summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
* Fix issue #62. Check for modifies clause and constructors in the enclosingGravatar qunyanm2015-03-31
| | | | class when determining whether a method named Main() is the program entry.
* Fix issue 61. Decreases are by default in ghost context. Therefore,Gravatar qunyanm2015-03-31
| | | | dontCareAboutCompilation flag should be set to false in the ResolveOpts.
* Fix issue #56. Convert parametered opaque type parameters into an IdentifierExprGravatar qunyanm2015-03-27
| | | | instead of FunctionCall.
* Fix issue #58. In TrSplitExpr(), add allocatedness for arguments to the inlinedGravatar qunyanm2015-03-26
| | | | call.
* Add LongLength methods to Set/Map in DafnyRuntime.csGravatar chrishaw2015-03-17
|
* MergeGravatar chrishaw2015-03-17
|\
* | Optimize datatype/tuple update expression to coalesce adjacent updates into ↵Gravatar chrishaw2015-03-17
| | | | | | | | a single update. Specifically, e[x1:=e1]...[xn:=en] is treated like a single update e[x1,...,xn := e1,...,en].
| * Fix issue #63. ForceSubstitutionOfQuantifiedVars for SetComprehesion.Gravatar qunyanm2015-03-17
|/
* Fix issue #55 and #64. When performing type argument substitution for functionGravatar qunyanm2015-03-16
| | | | | | F's corresponding F_FULL, make sure the function is indeed the FullVesion before the substitution. Also keep the TypeArgsSubst for the enclosingClass in the typeMap.
* Fixed merge issuesGravatar leino2015-03-13
|
* MergeGravatar leino2015-03-13
|\
* | Allow let-such-that expression to be compiled, provided that they provably ↵Gravatar leino2015-03-13
| | | | | | | | have a unique value
| * Allow trigger annotations in more statements and expressionsGravatar chrishaw2015-03-11
| |
| * Remove unneeded reference from DafnyExtension.csproj (accidentally added by ↵Gravatar chrishaw2015-03-11
| | | | | | | | my previous checkin)
| * Add DafnyOptions.txt, and have the Dafny Visual Studio extension read ↵Gravatar chrishaw2015-03-11
| | | | | | | | command-line options from DafnyOptions.txt during initialization. DafnyOptions.txt gets built into the VSIX file and is installed as part of the Dafny Visual Studio extension.
| * MergeGravatar qunyanm2015-03-11
| |\
| * | Fix issue #54 and #57. Resolve a formal's type before creating a substitute.Gravatar qunyanm2015-03-11
| | |
| | * Beefed up collection axioms (in particular, for maps) to improve the chance ↵Gravatar Rustan Leino2015-03-10
| |/ |/| | | | | of proving the existence check of let-such-that and assign-such-that
* | Fixed bug in resolution of illegal programs.Gravatar leino2015-03-10
| | | | | | | | Fixed comments in some test cases.
* | Removed some old and unused functionsGravatar leino2015-03-09
| |
* | This changeset changes the default visibility of a function/predicate body ↵Gravatar leino2015-03-09
| | | | | | | | | | | | | | | | | | | | outside the module that declares it. The body is now visible across the module boundary. To contain the knowledge of the body inside the module, mark the function/predicate as 'protected'. Semantics of 'protected': * The definition (i.e., body) of a 'protected' function is not visible outside the defining module * The idea is that inside the defining module, a 'protected' function may or may not be opaque. However, this will be easier to support once opaque/reveal are language primitives. Therefore, for the time being, {:opaque} is not allowed to be applied to 'protected' functions. * In order to extend the definition of a predicate in a refinement module, the predicate must be 'protected' * The 'protected' status of a function must be preserved in refinement modules
* | Added 'protected' keyword (syntax)Gravatar leino2015-03-07
|/
* Fix issue #60Gravatar qunyanm2015-03-06
|
* Stop pretty-print from emitting deprecated semi-colons.Gravatar qunyanm2015-03-05
|
* Fixed issue #59.Gravatar wuestholz2015-03-03
|
* Add imap display/update expressionsGravatar chrishaw2015-02-27
|
* Removed Util/VS2010 folder, which has been superseded by DafnyExtension ↵Gravatar Rustan Leino2015-02-27
| | | | years ago
* Add imap keyword to VS/emacs/vim/latex filesGravatar chrishaw2015-02-26
|
* Add imap type, which is like map but may have have infinite sizeGravatar chrishaw2015-02-26
|
* Add an explicit trigger so that Z3 doesn't choose an overly generous trigger ↵Gravatar Bryan Parno2015-02-09
| | | | on SeqIndex
* Make allowGlobals a bit more generousGravatar Bryan Parno2015-02-09
|
* Generate unique IDs hierarchically, to reduce changes to IDs when the ↵Gravatar leino2015-01-28
| | | | program is modified (to help with fine-grained caching)
* Did some refactoring to improve the name generation.Gravatar wuestholz2015-01-28
|
* Did some refactoring to improve the name generation.Gravatar wuestholz2015-01-28
|
* Did some refactoring to improve the name generation.Gravatar wuestholz2015-01-27
|
* Did some refactoring to improve the name generation.Gravatar wuestholz2015-01-27
|
* Did some refactoring to improve the name generation.Gravatar wuestholz2015-01-27
|
* Assume type properties of values that are created by a havoc assignment. ↵Gravatar Rustan Leino2015-01-27
| | | | | | Such assignments are part of assign-such-that statements, and thus this also fixes Issue 52.
* Fixed an encoding bug for newtypes (this fixes Issue #50)Gravatar Rustan Leino2015-01-27
|
* Minor change to a test caseGravatar wuestholz2015-01-27
|
* DafnyExtension: Made it use the more advanced on-demand re-verification by ↵Gravatar wuestholz2015-01-26
| | | | default.
* Updated test output after change in Boogie.Gravatar wuestholz2015-01-24
|
* Minor change to grammar to avoid missing tokenGravatar wuestholz2015-01-24
|
* Make sure to check that subrange types are not used as type parametersGravatar leino2015-01-23
|
* MergeGravatar leino2015-01-23
|\
* | Switched use of List(IToken) in UserDefinedType to NameSegment/ExprDotName, ↵Gravatar leino2015-01-23
| | | | | | | | | | | | | | so use the new name resolution machinery that handles modules and type parameters Included some inadvertently left-out test cases in dafny0/Modules0.dfy Fixed comparable-types tests
| * MergeGravatar Bryan Parno2015-01-13
| |\
| | * Minor change due to a change in BoogieGravatar wuestholz2015-01-13
| |/ |/|
* | When ambiguous references all resolve to the same declaration, don't complainGravatar leino2015-01-09
| |
| * Check the AllowGlobals flag in one additional location to avoid a contract ↵Gravatar Bryan Parno2015-01-08
|/ | | | failure