summaryrefslogtreecommitdiffstats
path: root/win/tclAppInit.c
Commit message (Collapse)AuthorAgeFilesLines
* Merge 8.6jan.nijtmans2022-02-011-5/+4
|\
| * Code cleanup in tclMain.c and tclAppInit.c: Make them Tcl-8.7-aware, usable ↵jan.nijtmans2022-02-011-9/+17
| | | | | | | | as more generic examples for extensions.
* | TIP #595 (for Tcl 8.7) part 1: just rename Tcl_StaticPackage to ↵jan.nijtmans2021-04-071-8/+12
| | | | | | | | Tcl_StaticLibrary, Tcl_PackageInitProc to Tcl_LibraryInitProc and Tcl_PackageUnloadProc to Tcl_LibraryUnloadProc. Adapt documentation, mentioning that the old names are now deprecated.
* | In configure/make Windows build, static builds include dde and registry ↵jan.nijtmans2020-12-071-2/+2
| | | | | | | | extension statically too implicitly
* | Deprecate the (internal) functions TclGuessPackageName/TclGetLoadedPackages ↵jan.nijtmans2020-11-061-2/+2
|\ \ | |/ | | | | functions, since they turn out to be useless. The can be removed in 9.0
| * Correct casing of "packageName" argument in Tcl_StaticPackage() call and ↵jan.nijtmans2020-11-061-2/+2
| | | | | | | | | | | | "load" command (which - actually - is not a packageName at all ...) UPDATE: I am moving this to a feature branch. Changing the arguments passed to Tcl_StaticPackage will be an incompatible break for any scripts currently searching the [info loaded] list for "dde" or "registry". Not a change we should put in a patchlevel without at least a ticket to record the explanation and a (POTENTIAL INCOMPATIBILITY) warning.
* | Merge 8.6 Eliminate some (correct but actually harmless) warnings only ↵jan.nijtmans2020-10-131-2/+2
|\ \ | |/ | | | | happening on win64
| * Fix warning, doing a static build on Windowsjan.nijtmans2020-10-131-2/+2
| |
| * Merge 8.5jan.nijtmans2020-10-091-1/+1
| |\
* | | Fix windows build (with older compilers)jan.nijtmans2020-04-141-1/+2
| | |
* | | TCL_UNUSED for win sources.dgp2020-03-091-2/+1
| | |
* | | Merge 8.7 and (hopefully) fix Travis build with C++jan.nijtmans2019-12-131-1/+3
|\ \ \
* | | | Merge tip-548 (since using the *WChar* functions leads to less type-casts in ↵jan.nijtmans2019-08-031-1/+1
|/ / / | | | | | | | | | | | | c++) Als finish implementaton for MacOSX
* | | Fix test for UNICODE in win/tclAppInit.c. No longer check for "tclsh ↵jan.nijtmans2018-10-121-1/+1
| | | | | | | | | | | | install", because tclsh is not compiled with that. Fix minizip.c compilation on systems without open64()
* | | Improvements for zipfs. Document that TclZipfs_AppHook only works on Windows ↵jan.nijtmans2018-09-261-1/+2
| | | | | | | | | | | | | | | in UNICODE mode. Also, remove this from the stub table because it should never be called this way. Prevent a possible crash on win32 startup.
* | | TIP #430 implementationjan.nijtmans2018-09-121-0/+2
|\ \ \
| * | | First pass on the Msys style windows build. Moving Zipfs features closer ↵hypnotoad2017-11-171-0/+2
| | | | | | | | | | | | | | | | together in the autoconf
* | | | merge 8.6sebres2018-08-291-1/+3
|\ \ \ \ | |/ / / |/| / / | |/ /
| * | small amend to [cae24931ed] (no _CRT_glob in both cases __GNUC__ || ↵sebres2018-08-291-5/+3
| | | | | | | | | | | | TCL_BROKEN_MAINARGS).
| * | partially cherry-picking of [5099a81b50], never reached 8.6, so for example ↵sebres2018-08-291-0/+4
| | | | | | | | | | | | build for MINGW breaks tests winpipe-8.1 etc, because "*" will be expanded.
* | | Finish libtommath's constification work. All done (hopefully). And make sure ↵jan.nijtmans2017-09-181-2/+2
|/ / | | | | | | that C preprocessor directives have the '#' as first character on the line.
* | Upon further review, due the order of #include of headers, we do not havedgp2015-03-131-0/+2
| | | | | | | | a history of a #define STRICT active when the #include "windows.h" in tclWinPort.h happens. Don't want to start something new now.
* | Disable command line globbing on MinGW compiles: this saves startup time, ↵jan.nijtmans2013-09-301-0/+1
|\ \ | |/ | | | | since the result of the glob is not used anywhere.
| * Disable command line globbing on MinGW compiles: this saves startup time, ↵jan.nijtmans2013-09-301-0/+1
| | | | | | | | since the result of the glob is not used anywhere.
* | Don't use deprecated stricmp/strnicmp any more, but underscored variant for ↵jan.nijtmans2013-06-141-10/+11
|\ \ | |/ | | | | | | non-GNU compilers. Some formatting - improved comments.
| * get rid of _ANSI_ARGS_jan.nijtmans2012-04-261-2/+2
| |\
* | | Use (preferred) Tcl_ObjSetVar2 in stead of Tcl_SetVar in tclAppInit.c, both ↵jan.nijtmans2013-02-171-3/+7
| | | | | | | | | | | | UNIX and Win.
* | | eliminate unnecessary TEXT() macrosjan.nijtmans2012-09-231-15/+15
| | |
* | | Adjust ckalloc/ckfree macros to greatly reduce number of explicit casts indkf2011-03-121-2/+2
| | | | | | | | | rest of Tcl source code. No ABI change. API change *should* be harmless.
* | | Now that we're no longer using SCM based on RCS, the RCS Keyword linesdgp2011-03-021-2/+0
|\ \ \ | |/ / | | | cause more harm than good. Purged them (except in zlib files).
| * | Now that we're no longer using SCM based on RCS, the RCS Keyword lines causedgp2011-03-021-2/+0
| |\ \ | | |/ | | | more harm than good. Purged them.
| | * Now that we're no longer using SCM based on RCS, the RCS Keyword lines causedgp2011-03-011-2/+0
| | | | | | | | | more harm than good. Purged them.
| | * * generic/tclEvent.c (Tcl_CreateThread): Replaced some calls todgp2007-03-191-2/+2
| | | | | | | | | | | | | | | | | | * generic/tclPkg.c (CheckVersion): Tcl_Alloc() with calls to * unix/tclUnixTime.c (SetTZIfNecessary): ckalloc(), which better * win/tclAppInit.c (setargv): supports memory debugging.
| | * Punt gracefully if exitToken was already destroyed.davygrvy2003-10-141-1/+7
| | |
| | * fix for [Bugs 733221, 733156]Joe Mistachkin2003-05-101-21/+70
| | |
* | | [FRQ 491789]: "setargv() doesn't support a unicode cmdline"nijtmans2010-11-181-43/+51
| | | | | | | | | | | | | | | now implemented for cygwin and mingw32 too. No longer disable tests Tcl_Main-1.4 and 1.6 on Windows, because those now work on all supported platforms.
* | | Bring compilation under mingw-w64 a bit closer to reality (but don't break ↵nijtmans2010-11-171-3/+3
| | | | | | | | | | | | mingw32 build!)
* | | Bring compilation under mingw-w64 a bit closer to realitynijtmans2010-11-161-28/+28
| | |
* | | [FRQ 491789]: "setargv() doesn't support a unicode cmdline" implemented for ↵nijtmans2010-11-041-6/+6
| | | | | | | | | | | | Tcl on MSVC++
* | | Make Tcl_SetPanicProc and Tcl_GetStringResult callable without stubs, just ↵nijtmans2010-09-231-2/+2
| | | | | | | | | | | | as Tcl_SetVar.
* | | Make compilable with -DUNICODE (not actived yet)nijtmans2010-09-231-74/+89
| | | | | | | | | | | | Many clean-ups in comments.
* | | For MINGW/CYGWIN, use GetCommandLineA explicitelynijtmans2010-05-031-2/+2
| | |
* | | * win/tclAppInit.c (Tcl_AppInit): Reverted Jan's change that addedKevin B Kenny2009-11-271-5/+1
| | | | | | | | | | | | | | | | | | | | | a call to Tcl_InitStubs. 'tclsh' and 'tcltest' are providers, not consumers of the Stubs table, and should not link with the Stubs library, but only with the main Tcl library. (In any case, the presence of Tcl_InitStubs broke the build.) [Bug 2902965]
* | | Fix [Bug 2902965] stub related changesnijtmans2009-11-261-2/+6
| | | | | | | | | | | | cause tclkit built to break
* | | * unix/tclAppInit.c: Repair broken build of the tcltest executable.dgp2009-11-191-13/+3
| | | | | | | | | | | | * win/tclAppInit.c: [Bug 2883850, 2900542].
* | | Style fixes (unfouling whitespace, sorting comments, removing useless casts, ↵dkf2009-01-091-2/+2
| | | | | | | | | | | | etc.)
* | | Get rid of pre-C89-isms (esp. CONST vs const).dkf2008-04-271-3/+3
|/ /
* | Complete the purge of K&R function definitions from manually-written code.dkf2007-04-161-7/+9
| |
* | * generic/tclBasic.c (Tcl_CreateMathFunc): Replaced somedgp2007-03-191-2/+2
| | | | | | | | | | | | | | * generic/tclEvent.c (Tcl_CreateThread): calls to Tcl_Alloc() * generic/tclObj.c (UpdateStringOfBignum): with calls to * unix/tclUnixTime.c (SetTZIfNecessary): ckalloc(), which better * win/tclAppInit.c (setargv): supports memory debugging.
* | * win/tclAppInit.c: WIN32 native console signal handler removed.davygrvy2006-01-051-141/+1
| | | | | | | | | | | | | | This was found to be interfering with TWAPI extension one. IMO, special services such as signal handlers should best be done with extensions to the core after discussions on c.l.t. about Roy Terry's tclsh children of a real windows service shell.