summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [ tcl-Feature Requests-2958832 ] Furthernijtmans2010-02-2511-125/+147
| | | | | | speed-up of ouster-hash function. Eliminate various unnecessary (ClientData) type casts.
* Convert many multi-line macros to use the best-practice do-while(0) formdkf2010-02-251-151/+183
|
* Correct silly error with missing semicolonsdkf2010-02-251-3/+3
|
* Fix some nasties with handling duplicate keys in list->dict->list conversions.dkf2010-02-244-7/+34
|
* Reduce ifdef-fery and size of activation record. More variables shared acrossdkf2010-02-242-603/+453
| | | | instructions than before.
* More tidying up (whitespace, spelling, useless parentheses, useless casts)dkf2010-02-2424-427/+465
|
* Pure whitespace changes, changing to follow Engineering Manual style.dkf2010-02-2420-137/+137
|
* [Bug 2957688]: clarified [socket -server] docsdkf2010-02-242-33/+43
|
* correct datenijtmans2010-02-231-1/+1
|
* Fix typos in ChangeLog.andreas_kupries2010-02-221-3/+3
|
* * generic/tclZlib.c (ZlibTransformInput): [Bug 2742041]: Added aandreas_kupries2010-02-222-3/+32
| | | | | | | | | hack to work around the general problem, early EOF recoginition based on the base-chgannel, instead of the data we have ready for reading in the transform. Long-term we need a proper general fix (likely tracking EOF on each level of the channel stack), with attendant complexity. Further: Z_BUF_ERROR can be ignored, and must be when feeding the zlib code with single characters.
* Remove unnecessary EXTERN's, which alreadynijtmans2010-02-2216-6297/+6462
| | | | | | | | are in the global stub table. Use @EXEEXT@ in stead of @EXT_SUFFIX@ Use -DBUILD_tcl in Makefile for CYGWIN Use EXTERN to control CYGWIN exported symbols Remove some unnecessary type casts.
* Added missing quotingdkf2010-02-221-1/+1
|
* Convert literal tabs in strings into \t sequences.dkf2010-02-221-59/+59
|
* Follow-up to Fix [Bug 2954959] expr abs(0.0) is -0.0nijtmans2010-02-218-39/+71
| | | | | | Some more tests, showing that the LONG implementation was not quite correct too, and a fix for that. Some more internal "const" additions
* * tests/regexp.test: Add test cases back ported frommdejong2010-02-212-1/+116
| | | | Jacl regexp work.
* Fix [Bug 2954959] expr abs(0.0) is -0.0nijtmans2010-02-213-6/+21
| | | | and added test cases for it.
* Make [string length] compiler handle more trivial cases.dkf2010-02-202-110/+112
|
* Small changes to align code style with Tcl Engineering Manualdkf2010-02-191-155/+169
|
* Correct compiler/linker flags for threaded builds on OpenBSD.stwo2010-02-193-3/+22
|
* [Bug 2954638]: Correct behaviour of manual page installer. Also added armouringdkf2010-02-192-27/+82
| | | | | to check that assumptions about the initial state are actually valid (e.g., look for existing input file).
* Return to using the classic hash function. Now with *extensive* notes in thedkf2010-02-174-41/+103
| | | | comments about why this function is preferred.
* Fix error in stack depth calculation for [dict update], correct misleadingdkf2010-02-173-7/+16
| | | | comment in description of opcode.
* missing commadgp2010-02-171-2/+2
|
* Change order of various struct members,nijtmans2010-02-162-63/+60
| | | | restoring potential binary incompatibility with Tcl 8.5
* Small fixes, namely:dkf2010-02-161-125/+158
| | | | | | Move some related pieces (e.g., for encoding handling) closer together Update comments to add more visual separation between bits and pieces Put parentheses round negative constants
* regendkf2010-02-161-6314/+6269
|
* Do not assume that all unix systems have the POSIX blkcnt_t type, since OpenBSDdkf2010-02-163-3/+13
| | | | apparently does not.
* Update literal table to use FNV hash function.dkf2010-02-162-82/+93
|
* Eliminate all internal Tcl_WinUtfToTCharnijtmans2010-02-156-27/+28
| | | | | and Tcl_WinTCharToUtf calls, needed for mslu support.
* reverted earlier rename from tcl*Stubs tonijtmans2010-02-1519-145/+372
| | | | | | | | tcl*ConstStubs, it's not necessary at all. tclEnsemble.c: Fix signed-unsigned mismatch make tclWinProcs "const" Add first part of mslu support, See [Feature Request #2819611]
* Fix [Bug 2950259] so that deleting an object by killing its namespace willdkf2010-02-155-12/+133
| | | | reliably call the object's destructor.
* Reduce code nesting in [namespace ensemble] implementation.dkf2010-02-141-139/+116
|
* Corrected a comment.dkf2010-02-141-3/+2
|
* Hive off the ensemble code into its own file.dkf2010-02-1310-3505/+3623
| | | | Split the [switch] compiler for sanity's sake.
* Add .lib, .exp and .res to .cvsignorenijtmans2010-02-132-0/+4
|
* use -pipe for gcc on win32 (mingw/cygwin)nijtmans2010-02-133-20/+9
|
* add tests for explicit backslash zero as argument to list commandmdejong2010-02-122-1/+26
|
* Correct datedkf2010-02-111-1/+1
|
* [Bug 2949740]: Do not try to put a NULL pipeline channel into binary mode.dkf2010-02-113-3/+11
|
* Tcl Bug 2826551 regexp bugs related to -all -line and -start and newlinesdkf2010-02-113-45/+320
|
* [Bug 2949397]: Prevent destructors from running on the two core class objectsdkf2010-02-112-2/+15
| | | | when the whole interpreter is being destroyed.
* Two more hash-iteration-order assumptions bite the dustdkf2010-02-101-3/+3
|
* Remove assumption of ordered results from [info procs]dkf2010-02-101-2/+2
|
* More zlib documentation improvements.dkf2010-02-103-24/+63
|
* Forgot the magic bias values. FNV is wildly magical...dkf2010-02-102-4/+4
|
* Improvements to zlib documentation (and formatting of it).dkf2010-02-103-10/+71
|
* Minor corrections as recommended by Joe English.dkf2010-02-091-4/+4
|
* Corrected case of trap patterndkf2010-02-091-2/+2
|
* Small cleanups of formatting issuesdkf2010-02-091-72/+88
|
generated by cgit v0.12 at 2024-11-09 05:50:55 (GMT)