summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Merge 8.6jan.nijtmans2020-09-2814-89/+80
|\
| * Make Tcl compile warning-free using -Wshadowjan.nijtmans2020-09-2814-110/+103
| |
* | Merge 8.6jan.nijtmans2020-09-261-10/+18
|\ \ | |/
| * Merge 8.5jan.nijtmans2020-09-261-6/+14
| |\
| | * Tweak xcode labels for Travis buildjan.nijtmans2020-09-261-5/+5
| | |
| | * Build Travis with Xcode 12 and 12u (Universal Apps)jan.nijtmans2020-09-251-10/+18
| | |
* | | Merge 8.6jan.nijtmans2020-09-256-15/+15
|\ \ \ | |/ /
| * | It's "utf-8", not "utf8" or "UTF8"jan.nijtmans2020-09-257-17/+17
| | |
* | | merge-markjan.nijtmans2020-09-250-0/+0
| | |
* | | Merge 8.6jan.nijtmans2020-09-253-25/+11
|\ \ \ | |/ /
| * | When sourcing script files (even when simulating that through open|read), ↵jan.nijtmans2020-09-253-1/+4
| |\ \ | | |/ | | | | | | | | | always set -eofchar \032 just like the source command does. Possible security issue: this could make it possible to evaluate hidden content at the end of pkgIndex files.
| | * When sourcing script files (even when simulating that through open|read), ↵jan.nijtmans2020-09-252-0/+3
| | | | | | | | | | | | | | | always set -eofchar \032 just like the source command does. Possible security issue: this could make it possible to evaluate hidden content at the end of pkgIndex files.
| * | Merge 8.5jan.nijtmans2020-09-251-23/+9
| |\ \ | | |/
| | * Add gcc-10 build to Travisjan.nijtmans2020-09-241-22/+8
| | |
| | * Let's see if we can recover from an occasional hickup like happened here: ↵jan.nijtmans2020-09-231-1/+1
| | | | | | | | | | | | [https://travis-ci.org/github/tcltk/tcl/jobs/729226819]
* | | When writing script files (like pkgIndex.tcl), always use -translation lf, ↵jan.nijtmans2020-09-224-0/+9
| | | | | | | | | | | | | | | so they don't cause problems on non-windows. When reading script files, always use -eofchar \032, as this might be left by Windows editors.
* | | Merge 8.6jan.nijtmans2020-09-202-7/+11
|\ \ \ | |/ /
| * | Improve TclInvalidateStringRep() macro such that (objPtr) is only evaluated ↵jan.nijtmans2020-09-202-7/+10
| | | | | | | | | | | | once. Addation brackets in TclHasStringRep() macro
* | | Merge 8.6dgp2020-09-201-16/+14
|\ \ \ | |/ /
| * | Silence compiler warning -- fix safety of macro.dgp2020-09-201-1/+1
| | |
| * | Make the check to avoid generating a string representation in [uplevel] a ↵pooryorick2020-09-202-16/+17
| | | | | | | | | | | | little less intrusive.
* | | Merge 8.6jan.nijtmans2020-09-204-7/+6
|\ \ \ | |/ /
| * | Merge 8.5jan.nijtmans2020-09-204-7/+6
| |\ \ | | |/
| | * Backport many (formatting) changes in tools/*. Nothing functional.jan.nijtmans2020-09-2018-189/+194
| | | | | | | | | testest.tcl: Use more uppercase hex.
* | | merge mark (safety)dgp2020-09-190-0/+0
|\ \ \ | |/ /
| * | Fix for [b9ecf3ce98], [uplevel] unnecessarily generates string representation.pooryorick2020-09-192-11/+51
| | |
* | | Fix for [b9ecf3ce98], [uplevel] unnecessarily generates string representation.pooryorick2020-09-192-11/+51
| | |
* | | Eliminate many usages of Tcl_NewObj (-> TclNewObj) and Tcl_NewIntObj (-> ↵jan.nijtmans2020-09-1750-269/+318
| | | | | | | | | | | | TclNewIntObj or Tcl_NewWideIntObj)
* | | Change version number of http package from 2.9.5 -> 2.10.0a1, since it's ↵jan.nijtmans2020-09-1711-24/+24
|\ \ \ | |/ / | | | | | | | | | different from the 2.9.5 version in Tcl 8.6.x. Use "$idx >= 0" check in stead of "$idx > -1" everywhere.
| * | Tcl 8.6 should not be tested with "package prefer latest" any more: All ↵jan.nijtmans2020-09-171-1/+0
| | | | | | | | | | | | included packages are supposed to be stable. Not to be merged to 8.7 (which still contains unstable packages)
* | | Merge 8.6jan.nijtmans2020-09-171-5/+14
|\ \ \ | |/ /
| * | Fix [835c93c000]: TIP #525 only implemented for non-singleproc casejan.nijtmans2020-09-171-5/+14
| |\ \ | | |/
| | * Fix [835c93c000]: TIP #525 only implemented for non-singleproc casejan.nijtmans2020-09-171-8/+17
| | |\
| | | * Proposed solution for [835c93c000]: TIP #525 only implemented for ↵jan.nijtmans2020-09-161-5/+14
| | |/ | |/| | | | | | | non-singleproc case
* | | Little tweak to makeHeader.tcl, not depending on lsearch returning -1 any morejan.nijtmans2020-09-161-1/+1
| | |
* | | one more testdgp2020-09-151-0/+3
|\ \ \
| * | | Add test for [string replace] troubles.dgp2020-09-151-0/+3
| | | |
* | | | Fix [b5777d3d32]: Crash corner-case in TIP #502 implementationjan.nijtmans2020-09-155-18/+62
|\ \ \ \ | |/ / /
| * | | Protect INST_STR_REPLACE toojan.nijtmans2020-09-041-0/+3
| | | |
| * | | Tests string-12.2[45].* and fixes to INST_STRING_RANGE.dgp2020-09-022-2/+24
| | | |
| * | | Fix for test lindex-18.0dgp2020-09-022-16/+21
| | | |
| * | | Test lindex-18.0 demonstrates same issue with INST_LIST_INDEXdgp2020-09-021-0/+8
| | | |
| * | | Different fix where existing protection tools for nesting bytecode executiondgp2020-09-021-0/+3
| | | | | | | | | | | | | | | | calls are used. This solution suggests there may be many more places needing protection. Any routine that gets passed "interp" is a possibility.
| * | | Add test-case util-9.59, which demonstrates bug [b5777d3d32]jan.nijtmans2020-08-251-0/+3
| | | |
* | | | Merge 8.6jan.nijtmans2020-09-143-18/+2
|\ \ \ \ | | |/ / | |/| |
| * | | Unbreak winFCmd-1.38 testcase on Windowsjan.nijtmans2020-09-141-1/+1
| | | |
| * | | Add .bmp to .gitattributes. Simplify .fossil-settings/binary-globjan.nijtmans2020-09-142-7/+11
| | | |
| * | | Eliminate many "register" keywords (which do nothing with modern compilers)jan.nijtmans2020-09-1419-268/+268
| | | | | | | | | | | | Eliminate many unnecessary type-casts to (unsigned)
* | | | Merge 8.6jan.nijtmans2020-09-1412-20/+20
|\ \ \ \ | |/ / /
| * | | Prevent the usage of the term safe/unsafe child. Suggested by Keith Nash. ↵jan.nijtmans2020-09-1416-33/+33
| | | | | | | | | | | | | | | | | | | | Thanks! More usage of $index<0 in stead of $index==-1 consistantly
generated by cgit v0.12 at 2025-01-11 23:30:42 (GMT)