ambiguities and treats state, transition, and
  4069 temporal levels more uniformly; introduces INCOMPATIBILITIES due to
  4070 changed syntax and (many) tactics;
  4071 
  4072 * HOL/inductive: Now also handles more general introduction rules such
  4073   as "ALL y. (y, x) : r --> y : acc r ==> x : acc r"; monotonicity
  4074   theorems are now maintained within the theory (maintained via the
  4075   "mono" attribute);
  4076 
  4077 * HOL/datatype: Now also handles arbitrarily branching datatypes
  4078   (using function types) such as
  4079 
  4080   datatype 'a tree = Atom 'a | Branch "nat => 'a tree"
  4081 
  4082 * HOL/record: record_simproc (part of the default simpset) takes care
  4083 of selectors applied to updated records; record_split_tac is no longer
  4084 part of the default claset; update_defs may now be removed from the
  4085 simpset in many cases; COMPATIBILITY: old behavior achieved by
  4086 
  4087   claset_ref () := claset() addSWrapper record_split_wrapper;
  4088   Delsimprocs [record_simproc]
  4089 
  4090 * HOL/typedef: fixed type inference for representing set; type
  4091 arguments now have to occur explicitly on the rhs as type constraints;
  4092 
  4093 * HOL/recdef (TFL): 'congs' syntax now expects comma separated list of theorem
  4094 names rather than an ML expression;
  4095 
  4096 * HOL/defer_recdef (TFL): like recdef but the well-founded relation can be
  4097 supplied later.  Program schemes can be defined, such as
  4098     "While B C s = (if B s then While B C (C s) else s)"
  4099 where the well-founded relation can be chosen after B and C have been given.
  4100 
  4101 * HOL/List: the constructors of type list are now Nil and Cons;
  4102 INCOMPATIBILITY: while [] and infix # syntax is still there, of
  4103 course, ML tools referring to List.list.op # etc. have to be adapted;
  4104 
  4105 * HOL_quantifiers flag superseded by "HOL" print mode, which is
  4106 disabled by default; run isabelle with option -m HOL to get back to
  4107 the original Gordon/HOL-style output;
  4108 
  4109 * HOL/Ord.thy: new bounded quantifier syntax (input only): ALL x<y. P,
  4110 ALL x<=y. P, EX x<y. P, EX x<=y. P;
  4111 
  4112 * HOL basic syntax simplified (more orthogonal): all variants of
  4113 All/Ex now support plain / symbolic / HOL notation; plain syntax for
  4114 Eps operator is provided as well: "SOME x. P[x]";
  4115 
  4116 * HOL/Sum.thy: sum_case has been moved to HOL/Datatype;
  4117 
  4118 * HOL/Univ.thy: infix syntax <*>, <+>, <**>, <+> eliminated and made
  4119 thus available for user theories;
  4120 
  4121 * HOLCF/IOA/Sequents: renamed 'Cons' to 'Consq' to avoid clash with
  4122 HOL/List; hardly an INCOMPATIBILITY since '>>' syntax is used all the
  4123 time;
  4124 
  4125 * HOL: new tactic smp_tac: int -> int -> tactic, which applies spec
  4126 several times and then mp;
  4127 
  4128 
  4129 *** LK ***
  4130 
  4131 * the notation <<...>> is now available as a notation for sequences of
  4132 formulas;
  4133 
  4134 * the simplifier is now installed
  4135 
  4136 * the axiom system has been generalized (thanks to Soren Heilmann)
  4137 
  4138 * the classical reasoner now has a default rule database
  4139 
  4140 
  4141 *** ZF ***
  4142 
  4143 * new primrec section allows primitive recursive functions to be given
  4144 directly (as in HOL) over datatypes and the natural numbers;
  4145 
  4146 * new tactics induct_tac and exhaust_tac for induction (or case
  4147 analysis) over datatypes and the natural numbers;
  4148 
  4149 * the datatype declaration of type T now defines the recursor T_rec;
  4150 
  4151 * simplification automatically does freeness reasoning for datatype
  4152 constructors;
  4153 
  4154 * automatic type-inference, with AddTCs command to insert new
  4155 type-checking rules;
  4156 
  4157 * datatype introduction rules are now added as Safe Introduction rules
  4158 to the claset;
  4159 
  4160 * the syntax "if P then x else y" is now available in addition to
  4161 if(P,x,y);
  4162 
  4163 
  4164 *** Internal programming interfaces ***
  4165 
  4166 * tuned simplifier trace output; new flag debug_simp;
  4167 
  4168 * structures Vartab / Termtab (instances of TableFun) offer efficient
  4169 tables indexed by indexname_ord / term_ord (compatible with aconv);
  4170 
  4171 * AxClass.axclass_tac lost the theory argument;
  4172 
  4173 * tuned current_goals_markers semantics: begin / end goal avoids
  4174 printing empty lines;
  4175 
  4176 * removed prs and prs_fn hook, which was broken because it did not
  4177 include \n in its semantics, forcing writeln to add one
  4178 uncoditionally; replaced prs_fn by writeln_fn; consider std_output:
  4179 string -> unit if you really want to output text without newline;
  4180 
  4181 * Symbol.output subject to print mode; INCOMPATIBILITY: defaults to
  4182 plain output, interface builders may have to enable 'isabelle_font'
  4183 mode to get Isabelle font glyphs as before;
  4184 
  4185 * refined token_translation interface; INCOMPATIBILITY: output length
  4186 now of type real instead of int;
  4187 
  4188 * theory loader actions may be traced via new ThyInfo.add_hook
  4189 interface (see src/Pure/Thy/thy_info.ML); example application: keep
  4190 your own database of information attached to *whole* theories -- as
  4191 opposed to intra-theory data slots offered via TheoryDataFun;
  4192 
  4193 * proper handling of dangling sort hypotheses (at last!);
  4194 Thm.strip_shyps and Drule.strip_shyps_warning take care of removing
  4195 extra sort hypotheses that can be witnessed from the type signature;
  4196 the force_strip_shyps flag is gone, any remaining shyps are simply
  4197 left in the theorem (with a warning issued by strip_shyps_warning);
  4198 
  4199 
  4200 
  4201 New in Isabelle98-1 (October 1998)
  4202 ----------------------------------
  4203 
  4204 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  4205 
  4206 * several changes of automated proof tools;
  4207 
  4208 * HOL: major changes to the inductive and datatype packages, including
  4209 some minor incompatibilities of theory syntax;
  4210 
  4211 * HOL: renamed r^-1 to 'converse' from 'inverse'; 'inj_onto' is now
  4212 called `inj_on';
  4213 
  4214 * HOL: removed duplicate thms in Arith:
  4215   less_imp_add_less  should be replaced by  trans_less_add1
  4216   le_imp_add_le      should be replaced by  trans_le_add1
  4217 
  4218 * HOL: unary minus is now overloaded (new type constraints may be
  4219 required);
  4220 
  4221 * HOL and ZF: unary minus for integers is now #- instead of #~.  In
  4222 ZF, expressions such as n#-1 must be changed to n#- 1, since #-1 is
  4223 now taken as an integer constant.
  4224 
  4225 * Pure: ML function 'theory_of' renamed to 'theory';
  4226 
  4227 
  4228 *** Proof tools ***
  4229 
  4230 * Simplifier:
  4231   1. Asm_full_simp_tac is now more aggressive.
  4232      1. It will sometimes reorient premises if that increases their power to
  4233         simplify.
  4234      2. It does no longer proceed strictly from left to right but may also
  4235         rotate premises to achieve further simplification.
  4236      For compatibility reasons there is now Asm_lr_simp_tac which is like the
  4237      old Asm_full_simp_tac in that it does not rotate premises.
  4238   2. The simplifier now knows a little bit about nat-arithmetic.
  4239 
  4240 * Classical reasoner: wrapper mechanism for the classical reasoner now
  4241 allows for selected deletion of wrappers, by introduction of names for
  4242 wrapper functionals.  This implies that addbefore, addSbefore,
  4243 addaltern, and addSaltern now take a pair (name, tactic) as argument,
  4244 and that adding two tactics with the same name overwrites the first
  4245 one (emitting a warning).
  4246   type wrapper = (int -> tactic) -> (int -> tactic)
  4247   setWrapper, setSWrapper, compWrapper and compSWrapper are replaced by
  4248   addWrapper, addSWrapper: claset * (string * wrapper) -> claset
  4249   delWrapper, delSWrapper: claset *  string            -> claset
  4250   getWrapper is renamed to appWrappers, getSWrapper to appSWrappers;
  4251 
  4252 * Classical reasoner: addbefore/addSbefore now have APPEND/ORELSE
  4253 semantics; addbefore now affects only the unsafe part of step_tac
  4254 etc.; this affects addss/auto_tac/force_tac, so EXISTING PROOFS MAY
  4255 FAIL, but proofs should be fixable easily, e.g. by replacing Auto_tac
  4256 by Force_tac;
  4257 
  4258 * Classical reasoner: setwrapper to setWrapper and compwrapper to
  4259 compWrapper; added safe wrapper (and access functions for it);
  4260 
  4261 * HOL/split_all_tac is now much faster and fails if there is nothing
  4262 to split.  Some EXISTING PROOFS MAY REQUIRE ADAPTION because the order
  4263 and the names of the automatically generated variables have changed.
  4264 split_all_tac has moved within claset() from unsafe wrappers to safe
  4265 wrappers, which means that !!-bound variables are split much more
  4266 aggressively, and safe_tac and clarify_tac now split such variables.
  4267 If this splitting is not appropriate, use delSWrapper "split_all_tac".
  4268 Note: the same holds for record_split_tac, which does the job of
  4269 split_all_tac for record fields.
  4270 
  4271 * HOL/Simplifier: Rewrite rules for case distinctions can now be added
  4272 permanently to the default simpset using Addsplits just like
  4273 Addsimps. They can be removed via Delsplits just like
  4274 Delsimps. Lower-case versions are also available.
  4275 
  4276 * HOL/Simplifier: The rule split_if is now part of the default
  4277 simpset. This means that the simplifier will eliminate all occurrences
  4278 of if-then-else in the conclusion of a goal. To prevent this, you can
  4279 either remove split_if completely from the default simpset by
  4280 `Delsplits [split_if]' or remove it in a specific call of the
  4281 simplifier using `... delsplits [split_if]'.  You can also add/delete
  4282 other case splitting rules to/from the default simpset: every datatype
  4283 generates suitable rules `split_t_case' and `split_t_case_asm' (where
  4284 t is the name of the datatype).
  4285 
  4286 * Classical reasoner / Simplifier combination: new force_tac (and
  4287 derivatives Force_tac, force) combines rewriting and classical
  4288 reasoning (and whatever other tools) similarly to auto_tac, but is
  4289 aimed to solve the given subgoal completely.
  4290 
  4291 
  4292 *** General ***
  4293 
  4294 * new top-level commands `Goal' and `Goalw' that improve upon `goal'
  4295 and `goalw': the theory is no longer needed as an explicit argument -
  4296 the current theory context is used; assumptions are no longer returned
  4297 at the ML-level unless one of them starts with ==> or !!; it is
  4298 recommended to convert to these new commands using isatool fixgoal
  4299 (backup your sources first!);
  4300 
  4301 * new top-level commands 'thm' and 'thms' for retrieving theorems from
  4302 the current theory context, and 'theory' to lookup stored theories;
  4303 
  4304 * new theory section 'locale' for declaring constants, assumptions and
  4305 definitions that have local scope;
  4306 
  4307 * new theory section 'nonterminals' for purely syntactic types;
  4308 
  4309 * new theory section 'setup' for generic ML setup functions
  4310 (e.g. package initialization);
  4311 
  4312 * the distribution now includes Isabelle icons: see
  4313 lib/logo/isabelle-{small,tiny}.xpm;
  4314 
  4315 * isatool install - install binaries with absolute references to
  4316 ISABELLE_HOME/bin;
  4317 
  4318 * isatool logo -- create instances of the Isabelle logo (as EPS);
  4319 
  4320 * print mode 'emacs' reserved for Isamode;
  4321 
  4322 * support multiple print (ast) translations per constant name;
  4323 
  4324 * theorems involving oracles are now printed with a suffixed [!];
  4325 
  4326 
  4327 *** HOL ***
  4328 
  4329 * there is now a tutorial on Isabelle/HOL (do 'isatool doc tutorial');
  4330 
  4331 * HOL/inductive package reorganized and improved: now supports mutual
  4332 definitions such as
  4333 
  4334   inductive EVEN ODD
  4335     intrs
  4336       null "0 : EVEN"
  4337       oddI "n : EVEN ==> Suc n : ODD"
  4338       evenI "n : ODD ==> Suc n : EVEN"
  4339 
  4340 new theorem list "elims" contains an elimination rule for each of the
  4341 recursive sets; inductive definitions now handle disjunctive premises
  4342 correctly (also ZF);
  4343 
  4344 INCOMPATIBILITIES: requires Inductive as an ancestor; component
  4345 "mutual_induct" no longer exists - the induction rule is always
  4346 contained in "induct";
  4347 
  4348 
  4349 * HOL/datatype package re-implemented and greatly improved: now
  4350 supports mutually recursive datatypes such as
  4351 
  4352   datatype
  4353     'a aexp = IF_THEN_ELSE ('a bexp) ('a aexp) ('a aexp)
  4354             | SUM ('a aexp) ('a aexp)
  4355             | DIFF ('a aexp) ('a aexp)
  4356             | NUM 'a
  4357   and
  4358     'a bexp = LESS ('a aexp) ('a aexp)
  4359             | AND ('a bexp) ('a bexp)
  4360             | OR ('a bexp) ('a bexp)
  4361 
  4362 as well as indirectly recursive datatypes such as
  4363 
  4364   datatype
  4365     ('a, 'b) term = Var 'a
  4366                   | App 'b ((('a, 'b) term) list)
  4367 
  4368 The new tactic  mutual_induct_tac [<var_1>, ..., <var_n>] i  performs
  4369 induction on mutually / indirectly recursive datatypes.
  4370 
  4371 Primrec equations are now stored in theory and can be accessed via
  4372 <function_name>.simps.
  4373 
  4374 INCOMPATIBILITIES:
  4375 
  4376   - Theories using datatypes must now have theory Datatype as an
  4377     ancestor.
  4378   - The specific <typename>.induct_tac no longer exists - use the
  4379     generic induct_tac instead.
  4380   - natE has been renamed to nat.exhaust - use exhaust_tac
  4381     instead of res_inst_tac ... natE. Note that the variable
  4382     names in nat.exhaust differ from the names in natE, this
  4383     may cause some "fragile" proofs to fail.
  4384   - The theorems split_<typename>_case and split_<typename>_case_asm
  4385     have been renamed to <typename>.split and <typename>.split_asm.
  4386   - Since default sorts of type variables are now handled correctly,
  4387     some datatype definitions may have to be annotated with explicit
  4388     sort constraints.
  4389   - Primrec definitions no longer require function name and type
  4390     of recursive argument.
  4391 
  4392 Consider using isatool fixdatatype to adapt your theories and proof
  4393 scripts to the new package (backup your sources first!).
  4394 
  4395 
  4396 * HOL/record package: considerably improved implementation; now
  4397 includes concrete syntax for record types, terms, updates; theorems
  4398 for surjective pairing and splitting !!-bound record variables; proof
  4399 support is as follows:
  4400 
  4401   1) standard conversions (selectors or updates applied to record
  4402 constructor terms) are part of the standard simpset;
  4403 
  4404   2) inject equations of the form ((x, y) = (x', y')) == x=x' & y=y' are
  4405 made part of standard simpset and claset via addIffs;
  4406 
  4407   3) a tactic for record field splitting (record_split_tac) is part of
  4408 the standard claset (addSWrapper);
  4409 
  4410 To get a better idea about these rules you may retrieve them via
  4411 something like 'thms "foo.simps"' or 'thms "foo.iffs"', where "foo" is
  4412 the name of your record type.
  4413 
  4414 The split tactic 3) conceptually simplifies by the following rule:
  4415 
  4416   "(!!x. PROP ?P x) == (!!a b. PROP ?P (a, b))"
  4417 
  4418 Thus any record variable that is bound by meta-all will automatically
  4419 blow up into some record constructor term, consequently the
  4420 simplifications of 1), 2) apply.  Thus force_tac, auto_tac etc. shall
  4421 solve record problems automatically.
  4422 
  4423 
  4424 * reorganized the main HOL image: HOL/Integ and String loaded by
  4425 default; theory Main includes everything;
  4426 
  4427 * automatic simplification of integer sums and comparisons, using cancellation;
  4428 
  4429 * added option_map_eq_Some and not_Some_eq to the default simpset and claset;
  4430 
  4431 * added disj_not1 = "(~P | Q) = (P --> Q)" to the default simpset;
  4432 
  4433 * many new identities for unions, intersections, set difference, etc.;
  4434 
  4435 * expand_if, expand_split, expand_sum_case and expand_nat_case are now
  4436 called split_if, split_split, split_sum_case and split_nat_case (to go
  4437 with add/delsplits);
  4438 
  4439 * HOL/Prod introduces simplification procedure unit_eq_proc rewriting
  4440 (?x::unit) = (); this is made part of the default simpset, which COULD
  4441 MAKE EXISTING PROOFS FAIL under rare circumstances (consider
  4442 'Delsimprocs [unit_eq_proc];' as last resort); also note that
  4443 unit_abs_eta_conv is added in order to counter the effect of
  4444 unit_eq_proc on (%u::unit. f u), replacing it by f rather than by
  4445 %u.f();
  4446 
  4447 * HOL/Fun INCOMPATIBILITY: `inj_onto' is now called `inj_on' (which
  4448 makes more sense);
  4449 
  4450 * HOL/Set INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
  4451   It and 'sym RS equals0D' are now in the default  claset, giving automatic
  4452   disjointness reasoning but breaking a few old proofs.
  4453 
  4454 * HOL/Relation INCOMPATIBILITY: renamed the relational operator r^-1
  4455 to 'converse' from 'inverse' (for compatibility with ZF and some
  4456 literature);
  4457 
  4458 * HOL/recdef can now declare non-recursive functions, with {} supplied as
  4459 the well-founded relation;
  4460 
  4461 * HOL/Set INCOMPATIBILITY: the complement of set A is now written -A instead of
  4462     Compl A.  The "Compl" syntax remains available as input syntax for this
  4463     release ONLY.
  4464 
  4465 * HOL/Update: new theory of function updates:
  4466     f(a:=b) == %x. if x=a then b else f x
  4467 may also be iterated as in f(a:=b,c:=d,...);
  4468 
  4469 * HOL/Vimage: new theory for inverse image of a function, syntax f-``B;
  4470 
  4471 * HOL/List:
  4472   - new function list_update written xs[i:=v] that updates the i-th
  4473     list position. May also be iterated as in xs[i:=a,j:=b,...].
  4474   - new function `upt' written [i..j(] which generates the list
  4475     [i,i+1,...,j-1], i.e. the upper bound is excluded. To include the upper
  4476     bound write [i..j], which is a shorthand for [i..j+1(].
  4477   - new lexicographic orderings and corresponding wellfoundedness theorems.
  4478 
  4479 * HOL/Arith:
  4480   - removed 'pred' (predecessor) function;
  4481   - generalized some theorems about n-1;
  4482   - many new laws about "div" and "mod";
  4483   - new laws about greatest common divisors (see theory ex/Primes);
  4484 
  4485 * HOL/Relation: renamed the relational operator r^-1 "converse"
  4486 instead of "inverse";
  4487 
  4488 * HOL/Induct/Multiset: a theory of multisets, including the wellfoundedness
  4489   of the multiset ordering;
  4490 
  4491 * directory HOL/Real: a construction of the reals using Dedekind cuts
  4492   (not included by default);
  4493 
  4494 * directory HOL/UNITY: Chandy and Misra's UNITY formalism;
  4495 
  4496 * directory HOL/Hoare: a new version of Hoare logic which permits many-sorted
  4497   programs, i.e. different program variables may have different types.
  4498 
  4499 * calling (stac rew i) now fails if "rew" has no effect on the goal
  4500   [previously, this check worked only if the rewrite rule was unconditional]
  4501   Now rew can involve either definitions or equalities (either == or =).
  4502 
  4503 
  4504 *** ZF ***
  4505 
  4506 * theory Main includes everything; INCOMPATIBILITY: theory ZF.thy contains
  4507   only the theorems proved on ZF.ML;
  4508 
  4509 * ZF INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
  4510   It and 'sym RS equals0D' are now in the default  claset, giving automatic
  4511   disjointness reasoning but breaking a few old proofs.
  4512 
  4513 * ZF/Update: new theory of function updates
  4514     with default rewrite rule  f(x:=y) ` z = if(z=x, y, f`z)
  4515   may also be iterated as in f(a:=b,c:=d,...);
  4516 
  4517 * in  let x=t in u(x), neither t nor u(x) has to be an FOL term.
  4518 
  4519 * calling (stac rew i) now fails if "rew" has no effect on the goal
  4520   [previously, this check worked only if the rewrite rule was unconditional]
  4521   Now rew can involve either definitions or equalities (either == or =).
  4522 
  4523 * case_tac provided for compatibility with HOL
  4524     (like the old excluded_middle_tac, but with subgoals swapped)
  4525 
  4526 
  4527 *** Internal programming interfaces ***
  4528 
  4529 * Pure: several new basic modules made available for general use, see
  4530 also src/Pure/README;
  4531 
  4532 * improved the theory data mechanism to support encapsulation (data
  4533 kind name replaced by private Object.kind, acting as authorization
  4534 key); new type-safe user interface via functor TheoryDataFun; generic
  4535 print_data function becomes basically useless;
  4536 
  4537 * removed global_names compatibility flag -- all theory declarations
  4538 are qualified by default;
  4539 
  4540 * module Pure/Syntax now offers quote / antiquote translation
  4541 functions (useful for Hoare logic etc. with implicit dependencies);
  4542 see HOL/ex/Antiquote for an example use;
  4543 
  4544 * Simplifier now offers conversions (asm_)(full_)rewrite: simpset ->
  4545 cterm -> thm;
  4546 
  4547 * new tactical CHANGED_GOAL for checking that a tactic modifies a
  4548 subgoal;
  4549 
  4550 * Display.print_goals function moved to Locale.print_goals;
  4551 
  4552 * standard print function for goals supports current_goals_markers
  4553 variable for marking begin of proof, end of proof, start of goal; the
  4554 default is ("", "", ""); setting current_goals_markers := ("<proof>",
  4555 "</proof>", "<goal>") causes SGML like tagged proof state printing,
  4556 for example;
  4557 
  4558 
  4559 
  4560 New in Isabelle98 (January 1998)
  4561 --------------------------------
  4562 
  4563 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  4564 
  4565 * changed lexical syntax of terms / types: dots made part of long
  4566 identifiers, e.g. "%x.x" no longer possible, should be "%x. x";
  4567 
  4568 * simpset (and claset) reference variable replaced by functions
  4569 simpset / simpset_ref;
  4570 
  4571 * no longer supports theory aliases (via merge) and non-trivial
  4572 implicit merge of thms' signatures;
  4573 
  4574 * most internal names of constants changed due to qualified names;
  4575 
  4576 * changed Pure/Sequence interface (see Pure/seq.ML);
  4577 
  4578 
  4579 *** General Changes ***
  4580 
  4581 * hierachically structured name spaces (for consts, types, axms, thms
  4582 etc.); new lexical class 'longid' (e.g. Foo.bar.x) may render much of
  4583 old input syntactically incorrect (e.g. "%x.x"); COMPATIBILITY:
  4584 isatool fixdots ensures space after dots (e.g. "%x. x"); set
  4585 long_names for fully qualified output names; NOTE: ML programs
  4586 (special tactics, packages etc.) referring to internal names may have
  4587 to be adapted to cope with fully qualified names; in case of severe
  4588 backward campatibility problems try setting 'global_names' at compile
  4589 time to have enrything declared within a flat name space; one may also
  4590 fine tune name declarations in theories via the 'global' and 'local'
  4591 section;
  4592 
  4593 * reimplemented the implicit simpset and claset using the new anytype
  4594 data filed in signatures; references simpset:simpset ref etc. are
  4595 replaced by functions simpset:unit->simpset and
  4596 simpset_ref:unit->simpset ref; COMPATIBILITY: use isatool fixclasimp
  4597 to patch your ML files accordingly;
  4598 
  4599 * HTML output now includes theory graph data for display with Java
  4600 applet or isatool browser; data generated automatically via isatool
  4601 usedir (see -i option, ISABELLE_USEDIR_OPTIONS);
  4602 
  4603 * defs may now be conditional; improved rewrite_goals_tac to handle
  4604 conditional equations;
  4605 
  4606 * defs now admits additional type arguments, using TYPE('a) syntax;
  4607 
  4608 * theory aliases via merge (e.g. M=A+B+C) no longer supported, always
  4609 creates a new theory node; implicit merge of thms' signatures is
  4610 restricted to 'trivial' ones; COMPATIBILITY: one may have to use
  4611 transfer:theory->thm->thm in (rare) cases;
  4612 
  4613 * improved handling of draft signatures / theories; draft thms (and
  4614 ctyps, cterms) are automatically promoted to real ones;
  4615 
  4616 * slightly changed interfaces for oracles: admit many per theory, named
  4617 (e.g. oracle foo = mlfun), additional name argument for invoke_oracle;
  4618 
  4619 * print_goals: optional output of const types (set show_consts and
  4620 show_types);
  4621 
  4622 * improved output of warnings (###) and errors (***);
  4623 
  4624 * subgoal_tac displays a warning if the new subgoal has type variables;
  4625 
  4626 * removed old README and Makefiles;
  4627 
  4628 * replaced print_goals_ref hook by print_current_goals_fn and result_error_fn;
  4629 
  4630 * removed obsolete init_pps and init_database;
  4631 
  4632 * deleted the obsolete tactical STATE, which was declared by
  4633     fun STATE tacfun st = tacfun st st;
  4634 
  4635 * cd and use now support path variables, e.g. $ISABELLE_HOME, or ~
  4636 (which abbreviates $HOME);
  4637 
  4638 * changed Pure/Sequence interface (see Pure/seq.ML); COMPATIBILITY:
  4639 use isatool fixseq to adapt your ML programs (this works for fully
  4640 qualified references to the Sequence structure only!);
  4641 
  4642 * use_thy no longer requires writable current directory; it always
  4643 reloads .ML *and* .thy file, if either one is out of date;
  4644 
  4645 
  4646 *** Classical Reasoner ***
  4647 
  4648 * Clarify_tac, clarify_tac, clarify_step_tac, Clarify_step_tac: new
  4649 tactics that use classical reasoning to simplify a subgoal without
  4650 splitting it into several subgoals;
  4651 
  4652 * Safe_tac: like safe_tac but uses the default claset;
  4653 
  4654 
  4655 *** Simplifier ***
  4656 
  4657 * added simplification meta rules:
  4658     (asm_)(full_)simplify: simpset -> thm -> thm;
  4659 
  4660 * simplifier.ML no longer part of Pure -- has to be loaded by object
  4661 logics (again);
  4662 
  4663 * added prems argument to simplification procedures;
  4664 
  4665 * HOL, FOL, ZF: added infix function `addsplits':
  4666   instead of `<simpset> setloop (split_tac <thms>)'
  4667   you can simply write `<simpset> addsplits <thms>'
  4668 
  4669 
  4670 *** Syntax ***
  4671 
  4672 * TYPE('a) syntax for type reflection terms;
  4673 
  4674 * no longer handles consts with name "" -- declare as 'syntax' instead;
  4675 
  4676 * pretty printer: changed order of mixfix annotation preference (again!);
  4677 
  4678 * Pure: fixed idt/idts vs. pttrn/pttrns syntactic categories;
  4679 
  4680 
  4681 *** HOL ***
  4682 
  4683 * HOL: there is a new splitter `split_asm_tac' that can be used e.g.
  4684   with `addloop' of the simplifier to faciliate case splitting in premises.
  4685 
  4686 * HOL/TLA: Stephan Merz's formalization of Lamport's Temporal Logic of Actions;
  4687 
  4688 * HOL/Auth: new protocol proofs including some for the Internet
  4689   protocol TLS;
  4690 
  4691 * HOL/Map: new theory of `maps' a la VDM;
  4692 
  4693 * HOL/simplifier: simplification procedures nat_cancel_sums for
  4694 cancelling out common nat summands from =, <, <= (in)equalities, or
  4695 differences; simplification procedures nat_cancel_factor for
  4696 cancelling common factor from =, <, <= (in)equalities over natural
  4697 sums; nat_cancel contains both kinds of procedures, it is installed by
  4698 default in Arith.thy -- this COULD MAKE EXISTING PROOFS FAIL;
  4699 
  4700 * HOL/simplifier: terms of the form
  4701   `? x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x)'  (or t=x)
  4702   are rewritten to
  4703   `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t)',
  4704   and those of the form
  4705   `! x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x) --> R(x)'  (or t=x)
  4706   are rewritten to
  4707   `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t) --> R(t)',
  4708 
  4709 * HOL/datatype
  4710   Each datatype `t' now comes with a theorem `split_t_case' of the form
  4711 
  4712   P(t_case f1 ... fn x) =
  4713      ( (!y1 ... ym1. x = C1 y1 ... ym1 --> P(f1 y1 ... ym1)) &
  4714         ...
  4715        (!y1 ... ymn. x = Cn y1 ... ymn --> P(f1 y1 ... ymn))
  4716      )
  4717 
  4718   and a theorem `split_t_case_asm' of the form
  4719 
  4720   P(t_case f1 ... fn x) =
  4721     ~( (? y1 ... ym1. x = C1 y1 ... ym1 & ~P(f1 y1 ... ym1)) |
  4722         ...
  4723        (? y1 ... ymn. x = Cn y1 ... ymn & ~P(f1 y1 ... ymn))
  4724      )
  4725   which can be added to a simpset via `addsplits'. The existing theorems
  4726   expand_list_case and expand_option_case have been renamed to
  4727   split_list_case and split_option_case.
  4728 
  4729 * HOL/Arithmetic:
  4730   - `pred n' is automatically converted to `n-1'.
  4731     Users are strongly encouraged not to use `pred' any longer,
  4732     because it will disappear altogether at some point.
  4733   - Users are strongly encouraged to write "0 < n" rather than
  4734     "n ~= 0". Theorems and proof tools have been modified towards this
  4735     `standard'.
  4736 
  4737 * HOL/Lists:
  4738   the function "set_of_list" has been renamed "set" (and its theorems too);
  4739   the function "nth" now takes its arguments in the reverse order and
  4740   has acquired the infix notation "!" as in "xs!n".
  4741 
  4742 * HOL/Set: UNIV is now a constant and is no longer translated to Compl{};
  4743 
  4744 * HOL/Set: The operator (UN x.B x) now abbreviates (UN x:UNIV. B x) and its
  4745   specialist theorems (like UN1_I) are gone.  Similarly for (INT x.B x);
  4746 
  4747 * HOL/record: extensible records with schematic structural subtyping
  4748 (single inheritance); EXPERIMENTAL version demonstrating the encoding,
  4749 still lacks various theorems and concrete record syntax;
  4750 
  4751 
  4752 *** HOLCF ***
  4753 
  4754 * removed "axioms" and "generated by" sections;
  4755 
  4756 * replaced "ops" section by extended "consts" section, which is capable of
  4757   handling the continuous function space "->" directly;
  4758 
  4759 * domain package:
  4760   . proves theorems immediately and stores them in the theory,
  4761   . creates hierachical name space,
  4762   . now uses normal mixfix annotations (instead of cinfix...),
  4763   . minor changes to some names and values (for consistency),
  4764   . e.g. cases -> casedist, dists_eq -> dist_eqs, [take_lemma] -> take_lemmas,
  4765   . separator between mutual domain defs: changed "," to "and",
  4766   . improved handling of sort constraints;  now they have to
  4767     appear on the left-hand side of the equations only;
  4768 
  4769 * fixed LAM <x,y,zs>.b syntax;
  4770 
  4771 * added extended adm_tac to simplifier in HOLCF -- can now discharge
  4772 adm (%x. P (t x)), where P is chainfinite and t continuous;
  4773 
  4774 
  4775 *** FOL and ZF ***
  4776 
  4777 * FOL: there is a new splitter `split_asm_tac' that can be used e.g.
  4778   with `addloop' of the simplifier to faciliate case splitting in premises.
  4779 
  4780 * qed_spec_mp, qed_goal_spec_mp, qed_goalw_spec_mp are available, as
  4781 in HOL, they strip ALL and --> from proved theorems;
  4782 
  4783 
  4784 
  4785 New in Isabelle94-8 (May 1997)
  4786 ------------------------------
  4787 
  4788 *** General Changes ***
  4789 
  4790 * new utilities to build / run / maintain Isabelle etc. (in parts
  4791 still somewhat experimental); old Makefiles etc. still functional;
  4792 
  4793 * new 'Isabelle System Manual';
  4794 
  4795 * INSTALL text, together with ./configure and ./build scripts;
  4796 
  4797 * reimplemented type inference for greater efficiency, better error
  4798 messages and clean internal interface;
  4799 
  4800 * prlim command for dealing with lots of subgoals (an easier way of
  4801 setting goals_limit);
  4802 
  4803 
  4804 *** Syntax ***
  4805 
  4806 * supports alternative (named) syntax tables (parser and pretty
  4807 printer); internal interface is provided by add_modesyntax(_i);
  4808 
  4809 * Pure, FOL, ZF, HOL, HOLCF now support symbolic input and output; to
  4810 be used in conjunction with the Isabelle symbol font; uses the
  4811 "symbols" syntax table;
  4812 
  4813 * added token_translation interface (may translate name tokens in
  4814 arbitrary ways, dependent on their type (free, bound, tfree, ...) and
  4815 the current print_mode); IMPORTANT: user print translation functions
  4816 are responsible for marking newly introduced bounds
  4817 (Syntax.mark_boundT);
  4818 
  4819 * token translations for modes "xterm" and "xterm_color" that display
  4820 names in bold, underline etc. or colors (which requires a color
  4821 version of xterm);
  4822 
  4823 * infixes may now be declared with names independent of their syntax;
  4824 
  4825 * added typed_print_translation (like print_translation, but may
  4826 access type of constant);
  4827 
  4828 
  4829 *** Classical Reasoner ***
  4830 
  4831 Blast_tac: a new tactic!  It is often more powerful than fast_tac, but has
  4832 some limitations.  Blast_tac...
  4833   + ignores addss, addbefore, addafter; this restriction is intrinsic
  4834   + ignores elimination rules that don't have the correct format
  4835         (the conclusion MUST be a formula variable)
  4836   + ignores types, which can make HOL proofs fail
  4837   + rules must not require higher-order unification, e.g. apply_type in ZF
  4838     [message "Function Var's argument not a bound variable" relates to this]
  4839   + its proof strategy is more general but can actually be slower
  4840 
  4841 * substitution with equality assumptions no longer permutes other
  4842 assumptions;
  4843 
  4844 * minor changes in semantics of addafter (now called addaltern); renamed
  4845 setwrapper to setWrapper and compwrapper to compWrapper; added safe wrapper
  4846 (and access functions for it);
  4847 
  4848 * improved combination of classical reasoner and simplifier:
  4849   + functions for handling clasimpsets
  4850   + improvement of addss: now the simplifier is called _after_ the
  4851     safe steps.
  4852   + safe variant of addss called addSss: uses safe simplifications
  4853     _during_ the safe steps. It is more complete as it allows multiple
  4854     instantiations of unknowns (e.g. with slow_tac).
  4855 
  4856 *** Simplifier ***
  4857 
  4858 * added interface for simplification procedures (functions that
  4859 produce *proven* rewrite rules on the fly, depending on current
  4860 redex);
  4861 
  4862 * ordering on terms as parameter (used for ordered rewriting);
  4863 
  4864 * new functions delcongs, deleqcongs, and Delcongs. richer rep_ss;
  4865 
  4866 * the solver is now split into a safe and an unsafe part.
  4867 This should be invisible for the normal user, except that the
  4868 functions setsolver and addsolver have been renamed to setSolver and
  4869 addSolver; added safe_asm_full_simp_tac;
  4870 
  4871 
  4872 *** HOL ***
  4873 
  4874 * a generic induction tactic `induct_tac' which works for all datatypes and
  4875 also for type `nat';
  4876 
  4877 * a generic case distinction tactic `exhaust_tac' which works for all
  4878 datatypes and also for type `nat';
  4879 
  4880 * each datatype comes with a function `size';
  4881 
  4882 * patterns in case expressions allow tuple patterns as arguments to
  4883 constructors, for example `case x of [] => ... | (x,y,z)#ps => ...';
  4884 
  4885 * primrec now also works with type nat;
  4886 
  4887 * recdef: a new declaration form, allows general recursive functions to be
  4888 defined in theory files.  See HOL/ex/Fib, HOL/ex/Primes, HOL/Subst/Unify.
  4889 
  4890 * the constant for negation has been renamed from "not" to "Not" to
  4891 harmonize with FOL, ZF, LK, etc.;
  4892 
  4893 * HOL/ex/LFilter theory of a corecursive "filter" functional for
  4894 infinite lists;
  4895 
  4896 * HOL/Modelcheck demonstrates invocation of model checker oracle;
  4897 
  4898 * HOL/ex/Ring.thy declares cring_simp, which solves equational
  4899 problems in commutative rings, using axiomatic type classes for + and *;
  4900 
  4901 * more examples in HOL/MiniML and HOL/Auth;
  4902 
  4903 * more default rewrite rules for quantifiers, union/intersection;
  4904 
  4905 * a new constant `arbitrary == @x.False';
  4906 
  4907 * HOLCF/IOA replaces old HOL/IOA;
  4908 
  4909 * HOLCF changes: derived all rules and arities
  4910   + axiomatic type classes instead of classes
  4911   + typedef instead of faking type definitions
  4912   + eliminated the internal constants less_fun, less_cfun, UU_fun, UU_cfun etc.
  4913   + new axclasses cpo, chfin, flat with flat < chfin < pcpo < cpo < po
  4914   + eliminated the types void, one, tr
  4915   + use unit lift and bool lift (with translations) instead of one and tr
  4916   + eliminated blift from Lift3.thy (use Def instead of blift)
  4917   all eliminated rules are derived as theorems --> no visible changes ;
  4918 
  4919 
  4920 *** ZF ***
  4921 
  4922 * ZF now has Fast_tac, Simp_tac and Auto_tac.  Union_iff is a now a default
  4923 rewrite rule; this may affect some proofs.  eq_cs is gone but can be put back
  4924 as ZF_cs addSIs [equalityI];
  4925 
  4926 
  4927 
  4928 New in Isabelle94-7 (November 96)
  4929 ---------------------------------
  4930 
  4931 * allowing negative levels (as offsets) in prlev and choplev;
  4932 
  4933 * super-linear speedup for large simplifications;
  4934 
  4935 * FOL, ZF and HOL now use miniscoping: rewriting pushes
  4936 quantifications in as far as possible (COULD MAKE EXISTING PROOFS
  4937 FAIL); can suppress it using the command Delsimps (ex_simps @
  4938 all_simps); De Morgan laws are also now included, by default;
  4939 
  4940 * improved printing of ==>  :  ~:
  4941 
  4942 * new object-logic "Sequents" adds linear logic, while replacing LK
  4943 and Modal (thanks to Sara Kalvala);
  4944 
  4945 * HOL/Auth: correctness proofs for authentication protocols;
  4946 
  4947 * HOL: new auto_tac combines rewriting and classical reasoning (many
  4948 examples on HOL/Auth);
  4949 
  4950 * HOL: new command AddIffs for declaring theorems of the form P=Q to
  4951 the rewriter and classical reasoner simultaneously;
  4952 
  4953 * function uresult no longer returns theorems in "standard" format;
  4954 regain previous version by: val uresult = standard o uresult;
  4955 
  4956 
  4957 
  4958 New in Isabelle94-6
  4959 -------------------
  4960 
  4961 * oracles -- these establish an interface between Isabelle and trusted
  4962 external reasoners, which may deliver results as theorems;
  4963 
  4964 * proof objects (in particular record all uses of oracles);
  4965 
  4966 * Simp_tac, Fast_tac, etc. that refer to implicit simpset / claset;
  4967 
  4968 * "constdefs" section in theory files;
  4969 
  4970 * "primrec" section (HOL) no longer requires names;
  4971 
  4972 * internal type "tactic" now simply "thm -> thm Sequence.seq";
  4973 
  4974 
  4975 
  4976 New in Isabelle94-5
  4977 -------------------
  4978 
  4979 * reduced space requirements;
  4980 
  4981 * automatic HTML generation from theories;
  4982 
  4983 * theory files no longer require "..." (quotes) around most types;
  4984 
  4985 * new examples, including two proofs of the Church-Rosser theorem;
  4986 
  4987 * non-curried (1994) version of HOL is no longer distributed;
  4988 
  4989 
  4990 
  4991 New in Isabelle94-4
  4992 -------------------
  4993 
  4994 * greatly reduced space requirements;
  4995 
  4996 * theory files (.thy) no longer require \...\ escapes at line breaks;
  4997 
  4998 * searchable theorem database (see the section "Retrieving theorems" on
  4999 page 8 of the Reference Manual);
  5000 
  5001 * new examples, including Grabczewski's monumental case study of the
  5002 Axiom of Choice;
  5003 
  5004 * The previous version of HOL renamed to Old_HOL;
  5005 
  5006 * The new version of HOL (previously called CHOL) uses a curried syntax
  5007 for functions.  Application looks like f a b instead of f(a,b);
  5008 
  5009 * Mutually recursive inductive definitions finally work in HOL;
  5010 
  5011 * In ZF, pattern-matching on tuples is now available in all abstractions and
  5012 translates to the operator "split";
  5013 
  5014 
  5015 
  5016 New in Isabelle94-3
  5017 -------------------
  5018 
  5019 * new infix operator, addss, allowing the classical reasoner to
  5020 perform simplification at each step of its search.  Example:
  5021         fast_tac (cs addss ss)
  5022 
  5023 * a new logic, CHOL, the same as HOL, but with a curried syntax
  5024 for functions.  Application looks like f a b instead of f(a,b).  Also pairs
  5025 look like (a,b) instead of <a,b>;
  5026 
  5027 * PLEASE NOTE: CHOL will eventually replace HOL!
  5028 
  5029 * In CHOL, pattern-matching on tuples is now available in all abstractions.
  5030 It translates to the operator "split".  A new theory of integers is available;
  5031 
  5032 * In ZF, integer numerals now denote two's-complement binary integers.
  5033 Arithmetic operations can be performed by rewriting.  See ZF/ex/Bin.ML;
  5034 
  5035 * Many new examples: I/O automata, Church-Rosser theorem, equivalents
  5036 of the Axiom of Choice;
  5037 
  5038 
  5039 
  5040 New in Isabelle94-2
  5041 -------------------
  5042 
  5043 * Significantly faster resolution;
  5044 
  5045 * the different sections in a .thy file can now be mixed and repeated
  5046 freely;
  5047 
  5048 * Database of theorems for FOL, HOL and ZF.  New
  5049 commands including qed, qed_goal and bind_thm store theorems in the database.
  5050 
  5051 * Simple database queries: return a named theorem (get_thm) or all theorems of
  5052 a given theory (thms_of), or find out what theory a theorem was proved in
  5053 (theory_of_thm);
  5054 
  5055 * Bugs fixed in the inductive definition and datatype packages;
  5056 
  5057 * The classical reasoner provides deepen_tac and depth_tac, making FOL_dup_cs
  5058 and HOL_dup_cs obsolete;
  5059 
  5060 * Syntactic ambiguities caused by the new treatment of syntax in Isabelle94-1
  5061 have been removed;
  5062 
  5063 * Simpler definition of function space in ZF;
  5064 
  5065 * new results about cardinal and ordinal arithmetic in ZF;
  5066 
  5067 * 'subtype' facility in HOL for introducing new types as subsets of existing
  5068 types;
  5069 
  5070 :mode=text:wrap=hard:maxLineLen=72:
  5071 $Id$