summaryrefslogtreecommitdiffstats
path: root/generic/tclPathObj.c
Commit message (Collapse)AuthorAgeFilesLines
* Plug memory leak handling circular path values.dgp2018-04-251-8/+6
|
* merge 8.7dgp2017-12-061-3/+2
|\
| * [ce3a211dcb] Failed file normalize when tail is empty string.dgp2017-12-061-3/+2
| |\
| | * [ce3a211dcb] Failed file normalize when tail is empty string.dgp2017-12-061-3/+2
| | |\
| | | * [ce3a211dcb] Failed file normalize when tail is empty string.dgp2017-12-061-3/+2
| | | |
* | | | More TIP 445 conversion of the "path" Tcl_ObjType.dgp2017-09-221-35/+5
| | | |
* | | | Partial conversion of the "path" Tcl_ObjType to TIP 445 contributed bydgp2017-09-221-130/+64
| | | | | | | | | | | | | | | | | | | | Cyan Ogilvie. This reworking also eliminates circular references in the "path" values. Test suite indicates the change works. Don't know what, if anything, was lost compared to the original Darley design.
* | | | merge trunkdgp2017-07-061-33/+31
|\ \ \ \ | |/ / /
| * | | merge 8.6dgp2017-07-061-39/+43
| |\ \ \ | | |/ /
| | * | Pull out of the loop a block of code that can only run in first iteration.dgp2017-07-061-18/+19
| | | |
| | * | Alternative fix for memleaks in fs path join machinery.dgp2017-07-061-17/+16
| | | |
| * | | merge core-8-6-branchjan.nijtmans2017-07-031-32/+26
| |\ \ \
| | * \ \ tclPathObj: fixed TclJoinPath - usage of released object and object leakage.sebres2017-07-031-32/+26
| | |\ \ \ | | | |/ / | | |/| / | | | |/ closes [adb198c256df8c4192838cc3c1112fb2821314e9]
| | | * tclPathObj: fixed TclJoinPath (backported from 8.6) - usage of released ↵sebres2017-07-031-84/+70
| | | | | | | | | | | | | | | | | | | | object and object leakage. closes [adb198c256df8c4192838cc3c1112fb2821314e9]
* | | | merge trunkdgp2017-05-081-1/+18
|\ \ \ \ | |/ / /
| * | | Revert the colorful debug garbage mistakenly committed.dgp2017-05-011-8/+2
| | | |
| * | | [f9fe90d0fa] [file join] normalization. See filesystem-1.52*dgp2017-05-011-3/+19
| |\ \ \ | | |/ /
| | * | [f9fe90d0fa] [file join] normalization. See filesystem-1.52*dgp2017-05-011-1/+11
| | |\ \ | | | |/
| | | * [f9fe90d0fa] [file join] normalization. See filesystem-1.52*dgp2017-05-011-1/+11
| | | |
| * | | [f34cf83dd0] An optimization was being taken in a case where it produced the ↵dgp2017-04-281-0/+7
| |\ \ \ | | |/ / | | | | | | | | wrong result, failing to collapse multiple /// into /. Testing on Windows where path expectations may vary would be a good idea, but since this is just an optimization avoidance, I suspect we're ok.
| | * | [f34cf83dd0] An optimization was being taken in a case where it produced the ↵dgp2017-04-281-0/+7
| | |\ \ | | | |/ | | | | | | | | wrong result, failing to collapse multiple /// into /. Testing on Windows where path expectations may vary would be a good idea, but since this is just an optimization avoidance, I suspect we're ok.
| | | * [f34cf83dd0] An optimization was being taken in a case where it produceddgp2017-04-281-0/+7
| | | | | | | | | | | | | | | | | | | | the wrong result, failing to collapse multiple /// into /. Testing on Windows where path expectations may vary would be a good idea, but since this is just an optimization avoidance, I suspect we're ok.
| * | | In TclGetNumberFromObj() macro (tclExecute.c): Don't fill in type if ↵jan.nijtmans2017-02-031-1/+1
| | | | | | | | | | | | | | | | | | | | TCL_ERROR is returned: The caller doesn't do anything with this. Don't access (non-const) variable tclEmptyStringRep any more, use its value (&tclEmptyString) directly. Only keep it in tclPkg.c, for error checking.
* | | | Reduce direct use of the tclEmptyStringRep.dgp2016-10-121-2/+1
| | | |
* | | | merge trunkdgp2016-07-091-27/+31
|\ \ \ \ | |/ / /
| * | | Merge core-8-6-branch: Bug [a47641a031]. TclJoinPath was calling ↵jan.nijtmans2016-07-081-3/+7
| |\ \ \ | | |/ / | | | | | | | | TclNewFSPathObj with a first argument that was not an absolute path. Added a check for that. Fixes Windows test failures fileSystem-1.{3,4}
| | * | Bug [a47641a031]. TclJoinPath was calling TclNewFSPathObj with aashok2016-07-081-3/+7
| | | | | | | | | | | | | | | | first argument that was not an absolute path. Added a check for that. Fixes Windows test failures fileSystem-1.{3,4}
| * | | Micro-optimization: Use TclGetStringFromObj in stead of Tcl_GetStringFromObj ↵jan.nijtmans2016-07-081-26/+26
| |/ / | | | | | | | | | in many places where possible.
* | | Use static name for a static struct.dgp2016-04-111-19/+19
|/ /
* | [3479689] Plug memory leak due to incomplete bug fix.dgp2014-07-111-12/+6
| |
* | Eliminate all usage of WIN32 and __WIN32__ macros: Some compilers (e.g. ↵jan.nijtmans2014-02-101-3/+3
| | | | | | | | | | Clang/LLVM) don't define it, and _WIN32 is much more portable anyway. See: [http://nadeausoftware.com/articles/2012/01/c_c_tip_how_use_compiler_predefined_macros_detect_operating_system#WindowsCygwinnonPOSIXandMinGW]
* | Eliminate all Tcl_ConvertToType calls and all direct calls to ↵jan.nijtmans2013-02-041-2/+2
| | | | | | | | typePtr->setFromAnyProc (except the call from inside the Tcl_ConvertToType function) from the Tcl core.
* | merge core-8-5-branchjan.nijtmans2013-01-311-2/+2
|\ \ | |/
| * Use twoPtrValue.ptr1 in stead of otherValuePtr everywhere. This is exactly ↵jan.nijtmans2013-01-311-2/+2
| | | | | | | | the same field, but it allows twoPtrValue.ptr2 to be used for other purposes.
* | 3559678 Fix bad filename normalization when the last component is the empty ↵dgp2012-08-201-2/+7
|\ \ | |/ | | | | string.
| * 3559678 Fix bad filename normalization when the last component is the empty ↵dgp2012-08-201-2/+7
| | | | | | | | string.
* | more result generation conversiondkf2012-08-031-9/+7
| |
* | Simplify tclFileSystem.h. Define structs where used.dgp2012-06-281-12/+10
|\ \ | |/
| * Simplify tclFileSystem.h. Define structs where used.dgp2012-06-281-12/+10
| |
* | fix some gcc 64-bit warningsjan.nijtmans2012-06-261-3/+0
|\ \ | |/ | | | | quoting improvements eliminate unused variable
| * fix some gcc 64-bit warningsjan.nijtmans2012-06-261-3/+0
| | | | | | | | quoting improvements eliminate unused variable
* | 3024359 Make sure that the per-thread cache of the list of file systemsdgp2012-06-251-4/+13
|\ \ | |/ | | | | | | currently registered is only updated at times when no active loops are traversing it. Also reduce the amount of epoch storing and checking to where it can make a difference.
| * Simplify bug fix so that active claims on the FilesystemRecord list of a threaddgp2012-06-251-2/+1
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | prevent any overwriting of that per-thread cache. This keeps active traversals of the list valid. The possible downside is that this may result in some delay in noticing new epochs and result in somewhat greater likelihood we will cache things in a "path" value that are out of date. Since the system has to deal with out of date cached data anyway, this should have no correctness affects, measured against the status quo. In multi-threaded operation the possibility of caching and/or retrieving outdated information can never be eliminated. Checkin also includes merge of 8.5.
| * \ merge 8.5dgp2012-06-211-41/+24
| |\ \ | | |/
| * | Only record the filesystemEpoch when it actually marks the validity ofdgp2012-06-211-6/+15
| | | | | | | | | something we are caching.
* | | Stop storing FilesystemRecord in the intrep of a "path". We never use it.dgp2012-06-211-41/+23
|\ \ \ | | |/ | |/| Store the Tcl_Filesystem instead, which is what we actually need.
| * | Stop storing FilesystemRecord in the intrep of a "path". We never use it.dgp2012-06-211-41/+24
| |/ | | | | Store the Tcl_Filesystem instead, which is what we actually need.
| |
| \
*-. \ Remove dead code that complicates fs path values but adds no value.dgp2012-06-201-42/+9
|\ \ \ | | |/
| | * Purge more dead fs path code.dgp2012-06-201-17/+3
| |/
| * Remove dead code that complicates fs path values but adds no value.dgp2012-06-201-26/+7
| |
generated by cgit v0.12 at 2024-11-17 00:26:24 (GMT)