Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
History for generic/tclUtf.c
|
2024-12-12
| ||
| 20:31 | Convert masks to an enum file: [f0b16961ba] check-in: [a5b1b09552] user: dkf branch: main, size: 67784 | |
|
2024-08-29
| ||
| 08:23 | Merge [f6c559b364487a9d]: Space before tab is an extremely unlikely to be correct indentation pattern. file: [0b521f749d] check-in: [8a11a89576] user: pooryorick branch: unchained, size: 68188 | |
|
2024-08-28
| ||
| 16:33 | Merge [3a60363aa10ef993]: Clean up a lot of whitespace issues. file: [eadac966f9] check-in: [627781a6db] user: pooryorick branch: unchained, size: 68188 | |
|
2024-08-27
| ||
| 21:31 | Merge [835b1a5566c28af4]: More whitespace cleanup. file: [e5098407b1] check-in: [fc7d06c8b8] user: pooryorick branch: unchained, size: 68149 | |
|
2024-08-26
| ||
| 11:01 | Merge [9b8a66aff6923a69]: Fix [6811a00819]: lsearch performance degradation on Tcl 8.6.11 release (thanks, Sergey!). file: [2c29aac647] check-in: [9f27ea7880] user: pooryorick branch: unchained, size: 68151 | |
| 09:45 | Merge [1fbff64078d7fb02]: Fix indentation/brace usage style issues. file: [866cfe06c0] check-in: [bd3faf4a67] user: pooryorick branch: unchained, size: 68161 | |
|
2024-08-25
| ||
| 13:22 | Merge [123c58d051c8c1ab]: "merge release", "Remove private characters from regexp control table, but add them back in [:cntrl:] class (so no change in regexp handling). Eliminated (size_t) type-casts. Backported from 8.7", and "Fix [d63061a1ac]: PRIVATE != CONTROL in Unicode". file: [ed0e0b3980] check-in: [23e01ffee0] user: pooryorick branch: unchained, size: 68167 | |
| 12:24 | Merge [ca14b3054d49c1f4]: TIP #652. file: [a6990b72fd] check-in: [cc54f1c099] user: pooryorick branch: unchained, size: 68188 | |
| 08:04 | Merge [aae56979fd434db3]: Optimize Tcl_UniCharIsControl(). Don't worry about range >= U+F0000, that's for TCL_UTF_MAX>3, which is unsupported for 8.6. file: [3c786ed8ad] check-in: [48dd7bab13] user: pooryorick branch: unchained, size: 68963 | |
| 07:33 | Merge [7b36a6fd46bbe54a]: Update the pkga.c/pkgua.c examples, making sure there's no misunderstanding that Tcl_UtfNcmp() expects Tcl_UniChar lengths, not byte lengths file: [4de4e04e34] check-in: [b693580597] user: pooryorick branch: unchained, size: 69062 | |
| 07:30 | Merge [f4d16ada5190a2d1]: Fix [4e38c347a4]: Changed contract for Tcl_UtfN(case)cmp in Tcl 8.7. file: [adf09b4a26] check-in: [3e36f3fe56] user: pooryorick branch: unchained, size: 69052 | |
| 05:36 | Merge [c83fa73b65ca3849]: Fix [abd489a1c]: TclStringCmp() calls functions through pointer to incorrect type. file: [d616d227dd] check-in: [c9ab508174] user: pooryorick branch: unchained, size: 67163 | |
| 05:32 | Merge [18e32ec525fe3202]: Unneeded #undef's. Testcase/comment cleanup. file: [344117b614] check-in: [ef90889292] user: pooryorick branch: unchained, size: 67073 | |
|
2024-08-20
| ||
| 20:04 | Merge [d24c3420fa4add26], TIP #665: Remove support for internal utf-16. file: [d0ebfb90da] check-in: [77d9dffc85] user: pooryorick branch: unchained, size: 67308 | |
|
2024-07-20
| ||
| 15:59 | Merge trunk [91c2f411e7e71552], TIP #636 (tip-636-tcl9-644), resolve all conflicts including those resulting from conflicting Tcl_Obj interface implementations, and relicense under GNU Affero General Public License. All tests pass. file: [30e5b0466c] check-in: [d8a3450f12] user: pooryorick branch: unchained, size: 67307 | |
|
2024-06-12
| ||
| 08:47 | Merge [fab05e2102] (core-8-branch): More Tcl_Size. file: [ada959f162] check-in: [bcf6945c28] user: pooryorick branch: unchained, size: 66937 | |
|
2024-06-04
| ||
| 15:35 | Space before tab is an extremely unlikely to be correct indentation pattern file: [29f12485b0] check-in: [f6c559b364] user: dkf branch: main, size: 67783 | |
|
2024-05-13
| ||
| 15:41 |
Clean up a lot of small whitespace issues
This is the dullest commit ever. Sorry. file: [c09a184b21] check-in: [3a60363aa1] user: dkf branch: main, size: 67817 | |
|
2024-05-10
| ||
| 15:40 | More whitespace cleanup file: [d80dbf635c] check-in: [835b1a5566] user: dkf branch: main, size: 67778 | |
|
2024-04-23
| ||
| 16:06 | Hmm, something is broken. Leaving this here but this commit is definitely wrong somehow file: [3b06de4a78] check-in: [b356268e32] user: dkf branch: tidy-indentation, size: 67824 | |
|
2024-03-20
| ||
| 20:29 | Fix [6811a00819]: lsearch performance degradation on Tcl 8.6.11 release (thanks, Sergey!) file: [5e9e721e61] check-in: [9b8a66aff6] user: jan.nijtmans branch: main, size: 67780 | |
| 20:09 | Fix [6811a00819]: lsearch performance degradation on Tcl 8.6.11 release (thanks, Sergey!) file: [e9392be0f4] check-in: [3186ba81a2] user: jan.nijtmans branch: core-8-branch, size: 76905 | |
| 17:56 | optimize TclUtfToUCS4 for single code units (non high surrogates), especially for ascii; fixes performance regression [6811a0081940b76c] file: [7e5292dabd] check-in: [ad35def80d] user: sebres branch: core-8-6-branch, size: 64804 | |
|
2024-03-19
| ||
| 14:56 | Fix indentation/brace usage style issues file: [f42d134067] check-in: [1fbff64078] user: dkf branch: main, size: 67790 | |
|
2024-03-03
| ||
| 21:48 | Merge 8.7 file: [4eb4e6aa0e] check-in: [123c58d051] user: jan.nijtmans branch: main, size: 67796 | |
| 21:42 | Merge 8.6 file: [398f88309e] check-in: [2490edbb83] user: jan.nijtmans branch: core-8-branch, size: 76938 | |
| 21:30 | Fix [d63061a1ac]: PRIVATE != CONTROL in Unicode file: [2096c6cc8b] check-in: [0480bdc823] user: jan.nijtmans branch: core-8-6-branch, size: 64805 | |
|
2024-02-11
| ||
| 22:05 | TIP #652 file: [d706e0f20e] check-in: [2e1b89ace3] user: jan.nijtmans branch: core-8-branch, size: 76959 | |
|
2024-02-07
| ||
| 14:47 | Implementation of TIP 652. file: [167a5a52d8] check-in: [f76bde9ba9] user: pooryorick branch: tip-652, size: 67817 | |
|
2024-01-30
| ||
| 17:07 | Tweaking indentation of code; really unimportant... file: [902d411682] check-in: [bb72806960] user: dkf branch: dkf-indent-tweak, size: 68630 | |
|
2024-01-22
| ||
| 15:43 | Merge 8.7 file: [b00de11567] check-in: [aae56979fd] user: jan.nijtmans branch: main, size: 68592 | |
|
2024-01-21
| ||
| 16:31 | Merge 8.6 file: [37b7a3481d] check-in: [9ebdb88b20] user: jan.nijtmans branch: core-8-branch, size: 77734 | |
| 16:26 | Optimize Tcl_UniCharIsControl(). Don't worry about range >= U+F0000, that's for TCL_UTF_MAX>3, which is unsupported for 8.6. file: [669f7ea6fc] check-in: [8e673f54d3] user: jan.nijtmans branch: core-8-6-branch, size: 64826 | |
|
2024-01-15
| ||
| 10:46 | Bug [d63061a1ac]: "PRIVATE != CONTROL in Unicode". Leave out "Co" (private-use characters) from "string is control". file: [e11c258454] check-in: [5219e3ff14] user: jan.nijtmans branch: bug-d63061a1ac, size: 77741 | |
|
2024-01-12
| ||
| 13:12 | Merge 8.7 file: [33bb3d3f9c] check-in: [7b36a6fd46] user: jan.nijtmans branch: main, size: 68691 | |
| 11:53 | Leave out Tcl_UtfNcmp/Tcl_UtfNcasecmp from -DTCL_NO_DEPRECATED builds, because it's part of the UTF16 compatibility layer file: [aa6e8b1e91] check-in: [2d72b799ff] user: jan.nijtmans branch: core-8-branch, size: 77833 | |
|
2024-01-10
| ||
| 21:01 | Fix [4e38c347a4] Changed contract for Tcl_UtfN(case)cmp in Tcl 8.7 file: [d1d67541a6] check-in: [45db2932ba] user: jan.nijtmans branch: core-8-branch, size: 77661 | |
| 15:12 | Fix [4e38c347a4]: Changed contract for Tcl_UtfN(case)cmp in Tcl 8.7 file: [7ad9a56545] check-in: [f4d16ada51] user: jan.nijtmans branch: main, size: 68681 | |
| 12:41 | TIP 685 implementation: rename "string is unicode" to "string is transferable". Also rename underlying C function "Tcl_UniCharIsUnicode" to "Tcl_UniCharIsTransferable". file: [74f7693140] check-in: [5018317bb2] user: oehhar branch: tip-685, size: 75746 | |
|
2023-12-29
| ||
| 23:54 | Fix [abd489a1c]: TclStringCmp() calls functions through pointer to incorrect type file: [8bad15be72] check-in: [c83fa73b65] user: jan.nijtmans branch: main, size: 66792 | |
| 14:57 | Unneeded #undef's. Testcase/comment cleanup file: [d2bc395e71] check-in: [18e32ec525] user: jan.nijtmans branch: main, size: 66702 | |
| 12:39 | Merge 8.6 file: [d88d36db90] check-in: [b307c14710] user: jan.nijtmans branch: core-8-branch, size: 75723 | |
|
2023-10-13
| ||
| 13:27 | Proposed fix for [abd489a1c]: TclStringCmp() calls functions through pointer to incorrect type. Modified, swapping the wrapping-order file: [03af0b6919] check-in: [6d9d5ac6a7] user: jan.nijtmans branch: bug-abd489a1c, size: 64925 | |
|
2023-09-12
| ||
| 10:11 | Merge 8.7 file: [c6ac1dcada] check-in: [d24c3420fa] user: jan.nijtmans branch: main, size: 66944 | |
|
2023-05-01
| ||
| 20:28 | More progress file: [80594eb5a9] check-in: [f3015fa249] user: jan.nijtmans branch: tip-665, size: 72531 | |
| 19:42 | Remove internal use of TCL_UTF_MAX=3 as much as possible, without compromizing existing TIP's file: [8fb9ac9c00] check-in: [86503e53c7] user: jan.nijtmans branch: tip-665, size: 72576 | |
|
2023-04-28
| ||
| 17:34 | Remove all code related to TCL_UTF_MAX=3. file: [35eabcd52d] check-in: [d0c75e3b88] user: pooryorick branch: unchained, size: 66945 | |
| 06:43 | Limit memset() to "TCL_UTF_MAX=3' builds. file: [76ecb3cdd4] check-in: [c72b11eac7] user: pooryorick branch: main, size: 71869 | |
|
2023-04-27
| ||
| 20:34 | memset(0xff) instead of memset(0) to accomodate tests that fill buffer with 0xff. file: [0c038d5268] check-in: [1e52f2b6c3] user: pooryorick branch: bug-f5eadcbf9a6b1b4c, size: 71815 | |
|
2023-04-25
| ||
| 20:34 | Fix for issue [f5eadcbf9a], passing pointer to uninitialized memory leads Tcl_UniCharToUtf() to corrupt data. file: [e85972715f] check-in: [db9f715fd5] user: pooryorick branch: bug-f5eadcbf9a6b1b4c, size: 71809 | |
|
2023-04-14
| ||
| 15:01 | Merge trunk file: [b1e70b23d0] check-in: [e36ab1e9ba] user: apnadkarni branch: tip-660, size: 71148 | |
|
2023-04-12
| ||
| 14:25 | Correct spelling errors in comments and documentation, but also a non-comment corrections in history.tcl and tcltest.test. file: [bee4197cdf] check-in: [d65da06a77] user: pooryorick branch: main, size: 71163 | |
| 13:30 | Correct spelling errors in comments and documentation, but also a non-comment corrections in history.tcl and tcltest.test. file: [784c46905b] check-in: [aca8de0aeb] user: pooryorick branch: core-8-branch, size: 75938 | |
| 09:35 | Correct spelling errors in comments and documentation, but also non-comment corrections in history.tcl and tcltest.test. file: [0c3f0782c0] check-in: [ee3df4e647] user: pooryorick branch: core-8-6-branch, size: 63167 | |
|
2023-04-04
| ||
| 20:50 | more progress file: [6e7ba5c478] check-in: [4e36871191] user: jan.nijtmans branch: optional-signed-size, size: 71215 | |
|
2023-04-02
| ||
| 21:49 | Merge 9.0. Add some more utility macro's file: [4954dba7d7] check-in: [4bedba476d] user: jan.nijtmans branch: optional-signed-size, size: 71147 | |
|
2023-03-30
| ||
| 18:01 | TIP 660. No compiler warnings. Tests suite pass on Win and Ubuntu file: [611210fab5] check-in: [eb81a25271] user: apnadkarni branch: tip-660, size: 71148 | |
|
2023-03-22
| ||
| 19:30 | Forgot one line in previous commit, and indenting file: [721bf4662a] check-in: [16b3efee0d] user: jan.nijtmans branch: main, size: 71163 | |
| 18:13 | Let's get in the 'readability' changes from the 'unchained' branch, without the need for all those partial merges. file: [73982f88a3] check-in: [3f420a07da] user: jan.nijtmans branch: main, size: 71044 | |
| 18:00 | Merge 8.7 file: [dbf3095366] check-in: [f4cf3c36b4] user: jan.nijtmans branch: main, size: 71112 | |
| 17:57 | Merge 8.6 file: [edad2ed8f4] check-in: [6063e17e15] user: jan.nijtmans branch: core-8-branch, size: 75938 | |
| 17:50 | Fix [0265750233]: invalid read in cmdAH-4.3.13.C1.solo.utf-8.tcl8 file: [d8c3dab8b5] check-in: [2ffcb8bcf4] user: jan.nijtmans branch: core-8-6-branch, size: 63167 | |
| 17:34 | Fix [0265750233]: invalid read in cmdAH-4.3.13.C1.solo.utf-8.tcl8. file: [0850778eea] check-in: [9f7e05419e] user: jan.nijtmans branch: mistake, size: 63167 | |
| 16:36 | Proposed fix for [0265750233]: invalid read in cmdAH-4.3.13.C1.solo.utf-8.tcl8. file: [63ea89e358] check-in: [4055888f8f] user: jan.nijtmans branch: bug-0265750233, size: 75916 | |
|
2023-03-09
| ||
| 21:50 | Merge 8.7 file: [9e14056004] check-in: [6c92d05c1f] user: jan.nijtmans branch: mistake, size: 71120 | |
| 21:35 |
Additional fix for [f3cb2a32d6]: uninitialized value in format-2.18
This commit causes a lot of breackage .... file: [2cf67fb76e] check-in: [8d939a1c22] user: jan.nijtmans branch: mistake, size: 75928 | |
|
2023-02-01
| ||
| 08:10 | (cherry-pick) Make Tcl_UniCharToUtf more readable and add test to exercise surrogate handling. (test-case was still missing, which cannot be used in Tcl 8.6) file: [bac1ff6f63] check-in: [3b953fea77] user: jan.nijtmans branch: core-8-branch, size: 75898 | |
| 07:32 | (cherry-pick) Make Tcl_UniCharToUtf more readable and add test to exercise surrogate handling. file: [5b2e8be87b] check-in: [f4c704bc57] user: jan.nijtmans branch: core-8-branch, size: 76021 | |
| 07:29 | (Cherry-pick) Make Tcl_UniCharToUtf more readable. file: [0db541f666] check-in: [b4571ae045] user: jan.nijtmans branch: core-8-6-branch, size: 63147 | |
| 00:34 | Merge trunk. file: [019b30b09c] check-in: [e052e25beb] user: pooryorick branch: unchained, size: 71123 | |
|
2023-01-31
| ||
| 23:04 | Make Tcl_UniCharToUtf more readable and add test to exercise surrogate handling. file: [2b4eaf61a3] check-in: [fcdc24c850] user: pooryorick branch: main, size: 71072 | |
| 22:15 | Fix error introduced in [3e5e37f83b058f3d] for Tcl_UniCharToUtf, and add test. file: [4b1119fbfe] check-in: [df3390187d] user: pooryorick branch: py_easier_to_read, size: 71245 | |
|
2023-01-30
| ||
| 19:37 | Update code comments for Tcl_UniCharToUtf(). file: [0803bdaf9e] check-in: [aad66f6a63] user: pooryorick branch: unchained, size: 71205 | |
| 11:59 |
A few more readability changes to Tcl_UniCharToUtf()
jn: Please, don't do that here. Tcl_UniCharToUtf() is shared between 8.6, 8.7 and 9.0. So if you want to make it easier to read, it should be done on all 3 branches. I know you only care about "trunk", but it makes maintenance on 8.6/8.7/9.0 harder than it already is. I don't want to spend time on reviewing such kind of changes, and no-one else is doing it. Thanks for understanding (I hope)! file: [8e2aa1cbcf] check-in: [b8524737fc] user: pooryorick branch: py_easier_to_read, size: 71246 | |
| 11:22 | Make Tcl_UniCharToUtf() a little easier to read. file: [87063883cc] check-in: [3e5e37f83b] user: pooryorick branch: py_easier_to_read, size: 71246 | |
|
2022-11-16
| ||
| 20:52 | one more (Tcl_UniCharToUtf), and adapt documentation file: [5d60e82b79] check-in: [89bee74fab] user: jan.nijtmans branch: main, size: 71201 | |
| 20:39 | Change 5 functions signatures from int -> size_t. Those should have been part of TIP #494 (Thanks, Gustaf!) file: [1b4b9d0cc8] check-in: [2571961f21] user: jan.nijtmans branch: main, size: 71198 | |
|
2022-09-21
| ||
| 18:46 | various break-fix measures file: [69d5c7f407] check-in: [eb68153185] user: bch branch: bch_sign_and_width, size: 71189 | |
| 17:50 | merge [trunk] file: [3402756a9b] check-in: [ab16261020] user: bch branch: bch_sign_and_width, size: 71191 | |
|
2022-07-12
| ||
| 09:22 | Merge 8.7 file: [363f78a471] check-in: [db6106323b] user: jan.nijtmans branch: main, size: 71183 | |
| 09:03 | Merge 8.7. Clean-up tclWinConsole.c the same way file: [191f492913] check-in: [da4d986fd1] user: jan.nijtmans branch: core-8-branch, size: 76027 | |
| 08:20 | Code cleanup (use {} in if/else statemenets) file: [cb8804bd23] check-in: [ba232277ee] user: jan.nijtmans branch: core-8-6-branch, size: 63153 | |
|
2022-04-20
| ||
| 20:24 | merge [trunk] file: [6e83bd1cbf] check-in: [3550542d19] user: bch branch: bch_sign_and_width, size: 71095 | |
|
2022-04-08
| ||
| 09:37 | Merge 9.0 file: [e9652741f4] check-in: [a1d061ebc8] user: jan.nijtmans branch: tip-619, size: 71191 | |
|
2022-04-01
| ||
| 13:12 | Merge 9.0 file: [05030a04f9] check-in: [4ce1a22a7f] user: jan.nijtmans branch: tip-619, size: 71336 | |
| 10:29 | Add UTF-16 versions of Tcl_GetCharLength/Tcl_GetRange/Tcl_GetUniChar to the stub table. Should have been part of TIP #542. Needed for Tk's "glyph_indexing_2" branch file: [163a4e6715] check-in: [9e1c0d7636] user: jan.nijtmans branch: main, size: 71087 | |
|
2022-03-29
| ||
| 22:17 | Merge 9.0. Fix CONTINUATION macro, and testcases file: [b76b499796] check-in: [38c85a4afd] user: jan.nijtmans branch: tip-619, size: 71295 | |
|
2022-03-24
| ||
| 11:34 | Implement PANIC when the UTF16 compatibility layer is used in combination with -DTCL_NO_DEPRECATED file: [bb16281435] check-in: [c9d9aaf618] user: jan.nijtmans branch: full-utf-for-87, size: 76035 | |
| 10:48 | When compiled with TCL_NO_DEPRECATED, remove the UTF16 compatibility layer. So, we make sure that it is never used internally for the Core. This means that extensions using the compatibility layer won't work any more in this mode; extensions should be compiled using TCL_UTF_MAX=4 then they work again. file: [220088a79b] check-in: [a68880dccd] user: jan.nijtmans branch: full-utf-for-87, size: 75936 | |
|
2022-03-23
| ||
| 14:32 | Fix Tcl_UniCharAtIndex() for UTF-16 compabitility layer file: [296a4ff68d] check-in: [01f2ee09b4] user: jan.nijtmans branch: full-utf-for-87, size: 75870 | |
|
2022-03-22
| ||
| 23:18 | Simplyfy Tcl_UtfAtIndex file: [83af4628be] check-in: [d395ff5ce6] user: jan.nijtmans branch: tip-622, size: 71046 | |
| 15:38 | Feature-complete file: [a481b61bb7] check-in: [d2ae10faca] user: jan.nijtmans branch: full-utf-for-87, size: 75894 | |
| 11:25 | More progress file: [1964d095c7] check-in: [8a7b81816e] user: jan.nijtmans branch: full-utf-for-87, size: 74396 | |
| 08:16 | Add UTF-16 versions of Tcl_UniCharLength/Tcl_NumUtfChars/Tcl_UtfAtIndex. Needed for Tk's glyph_indexing_2, and possibly other extensions sticking at TCL_UTF_MAX=3 file: [f7f920bff9] check-in: [d9384cad48] user: jan.nijtmans branch: tip-622, size: 71343 | |
|
2022-03-16
| ||
| 23:08 | Handle Tcl_UtfAtIndex file: [3e516c1808] check-in: [4d35f48f7c] user: jan.nijtmans branch: full-utf-for-87, size: 74374 | |
|
2022-03-14
| ||
| 16:06 | More progress file: [1faf708527] check-in: [48ab8b7472] user: jan.nijtmans branch: full-utf-for-87, size: 73884 | |
|
2022-03-11
| ||
| 22:43 | 2 more functions file: [c549181e96] check-in: [f3580d9f09] user: jan.nijtmans branch: full-utf-for-87, size: 73755 | |
| 22:11 | Handle TclUniCharNcmp() file: [c920f06a63] check-in: [715b107f62] user: jan.nijtmans branch: full-utf-for-87, size: 69809 | |
|
2022-03-08
| ||
| 22:26 | initial work at using signed-storage for cases where values can actually be signed. file: [3b51e7483c] check-in: [5030ee02cf] user: bch branch: bch_sign_and_width, size: 69392 | |
|
2022-03-03
| ||
| 13:05 | TIP #619 implementation. tests not working yet file: [14dbaef4ca] check-in: [e791d2994c] user: jan.nijtmans branch: tip-619, size: 69633 | |
|
2022-02-24
| ||
| 22:20 | 3 more files with TCL_UTF_MAX checks file: [2a18863905] check-in: [70eb0efb69] user: jan.nijtmans branch: core-8-branch, size: 69151 | |
| 21:44 | Merge 8.7 file: [2ff9e1f3ce] check-in: [0907ba571c] user: jan.nijtmans branch: main, size: 69384 | |
|
2022-02-03
| ||
| 13:13 | TIP #617: Tcl_WCharLen/Tcl_Char16Len file: [5fc32a6ed7] check-in: [2b0167c4a2] user: jan.nijtmans branch: tip-617, size: 69159 | |
|
2021-04-30
| ||
| 08:49 | Merge 8.7. Remove "string bytelength" completely. Also fix some TIP #595 leftover testcases, which were skipped file: [8f2b0ccaa6] check-in: [ef90dfbe67] user: jan.nijtmans branch: main, size: 68823 | |
|
2021-03-17
| ||
| 12:04 | Merge 8.7 file: [95bc7c89a5] check-in: [0d01d35ca0] user: jan.nijtmans branch: tip-597, size: 68595 | |
|
2021-03-15
| ||
| 12:31 | Merge 8.7 (this is the TIP #575 implementation for Tcl 9.0) file: [87a9e182f7] check-in: [3fed9809e1] user: jan.nijtmans branch: main, size: 68048 | |
| 11:52 | Implement TIP #575: Switchable Tcl_UtfCharComplete()/Tcl_UtfNext()/Tcl_UtfPrev() file: [f559d46094] check-in: [4abf5fb992] user: jan.nijtmans branch: core-8-branch, size: 67820 | |
|
2021-03-11
| ||
| 12:19 | Backport Tcl_UtfCharComplete() functionality from 8.6 for TCL_UTF_MAX>3. This makes Tcl_UtfCharComplete() usable to protect Tcl_UtfNext() calls for overflow. No change for TCL_UTF_MAX=3 (default build) file: [0225383322] check-in: [0f1dccacba] user: jan.nijtmans branch: core-8-5-branch, size: 55274 | |
|
2021-03-10
| ||
| 16:17 | Merge 8.7 file: [2ecfd1c249] check-in: [3c64a79d6a] user: jan.nijtmans branch: main, size: 68564 | |
| 16:16 | Merge 8.6 file: [309a58a3e8] check-in: [9491870854] user: jan.nijtmans branch: core-8-branch, size: 68336 | |
| 16:12 | Repair Tcl_UniCharNcasecmp() in the same way as Tcl_UniCharNcmp() for fix [4c591fa487]. Also put back minor optimization for big-endian machines removed in the previous commit file: [7d9aff2106] check-in: [1b0eda88cc] user: jan.nijtmans branch: core-8-6-branch, size: 63157 | |
| 15:48 | Merge 8.7 file: [90ae9091cd] check-in: [62a4501023] user: jan.nijtmans branch: main, size: 68085 | |
| 15:47 | Merge 8.6 file: [dbc23d8625] check-in: [bcaebc9bb7] user: jan.nijtmans branch: core-8-branch, size: 67857 | |
| 15:39 | Fix [4c591fa487]: [string compare] EIAS violation file: [9f093c66fb] check-in: [6a5a5e21f1] user: jan.nijtmans branch: core-8-6-branch, size: 62676 | |
| 12:55 | TIP #597 implementation: "string is unicode" and new wtf-8 encoding file: [37c46c7c61] check-in: [fc3656894a] user: jan.nijtmans branch: tip-597, size: 68625 | |
|
2021-03-03
| ||
| 14:31 | Backport improvements in UTF-8 handling for Tcl_UtfPrev/Tcl_UtfNext from 8.7 (through 8.6). No change for TCL_UTF_MAX=3. Adapt test-cases accordingly file: [25aa4d36c7] check-in: [4c185260b8] user: jan.nijtmans branch: core-8-5-branch, size: 54494 | |
|
2021-03-02
| ||
| 11:02 | Merge 8.7 file: [bec813293d] check-in: [981b059a3f] user: jan.nijtmans branch: main, size: 68078 | |
| 10:53 | Using 0xFC00 is more readable here than ~0x3FF. It's sufficient becauwe ch1 and ch2 are only 16-bit. Backported from 8.7 file: [1a3a67aab3] check-in: [6064dfa0b2] user: jan.nijtmans branch: core-8-6-branch, size: 62668 | |
| 10:38 | Merge 8.6 file: [e34c706dc3] check-in: [b421cf94ac] user: jan.nijtmans branch: core-8-branch, size: 67850 | |
| 10:10 | Backport some UTF-8-related changed from 8.7 to 8.6, only for TCL_UTF_MAX > 3. No change for TCL_UTF_MAX=3. Also adapt test-cases accordingly, and add comments why the changes were done. file: [85fb86571b] check-in: [e0cba87ba8] user: jan.nijtmans branch: core-8-6-branch, size: 62668 | |
|
2021-02-16
| ||
| 08:10 | Fix Tcl_UtfPrev for TCL_UTF_MAX>3, so it can jump back over Emoji. Backported from 8.7, no change for TCL_UTF_MAX=3. This way, the previous fix can be slightly more simplified, and working for TCL_UTF_MAX>3 too. file: [b7dbe89f73] check-in: [5437791423] user: jan.nijtmans branch: core-8-6-branch, size: 62083 | |
|
2020-12-13
| ||
| 16:59 | Merge 8.7 file: [d677f79a41] check-in: [0631cdce81] user: jan.nijtmans branch: tip-575, size: 67079 | |
|
2020-12-08
| ||
| 15:42 | Merge 8.7 file: [c8ad303c5a] check-in: [71493b571b] user: jan.nijtmans branch: main, size: 67689 | |
| 15:31 | Add -finput-charset=UTF-8 and -fextended-identifiers to gcc (and clang). All C sources can now use UTF-8, as far as gcc/clang/msvc support it. Not used yet file: [b10b993d1b] check-in: [4254aa305b] user: jan.nijtmans branch: core-8-branch, size: 67461 | |
|
2020-11-18
| ||
| 14:29 | Merge 8.7 file: [1ddca5273e] check-in: [0e52f34ca2] user: jan.nijtmans branch: main, size: 67690 | |
| 13:51 | More usage of TclUtfToUCS4/TclUniCharToUCS4 in stead of it's UniChar variants: This handles surrogate pairs better. file: [be48f7aa0c] check-in: [d843858583] user: jan.nijtmans branch: core-8-branch, size: 67462 | |
|
2020-11-05
| ||
| 17:06 | Merge 8.7 file: [4714ced6aa] check-in: [23c0c45dd0] user: jan.nijtmans branch: tip-575, size: 67080 | |
|
2020-06-04
| ||
| 15:15 | Merge 8.7. Use more TCL_INDEX_NONE in documentation/headers/code. file: [5acf707511] check-in: [baad702302] user: jan.nijtmans branch: trunk, size: 67436 | |
|
2020-05-25
| ||
| 11:53 | Finish implementation of "string nextchar|nextword|prevchar|prevword". Not thourougly test yet, but seems OK at first sight. file: [aebc1a920f] check-in: [0a46907d56] user: jan.nijtmans branch: tip-575, size: 67687 | |
| 09:46 | Merge 8.7 file: [a394c3ef5f] check-in: [383de70ed5] user: jan.nijtmans branch: trunk, size: 67442 | |
| 09:32 | Fix compiled "string is <class>" for characters > U+FFFF. Add testcase exposing this bug. file: [1279a196e3] check-in: [2a3709ca18] user: jan.nijtmans branch: core-8-branch, size: 67208 | |
| 09:02 | Fix compiled "string is <class>" for TCL_UTF_MAX=4 build, for characters > U+FFFF. file: [935c0f05f5] check-in: [1eec2e52c3] user: jan.nijtmans branch: core-8-6-branch, size: 62059 | |
|
2020-05-22
| ||
| 21:28 | Split more "string" functions. New helper function TclUniCharToUCS4(), not used yet but that's the next step. file: [1c57261714] check-in: [ec03d6f62d] user: jan.nijtmans branch: tip-575, size: 67433 | |
| 14:36 | Merge 8.7 Add function Tcl_UniCharFold(). It's the same as Tcl_UniCharToLower() for now, but that will change. file: [9b9a0643d1] check-in: [c338858460] user: jan.nijtmans branch: tip-575, size: 67050 | |
|
2020-05-20
| ||
| 19:12 | Merge 8.7 file: [d5326119f4] check-in: [4d5fc5a4f3] user: jan.nijtmans branch: trunk, size: 67059 | |
| 19:09 | Adapt some comments, which are not correct for Tcl 8.7 any more file: [daab2dad3d] check-in: [1203d7b979] user: jan.nijtmans branch: core-8-branch, size: 66825 | |
|
2020-05-18
| ||
| 20:23 | Tiny fix for TCL_UTF_MAX=4 build only: Since Tcl_UtfNext() verifies 4 bytes for lead bytes F0-F5, Tcl_UtfCharComplete() should guarantee that those 4 bytes are available, not 3. file: [2fd4a1f9d9] check-in: [f299c3c546] user: jan.nijtmans branch: core-8-6-branch, size: 61648 | |
| 11:51 | Merge 8.7 file: [989f7f2b9d] check-in: [c933eb1f2d] user: jan.nijtmans branch: tip-573, size: 62334 | |
| 10:10 | Adapt Tcl_UtfPrev()/Tcl_UtfNext() to be consistant with Tcl_UtfToUniChar() file: [7ae9dd89ae] check-in: [6aa676e6bc] user: jan.nijtmans branch: tip-573, size: 61033 | |
|
2020-05-13
| ||
| 22:07 | Merge 8.7. Further progress with TIP implementation. file: [5f36d4015f] check-in: [afe207f116] user: jan.nijtmans branch: tip-575, size: 66452 | |
| 16:51 | Fix [ed29806baf]: Tcl_UtfToUniChar reads more than TCL_UTF_MAX bytes file: [163cc43ab2] check-in: [2a7beb9d18] user: jan.nijtmans branch: core-8-6-branch, size: 61648 | |
|
2020-05-12
| ||
| 21:22 | Merge 8.7 file: [0f33a04699] check-in: [ee9f526818] user: jan.nijtmans branch: trunk, size: 67067 | |
| 21:17 | Little tweak to Tcl_UniCharAtIndex(): Protect against negative index, return -1 in that case. file: [c0daf522a8] check-in: [fba6de9b22] user: jan.nijtmans branch: core-8-branch, size: 66833 | |
| 19:47 | Merge 8.7 file: [9cc185dddb] check-in: [d513c9b2fd] user: jan.nijtmans branch: trunk, size: 67059 | |
| 19:41 | Merge 8.6 file: [5fc434c5c5] check-in: [1d587617b0] user: jan.nijtmans branch: core-8-branch, size: 66794 | |
| 19:20 | Revert implementation of Tcl_UniCharAtIndex() change done in this commit: [6596c4af31e29b5d]. Just look at the Tcl_UtfAtIndex() implementation for TCL_UTF_MAX=4: It's not the same. There are no test-cases for Tcl_UniCharAtIndex(), see [f45d0dc1a7], not really worth to write one, since the implementation of this function didn't change in 20 years. file: [c08942135d] check-in: [0418b67de2] user: jan.nijtmans branch: core-8-6-branch, size: 62224 | |
| 18:59 | First, experimental implementation of TIP #575. Barely tested, will fail. WIP file: [e28fe4a740] check-in: [d15b9cb99e] user: jan.nijtmans branch: tip-575, size: 66518 | |
| 11:08 | Merge testcase cleanup. Make Tcl_UtfPrev() behave the same for any TCL_UTF_MAX value, since we didn't figure out yet how it should behave for TCL_UTF_MAX>3. file: [c2b8a025c1] check-in: [b5d4bf440b] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 61613 | |
| 07:26 | Fix "knownBug" utf-4.11. Turns out a few other testcases where still not correct, now they are. Make next/prev behavior the same for all TCL_UTF_MAX values, since the exact behavior for TCL_UTF_MAX>3 should be worked out further for Tcl 8.7 first, then everything agreed upon can be backported. file: [6bc82beea3] check-in: [8aa5fcb56e] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 61680 | |
|
2020-05-11
| ||
| 12:25 | Merge 8.6. Mark testcase utf-4.11 as "knownBug": this one still doesn't give the right answer. Add testcase 4.14 with similar corner-case, this one is OK. file: [bff485763b] check-in: [aeb46f9ed0] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 62194 | |
| 11:26 | Merge 8.7 file: [6fd45d6640] check-in: [89e2216acf] user: jan.nijtmans branch: trunk, size: 66804 | |
| 10:18 | Merge 8.6 file: [0fec4430b8] check-in: [097c064eb4] user: jan.nijtmans branch: core-8-branch, size: 66580 | |
| 10:03 | Tweak the Tcl_UtfPrev() implementation for TCL_UTF_MAX=4. This fixes 10 testcases in 4 groups (utf-7.10, utf-7.15, utf-7.40 and utf-7.48) , where Tcl_UtfPrev() didn't jump to the beginning of the UTF-8 character, even though there was no limitation which prevented that. So, this is actually a bug-fix for the TIP #389 implementation. file: [a4178ad1d0] check-in: [ae6c7e8b86] user: jan.nijtmans branch: core-8-6-branch, size: 62210 | |
| 07:42 | Merge 8.7 file: [3ceb6a9145] check-in: [fe9bc8500a] user: jan.nijtmans branch: trunk, size: 66748 | |
| 07:41 | Merge 8.6 file: [9fed91ce86] check-in: [6a7f9c3f67] user: jan.nijtmans branch: core-8-branch, size: 66524 | |
| 07:39 | occurance -> occurrence. file: [0834b8cf55] check-in: [1f06b263bc] user: jan.nijtmans branch: core-8-6-branch, size: 62212 | |
|
2020-05-10
| ||
| 20:58 | Demonstration for documentation bug, and suggestion for improved wording. More explanation will follow in the ticket. file: [194cf0d34f] check-in: [94b8ef9338] user: jan.nijtmans branch: bug-81242a48c8, size: 66293 | |
| 20:15 | Merge 8.7 file: [1b799deb71] check-in: [80b28f70b4] user: jan.nijtmans branch: trunk, size: 66747 | |
| 20:11 | Merge 8.6 file: [7f66271ea1] check-in: [2cff341b8c] user: jan.nijtmans branch: core-8-branch, size: 66523 | |
| 19:28 | Tweak Invalid() function: No need for "return 0" twice in the function. For start bytes F0-F4, case TCL_UTF_MAX=4, Tcl_UtfToUniChar() reads 3 bytes but only advances 1 byte. So Tcl_UtfCharComplete() must make sure 3 bytes are available, not 1. Adapt Tcl_UtfCharComplete() accordingly. No change for TCL_UTF_MAX=[3|6] file: [c5862ed2d3] check-in: [1c924d98e0] user: jan.nijtmans branch: core-8-6-branch, size: 62211 | |
|
2020-05-08
| ||
| 15:19 | Rebase to latest core-8-6-branch. file: [1102147264] check-in: [03e4d0a22a] user: jan.nijtmans branch: bug-31aa44375d, size: 61282 | |
|
2020-05-07
| ||
| 22:06 | merge 8.7 file: [9b16395cee] check-in: [b7293ef281] user: dgp branch: trunk, size: 66759 | |
| 21:59 | Merge 8.6 file: [5f74755db2] check-in: [6f4a6b90ef] user: dgp branch: core-8-branch, size: 66535 | |
| 21:47 | Merge changes from parent branch file: [467388c349] check-in: [2742b2b00d] user: dgp branch: bug-31aa44375d, size: 61305 | |
| 21:28 | Merge 8.6 file: [d282366dbc] check-in: [c59e41ca4e] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 61589 | |
| 20:44 | merge 8.5 file: [f2c0d97c09] check-in: [46d4cccefe] user: dgp branch: core-8-6-branch, size: 61720 | |
| 20:31 | Same trouble with Tcl_UtfToUniCharDstring. Test and fix. file: [6406b6da69] check-in: [1a4edbc67e] user: dgp branch: core-8-5-branch, size: 54068 | |
| 19:22 | merge 8.5 file: [a7f7399b26] check-in: [488d1d5841] user: dgp branch: core-8-6-branch, size: 61723 | |
| 19:08 | Fix. Note that just because we get one positive detection of an incomplete character, we cannot conclude that the next byte also will be, or can by taken as a single byte. At least we cannot when TCL_UTF_MAX > 3 so that we have room for valid two-byte sequences after incomplete sequence detection. No need for conditional code, just use an algorithm that always works. file: [633872436b] check-in: [899e66a3c0] user: dgp branch: bug-b2816a3afe, size: 54075 | |
| 18:23 | New approach to fixing the regression reported in [31aa44375d] builds on recent reforms. Older efforts aborted. file: [eeafc020ea] check-in: [ad31bd7310] user: dgp branch: bug-31aa44375d, size: 61319 | |
| 14:15 | For TCL_UTF_MAX==4: Make sure that Tcl_UtfNext()/Tcl_UtfPrev() never move more than 3 bytes. This is more consistant with what Tcl 8.7 does too. For TCL_UTF_MAX==6: Make sure that Tcl_UtfNext()/Tcl_UtfPrev() never move more than 4 bytes. For TCL_UTF_MAX==3: No change. Introduce ucs2_utf16 test constraint, since many test results now become the same for ucs2 and utf16. file: [4c78350590] check-in: [116d4a8943] user: jan.nijtmans branch: core-8-6-branch, size: 61728 | |
| 13:37 | Merge 8.6. Some more tweaks to Tcl_UtfPrev(), so it cannot jump back 4 bytes in "utf16" build any more. file: [9d407eb301] check-in: [6f4bdb20da] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 61597 | |
| 11:09 | Merge 8.7 file: [32dd82e7e4] check-in: [3516a881cf] user: jan.nijtmans branch: trunk, size: 66773 | |
| 10:56 | Merge 8.6 file: [80c3a3b4a5] check-in: [2de70b5bd1] user: jan.nijtmans branch: core-8-branch, size: 66549 | |
| 10:09 | Optimize Tcl_UtfToUniCharDString() file: [73e3d5ab08] check-in: [806e1e868c] user: jan.nijtmans branch: core-8-6-branch, size: 61725 | |
| 09:31 | Tighten optimization in Tcl_UtfToUniCharDString(), just as in Tcl_NumUtfChars(). Don't use "-1" in the Tcl_NumUtfChars() calculation, since that raises more questions than it solves, but that's easy to be remedied as well: Juse use >= in stead of > in the comparation. Great idea, Don! Backport more code formatting from Tcl 8.6 (e.g. use of CONST, which makes no sense any more in c-files) file: [d9de658c5f] check-in: [49fb3b2f1a] user: jan.nijtmans branch: core-8-5-branch, size: 54086 | |
|
2020-05-06
| ||
| 21:59 | merge 8.7 file: [547abbcb04] check-in: [f21991ce1e] user: dgp branch: trunk, size: 66348 | |
| 21:52 | merge 8.6 file: [8ca0195fe0] check-in: [4d08cde908] user: dgp branch: core-8-branch, size: 66124 | |
| 21:42 | merge 8.5 file: [9aa5c57149] check-in: [62362d0caa] user: dgp branch: core-8-6-branch, size: 61525 | |
| 21:08 | Tighten optimization in Tcl_NumUtfChars. Explain in comments. file: [6ec6093330] check-in: [dabb52db36] user: dgp branch: core-8-5-branch, size: 54053 | |
| 19:55 | merge 8.7 file: [f36b3644ff] check-in: [541cffe991] user: dgp branch: trunk, size: 65736 | |
| 19:48 | merge 8.6 file: [85ae6af27c] check-in: [bf737b27ba] user: dgp branch: core-8-branch, size: 65547 | |
| 19:31 | merge 8.5 file: [cafc8c7a54] check-in: [01956c0799] user: dgp branch: core-8-6-branch, size: 60948 | |
| 19:22 | Restore safe calls of Invalid(). file: [fd1606ce1a] check-in: [8d0f9fd43b] user: dgp branch: core-8-5-branch, size: 53477 | |
| 16:58 | The routine Invalid() has been revised to do something different. Update the comments to describe what it does now, and cautions that callers take into account. file: [2f13e3001c] check-in: [1835c80d8f] user: dgp branch: core-8-5-branch, size: 53144 | |
| 13:31 | Merge 8.7 file: [b4ca240e7a] check-in: [c990c1d146] user: jan.nijtmans branch: trunk, size: 64980 | |
| 13:22 | Merge 8.6 file: [0103a8bc49] check-in: [918cfd8094] user: jan.nijtmans branch: core-8-branch, size: 64791 | |
| 13:14 | Merge 8.5. More usage of UCHAR() macro. file: [828ebbe4fc] check-in: [3f3c0fda44] user: jan.nijtmans branch: core-8-6-branch, size: 60192 | |
| 13:03 |
Change Invalid() parameter type to "const char *". Also call Invalid() first in Tcl_UtfNext(), so if src[1] is invalid src[2] doesn't need to be checked any more.
Note: This order change, calling Invalid() first was wrong, and is corrected in later commits. Thanks, Don, for noticing this! file: [2ffd834b54] check-in: [31c95595b2] user: jan.nijtmans branch: core-8-5-branch, size: 52721 | |
| 09:44 | Merge 8.7 file: [ee720bf978] check-in: [97679b4f1d] user: jan.nijtmans branch: tip-573, size: 64914 | |
|
2020-05-05
| ||
| 16:00 | More usage of TclUtfToUCS4(), so we can use the whole Unicode range better in TCL_UTF_MAX>3 builds. file: [ccbd485d3f] check-in: [26e57ca148] user: jan.nijtmans branch: core-8-6-branch, size: 60187 | |
| 13:23 | Add 4 test-cases that could fool Tcl_UtfPrev (but ... actually they don't). Make sure that Tcl_UtfPrev() never reads more than 3 trail bytes (or 4 when TCL_UTF_MAX > 4). Those are the same limits as for Tcl_UtfNext() and Tcl_UtfToUniChar() file: [4f3603d99e] check-in: [50e98246d7] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 60100 | |
| 11:54 | Merge 8.7 file: [c5ed1ee11f] check-in: [95e9950dad] user: jan.nijtmans branch: trunk, size: 64975 | |
| 11:44 | Fix Tcl_UtfPrev() such that it can never go back more than TCL_UTF_MAX bytes. Already done correctly on core-8-6-branch, but this was never forwarded to core-8-branch. file: [9315ba8c93] check-in: [21adba4503] user: jan.nijtmans branch: core-8-branch, size: 64786 | |
| 10:06 | Merge 8.6 file: [97b7eecfe0] check-in: [5d3edce6f4] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 60084 | |
| 09:51 | Merge 8.7 file: [e11eca8c98] check-in: [bd1dcc78c1] user: jan.nijtmans branch: trunk, size: 64882 | |
| 08:16 | Merge 8.6 file: [de2ccc2bf4] check-in: [7d127f6d27] user: jan.nijtmans branch: core-8-branch, size: 64693 | |
| 07:39 | Merge 8.5 file: [98970568cf] check-in: [822925b9b4] user: jan.nijtmans branch: core-8-6-branch, size: 60215 | |
| 07:29 | Properly protect "Invalid" function against lead bytes 0x80-0xBF. This fixes "knownBug" testcase utf-6.93.1. Rename tip389 selector to utf16, since that's what it actually is, in contrast to ucs2 and ucs4. file: [993152b84c] check-in: [b0b773f640] user: jan.nijtmans branch: core-8-5-branch, size: 52716 | |
|
2020-05-04
| ||
| 14:17 | More progress/simplification file: [e972014e95] check-in: [a17d905779] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 60105 | |
| 13:01 | Merge 8.6 file: [60207ec948] check-in: [1256ace951] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 60870 | |
| 12:31 | New internal function TclGetUCS4() only available when TCL_UTF_MAX=4. This fixes all "knownBug" testcases related to tip389. file: [2283dbc19a] check-in: [41517f0841] user: jan.nijtmans branch: core-8-6-branch, size: 60226 | |
| 11:08 | Merge 8.6 file: [05701644db] check-in: [76cc0911f3] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 60847 | |
| 08:35 | (partial) fix for [9d0cb35bb2]: Various issues with core-8-6-branch, TCL_UTF_MAX=4. (even though TCL_UTF_MAX=4 is unsupported, it would be nice to make it work) Marked various test-cases as "knownBug", those work correctly in core-8-branch (8.7). The fix there could be backported. Low prio. file: [fc6d7ce227] check-in: [af513d6a16] user: jan.nijtmans branch: core-8-6-branch, size: 60073 | |
|
2020-05-03
| ||
| 22:27 | Merge 8.6 file: [ab4526735f] check-in: [0de2fe18d1] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 60805 | |
| 22:16 | Re-join utf-6.93.0 and utf-6.93.1 (please disregard comment in previous commit, it was not correct). Perfectionalize TclUtfToUCS4()/TclUCS4Complete() and new (internal) function TclUCS4ToUtf(). They can help preventing bugs regarding splitting/joining surrogates. Used them in a few more places. file: [aa796ca523] check-in: [161196f054] user: jan.nijtmans branch: core-8-6-branch, size: 60121 | |
|
2020-05-02
| ||
| 22:48 | Join test-cases utf-6.93.0 and utf-6.93.1, which MUST give the same answer always for whatever testConstraints. Fix one invalid use of TclUCS4Complete(), and let TclUtfToUCS4() handle (invalid) 4-byte sequences. Test-case cleanup (removal of unnecessary quoting) file: [b74d0eff81] check-in: [e145652a1f] user: jan.nijtmans branch: core-8-6-branch, size: 58821 | |
| 21:54 | Seems almost correct. Still problem with "string index" for TCL_UTF_MAX>3 file: [21368c5c14] check-in: [20993bd6c0] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 59505 | |
| 10:15 | More fixes for [ed29806baf]. Not working yet. WIP file: [09e78bf6c7] check-in: [f670d1a41f] user: jan.nijtmans branch: bug-ed29806baf-8.6, size: 59677 | |
|
2020-05-01
| ||
| 15:29 | Rebase to core-8-6-branch (I don't think it's solved yet, but let's see) file: [fa452f53e1] check-in: [04631bfc72] user: jan.nijtmans branch: bug-ed29806baf, size: 58569 | |
| 14:44 | Merge 8.7 file: [e1d0eb44d9] check-in: [056c1aad28] user: jan.nijtmans branch: trunk, size: 64893 | |
| 14:20 | Fix first part of [ed29806baf]: Tcl_UtfToUniChar reads more than TCL_UTF_MAX bytes. Tcl_UtfToUniChar() now never reads more than TCL_UTF_MAX bytes any more. Since the UtfToUtf encoder/decoder now uses TclUtfToUCS4() it doesn't join 2 surrogates as 2 x 3-byte sequences any more. Actually, it shouldn't, because such sequences are invalid UTF-8. Therefore, added the ucs2 constraint to testcase encoding-15.4. Let's see how TIP #573 goes, this TIP should make this change official. Other callers of Tcl_UtfToUniChar() needs to be revised for the same problem. Most callers will need to change Tcl_UtfToUniChar() -> TclUtfToUCS4() and Tcl_UtfCharComplete() -> TclUCS4Complete(), but that's not done yet. file: [016c050892] check-in: [1d9487bc7e] user: jan.nijtmans branch: core-8-branch, size: 64704 | |
| 13:38 | Fix first part of [ed29806baf]: Tcl_UtfToUniChar reads more than TCL_UTF_MAX bytes. Tcl_UtfToUniChar() now never reads more than TCL_UTF_MAX bytes any more. The UtfToUtf encoder/decoder is adapted to do attitional checks (more tricky than in Tcl 8.7, since we want compatibility with earlier 8.6 releases). Other callers of Tcl_UtfToUniChar() needs to be revised for the same problem. Most callers will need to change Tcl_UtfToUniChar() -> TclUtfToUCS4() and Tcl_UtfCharComplete() -> TclUCS4Complete(), but that's not done yet. file: [f5fdfb6efc] check-in: [5f2bb912cf] user: jan.nijtmans branch: core-8-6-branch, size: 58821 | |
|
2020-04-30
| ||
| 21:23 | merge 8.7 file: [3f6b7394c0] check-in: [25e735d05d] user: dgp branch: bug-e617e8a71a, size: 68058 | |
| 19:25 | merge 8.7 file: [70acfc1299] check-in: [948dd75b36] user: dgp branch: trunk, size: 64940 | |
| 19:24 | merge 8.6 file: [f60b307a2e] check-in: [c1648c2e63] user: dgp branch: core-8-branch, size: 64751 | |
| 18:58 | merge 8.5 file: [6f43f892c5] check-in: [c4bcfe94fe] user: dgp branch: core-8-6-branch, size: 58884 | |
| 17:21 | Work In Progress. Much trickiness about sorting out expectations. file: [74a6d3dafe] check-in: [581d6b939a] user: dgp branch: bug-e617e8a71a, size: 67780 | |
| 15:21 | Add comments so I'll know again later why this is here. file: [c0c5a180c4] check-in: [de47f8d8ee] user: dgp branch: core-8-5-branch, size: 52727 | |
| 14:54 | First, prove that bug [ed29806baf] is present in 8.7 too. Let's see what test-cases fail when we no longer check the validity of the 3th trail byte. file: [e043fcdbc0] check-in: [5510c6045d] user: jan.nijtmans branch: bug-ed29806baf-8.7, size: 64539 | |
| 14:20 | Merge 8.7 file: [25e75d1724] check-in: [a65a5fe9ec] user: jan.nijtmans branch: trunk, size: 64767 | |
| 14:16 | Backout the quick-fix workaround in Tcl_NumUtfChars as a way to monitor progress on this issue. file: [3b2d18dec3] check-in: [b8aafb8d10] user: dgp branch: bug-ed29806baf, size: 58443 | |
| 12:51 | Let's not get out the src[3] check yet. file: [bb95bef2f4] check-in: [577f5c5de8] user: jan.nijtmans branch: core-8-branch, size: 64578 | |
| 12:45 | Merge 8.6 file: [bd04a59c24] check-in: [7ce14c8063] user: jan.nijtmans branch: core-8-branch, size: 64582 | |
| 12:09 | Partial fix for [ed29806ba]: Tcl_UtfToUniChar reads more than TCL_UTF_MAX bytes. file: [4907cb1886] check-in: [98f5334275] user: jan.nijtmans branch: core-8-6-branch, size: 58711 | |
|
2020-04-29
| ||
| 22:22 | merge 8.7 file: [054d4aae77] check-in: [7db8756066] user: dgp branch: bug-e617e8a71a, size: 66345 | |
| 20:44 | Backport many UNICODE_OUT_OF_RANGE() calls. This should fix [69634d51fb74551b] for Tcl 8.5 (with TCL_UTF_MAX=4) too. Also fix some comments which were not up to date. No change at all in behavior for TCL_UTF_MAX=3. file: [16c3f0784c] check-in: [fe05235530] user: jan.nijtmans branch: core-8-5-branch, size: 52554 | |
| 19:23 | Merge 8.7 file: [a4101aad1d] check-in: [9f5d7846b9] user: jan.nijtmans branch: trunk, size: 64662 | |
| 19:20 | Merge-mark 8.6 (Use of UNICODE_OUT_OF_RANGE() macro already was in 8.7). Quick exit from Tcl_UtfToChar16()/Tcl_UtfToUniChar() when lead-byte is 0xF5 - 0xF7. file: [f06fe486d5] check-in: [24f0963165] user: jan.nijtmans branch: core-8-branch, size: 64473 | |
| 19:00 | Add UNICODE_OUT_OF_RANGE() calls to UCS4ToTitle() and friends. Backported from 8.7. This fixes [69634d51fb]: handling out of range UCS4 values (at least, it's fixed in 8.6 now the same way as it's fixed in 8.7). file: [09fa94833f] check-in: [8daf5a10dd] user: jan.nijtmans branch: core-8-6-branch, size: 58601 | |
| 17:39 | First attempt at extending routine to deal with surrogate pairs. Exposes problems with the interface. file: [c223a30f70] check-in: [f6001c20fb] user: dgp branch: bug-e617e8a71a, size: 66436 | |
|
2020-04-28
| ||
| 21:45 | Merge 8.7 file: [0d52a887c6] check-in: [b11d56197b] user: jan.nijtmans branch: rfe-f443140a85, size: 63972 | |
| 14:37 | Implementation for TIP #573: Surrogates are invalid file: [9457d8b6d7] check-in: [749d917ed5] user: jan.nijtmans branch: tip-573, size: 64692 | |
| 09:46 | Change test expectations to what desired. Mark failing tests with "knownBug". 10 test-cases are affected. Still work do to .... file: [aaa764e1d8] check-in: [4641000dd1] user: jan.nijtmans branch: rfe-f443140a85, size: 63975 | |
|
2020-04-27
| ||
| 20:49 | First shot at implementatio for [f443140a85]. Far from correct yet, since Tcl_UtfPrev() gives strange results. Tcl_UtfNext() looks OK at first sight. Further testing needed. file: [765a2c48f8] check-in: [b287415a79] user: jan.nijtmans branch: rfe-f443140a85, size: 63920 | |
| 12:59 | merge 8.7 file: [33da1294ac] check-in: [94ed8d8b2b] user: dgp branch: trunk, size: 64753 | |
| 12:57 | merge 8.6 file: [1f829b2e0d] check-in: [60de0f0a95] user: dgp branch: core-8-branch, size: 64564 | |
| 12:26 | Use lossless internal routines to cover extended characters. file: [652d56928a] check-in: [ef84b2a306] user: dgp branch: bug-45ca2338cd, size: 58446 | |
| 01:29 | Possible fix for [string to*] writing out a high surrogate at end of string. file: [f35edc2a16] check-in: [cd98f1d62d] user: dgp branch: bug-45ca2338cd, size: 58102 | |
|
2020-04-26
| ||
| 15:21 | Merge 8.7 file: [8b59a487f5] check-in: [631a21f090] user: jan.nijtmans branch: trunk, size: 65642 | |
| 15:11 | Cherry-pick Tcl_UniCharAtIndex() implementation from [6596c4af31], but adapted to the needs of TIPs 389/542. file: [792691dbab] check-in: [25b3737625] user: jan.nijtmans branch: core-8-branch, size: 65453 | |
| 13:25 | Remove the function Tcl_UniCharAtIndex() completely from the core. Meant as a demonstration for ticket [f45d0dc1a7]. file: [30c86e86e2] check-in: [90be2283d0] user: jan.nijtmans branch: bug-f45d0dc1a7, size: 58038 | |
|
2020-04-25
| ||
| 22:50 | Merge 8.7 file: [c702e30a43] check-in: [dfd3e79a8c] user: jan.nijtmans branch: trunk, size: 66113 | |
| 22:32 | Merge 8.6 file: [14d53fbe21] check-in: [30a29680e1] user: jan.nijtmans branch: core-8-branch, size: 65888 | |
| 22:16 | encoding-12.6 only works for "ucs2" for now. Don't use (deprecated) INLINE and CONST file: [80e1a29677] check-in: [aaeafee98b] user: jan.nijtmans branch: core-8-6-branch, size: 57965 | |
| 16:04 | Close utf-next-regressions file: [23810d5a63] check-in: [aea91e2e26] user: dgp branch: core-8-6-branch, size: 57972 | |
| 14:37 | merge 8.6 file: [d53a8ca4ee] check-in: [2e6bf5602d] user: dgp branch: dgp-alternative, size: 59653 | |
|
2020-04-24
| ||
| 22:40 | Quickfix to Tcl_NumUtfChars(). Barely used in Tcl core. Still needs a better look. Mark two new tests as knownBug. Needs a further look as well. file: [55a7e499c6] check-in: [1c7c61d459] user: jan.nijtmans branch: core-8-6-branch, size: 59653 | |
| 21:54 | WIP merging 8.6. file: [844569e750] check-in: [c18876475f] user: dgp branch: dgp-alternative, size: 59620 | |
| 21:17 | merge 8.6 file: [a0bf1d8545] check-in: [a68982ebc3] user: dgp branch: mistake, size: 59385 | |
| 21:08 |
Merge 8.6; still have to sort out origins of failing tests.`
Botched this merge. Eliminated the things the branch was created to preserve in the first place. file: [40de756444] check-in: [137f33c330] user: dgp branch: mistake, size: 59156 | |
| 21:01 | Merge 8.5 (but this time correct) file: [4f0c01adfc] check-in: [38dbfc8a67] user: jan.nijtmans branch: core-8-6-branch, size: 59385 | |
| 20:51 | Merge 8.5. Failing tests need examination and adjustment. file: [f10ef7bc58] check-in: [12254b0693] user: dgp branch: core-8-6-branch, size: 59156 | |
| 17:27 | merge 8.5 file: [cce3cda4a4] check-in: [4285f56a71] user: dgp branch: dgp-27944a3661, size: 53299 | |
| 17:18 | Revert the parts of [76213b3f72] that converted callers of Tcl_UtfToUniChar into callers of Tcl_UtfNext. With this reversion, any future divergence between those two will not harm these callers. Retain the tests, and retain the new implementation of Tcl_UtfNext itself and its new macro form. file: [8d925de3c1] check-in: [870aba745a] user: dgp branch: core-8-5-branch, size: 52773 | |
| 13:32 | Merge 8.7 file: [3d047895f9] check-in: [33a6f75e9c] user: jan.nijtmans branch: trunk, size: 67794 | |
| 13:31 | Merge 8.6 file: [48cc25301d] check-in: [0b05e92ef6] user: jan.nijtmans branch: core-8-branch, size: 67576 | |
| 13:15 | Merge 8.5. Fix regression in Tcl_UtfComplete(), actually already present for longer time but masked by error in TclUtfNext() macro. Adapt expectations accordingly ("/xA0/xA0" should really have length 2 ....) file: [6e756078e9] check-in: [8cf6bd3ab7] user: jan.nijtmans branch: core-8-6-branch, size: 59620 | |
| 10:49 | Merge 8.7 file: [4ebccd3c84] check-in: [aa56ee88d9] user: jan.nijtmans branch: trunk, size: 67773 | |
|
2020-04-23
| ||
| 22:10 | Second attempt file: [3bb45d9524] check-in: [312fb13ea5] user: jan.nijtmans branch: utf-next-for-8.7, size: 67555 | |
| 21:50 | First attempt to merge Tcl_UtfNext()/Tcl_UtfPrev() improvements (check for invalid byte sequences) to 8.7 file: [05f74e48c9] check-in: [891db10cfb] user: jan.nijtmans branch: utf-next-for-8.7, size: 67717 | |
| 20:22 | Demonstrate that the failing tests on the 8.6 branch tip can equally well be solved by backing out the recent changes associated with [27944a3661]. file: [83043039a5] check-in: [2c7b1e958d] user: dgp branch: dgp-alternative, size: 59598 | |
| 19:25 | merge 8.5 file: [0b1e5cd94d] check-in: [92946e427a] user: dgp branch: dgp-27944a3661, size: 53214 | |
| 19:07 |
Fix regression in Tcl_NumUtfChars, caused by this commit: [6596c4af31e29b5d]. Expectations of failing tests was adapted later, that's why this was missed.
Lesson: Tcl_UtfNext() is _not_ just an optimized replacement for Tcl_UtfToUniChar(). Sorry, but this change it just to dangerous!
Tcl_UniCharAtIndex() and Tcl_UtfAtIndex() most likely have the same regression when fead with invalid byte-sequences, therefore reverted those too.
HOLD ON! These regressions are equally the result of [5c322bbd51]. It takes both changes to cause the failing tests. We need to argue about which change was the wrong one. file: [6e58a843ab] check-in: [3fd0be11ba] user: jan.nijtmans branch: utf-next-regresions, size: 57833 | |
| 19:04 | Argument conditions for Invalid() call were not always satisfied. file: [7529c5d123] check-in: [0200ddd3d4] user: dgp branch: core-8-5-branch, size: 52688 | |
| 18:22 | Revised Tcl_UtfCharComplete() to be a proper safety filter for the revised needs of callers of Tcl_UtfNext(). file: [00df497402] check-in: [fecfc37392] user: dgp branch: dgp-27944a3661, size: 53186 | |
| 14:51 | Merge the two modes of Tcl_UtfNext into a single loop. file: [3a7cbf9366] check-in: [c6dba8e537] user: dgp branch: dgp-27944a3661, size: 52934 | |
| 14:02 |
Revise the totalBytes array so that it stores the number of bytes in a valid
byte sequence beginning with the byte value of the index. This is harmless
to existing uses of the array. Trail bytes do not lead a valid byte sequence
so their entries are changed to 0.
CORRECTION: That's not quite true. The change revises the results of tests utf-6.95 and utf-6.116 through changing results of Tcl_UtfCharComplete, but that's the point of this path. file: [2387b5d5bf] check-in: [4f1fbe0e8b] user: dgp branch: dgp-27944a3661, size: 52865 | |
| 11:31 | Fix [27944a3661]: Taming test utf-6.88 Fix [c11e0c5ce4]: Regression in Tcl_UtfCharComplete Fix [1b1f5f0b53]: Tcl_UtfNext incompatibility in Tcl 8.6.10 file: [9220af0032] check-in: [5c322bbd51] user: jan.nijtmans branch: core-8-6-branch, size: 59735 | |
|
2020-04-22
| ||
| 22:13 | (cherry-pick): Update documentation of Tcl_UtfPrev/Tcl_UtfNext back to how it was. Will be updated later, when implementation is ready and agreed upon. file: [2179ac8253] check-in: [dabf4dd28a] user: jan.nijtmans branch: core-8-6-branch, size: 60052 | |
| 20:10 | Adapt implementation and tests to vary behavior with TCL_UTF_MAX. file: [d32185c5a9] check-in: [de68f922b4] user: dgp branch: dgp-27944a3661, size: 52865 | |
| 19:22 | Place first-draft implementation of the proposed change to handling of trail bytes by Tcl_UtfNext on branch dgp-27944a3661 for examination. file: [a8ffe9df20] check-in: [853c8b207f] user: dgp branch: dgp-27944a3661, size: 52832 | |
| 14:49 | Merge 8.6 file: [1865dc3e3c] check-in: [43178b7bb4] user: jan.nijtmans branch: bug-c11e0c5ce4, size: 59746 | |
| 08:10 | Merge 8.7 file: [e41afeed1d] check-in: [e29ce0fdea] user: jan.nijtmans branch: trunk, size: 63697 | |
| 07:59 | Fix [27944a3661]: Taming test utf-6.88. Long-standing bug in Tcl_UtfNext(). Corner-case when the pointer doesn't advance to the start-byte of the next UTF-8 character. file: [6341278860] check-in: [aff418be86] user: jan.nijtmans branch: core-8-branch, size: 63472 | |
| 07:41 | Merge 8.5 file: [457a8304d3] check-in: [57148adcab] user: jan.nijtmans branch: core-8-6-branch, size: 61408 | |
|
2020-04-21
| ||
| 10:24 | Merge 8.7 file: [4f963f9768] check-in: [1d9d4a348c] user: jan.nijtmans branch: trunk, size: 63479 | |
| 09:49 | More test cleanup file: [519c9098ce] check-in: [926d62f948] user: jan.nijtmans branch: core-8-branch, size: 63254 | |
| 07:26 | Wrong indent in comment file: [7077e25dc4] check-in: [6a12cc0c59] user: jan.nijtmans branch: bug-27944a3661, size: 63474 | |
| 07:18 | Merge 8.7 file: [63d967a3f3] check-in: [8895c470f0] user: jan.nijtmans branch: bug-27944a3661, size: 63473 | |
| 07:03 | Add more test-cases for TCL_UTF_MAX>3 file: [21aa145451] check-in: [12d578a6ad] user: jan.nijtmans branch: bug-c11e0c5ce4, size: 59375 | |
|
2020-04-20
| ||
| 22:00 | Merge 8.7 file: [8563f71eb0] check-in: [e767cf6856] user: jan.nijtmans branch: trunk, size: 63477 | |
| 21:54 | Teach Tcl_UtfPrev() that 0xC1 is _always_ an invallid byte. Test-case utf-7.34. Make sure that Tcl_UtfCharComplete(src, TCL_UTF_MAX) always returns 1, for whatever bytes, since that's the maximum number of bytes Tcl_UtfToUniChar() can read in a single call. file: [94899e2693] check-in: [5f382f2056] user: jan.nijtmans branch: core-8-branch, size: 63252 | |
| 21:07 | Change a few variables from type "int" to "size_t". Always test TCL_UTF_MAX for <= 3 or > 3, because that's the only 2 flavours we really have. file: [53728707d7] check-in: [d515532940] user: jan.nijtmans branch: trunk, size: 63182 | |
| 15:20 | Proposed fix for [c11e0c5ce4]: Regression in Tcl_UtfCharComplete. file: [123d9496d9] check-in: [21362ee023] user: jan.nijtmans branch: bug-c11e0c5ce4, size: 59372 | |
| 11:56 | (cherry-pick): Proposed fix for [27944a3661]: Taming test utf-6.88. file: [8f5fe9844f] check-in: [c0d6e94d4e] user: jan.nijtmans branch: bug-c11e0c5ce4, size: 59895 | |
| 05:17 | Apply fix for [2738427] and adjust tests. file: [71c4cd01eb] check-in: [27cfa0775d] user: dgp branch: dgp-utf-explore, size: 52919 | |
| 05:09 | Apply fix for [493dccc2de] and adjust tests. file: [6454d41361] check-in: [7db642d35a] user: dgp branch: dgp-utf-explore, size: 52776 | |
| 04:58 | Apply fix for [5e6346a252] and adjust tests. file: [385e9d7fe5] check-in: [32bc8e9f0c] user: dgp branch: dgp-utf-explore, size: 51958 | |
| 04:40 | Apply first fix of [c61818e4c9] and adjust tests. file: [d71ee01f31] check-in: [4227206dc3] user: dgp branch: dgp-utf-explore, size: 48435 | |
| 04:02 | Apply the overlong rejection backport fix again, but this time don't attempt to include surrogate support. This was the first attempt on the 8.5 branch to leave behind the long outdated FSS-UTF approach that suggested support for 5- and 6-byte sequence support. Clearly there was no useful support for a TCL_UTF_MAX=4 build before this, and a TCL_UTF_MAX=6 build could have only been nonsense. There's no history worth reclaiming compatibility with in the custom builds. file: [68e09348df] check-in: [daafc7f0aa] user: dgp branch: dgp-utf-explore, size: 46751 | |
|
2020-04-19
| ||
| 20:49 |
Merge 8.5, and add the fix for [27944a3661] here too. Getting closer to what test-results we expect.
Overlong sequences not handled yet, so that's where differences are expected file: [1baf563b6a] check-in: [7c70d7b16d] user: jan.nijtmans branch: core-8-5-orig, size: 51337 | |
| 09:57 | Proposed fix for [27944a3661]: Taming test utf-6.88. This fix is not optimized, it still uses TclUtfToUniChar() in its implementation. But optimizing work is on its way, hopefully, coming through 8.5 .. 8.6 .. and up. file: [918a834bf7] check-in: [74d3f929c8] user: jan.nijtmans branch: bug-27944a3661, size: 63296 | |
|
2020-04-18
| ||
| 13:53 |
Missing Tcl_UniChar initializations allow 4-byte Tcl_UtfToUniChar to act on
uninitialized mem and get into big trouble.
This bug fix is abandoned. A better solution was adopted instead. file: [c949351d34] check-in: [517b1c87da] user: dgp branch: bug-c574e50a3b, size: 49343 | |
| 13:47 | Fix [c574e50a3b30e76f]: CRASH: utf-2.[89] in 8.5 built with TCL_UTF_MAX=4 file: [2aff7aacbb] check-in: [dda19888bd] user: jan.nijtmans branch: bug-c574e50a3b, size: 52667 | |
| 12:46 | Update documentation of Tcl_UtfPrev/Tcl_UtfNext back to how it was. Will be updated later, when implementation is ready and agreed upon. file: [b85447eb6b] check-in: [05486a901f] user: jan.nijtmans branch: core-8-5-branch, size: 55396 | |
|
2020-04-17
| ||
| 21:07 | [493dccc2de] Revise sequence validity check to reject out of range decodes too. file: [0adcc0196e] check-in: [2282b5ecbf] user: dgp branch: core-8-5-branch, size: 56758 | |
| 11:02 | Fix implementation of Tcl_UtfAtIndex() for TCL_UTF_MAX=6 (There's no test-case for this in the core-8-6-branch, but there is in core-8-branch). file: [5e9dae4c8a] check-in: [a14c9e965e] user: jan.nijtmans branch: core-8-6-branch, size: 61037 | |
| 10:32 | More test-cases. Mark test-case utf-2.11 as "knownBug", doesn't give the right answer for any TCL_UTC_MAX value. TODO: To Be Fixed! (Don ????) Fix build/testcase for TCL_UTF_MAX=6 (testcase is OK, "string length" implementation was not!) file: [59de7f446e] check-in: [d9a7cd6292] user: jan.nijtmans branch: core-8-6-branch, size: 61016 | |
| 09:17 | Original implementation of Tcl 8.5 before the Tcl_UtfPrev/Tcl_UtfNext reform. But: Add the (harmless) TclUtfNext/TclUtfPrev macro's and add the 0xC1 byte as being invalid. Then ... add the testcases from core-8-5-branch (and a few more), with adapted expectations according to the original implementation. Let's see where we get. file: [25c60f32b5] check-in: [11eb38dbe2] user: jan.nijtmans branch: core-8-5-orig, size: 51201 | |
| 05:49 | Merging forward the Utf changes. Needs some repair yet. file: [ef790bca37] check-in: [36303870fc] user: dgp branch: dgp-fixme, size: 67944 | |
| 05:25 |
merge 8.6
Abort this branch. Rebasing a new proposed fix on top of recent reforms makes more sense now. file: [ff91171232] check-in: [73a39418d6] user: dgp branch: bug-31aa44375d, size: 61171 | |
| 05:14 | When supporting 4-byte sequences even with TCL_UTF_MAX = 3, need to paramterize a few things differently. (utf-4.11 failures). file: [0a6751f73d] check-in: [5b5ab20558] user: dgp branch: dgp-fixme, size: 60887 | |
| 04:45 | When supporting 4-byte sequences, make sure the Overlong test does too, and make sure the test results reflect it. file: [9218232ed2] check-in: [e537b566a1] user: dgp branch: dgp-fixme, size: 60981 | |
| 04:08 | Bring the single-byte marker for invalid lead byte \xC1 into the complete table file: [2b01547609] check-in: [8d170b2c64] user: dgp branch: core-8-6-branch, size: 60980 | |
| 04:07 | Merge 8.6 file: [c2a23eb5a7] check-in: [f579c619ea] user: dgp branch: bug-31aa44375d, size: 61171 | |
| 03:54 | Fix the bad tests utf-2.11 and utf-6.88 that expected the wrong results. Also reconcile the merge from 8.5 to the new decoupling of bytesequence counts from indexed code unit couints. Docs still need an update. file: [a744ed0e1f] check-in: [7f56d3c4b1] user: dgp branch: core-8-6-branch, size: 60980 | |
|
2020-04-16
| ||
| 22:19 | Fix build for TCL_UTF_MAX=4. Mark some failing tests with "knownBug". Those still need to be fixed! file: [c25a1a5819] check-in: [660902f443] user: jan.nijtmans branch: core-8-6-branch, size: 61248 | |
| 22:06 | Fix more test-cases for TCL_UTF_MAX=3 file: [7949e6dbc9] check-in: [46026a9db7] user: jan.nijtmans branch: core-8-6-branch, size: 61247 | |
| 20:59 | Adjust test results and implementation for Tcl 8.6 current support of 4-byte sequences in a TCL_UTF_MAX=3 build. file: [d778cb9e8c] check-in: [010ec5fc7c] user: dgp branch: core-8-6-branch, size: 61159 | |
| 20:52 | merge 8.6 file: [18b0ad5b14] check-in: [660bca000f] user: dgp branch: bug-31aa44375d, size: 61227 | |
| 20:38 | Merge 8.5. Failing tests for now. To be remedied shortly. file: [a76bc31afc] check-in: [6596c4af31] user: dgp branch: core-8-6-branch, size: 61139 | |
| 20:28 | merge litter file: [0d78cf74a8] check-in: [7e421dce7c] user: dgp branch: core-8-5-branch, size: 56387 | |
| 20:02 | merge 8.6 file: [6245021e7b] check-in: [ad63873abe] user: dgp branch: bug-31aa44375d, size: 55797 | |
| 19:02 | More detailed comments. file: [bd309aa713] check-in: [f6a8c5432f] user: dgp branch: dgp-utf-next, size: 56394 | |
| 18:42 | compiler warning file: [39c439d18f] check-in: [f4baf7e7af] user: dgp branch: dgp-utf-next, size: 55038 | |
| 18:40 | More tests and fix for overlong handling in revised Tcl_UtfNext. file: [9bef9c5891] check-in: [29041a6d75] user: dgp branch: dgp-utf-next, size: 55021 | |
| 17:55 | merge 8.5 file: [2b00aa795e] check-in: [aadfd3448c] user: dgp branch: dgp-utf-next, size: 54973 | |
| 17:30 | Convert Overlong() to use a lookup table. file: [2632bc61a4] check-in: [5550e9ac96] user: dgp branch: bug-5e6346a252, size: 54724 | |
| 16:22 | When we reject overlong sequences, \xC1 is as invalid a lead byte as \xFF. file: [cb7af25b02] check-in: [ce6e2901e2] user: dgp branch: bug-5e6346a252, size: 54597 | |
| 15:48 | merge 8.5 file: [61fdf895aa] check-in: [877190c2e0] user: dgp branch: dgp-utf-next, size: 51450 | |
| 15:38 | merge 8.5 file: [0283bd7e47] check-in: [b59c18d809] user: dgp branch: bug-5e6346a252, size: 54614 | |
|
2020-04-15
| ||
| 22:39 | Refactor the Overlong test into a utility routine. file: [ac914a5f27] check-in: [9cf8c75384] user: dgp branch: bug-5e6346a252, size: 54616 | |
| 22:28 | Use test existence to shorten comment. file: [0dcc7991ab] check-in: [32c044a1f8] user: dgp branch: bug-5e6346a252, size: 54087 | |
| 21:48 | Rework Tcl_UtfPrev so it properly handles overlong sequences. file: [465d171fd5] check-in: [bd5edfc3bc] user: dgp branch: bug-5e6346a252, size: 54308 | |
| 20:03 | Merge 8.7 file: [cbbad6543c] check-in: [94f0de486e] user: jan.nijtmans branch: trunk, size: 63182 | |
| 19:54 | Merge 8.6 file: [413cb0fda4] check-in: [9adf3a7df5] user: jan.nijtmans branch: core-8-branch, size: 63075 | |
| 19:38 | Merge 8.5 file: [10fc6f514a] check-in: [58b4d49d41] user: jan.nijtmans branch: core-8-6-branch, size: 55681 | |
| 19:33 | New test command "testutfnext", not used yet in actual test-cases. Being merged up to higher branches. (Thanks, Don!) file: [5e70ee5041] check-in: [4eb84fae1d] user: jan.nijtmans branch: core-8-5-branch, size: 51201 | |
|
2020-04-14
| ||
| 21:32 | typo file: [f5da4a5eb4] check-in: [3e47eb4000] user: dgp branch: dgp-utf-next, size: 51452 | |
| 21:25 | Fix the bad logic in Tcl_UtfNext(). file: [fbf644d6c2] check-in: [eb95fbb884] user: dgp branch: dgp-utf-next, size: 51453 | |
| 21:16 | Replace calls of TclUtfToUniChar() with TclUtfNext() when caller has no decoding need. Failing test string-22.14 indicates something is still not quite right. Now that Tcl_NumUtfChars() is not paying decoding prices, we let it spend to properly protect against overflow. [2738427] file: [289613fbcc] check-in: [d0cc6cdb7f] user: dgp branch: dgp-utf-next, size: 51277 | |
| 20:03 | The function of Tcl_UtfNext() is to advance a pointer. There's nothing inherent in that task that requires decoding of the characters, but the implementation does that. Let's try a simpler solution for callers that do not need the content decoded. file: [4f6160fc46] check-in: [9e87b14c18] user: dgp branch: dgp-utf-next, size: 51410 | |
| 14:30 | Merge 8.7 file: [9d84961fe0] check-in: [bd5cc21d39] user: jan.nijtmans branch: trunk, size: 63184 | |
| 10:17 | Merge 8.6 file: [68226143a0] check-in: [250fd8a281] user: jan.nijtmans branch: core-8-branch, size: 63077 | |
| 09:19 | Merge 8.6 file: [aecd727c2e] check-in: [76244d6280] user: jan.nijtmans branch: bug-31aa44375d, size: 55799 | |
| 08:44 | Fix unit-test, change expectations according to current 8.6 branch (not handling [1b1f5f0b53] yet, doing that in separate branch) file: [1770c1c337] check-in: [e64b5a4785] user: jan.nijtmans branch: bug-c61818e4c9, size: 55683 | |
| 00:09 | Create separate tables to serve Tcl_UtfPrev and Tcl_UtfComplete. file: [dec5b78a2e] check-in: [c47ec8e3a3] user: dgp branch: bug-c61818e4c9, size: 55751 | |
|
2020-04-13
| ||
| 21:45 | merge 8.6 file: [95b1991b1b] check-in: [9eccb6d91f] user: dgp branch: bug-31aa44375d, size: 55096 | |
| 21:36 | Merge 8.5. Failing tests highlight ticket [1b1f5f0b53]. file: [ddad3919a5] check-in: [50b6d532c4] user: dgp branch: core-8-6-branch, size: 55085 | |
| 18:42 | Make the comments describing Tcl_UtfPrev more complete and precise. file: [095e0f9527] check-in: [0db5bfbfeb] user: dgp branch: dgp-utf-prev-alt, size: 51203 | |
|
2020-04-10
| ||
| 15:22 | Merge 8.7 file: [196953aa7e] check-in: [894e2ce40e] user: jan.nijtmans branch: trunk, size: 60900 | |
| 13:02 | Merge 8.6 file: [c4d166b8d8] check-in: [cd1ce15d26] user: jan.nijtmans branch: core-8-branch, size: 60794 | |
| 12:19 | Since Tcl_UtfCharComplete() now guarantees that at least 3 more bytes are available for header bytes 0x80-0xBF, check those 3 bytes first in Tcl_UtfToUniChar() before doing other checks (that might point to uninitialized memory in non-confirming extensions) file: [5c4cc96207] check-in: [99ac71af15] user: jan.nijtmans branch: core-8-6-branch, size: 53400 | |
|
2020-04-09
| ||
| 19:22 | merge 8.5 file: [cc540eb927] check-in: [d3bf59e09a] user: dgp branch: dgp-utf-prev-alt, size: 49569 | |
| 17:30 | Guarantee TclNeedSpace and TclFindElement have common definition of whitespace by having both call the same routine. Create a macro form to contain performance costs and adapt callers. file: [321be2df7e] check-in: [da9a473a26] user: dgp branch: core-8-5-branch, size: 49519 | |
|
2020-04-08
| ||
| 17:31 | Apply better bug fix that does not create new bugs this time. file: [78ac68637f] check-in: [7ee618b9ca] user: dgp branch: dgp-utf-prev-alt, size: 49568 | |
| 16:43 | Restore the original Tcl_UtfPrev routine. Fails a different set of tests. Many fewer. file: [c14653a813] check-in: [6c34d40ef3] user: dgp branch: dgp-utf-prev-alt, size: 49572 | |
| 15:37 | introduces new utf-internal definition UTF_TO_UNI_MAX, as maximal value that Tcl_UtfToUniChar could return depending to UTF_TO_UNI_MAX and tcl version (here 3 or 4), restricting search cycle of Tcl_UtfPrev now. file: [63dd626b9c] check-in: [f10ba6b7ac] user: sebres branch: sebres-utf-prev-fix, size: 50307 | |
| 13:26 | partially revert [e8bfb4c2ba884b1b], a prevention added previously in string trim functions (compare char lengths) is not needed with new version of Tcl_UtfPrev (leave other optimizations as appropriate, may be important in some obscure case not covered by current or future implementation of Tcl_UtfPrev/Tcl_UtfToUniChar) file: [d84096c4c4] check-in: [57aa5dae72] user: sebres branch: sebres-utf-prev-fix, size: 50138 | |
| 12:32 | Rewrite of Tcl_UtfPrev that restores the correct behavior (no inconsistency between Tcl_UtfPrev/Tcl_UtfToUniChar anymore), Tcl_UtfPrev strictly follows all rules of Tcl_UtfToUniChar now, also fixing all test cases now. file: [91d7ab8dba] check-in: [cdffcbaec9] user: sebres branch: sebres-utf-prev-fix, size: 50144 | |
|
2020-04-07
| ||
| 20:04 |
Rewrite of Tcl_UtfPrev that fixes most of the failing tests.
Remaining failures demonstrate that the interface of Tcl_UtfPrev cannot be satisfied robustly if we stick to the restriction of not reading (*src). file: [73e1a9bb8b] check-in: [ee25edfdb7] user: dgp branch: dgp-utf-prev-fix, size: 50382 | |
|
2020-04-06
| ||
| 09:22 | Make Tcl_UtfCharComplete() usable for both Tcl_UtfToUniChar() and Tcl_UtfToChar16(). Defect noticed by Don Porter. Thanks! Add test-cases, assuring correct handling of 4-byte UTF-8 sequences. Use "end-1", "end" and "end+1" in testcases related to Tcl_NumUtfChars(), that's more readable/maintainable than integers. file: [086b32b017] check-in: [20a619daf1] user: jan.nijtmans branch: core-8-branch, size: 60756 | |
| 07:53 | Revert commit [aed6634d2ccf2107], which backported part of TIP #389 (regarding internal handling of 4-byte UTF-8 sequences) to 8.6. file: [af4d711786] check-in: [48c83d55c8] user: jan.nijtmans branch: bug-31aa44375d, size: 53373 | |
|
2020-04-05
| ||
| 20:36 | Merge 8.6 file: [1953dc70c0] check-in: [b69fafa7f6] user: jan.nijtmans branch: core-8-branch, size: 60886 | |
| 20:28 | Partial fix for [31aa44375de2c87e]: Tcl_NumUtfChars regression in default 8.6 build. This commit brings Tcl_UtfCharComplete() into agreement with Tcl_UtfToUniChar(), whether it demands 1, 3 or 4 succeeding bytes. file: [ee669f4eb3] check-in: [166c0270e7] user: jan.nijtmans branch: core-8-6-branch, size: 53362 | |
|
2020-04-03
| ||
| 14:31 | merge 8.7 file: [4cb15b7aca] check-in: [cfeafcaff1] user: dgp branch: trunk, size: 60862 | |
| 14:28 | Fix broken build. file: [f7e34e9db0] check-in: [d76b30b67a] user: dgp branch: core-8-branch, size: 60638 | |
| 09:14 | Merge 8.7 file: [6d7160030f] check-in: [96a31010ce] user: jan.nijtmans branch: trunk, size: 60858 | |
| 09:13 | Simplify implementation of TclUtfToUCS4: The #undefined Tcl_UtfToUniChar() already does everything for use here (Unlike in Tcl 8.6, with has to live without TIP #542) file: [dfd49c138b] check-in: [4af836a404] user: jan.nijtmans branch: core-8-branch, size: 60634 | |
|
2020-04-02
| ||
| 23:26 | merge 8.7 file: [8d6c7ecaf2] check-in: [abac91f4a0] user: dgp branch: trunk, size: 61449 | |
| 22:18 | merge 8.6 file: [88d22c68cc] check-in: [f1db0ed67c] user: dgp branch: core-8-branch, size: 61225 | |
| 22:18 | Remove stray debug file: [252103a91a] check-in: [7fad0af377] user: dgp branch: core-8-6-branch, size: 53114 | |
| 22:05 | typo file: [9fcff62627] check-in: [e20f26ea10] user: dgp branch: core-8-6-branch, size: 53158 | |
| 20:36 | More callers. file: [ffc4318663] check-in: [b092a8788b] user: dgp branch: core-8-6-branch, size: 53156 | |
| 20:18 | New utility routine TclUtfToUCS4() to contain some complexity. Two callers adapted. file: [dd4033f180] check-in: [63e224f042] user: dgp branch: core-8-6-branch, size: 53667 | |
|
2020-03-18
| ||
| 14:06 | Merge 8.7 file: [611c4a948f] check-in: [1d0787097e] user: jan.nijtmans branch: trunk, size: 60180 | |
| 13:54 | Merge 8.6 file: [fe4cb5b540] check-in: [2b7c852b44] user: jan.nijtmans branch: core-8-branch, size: 59934 | |
| 12:51 | More uppercase HEX representations in source-code. file: [25e980fe33] check-in: [7d760e228e] user: jan.nijtmans branch: core-8-6-branch, size: 52183 | |
|
2020-01-20
| ||
| 14:19 | Merge 8.7 file: [feaee9d05d] check-in: [3d7b2dce60] user: jan.nijtmans branch: trunk, size: 60188 | |
|
2019-12-03
| ||
| 16:13 | Merge 8.7 file: [e638928518] check-in: [4105725743] user: jan.nijtmans branch: utf-max, size: 59942 | |
|
2019-12-02
| ||
| 20:48 | Merge 8.6 file: [fb044f8c43] check-in: [468df9021d] user: jan.nijtmans branch: core-8-branch, size: 59942 | |
| 20:26 | If TCL_UTF_MAX>=4, make Tcl_ParseBackslash combine two surrogates so they appear as one 4-byte UTF-8 byte sequence from the start. Add test-case for this. file: [dcaec0d214] check-in: [43032d7ba3] user: jan.nijtmans branch: core-8-6-branch, size: 52191 | |
|
2019-09-25
| ||
| 13:42 | Merge 8.7 file: [16bdd9b5d7] check-in: [be233f3e67] user: jan.nijtmans branch: utf-max, size: 59947 | |
|
2019-09-16
| ||
| 21:41 | Merge 8.7 file: [7ce80f4d77] check-in: [343db6648f] user: jan.nijtmans branch: trunk, size: 60198 | |
| 21:35 | Merge 8.6 file: [077dcce582] check-in: [5057c37bdf] user: jan.nijtmans branch: core-8-branch, size: 59947 | |
| 21:18 | Bugfix in Tcl_UtfPrev/Tcl_UtfNext: When handling 4-byte UTF-8 byte sequences, those should be able to move back/forward 4 bytes if TCL_UTF_MAX <= 4. Update comment accordingly. Bugfix in Tcl_UtfFindFirst/Tcl_UtfFindLast: Those functions should be able to find both the high surrogate (if asked for) as also the full character (combination of both surrogates) file: [83a9dd9fa9] check-in: [aed6634d2c] user: jan.nijtmans branch: core-8-6-branch, size: 52196 | |
|
2019-09-14
| ||
| 13:11 | Two paces where TCL_AUTO_LENGTH should be used file: [461130e810] check-in: [94f65c8701] user: jan.nijtmans branch: trunk, size: 60147 | |
| 13:07 | Merge 8.7 file: [4bf0afccb9] check-in: [9ea2eea22d] user: jan.nijtmans branch: trunk, size: 60117 | |
|
2019-08-15
| ||
| 08:59 | Merge 8.7 file: [f956e30ce4] check-in: [42a10393d8] user: jan.nijtmans branch: tip-548, size: 59896 | |
| 08:10 | Merge 8.7 file: [e68323bfe6] check-in: [f33e2933b5] user: jan.nijtmans branch: trunk, size: 61243 | |
|
2019-08-14
| ||
| 07:24 | Merge 8.7 file: [bfb184343a] check-in: [c367ba59b1] user: jan.nijtmans branch: no-register, size: 61014 | |
|
2019-08-12
| ||
| 20:38 | Merge branch tip-548. No longer define addtional stub-entries for functions that will be removed (because of deprecation) anyway file: [66aeb9ae4d] check-in: [50d822dbab] user: jan.nijtmans branch: utf-max, size: 59986 | |
|
2019-08-11
| ||
| 21:17 | Fix handling of length (size_t)-1 in tclMain.c. This should fix handling of command-line arguments with TCL_UTF_MAX=6, necessary to make tclsh run at all ... file: [2a1621d381] check-in: [33b7c8c229] user: jan.nijtmans branch: trunk, size: 61333 | |
|
2019-08-05
| ||
| 20:15 | Fix signature of TclWCharToUtfDString for TCL_UTF_MAX=6, and handling of length -1 file: [0b2a3ddc95] check-in: [1b13f758a5] user: jan.nijtmans branch: trunk, size: 61325 | |
|
2019-08-02
| ||
| 15:03 | Merge tip-548 file: [afbfcc9cdd] check-in: [d3a7842460] user: jan.nijtmans branch: utf-max, size: 60351 | |
| 14:26 | Merge 8.7 file: [116ccc2362] check-in: [b0fbdeb265] user: jan.nijtmans branch: trunk, size: 61304 | |
| 13:35 | Document that the *Backslash parsing functions output maximum 4 bytes, irrespectable of the TCL_UTF_MAX setting: It could be 4 for the "\Uxxxxxx" construct, but never more. Move <stddef.h> and <locale.h> to tclInt.h, so the can be removed from various other places. file: [50255cf53c] check-in: [1f393d7d01] user: jan.nijtmans branch: core-8-branch, size: 61104 | |
| 09:00 | Merge 8.7. Some formatting. file: [436c799582] check-in: [6305175e0c] user: jan.nijtmans branch: tip-548, size: 59986 | |
|
2019-08-01
| ||
| 21:55 | Protect Tcl_AToB() functions against NULL input file: [9f86858a88] check-in: [6e1922b861] user: jan.nijtmans branch: utf-max, size: 60358 | |
| 16:03 | Merge 8.7. Documentation improvements and code cleanup. Approaching finish. file: [bff0d2cb29] check-in: [57546481c1] user: jan.nijtmans branch: tip-548, size: 59996 | |
| 11:55 | Merge tip-548 file: [d2578eb0fc] check-in: [8819e7a6a3] user: jan.nijtmans branch: utf-max, size: 60180 | |
| 08:02 | Attempt to fix a179564826: Tk 8.6: prevent issues when encountering non-BMP Unicode characters file: [f55d7511cf] check-in: [f6eb4196ee] user: jan.nijtmans branch: bug-a179564826, size: 52162 | |
|
2019-07-31
| ||
| 19:47 | Merge 8.5 file: [9742291082] check-in: [98aa5b2f17] user: jan.nijtmans branch: core-8-6-branch, size: 52131 | |
| 19:39 | (cherry-pick from core-8-branch): Replace memcpy() calls with memmove() to avoid undefined behavior when source and destination overlap file: [c7c2794917] check-in: [382e19dea0] user: jan.nijtmans branch: core-8-5-branch, size: 49518 | |
|
2019-07-17
| ||
| 15:38 | Eliminate "register" keyword _everywhere_ in Tcl. This keyword is deprecated in C++ (removed in C++17, even), and essentially does nothing with most modern compilers. file: [5845f19aa9] check-in: [f074bda87c] user: jan.nijtmans branch: no-register, size: 61024 | |
|
2019-07-11
| ||
| 07:18 | Rename UTF-related functions to "WChar" and "Char16" variants, more intuitive because they represent wchar_t and char16_t (since C++11) types in modern compilers. file: [2893b6395e] check-in: [070bfd62cb] user: jan.nijtmans branch: tip-548, size: 60110 | |
|
2019-07-07
| ||
| 20:38 | Merge 8.7, and a few tweaks: Only provide Tcl_WinUtfToTChar on Tcl 8.x, not on 9.0 any more file: [30ac31a78d] check-in: [49e4bfa90e] user: jan.nijtmans branch: tip-548, size: 60070 | |
|
2019-07-06
| ||
| 23:09 | Fix UNIX/Mac build file: [b69a96ffb1] check-in: [dde79eb6a8] user: jan.nijtmans branch: tip-548, size: 60012 | |
|
2019-07-05
| ||
| 09:03 | Improvement: always export both 16-bit and 32-bit UTF function file: [94fa21f436] check-in: [27f2c4cf5e] user: jan.nijtmans branch: tip-548, size: 60003 | |
|
2019-06-03
| ||
| 19:48 | TIP #548: Deprecate Tcl_WinUtfToTChar() and Tcl_WinTCharToUtf() and provide more flexible replacement functions file: [e71bceebf1] check-in: [49785ba3b0] user: jan.nijtmans branch: tip-548, size: 60925 | |
|
2019-05-22
| ||
| 21:50 | More simplifications, taking deprecations into account file: [580294f8cd] check-in: [b95de9a625] user: jan.nijtmans branch: utf-max, size: 60174 | |
| 07:32 | More WIP: eliminate all usage of (platform-specific) Tcl_WinTCharToUtf()/Tcl_WinUtfToTChar() to its proposed portable replacements: Tcl_Utf16ToUtfDString()/Tcl_UtfToUtf16DString() This allows for Tcl_WinTCharToUtf()/Tcl_WinUtfToTChar() to be declared deprecated. file: [0cc57f9f6f] check-in: [a33e22b6ba] user: jan.nijtmans branch: utf-max, size: 64817 | |
|
2019-05-10
| ||
| 07:50 | Merge trunk file: [babc23af02] check-in: [f0009090a2] user: jan.nijtmans branch: regexp-api-64bit, size: 61314 | |
| 07:46 | merge 8.7 file: [1d09eea1b6] check-in: [f3302db091] user: jan.nijtmans branch: utf-max, size: 64767 | |
|
2019-04-19
| ||
| 00:38 | merge 8.7 file: [5965de89d4] check-in: [ddaa30125b] user: dkf branch: trunk, size: 61315 | |
|
2019-04-17
| ||
| 19:23 | Replace memcpy() calls with memmove() to avoid undefined behavior when source and destination overlap. file: [840b851393] check-in: [0b45548847] user: dgp branch: core-8-branch, size: 61114 | |
|
2019-03-31
| ||
| 22:03 | Enhance documentations. Move TCL_INDEX_NONE from tclInt.h to tcl.h, since it's too useful. file: [297ca17adf] check-in: [b1a506218e] user: jan.nijtmans branch: regexp-api-64bit, size: 61310 | |
|
2019-03-28
| ||
| 22:49 | Merge 8.7 file: [8450344d93] check-in: [eebb1e7ee1] user: jan.nijtmans branch: utf-max, size: 64763 | |
|
2019-03-24
| ||
| 16:52 | Merge 8.7 file: [e8c33d8e06] check-in: [c06a872943] user: jan.nijtmans branch: trunk, size: 61311 | |
| 16:50 | Since only bytes 0xF0 - 0xF4 can be the first byte of a valid 4-byte UTF-8 byte sequence, account for that in Tcl_UtfCharComplete(). file: [d96017f3d2] check-in: [9c09af3627] user: jan.nijtmans branch: core-8-branch, size: 61110 | |
| 16:46 | Since only bytes 0xF0 - 0xF4 can be the first byte of a valid 4-byte UTF-8 byte sequence, account for that in Tcl_UtfCharComplete(). Only effective when TCL_UTF_MAX>3 file: [3a932cdb85] check-in: [8e7ac039ab] user: jan.nijtmans branch: core-8-6-branch, size: 52115 | |
| 16:43 | Since only bytes 0xF0 - 0xF4 can be the first byte of a valid 4-byte UTF-8 byte sequence, account for that in Tcl_UtfCharComplete(). Only effective when TCL_UTF_MAX>3 file: [1afff3cd02] check-in: [12e58fc74d] user: jan.nijtmans branch: core-8-5-branch, size: 49514 | |
| 13:05 | Merge 8.7 file: [1a683a59b0] check-in: [15927b5ba6] user: jan.nijtmans branch: utf-max, size: 64773 | |
|
2019-03-21
| ||
| 20:28 | Merge 8.7. Also fix invalid reference to TclUtfToWChar, causing build failure file: [12b4c2e88b] check-in: [7c63883789] user: jan.nijtmans branch: utf-max, size: 64783 | |
| 20:11 | Merge 8.7 file: [64a2c08a05] check-in: [871076a655] user: jan.nijtmans branch: trunk, size: 61321 | |
| 19:56 | Remove incorrect comment. Simplify handling of last bytes in Tcl_UniCharToUtfDString(), since TclUtfToUniChar() already turns out to handle cp1252 fall-back correctly. file: [872868a1fa] check-in: [33251a211f] user: jan.nijtmans branch: core-8-branch, size: 61120 | |
|
2019-03-20
| ||
| 22:54 | Merge 8.7 file: [b9c85dc62a] check-in: [3ea5d3e8a3] user: jan.nijtmans branch: utf-max, size: 64957 | |
| 22:51 | Merge 8.7 file: [29dc2e6ddb] check-in: [bb9b52ab82] user: jan.nijtmans branch: trunk, size: 61496 | |
| 22:45 | Fix Tcl_UtfToUniCharDString() function, handling invalid byte at the end of the string: Not quite correct for bytes between 0x80-0x9F, according to TIP file: [e3072ae62e] check-in: [3e8ada19f5] user: jan.nijtmans branch: core-8-branch, size: 61295 | |
|
2019-03-18
| ||
| 22:34 | Merge 8.7 file: [45d678a779] check-in: [48c676b649] user: jan.nijtmans branch: trunk, size: 61473 | |
| 22:32 | Comment Comment Tcl_UniCharToUtf() better, what happens handling surrogates. Add type cast in tclUtf.c, making actual check clearer file: [8c6f2eb584] check-in: [b02df08680] user: jan.nijtmans branch: core-8-branch, size: 61272 | |
| 20:07 | Add 4 new encodings, and add documentation. file: [da54a774af] check-in: [0ac59eb0c6] user: jan.nijtmans branch: utf-max, size: 64937 | |
|
2019-03-17
| ||
| 22:01 | More WIP. Seems to be *almost* working. file: [7e148c56a6] check-in: [ab13cbd74c] user: jan.nijtmans branch: utf-max, size: 64779 | |
|
2019-03-16
| ||
| 21:10 | Merge 8.7 Move up some stub entries related to Tcl_UniChar Use TCL_UTF_MAX=4 for full Unicode in stead of TCL_UTF_MAX=6 (TCL_UTF_MAX: 3 is default) file: [819d977c5c] check-in: [81502a66ed] user: jan.nijtmans branch: utf-max, size: 61340 | |
|
2019-03-14
| ||
| 20:59 | Merge 8.7. Fix 2 test-cases which were failing for TCL_UTF_MAX=6 file: [da5dbc33a1] check-in: [4032e7fe99] user: jan.nijtmans branch: utf-max, size: 61340 | |
|
2019-03-12
| ||
| 20:39 | Even better support for -DTCL_UTF_MAX=6. Ongoing improvements (TIP being planned) file: [9fd3389618] check-in: [fdcb2a7323] user: jan.nijtmans branch: utf-max, size: 61291 | |
|
2019-03-10
| ||
| 21:04 | Merge 8.7 file: [9ec20eb789] check-in: [316ceb7616] user: jan.nijtmans branch: trunk, size: 61463 | |
| 20:18 | re-implemente changes in win/tclWinFile.c (handling -DTCL_UTF_MAX=6) using 3 new utility functions. This allows to re-use code in more places: cleaner implementation more future-proof. file: [b626580e33] check-in: [9eb437a15d] user: jan.nijtmans branch: core-8-branch, size: 61262 | |
|
2019-03-08
| ||
| 23:54 | Merge trunk. Further WIP for TIP #497, far from finished .... file: [dbe3f15e63] check-in: [9fcbdde251] user: jan.nijtmans branch: tip-497, size: 53015 | |
| 22:51 | Merge 8.7 file: [5f3cb4c241] check-in: [fe02c8898e] user: jan.nijtmans branch: trunk, size: 56119 | |
|
2019-03-07
| ||
| 22:02 | Fixes for TCL_UTF_MAX=6, (gcc compiler warnings). Also make everything work on win32/win64. Patch adapted from Androwish (thanks, Werner!) file: [3bc47b8988] check-in: [650574e0fb] user: jan.nijtmans branch: utf-max-6, size: 55918 | |
|
2019-03-05
| ||
| 18:23 | merge 8.7 (TIP#527, New measurement facilities in TCL: New command timerate, performance test suite) file: [39861b85b0] check-in: [e41cbd042a] user: sebres branch: trunk, size: 56041 | |
|
2019-03-02
| ||
| 17:21 | Add build with -DTCL_UTF_MAX=6 to travis CI. Also fix 2 gcc compiler-warnings occurring with -DTCL_UTF_MAX=6 file: [d2bbeadb16] check-in: [9b2a385a0f] user: jan.nijtmans branch: core-8-branch, size: 55840 | |
| 16:53 | Merge 8.7 file: [94d7d839ce] check-in: [e766d23655] user: jan.nijtmans branch: trunk, size: 56015 | |
| 16:52 | Minor optimization in UTF-8 handling, and add some comments describing how Tcl_UniCharToUtf() handles surrogates. file: [a53e54f604] check-in: [6e3632ede5] user: jan.nijtmans branch: core-8-branch, size: 55814 | |
| 16:04 | Backport [bd94500678e837d7] from 8.7, preventing endless loops in UTF-8 conversions when handling surrogates. Only effective when compiling with -DTCL_UTF_MAX=4|6 (default: 3). Meant for benefit of Androwish. file: [3926a96e0c] check-in: [9e1984c250] user: jan.nijtmans branch: core-8-6-branch, size: 52121 | |
|
2019-03-01
| ||
| 20:25 | merge 8.7 file: [859e807c72] check-in: [bc57eb7213] user: dgp branch: trunk, size: 55336 | |
| 20:24 | A confusion about signed vs unsigned comparision caused Tcl_UtfToUniChar() to return the wrong answer (contents of random memory) for each single byte UTF-8 in the input. This commit fixes that bug. More commentary on https://core.tcl-lang.org/tcl/tktview/bd94500678 file: [3b7ff4047c] check-in: [81046b694f] user: dgp branch: core-8-branch, size: 55135 | |
|
2019-02-27
| ||
| 21:58 | Merge 8.7 file: [b6c50a53f1] check-in: [727e74f081] user: jan.nijtmans branch: trunk, size: 55314 | |
|
2019-02-26
| ||
| 19:37 | More use of (efficient) TclHasIntRep() macro. Also eliminate many (size_t) and (unsigned) type-casts, which don't make sense any more. file: [8b057c57be] check-in: [2c7db3fa01] user: jan.nijtmans branch: mistake, size: 54843 | |
|
2019-02-25
| ||
| 21:10 | Finish complete fix, all corner-cases correct now. Also spurious UTF-8 testcase failure (as seen on travis) fixed now. file: [5a69d8b0b8] check-in: [b3d886c84f] user: jan.nijtmans branch: bug-bd94500678, size: 55113 | |
|
2019-02-19
| ||
| 20:17 | Merge 8.7 file: [ad73841f17] check-in: [2b82daafb8] user: jan.nijtmans branch: trunk, size: 55045 | |
| 20:16 | Merge 8.6 file: [fc24fefce3] check-in: [1b17625b60] user: jan.nijtmans branch: core-8-branch, size: 54879 | |
| 19:38 | Minor optimizations file: [ada6318df4] check-in: [0e2621fc4b] user: jan.nijtmans branch: bug-bd94500678, size: 55015 | |
|
2019-02-18
| ||
| 20:48 | Proposed fix for [bd94500678]: SEGFAULT by conversion of unicode (out of BMP) to byte-array. file: [d4d424d4fc] check-in: [9f67c17d01] user: jan.nijtmans branch: bug-bd94500678, size: 55303 | |
|
2019-02-04
| ||
| 22:45 | Merge trunk file: [b86e29514a] check-in: [a6db8815ce] user: jan.nijtmans branch: tip-502-for-9, size: 55051 | |
|
2018-10-27
| ||
| 19:33 | Backport various minor issues from 8.6: - gcc compiler warning in tclDate.c - protect Tcl_UtfToUniCharDString() from ever reading more than "length" bytes from its input, not even in the case of invalid UTF-8. - update to latest tzdata - fix 2 failing test-cases on MacOSX file: [6468647908] check-in: [f339295ed5] user: jan.nijtmans branch: core-8-5-branch, size: 49520 | |
|
2018-10-09
| ||
| 18:52 | Merge trunk file: [83b59c8873] check-in: [2ec0fe1752] user: jan.nijtmans branch: tip-497, size: 58210 | |
|
2018-10-08
| ||
| 19:00 | TIP #494 implementation: More use of size_t in Tcl 9 file: [0c6eb3b4a5] check-in: [f3d49044c4] user: jan.nijtmans branch: trunk, size: 55044 | |
| 18:34 | Merge 8.6 Also fix startup problems on win32, when the encoding path contains invalid UTF-8 (reported by François Vogel) Various other code cleanup, e.g. remove empty.zip file, as this didn't work quite as expected. file: [f38fd5f3de] check-in: [59a19cdee3] user: jan.nijtmans branch: core-8-branch, size: 54885 | |
|
2018-10-06
| ||
| 19:20 | Use more TCL_AUTO_LENGTH, when appropriate file: [26a99cdd2c] check-in: [c643e8fe38] user: jan.nijtmans branch: memory-API, size: 55050 | |
|
2018-10-04
| ||
| 21:07 | merge trunk file: [9ca6287650] check-in: [3ca1d5cace] user: jan.nijtmans branch: memory-API, size: 55035 | |
|
2018-10-03
| ||
| 20:02 | Merge 8.6 file: [8ff5e7dab8] check-in: [252cdd0247] user: jan.nijtmans branch: core-8-branch, size: 54891 | |
| 19:24 | Tcl_UniCharToUtfDString: Don't allocate too much memory for this function. Tcl_UtfToUniCharDString: Don't allocate too much memory for this function. And make sure that we never access more than 'length' bytes from the string, not even when encountering invalid UTF-8. file: [5c56cf0280] check-in: [8ba821d1fd] user: jan.nijtmans branch: core-8-6-branch, size: 51782 | |
|
2018-07-05
| ||
| 21:25 | Merge trunk Handle TclCopyAndCollapse, *Trace* et al file: [5b1fa6fc07] check-in: [399b8e7649] user: jan.nijtmans branch: memory-API, size: 54784 | |
|
2018-06-28
| ||
| 21:42 | More API changes using size_t. Internal changes not complete yet (WIP) file: [e1a768ce0f] check-in: [1bfecd9172] user: jan.nijtmans branch: memory-API, size: 54788 | |
|
2018-06-27
| ||
| 19:09 | merge trunk file: [5d60dd06e5] check-in: [2cc2d71f0a] user: jan.nijtmans branch: memory-API, size: 54782 | |
|
2018-06-24
| ||
| 20:26 | Fix "string tolower" and friends for handling unpaired surrogates correctly. Also add test-cases for those situations. Various typo's in comments. file: [2f8a5b250e] check-in: [1cdc9199e9] user: jan.nijtmans branch: core-8-branch, size: 54640 | |
|
2018-06-18
| ||
| 15:59 | Merge 8.6. And add more documentation and test-cases regarding the behavior of Tcl_UniCharToUtf() file: [3e4218a8ae] check-in: [3cb0cedeb6] user: jan.nijtmans branch: core-8-branch, size: 54235 | |
| 15:54 | Fix [53cad613d8]: TIP 389 implementation makes Tk tests font-4.12 and font-4.15 fail. One more situation in which high surrogate causes problem file: [8a6cd287b8] check-in: [c45de6cdb4] user: jan.nijtmans branch: core-8-6-branch, size: 51617 | |
|
2018-05-25
| ||
| 06:55 | merge trunk file: [4d8ecc6ca8] check-in: [e9340634d6] user: jan.nijtmans branch: memory-API, size: 54132 | |
|
2018-05-23
| ||
| 20:09 | merge trunk file: [e99fd221b0] check-in: [1a6fcb9bdc] user: jan.nijtmans branch: memory-API, size: 54023 | |
|
2018-05-11
| ||
| 09:20 | Merge 8.5. This adds Emoji 11.0 support, when Tcl is compiled with TCL_UTF_MAX>3. Useful for Androwish, for example. file: [18caccb00b] check-in: [708287d936] user: jan.nijtmans branch: core-8-6-branch, size: 51380 | |
| 09:14 | Add emoji 11.0 to the set. Only active when compiled with TCL_UTF_MAX>3. Also prepare tooling for Unicode 11.0 (while being on it) file: [deb90da498] check-in: [ee9f293421] user: jan.nijtmans branch: core-8-5-branch, size: 49339 | |
|
2018-05-07
| ||
| 07:40 | Remove some tip389 restrictions in test-cases, which are no longer necessary. Eliminate gcc compiler warnings when compiling with -DTCL_UTF_MAX=6 Other code clean-up and comment improvements. No change in functionality. file: [ca1f89b62b] check-in: [385fda311b] user: jan.nijtmans branch: core-8-branch, size: 53998 | |
|
2018-05-01
| ||
| 19:02 | Start implementing TIP #497. regexp's now are >BMP-aware. WIP file: [150a0d9005] check-in: [47ace058d4] user: jan.nijtmans branch: tip-497, size: 57214 | |
| 18:41 | Implement special "string totitle" for Extended Georgian characters (new behavior in Unicode 11) file: [caf8cb360a] check-in: [827b7761e6] user: jan.nijtmans branch: core-8-branch, size: 54010 | |
|
2018-04-23
| ||
| 23:32 | Merge 8.6 (bug-fix and test-case for Tcl_UtfAtIndex with TCL_UTF_MAX=4) file: [5f0b67c7df] check-in: [567e61b329] user: jan.nijtmans branch: mistake, size: 53810 | |
| 23:23 | Bug-fix in Tcl_UtfAtIndex (for TCL_UTF_MAX=4 only). With test-case (in "string totitle") demonstrating the bug. file: [2e32d28e48] check-in: [3d8301e3c6] user: jan.nijtmans branch: core-8-6-branch, size: 51170 | |
|
2018-04-20
| ||
| 10:16 | TIP #389 implementation. file: [b4aa23654b] check-in: [e109760b1c] user: jan.nijtmans branch: core-8-branch, size: 53402 | |
|
2018-04-19
| ||
| 22:29 | Slightly improved (more fail-safe) surrogate handling for TCL_UTF_MAX>3. Backported from latest TIP 389 implementation. (to be used for androwish) file: [e8fa0b3a54] check-in: [686259e650] user: jan.nijtmans branch: core-8-6-branch, size: 50997 | |
|
2018-04-17
| ||
| 21:49 | Slightly better unmatched-surrogates handling. Unmatched High surrogates will still be silently removed, but Unmatched Low surrogates will pass through as-is now. Inspired by Kevin Kenny's remarks. Thanks! file: [1b3222a95c] check-in: [1997a15ffd] user: jan.nijtmans branch: tip-389, size: 54226 | |
|
2018-01-31
| ||
| 12:18 | Change Tcl_Token definition (int -> size_t). Many related code-changes. file: [2fdbede656] check-in: [e3a724b790] user: jan.nijtmans branch: memory-API, size: 51867 | |
|
2018-01-29
| ||
| 11:36 | merge trunk file: [81d5d091e6] check-in: [3faa71ab4f] user: jan.nijtmans branch: memory-API, size: 51864 | |
|
2018-01-10
| ||
| 08:27 | merge core-8-branch file: [feb7253c86] check-in: [b3fc2fbe3d] user: jan.nijtmans branch: tip-389, size: 53860 | |
| 08:25 | Fix 00a27923ee: (Tcl part, remaining is in Tk) text/entry dysfunctional when pasting an emoji on MacOSX. This changes the handling of incoming valid 4-byte UTF-8 sequences: Those are no longer split in 4 separate characters (as was done for invalid byte sequences) but replaced by a single ' replacement character' . file: [ae6a7420d8] check-in: [8481a52495] user: jan.nijtmans branch: core-8-branch, size: 51831 | |
|
2018-01-09
| ||
| 11:15 | (partial) fix for 00a27923ee: text/entry dysfunctional when pasting an emoji on MacOSX. Don't handle incoming valid 4-byte UTF-8 characters as invalid byte sequences (since they aren't), but as being the Unicode replacement character. file: [2e1cc8edda] check-in: [f0adfe7dac] user: jan.nijtmans branch: bug-00a27923ee, size: 50872 | |
|
2017-12-28
| ||
| 18:49 | Fix handling of surrogates (when TCL_UTF_MAX > 3) in Tcl_UtfNcmp()/Tcl_UtfNcasecmp()/TclUtfCasecmp(). Backported from core-8-branch, where this was fixed already. file: [06ae0e4cdb] check-in: [1ebc1bcaa5] user: jan.nijtmans branch: core-8-6-branch, size: 50632 | |
|
2017-12-01
| ||
| 15:03 | merge trunk file: [104e15d362] check-in: [e008d0adce] user: jan.nijtmans branch: memory-API, size: 51624 | |
| 11:33 | merge core-8-branch file: [335ddcbc61] check-in: [dabd924a87] user: jan.nijtmans branch: tip-389, size: 53620 | |
|
2017-11-30
| ||
| 13:14 | Fix [8e1e31eac0]: lsort treats NUL chars strangely. Also fix various initializations, which only make a difference when TCL_UTF_MAX == 4. Add new test-cases which demonstrate the fix. For TCL_UTF_MAX == 4, surrogates will now be handled as expected as well when sorting. file: [8c0c76d552] check-in: [b6438b69ad] user: jan.nijtmans branch: core-8-branch, size: 51591 | |
|
2017-11-29
| ||
| 12:28 | merge core-8-6-branch file: [143baa2d32] check-in: [e45dcdac38] user: jan.nijtmans branch: core-8-branch, size: 50421 | |
| 12:27 | Fix Tcl_UtfFindFirst()/Tcl_UtfFindLast(), which were broken by [83c0c569d6]. Not detected, because those functions aren't used anywhere in Tcl. So, added new test-cases, makeing sure this doesn't happen again. file: [047d99f7af] check-in: [d906b55e4b] user: jan.nijtmans branch: core-8-6-branch, size: 50495 | |
| 11:49 | Merge core-8-branch. Also, use a different value for TCL_STUB_MAGIC when TCL_UTF_MAX>4. file: [8697ae0536] check-in: [1916b6a72e] user: jan.nijtmans branch: tip-389, size: 53414 | |
| 11:05 | merge core-8-6-branch file: [32d38162fe] check-in: [8976a447aa] user: jan.nijtmans branch: core-8-branch, size: 50415 | |
| 11:04 | Update some functions in tclUtf.c to handle surrogate pairs when TCL_UTF_MAX == 4. Also update documentation to distinguish better between "Tcl_UniChar" and "Unicode character": Those are not necessary the same when TCL_UTF_MAX == 4. No change when TCL_UTF_MAX == 4 or TCL_UTF_MAX == 6. file: [a8a9b0c072] check-in: [83c0c569d6] user: jan.nijtmans branch: core-8-6-branch, size: 50489 | |
| 09:49 | Fix [8e1e31eac0]: lsort treats NUL chars strangely file: [3634cc925c] check-in: [e2a6110884] user: jan.nijtmans branch: tip-389, size: 53032 | |
| 08:59 | Treat invalid UTF-8 characters in the range 0x80-0x9F as cp1252: See https://en.wikipedia.org/wiki/UTF-8. To be added to TIP #389 file: [e2b60d7d03] check-in: [b2521a4844] user: jan.nijtmans branch: tip-389, size: 51439 | |
|
2017-11-20
| ||
| 12:58 | merge core-8-branch file: [4746e08149] check-in: [1a3cef7d5c] user: jan.nijtmans branch: tip-389, size: 50879 | |
|
2017-11-17
| ||
| 16:08 | merge core-8-branch. Fix some Tcl_UniChar initialization, in case TCL_UTF_MAX == 4 file: [e1dd3d468b] check-in: [37b4aab687] user: jan.nijtmans branch: tip-389, size: 50849 | |
|
2017-11-16
| ||
| 13:50 | Handle Tcl_UtfAtIndex/Tcl_UniCharAtIndex for extended index range. More field fixes. file: [f289098ae1] check-in: [78e8c0c5f6] user: jan.nijtmans branch: memory-API, size: 49361 | |
|
2017-11-07
| ||
| 12:15 | Somewhat simplified implementation of TIP #389, in which the "string length" if characters > U+FFFF is considered to be 2, not 1. file: [dbf9778521] check-in: [d224d38a6d] user: jan.nijtmans branch: tip-389, size: 50453 | |
|
2017-09-10
| ||
| 13:35 | merge novem file: [e3b0ac4884] check-in: [4fa9cc2903] user: jan.nijtmans branch: novem-more-memory-API, size: 49345 | |
|
2017-09-01
| ||
| 08:50 | merge trunk file: [812d0a432b] check-in: [3f9db43f3e] user: jan.nijtmans branch: tip-389-impl, size: 51044 | |
|
2017-08-29
| ||
| 09:19 | Merge trunk file: [a37dead009] check-in: [f2f6504adb] user: jan.nijtmans branch: novem, size: 49347 | |
|
2017-08-25
| ||
| 13:46 | Merge trunk file: [0b140acd38] check-in: [c841475a89] user: jan.nijtmans branch: tip-389-impl, size: 51586 | |
|
2017-08-18
| ||
| 22:06 | merge core-8-6-branch file: [719d2c9f48] check-in: [75da8b29f8] user: jan.nijtmans branch: trunk, size: 49382 | |
|
2017-07-03
| ||
| 12:32 | merge core-8-6-branch file: [acc04c27ac] check-in: [4467b7768e] user: jan.nijtmans branch: rfe-6c0d7aec67, size: 49456 | |
| 08:27 | 'inline static' -> 'static inline' and 'INLINE' -> 'inline', for consistancy. file: [e3eb4abccf] check-in: [5b95f585fa] user: jan.nijtmans branch: core-8-6-branch, size: 48209 | |
|
2017-06-13
| ||
| 12:18 | merge core-8-6-branch file: [bd5a76e562] check-in: [8eec477ed9] user: jan.nijtmans branch: rfe-6c0d7aec67, size: 49449 | |
|
2017-06-09
| ||
| 12:20 | merge trunk file: [51c99fdd3f] check-in: [ee2c5cf945] user: jan.nijtmans branch: tip-389-impl, size: 50991 | |
| 11:48 | merge novem file: [ca0264252a] check-in: [f687ce6a36] user: jan.nijtmans branch: novem-more-memory-API, size: 48062 | |
|
2017-06-08
| ||
| 12:59 | merge trunk file: [a2c1bb7c45] check-in: [7b30d63181] user: jan.nijtmans branch: novem, size: 48100 | |
| 12:38 | merge core-8-6-branch file: [fbcc907f13] check-in: [9053c4a54f] user: jan.nijtmans branch: trunk, size: 48135 | |
| 12:37 | Fix [2738427]: Tcl_NumUtfChars(...) no overflow check. file: [fc2a8e3d34] check-in: [6b5843fde9] user: jan.nijtmans branch: core-8-6-branch, size: 48216 | |
| 12:34 | Fix [2738427]: Tcl_NumUtfChars(...) no overflow check. file: [1f59446a7f] check-in: [e376f4b734] user: jan.nijtmans branch: core-8-5-branch, size: 46889 | |
| 08:26 | Better UTF-8 surrogate handling, only functional when TCL_UTF_MAX>3 file: [ab16f203f3] check-in: [5ae46a0093] user: jan.nijtmans branch: rfe-6c0d7aec67, size: 49268 | |
|
2017-06-06
| ||
| 14:46 | Merge trunk file: [4ac6e3f55e] check-in: [b03462a529] user: jan.nijtmans branch: tip-389-impl, size: 50671 | |
| 09:25 | merge trunk file: [5c1140810c] check-in: [6c9ae4580b] user: jan.nijtmans branch: novem, size: 47957 | |
| 09:24 | merge core-8-6-branch file: [5eabb6b0c9] check-in: [28a2ae3999] user: jan.nijtmans branch: trunk, size: 47992 | |
| 09:23 | Follow-up to [67aa9a2070]: Use uppercase consistantly, slight optimization in character tests, comment fixes. No change in functionality. file: [0164f065d3] check-in: [5b20178e5a] user: jan.nijtmans branch: core-8-6-branch, size: 48073 | |
| 09:02 | [67aa9a2070] Tcl_UtfToUniChar returns single byte for invalid UTF-8 input as documented. file: [3a98891648] check-in: [3998688718] user: jan.nijtmans branch: core-8-5-branch, size: 46746 | |
|
2017-05-31
| ||
| 12:51 | merge trunk file: [69395d50cc] check-in: [f07a85fee0] user: jan.nijtmans branch: tip-389-impl, size: 50309 | |
| 12:06 | merge core-8-6-branch file: [10fb9f5ff7] check-in: [c30eb96495] user: jan.nijtmans branch: trunk, size: 47983 | |
| 12:05 | Fix [67aa9a2070]: Security: Invalid UTF-8 can inject unexpected characters file: [f1b6c68212] check-in: [f1b9559259] user: jan.nijtmans branch: sebres-8-6-clock-speedup-cr1, size: 48064 | |
|
2017-05-29
| ||
| 15:38 | merge trunk file: [e7e769c463] check-in: [6d946ddf77] user: jan.nijtmans branch: tip-389-impl, size: 50354 | |
|
2016-08-30
| ||
| 13:57 | merge trunk file: [12de15edf6] check-in: [2a530e021b] user: jan.nijtmans branch: tip-389-impl, size: 49592 | |
| 13:12 | merge trunk file: [1e04544386] check-in: [11d4243035] user: jan.nijtmans branch: novem, size: 47810 | |
| 13:07 | Don't ever allow UTF-8 sequences of more than 4 characters to be generated or parsed, even when TCL_UTF_MAX>4: According to current Unicode standard, a byte string of >4 characters can never form a single UTF-8 character. And a few minor micro-optimizations related to UTF-8 handling. file: [3ed37d3c92] check-in: [2de1551609] user: jan.nijtmans branch: trunk, size: 47845 | |
| 13:00 | Don't ever allow UTF-8 sequences of more than 4 characters to be generated or parsed, even when TCL_UTF_MAX>4: According to current Unicode standard, a byte string of >4 characters can never form a single UTF-8 character. And a few minor micro-optimizations related to UTF-8 handling. file: [f38d3847d1] check-in: [c0a65532a7] user: jan.nijtmans branch: core-8-6-branch, size: 47926 | |
|
2016-04-08
| ||
| 14:25 | Merge trunk. Add new bitflags to tclStringRep.h (not used yet) file: [84f00ec8a7] check-in: [2d87e13575] user: jan.nijtmans branch: tip-389-impl, size: 50615 | |
|
2016-04-05
| ||
| 12:07 | merge trunk file: [24271b6f0f] check-in: [9edc83a71a] user: dgp branch: novem, size: 47851 | |
| 09:32 | Rename UtfCount() to TclUtfCount() and use it in more places. Suggested by pspjuth here: [e99a79a32650e7e5] file: [a677bf3b17] check-in: [9c4e4beddb] user: jan.nijtmans branch: trunk, size: 47886 | |
|
2016-03-22
| ||
| 16:09 | Update all Unicode tables to version 9.0 beta file: [63578ee9bf] check-in: [203b8dcd6c] user: jan.nijtmans branch: tip-389-impl, size: 50696 | |
|
2015-10-24
| ||
| 02:37 | merge changes from pspjuth that optimize conversion from unichar to utf and add optimized versions for reading a word from byte codes. file: [74420169a8] check-in: [e99a79a326] user: kbk branch: drh-micro-optimization, size: 47955 | |
|
2015-09-22
| ||
| 08:20 | merge trunk file: [157972ae30] check-in: [eb6c2fe41b] user: jan.nijtmans branch: novem, size: 47932 | |
|
2015-09-02
| ||
| 08:44 | merge trunk file: [35f36782ec] check-in: [50dc66790e] user: jan.nijtmans branch: tip-389-impl, size: 50698 | |
|
2015-09-01
| ||
| 18:54 | Various Unicode handling enhancements, when building with TCL_UTF_MAX > 3, inspired by androwish. No effect if TCL_UTF_MAX=3 (which is the default) file: [09f91effbd] check-in: [e2278643dc] user: jan.nijtmans branch: trunk, size: 47967 | |
|
2014-05-01
| ||
| 09:40 | Merge trunk. Update Unicode tables to Unicode 7.0 beta. file: [e3d12611ee] check-in: [13a1d81916] user: jan.nijtmans branch: tip-389-impl, size: 51750 | |
|
2013-09-26
| ||
| 13:13 |
merge novem
WARNING: No checks of build-ability done yet. file: [ece5863523] check-in: [2688d65077] user: dkf branch: novem-64bit-sizes, size: 47059 | |
|
2013-08-02
| ||
| 10:33 | merge trunk file: [8df87bb7d1] check-in: [396ccb299c] user: jan.nijtmans branch: novem, size: 47028 | |
|
2013-07-29
| ||
| 10:12 | Make sure that "string is space \u202f" will continue to return "1", even if in future Unicode this character (NARROW_NO_BREAK_SPACE) will cease to be a space. See: http://www.unicode.org/review/pri249/. Don't hardcode "tclWinError.o" for Cygwin file: [33f59c69c1] check-in: [a72287aa7d] user: jan.nijtmans branch: trunk, size: 47063 | |
| 09:29 | Make sure that "string is space \u202f" will continue to return "1", even if in future Unicode this character (NARROW_NO_BREAK_SPACE) will cease to be a space. See: http://www.unicode.org/review/pri249/ file: [c78a847600] check-in: [334ab96e5e] user: jan.nijtmans branch: core-8-5-branch, size: 46902 | |
|
2013-07-08
| ||
| 18:56 | Unbreak MSVC6 debug build (thanks Andreas Kupries!) file: [7f126daa4a] check-in: [b259e31b93] user: jan.nijtmans branch: novem, size: 46998 | |
| 18:56 | Unbreak MSVC6 debug build (thanks Andreas Kupries!) file: [6df9363dfd] check-in: [d369017148] user: jan.nijtmans branch: trunk, size: 47033 | |
| 18:55 | Unbreak MSVC6 debug build (thanks Andreas Kupries!) file: [cdb7d01000] check-in: [728fb2f25b] user: jan.nijtmans branch: core-8-5-branch, size: 46872 | |
|
2013-06-18
| ||
| 13:46 | merge trunk file: [2c2fb3838a] check-in: [e6a93ffe18] user: jan.nijtmans branch: tip-389-impl, size: 51727 | |
| 11:50 | merge trunk file: [9bb2674df5] check-in: [bc4d6bb1d4] user: jan.nijtmans branch: novem, size: 46991 | |
|
2013-06-17
| ||
| 13:29 | Use more portable TclIsSpaceProc() in stead of isspace(). file: [d724f20e08] check-in: [4bfe3111b1] user: jan.nijtmans branch: trunk, size: 47026 | |
| 13:23 | Use more portable TclIsSpaceProc() in stead of isspace(). Make sure that "string is space \u180e" continues to return 1 for whatever unicode version. file: [8bab297239] check-in: [5712054958] user: jan.nijtmans branch: core-8-5-branch, size: 46865 | |
|
2013-06-10
| ||
| 07:36 | Update Unicode tables to Unicode 6.3 beta. Merge trunk file: [928ed7508d] check-in: [9e9232f06e] user: jan.nijtmans branch: tip-389-impl, size: 51749 | |
|
2013-05-28
| ||
| 07:25 | merge trunk file: [fda007828f] check-in: [03d94f0a34] user: jan.nijtmans branch: novem, size: 47013 | |
|
2013-05-22
| ||
| 13:07 | [3613609]: Replace strcasecmp() with UTF-8-aware version. file: [4f663c3e2e] check-in: [89f027f118] user: dkf branch: trunk, size: 47048 | |
| 12:55 | Fixed the weird edge case. file: [27e7fb5619] check-in: [93dd8bb33b] user: dkf branch: bug-3613609, size: 46831 | |
|
2013-05-21
| ||
| 09:38 | Slight improvement: if cs = "\xC0\x80" and ct = "\x00", loop would continue after NUL-byte, this should not happen. file: [a8a8d1c5ce] check-in: [a765f37f78] user: jan.nijtmans branch: bug-3613609, size: 46535 | |
| 09:27 | Proposed solution for 3613609: lsort -nocase does not sort non-ASCII correctly file: [a1c47ea4d8] check-in: [66c30c4369] user: jan.nijtmans branch: bug-3613609, size: 46506 | |
|
2013-05-06
| ||
| 09:08 | Change Tcl_UtfNcmp and friend's signature to use size_t in stead of unsigned long. This is potentially binary incompatible on win64, but not on any other platform. It eliminates the need for special stub-wrappers on Cygwin64 for those functions. "novem" doesn't promise binary compatibility anyway. file: [a57017e4db] check-in: [9bb59c6083] user: jan.nijtmans branch: novem, size: 46130 | |
|
2013-02-25
| ||
| 16:03 | Merge trunk. Unicode 6.3 does not have that many spaces..... file: [af50b567c7] check-in: [0e5e775003] user: jan.nijtmans branch: tip-389-impl, size: 50866 | |
| 13:52 | merge trunk. Update all unicode tables to current state of Unicode 6.3 (not released yet) file: [7dacd44f65] check-in: [2fffdb3621] user: jan.nijtmans branch: novem, size: 46165 | |
| 13:38 | Merge trunk. Upgrade all tables to Unicode 6.3 (not released yet) file: [9f57826203] check-in: [adf88dd045] user: jan.nijtmans branch: tip-389-impl, size: 50907 | |
| 13:16 | For Unicode 6.3, mongolian vowel separator (U+180e) is nominated to change character class from Space to Control character. Make sure that "string is space" will continue to return 1 for this character. See TIP #413. file: [5ce9c3041c] check-in: [b553432c31] user: jan.nijtmans branch: trunk, size: 46165 | |
|
2012-11-29
| ||
| 21:25 | Purge remnants of support for compilers ignorant of C keyword 'inline'. file: [27ec49e846] check-in: [3b2b41c66a] user: dgp branch: novem, size: 46130 | |
|
2012-11-18
| ||
| 17:21 | ... and even more file: [a79e50140e] check-in: [7a9b06162b] user: dkf branch: novem-64bit-sizes, size: 46126 | |
|
2012-11-16
| ||
| 14:15 | More work done. Still I am finding places where int should be size_t and reverse. file: [95e82b324a] check-in: [d348e679f7] user: jan.nijtmans branch: novem-64bit-sizes, size: 46165 | |
| 12:38 |
merge "novem"
Everything compiles now, but it doesn't run yet. file: [c5f05186c5] check-in: [005a09e2be] user: jan.nijtmans branch: novem-64bit-sizes, size: 46172 | |
|
2012-10-09
| ||
| 01:48 |
merge trunk
Dont include U+0082 and U+0083 in the Tcl space set file: [4734351403] check-in: [227a4f0b70] user: jan.nijtmans branch: tip-318-update, size: 46130 | |
|
2012-09-23
| ||
| 16:48 | tip 318 update file: [0020a81e19] check-in: [f09c1bc377] user: jan.nijtmans branch: tip-318-update, size: 46195 | |
|
2012-09-21
| ||
| 09:27 | new Tcl_UniCharIsWhitespace function file: [bc71fd9e19] check-in: [7a11f2ec28] user: jan.nijtmans branch: tip-318-update, size: 46696 | |
|
2012-07-03
| ||
| 19:02 | Feature branch to explore making use of the Hoehrmann UTF-8 decoder. http://bjoern.hoehrmann.de/utf-8/decoder/dfa/ file: [22aafac1f8] check-in: [8e353afbe6] user: dgp branch: dgp-hoehrmann-decoder, size: 46264 | |
|
2012-02-03
| ||
| 21:25 | merge trunk file: [8252ed9abd] check-in: [ffb46b40ba] user: jan.nijtmans branch: tip-389-impl, size: 50721 | |
|
2012-01-26
| ||
| 21:58 | merge trunk file: [b92935e908] check-in: [702fdde8e2] user: jan.nijtmans branch: tip-389-impl, size: 50711 | |
|
2012-01-23
| ||
| 22:19 | merge trunk file: [8b3efc8560] check-in: [9eab8924a0] user: jan.nijtmans branch: tip-389-impl, size: 50555 | |
| 09:38 | merge trunk file: [d3c3ef5c3a] check-in: [41d9eb81af] user: dkf branch: dkf-utf16-branch, size: 45980 | |
|
2012-01-22
| ||
| 21:50 | [Frq 3473670]: Various Unicode-related file: [b14d5dc7be] check-in: [d772d08f8a] user: jan.nijtmans branch: trunk, size: 45979 | |
| 21:49 | [Frq 3473670]: Various Unicode-related speedups/robustness file: [821f56bcd2] check-in: [2ccfd0f771] user: jan.nijtmans branch: core-8-5-branch, size: 45948 | |
|
2012-01-16
| ||
| 13:35 | merge trunk file: [6b20262117] check-in: [0d8d161fe8] user: dkf branch: dkf-utf16-branch, size: 46056 | |
|
2012-01-14
| ||
| 09:30 | rfe-3473670: Various Unicode-related speedups/robustness file: [b972e779b5] check-in: [92168a99c1] user: jan.nijtmans branch: rfe-3473670, size: 46004 | |
|
2012-01-09
| ||
| 20:34 | [Bug 3464428] string is graph \u0120 is wrong file: [f103b9e575] check-in: [e9a619e9dc] user: jan.nijtmans branch: trunk, size: 46055 | |
| 20:31 | [Bug 3464428] string is graph \u0120 is wrong file: [89865539a7] check-in: [14fc5c19b7] user: jan.nijtmans branch: core-8-5-branch, size: 46024 | |
| 19:59 | [Bug 3464428] string is graph \u0120 is wrong file: [8c9e89c5d6] check-in: [a0c0feafe9] user: jan.nijtmans branch: core-8-4-branch, size: 46080 | |
|
2011-12-31
| ||
| 15:12 | merge trunk file: [d7f87d218d] check-in: [e0cf8ae638] user: dkf branch: dkf-utf16-branch, size: 46758 | |
|
2011-12-24
| ||
| 00:30 | [Bug 3464428] string is graph \u0120 is wrong file: [0877da7330] check-in: [0c1ac83954] user: jan.nijtmans branch: trunk, size: 46757 | |
| 00:15 | [Bug 3464428] string is graph \u0120 is wrong file: [ff4f29c06b] check-in: [005fc77cde] user: jan.nijtmans branch: core-8-5-branch, size: 46726 | |
|
2011-12-23
| ||
| 23:31 | [Bug 3464428] string is graph \u0120 is wrong file: [e5c5f74bb0] check-in: [13071df962] user: jan.nijtmans branch: core-8-4-branch, size: 46880 | |
|
2011-08-24
| ||
| 08:32 | fix tests utf-2.8 and utf-2.9 file: [3520ee4534] check-in: [39ae4108bf] user: jan.nijtmans branch: tip-389-impl, size: 50669 | |
| 07:50 | Upcoming TIP implementation: Full support for Unicode 6.0 file: [e61b09f86a] check-in: [5721cf9ae6] user: jan.nijtmans branch: tip-389-impl, size: 50601 | |
|
2011-07-27
| ||
| 10:40 | Start work towards being able to work with utf8 fully and utf16 and other things outside the BMP. file: [2a2f356f38] check-in: [f9f8c8425c] user: dkf branch: dkf-utf16-branch, size: 46774 | |
|
2011-04-28
| ||
| 16:02 | More isspace() callers. file: [4469252ff0] check-in: [41acfe91ea] user: dgp branch: trunk, size: 46773 | |
| 16:00 | More isspace() callers. file: [11a1eed14a] check-in: [88095bbde0] user: dgp branch: core-8-5-branch, size: 46742 | |
|
2011-03-02
| ||
| 20:27 | Now that we're no longer using SCM based on RCS, the RCS Keyword lines cause more harm than good. Purged them (except in zlib files). file: [b95ca708fe] check-in: [c64f310d38] user: dgp branch: trunk, size: 46789 | |
| 16:06 | Now that we're no longer using SCM based on RCS, the RCS Keyword lines cause more harm than good. Purged them. file: [f3b9f73e39] check-in: [79367df0f0] user: dgp branch: core-8-5-branch, size: 46758 | |
|
2011-03-01
| ||
| 15:38 | Now that we're no longer using SCM based on RCS, the RCS Keyword lines cause more harm than good. Purged them. file: [515557ce50] check-in: [90b4acd7bd] user: dgp branch: core-8-4-branch, size: 46896 | |
|
2010-11-17
| ||
| 16:32 | Next slice, big chunk of the generic IO layer plus a few miscellanea. file: [daefb09659] check-in: [40a5fd2632] user: andreask branch: activestate-nre-excised-variant-1-roll-forward, size: 46856 | |
|
2009-09-07
| ||
| 16:39 | merge updates from HEAD file: [2922d7afd9] check-in: [8c0a6a5799] user: dgp branch: dgp-refactor, size: 46860 | |
| 07:28 |
* generic/tclExecute.c: fix potential uninitialized variable use and * generic/tclFCmd.c: null dereference flagged by clang static * generic/tclProc.c: analyzer. * generic/tclTimer.c: * generic/tclUtf.c: * generic/tclExecute.c: silence false positives from clang static * generic/tclIO.c: analyzer about potential null dereference. * generic/tclScan.c: * generic/tclCompExpr.c:file: [69316ce61d] check-in: [e93f957325] user: das branch: trunk, size: 46856 | |
|
2009-02-11
| ||
| 17:27 | merge updates from HEAD file: [41d4f6bc48] check-in: [f07460d448] user: dgp branch: dgp-refactor, size: 46856 | |
| 15:28 |
* generic/tclStringObj.c: Changed type of the 'allocated' field of the String struct from size_t to int since only int values are ever stored in it.file: [c4aba6267e] check-in: [93efedde3f] user: dgp branch: trunk, size: 46852 | |
|
2008-05-11
| ||
| 04:22 | merge updates from HEAD file: [899279da02] check-in: [b084fd8e3a] user: dgp branch: dgp-refactor, size: 46829 | |
|
2008-04-27
| ||
| 22:21 | Get rid of pre-C89-isms (esp. CONST vs const). file: [d58ad8109d] check-in: [2d205c22fb] user: dkf branch: trunk, size: 46825 | |
|
2005-11-03
| ||
| 17:52 | merge updates from HEAD file: [7ee2da14dc] check-in: [d827b9cf1e] user: dgp branch: dgp-refactor, size: 46829 | |
|
2005-10-31
| ||
| 15:59 | Convert to using ANSI decls/definitions and using the (ANSI) assumption that NULL can be cast to any pointer type transparently. file: [231e655e04] check-in: [1e0170d2bf] user: dkf branch: trunk, size: 46825 | |
|
2005-09-12
| ||
| 15:40 | merge updates from HEAD file: [c7c9e665ca] check-in: [156f19bcaf] user: dgp branch: dgp-refactor, size: 47247 | |
|
2005-09-09
| ||
| 18:48 |
[kennykb-numerics-branch] Merge updates from HEAD.file: [b74363491e] check-in: [343239eeff] user: dgp branch: kennykb-numerics-branch, size: 47247 | |
|
2005-09-07
| ||
| 15:31 |
* generic/tclUtf.c (Tcl_UniCharToUtf): Corrected handling of negative * tests/utf.test (utf-1.5): Tcl_UniChar input value. Incorrect handling was producing byte sequences outside of Tcl's legal internal encoding. [Bug 1283976].file: [2830ca0d80] check-in: [c76f2a1966] user: dgp branch: trunk, size: 47243 | |
| 14:35 |
* generic/tclUtf.c (Tcl_UniCharToUtf): Corrected handling of negative * tests/utf.test (utf-1.5): Tcl_UniChar input value. Incorrect handling was producing byte sequences outside of Tcl's legal internal encoding. [Bug 1283976].file: [4d21dfc317] check-in: [8d8a47a587] user: dgp branch: core-8-4-branch, size: 46967 | |
|
2005-08-02
| ||
| 18:14 | merge updates from HEAD file: [642f661d22] check-in: [10feab7c07] user: dgp branch: kennykb-numerics-branch, size: 47152 | |
|
2005-07-26
| ||
| 04:11 | Merge updates from HEAD file: [6b3a173352] check-in: [8351a734a6] user: dgp branch: dgp-refactor, size: 47152 | |
|
2005-07-21
| ||
| 14:38 | Systematizing the formatting file: [459bdd81b3] check-in: [ac613e6b94] user: dkf branch: trunk, size: 47148 | |
|
2005-06-13
| ||
| 01:45 |
*** MERGE WITH HEAD *** (tag msofer-wcodes-20050611)file: [abb0e59dd0] check-in: [d666b09ed5] user: msofer branch: msofer-wcodes-branch, size: 47149 | |
|
2005-05-10
| ||
| 18:33 | Merged kennykb-numerics-branch back to the head; TIPs 132 and 232 file: [05ebb1f08c] check-in: [1cc2336920] user: kennykb branch: trunk, size: 47146 | |
|
2005-05-05
| ||
| 17:55 | Merged with HEAD file: [f8567f610d] check-in: [b77c9a87c6] user: kennykb branch: kennykb-numerics-branch, size: 47150 | |
|
2005-05-04
| ||
| 17:34 | merge updates from HEAD file: [4702b80d6d] check-in: [edf99c3880] user: dgp branch: dgp-refactor, size: 47146 | |
|
2005-05-03
| ||
| 18:07 |
* doc/DString.3: Eliminated use of identifier "string" in Tcl's * doc/Environment.3: public C API to avoid conflict/confusion with * doc/Eval.3: the std::string of C++. * doc/ExprLong.3, doc/ExprLongObj.3, doc/GetInt.3, doc/GetOpnFl.3: * doc/ParseCmd.3, doc/RegExp.3, doc/SetResult.3, doc/StrMatch.3: * doc/Utf.3, generic/tcl.decls, generic/tclBasic.c, generic/tclEnv.c: * generic/tclGet.c, generic/tclParse.c, generic/tclParseExpr.c: * generic/tclRegexp.c, generic/tclResult.c, generic/tclUtf.c: * generic/tclUtil.c, unix/tclUnixChan.c: * generic/tclDecls.h: `make genstubs`file: [bfdb1027d4] check-in: [83aa957ebe] user: dgp branch: trunk, size: 47142 | |
|
2003-10-16
| ||
| 02:28 | Merged updates from HEAD file: [0a87a53c0a] check-in: [44102608b1] user: dgp branch: dgp-refactor, size: 46872 | |
|
2003-10-08
| ||
| 14:24 | Made Tcl_NumUtfChars do the right thing with \u0000 when guessing the length because of a negative 'length' parameter. [Bug 769812] file: [4bafe028d7] check-in: [6b243da1f0] user: dkf branch: trunk, size: 46868 | |
| 14:21 | Made Tcl_NumUtfChars do the right thing with \u0000 when guessing the length because of a negative 'length' parameter. [Bug 769812] file: [d09ebf9cdd] check-in: [257a93c349] user: dkf branch: core-8-4-branch, size: 46872 | |
|
2003-03-06
| ||
| 23:27 |
* generic/TclUtf.c (Tcl_UniCharNcasecmp): Corrected failure to * tests/utf.test (utf-25.*): properly compare Unicode strings of different case in a case insensitive manner. [Bug 699042]file: [386fe22320] check-in: [a7fde7d55a] user: dgp branch: trunk, size: 46896 | |
| 23:24 |
* generic/TclUtf.c (Tcl_UniCharNcasecmp): Corrected failure to * tests/utf.test (utf-25.*): properly compare Unicode strings of different case in a case insensitive manner. [Bug 699042]file: [bb62f8f756] check-in: [8003bbacd1] user: dgp branch: core-8-4-branch, size: 46900 | |
|
2003-02-18
| ||
| 02:25 |
* generic/tclExecute.c (TclExecuteByteCode INST_STR_MATCH): * generic/tclCmdMZ.c (Tcl_StringObjCmd STR_MATCH): * generic/tclUtf.c (TclUniCharMatch): * generic/tclInt.decls: add private TclUniCharMatch function that * generic/tclIntDecls.h: does string match on counted unicode * generic/tclStubInit.c: strings. Tcl_UniCharCaseMatch has the * tests/string.test: failing that it can't handle strings or * tests/stringComp.test: patterns with embedded NULLs. Added tests that actually try strings/pats with NULLs. TclUniCharMatch should be TIPed and made public in the next minor version rev.file: [cac938b6c2] check-in: [28dcdcf39e] user: hobbs branch: trunk, size: 46835 | |
|
2002-11-12
| ||
| 02:26 |
* generic/tclUtf.c: make use of TclUtfToUniChar macro throughout the functions, and add extra optimization to Tcl_NumUtfChars for one-byte/char case.file: [78e68b4785] check-in: [af7f25d96a] user: hobbs branch: trunk, size: 42114 | |
|
2002-08-20
| ||
| 20:25 | merged with trunk at tag macosx-8-4-merge-2002-08-20-trunk file: [7443fa11a6] check-in: [354986d9c3] user: das branch: macosx-8-4-branch, size: 41944 | |
|
2002-08-05
| ||
| 03:24 |
* doc/CmdCmplt.3: Applied Patch 585105 to fully CONST-ify * doc/Concat.3: all remaining public interfaces of Tcl. * doc/CrtCommand.3: Notably, the parser no longer writes on * doc/CrtSlave.3: the string it is parsing, so it is no * doc/CrtTrace.3: longer necessary for Tcl_Eval() to be * doc/Eval.3: given a writable string. Also, the * doc/ExprLong.3: refactoring of the Tcl_*Var* routines * doc/LinkVar.3: by Miguel Sofer is included, so that the * doc/ParseCmd.3: "part1" argument for them no longer needs * doc/SetVar.3: to be writable either. * doc/TraceVar.3: * doc/UpVar.3: Compatibility support has been enhanced so * generic/tcl.decls that a #define of USE_NON_CONST will remove * generic/tcl.h all possible source incompatibilities with * generic/tclBasic.c the 8.3 version of the header file(s). * generic/tclCmdMZ.c The new #define of USE_COMPAT_CONST now does * generic/tclCompCmds.c what USE_NON_CONST used to do -- disable * generic/tclCompExpr.c only those new CONST's that introduce * generic/tclCompile.c irreconcilable incompatibilities. * generic/tclCompile.h * generic/tclDecls.h Several bugs are also fixed by this patch. * generic/tclEnv.c [Bugs 584051,580433] [Patches 585105,582429] * generic/tclEvent.c * generic/tclInt.decls * generic/tclInt.h * generic/tclIntDecls.h * generic/tclInterp.c * generic/tclLink.c * generic/tclObj.c * generic/tclParse.c * generic/tclParseExpr.c * generic/tclProc.c * generic/tclTest.c * generic/tclUtf.c * generic/tclUtil.c * generic/tclVar.c * mac/tclMacTest.c * tests/expr-old.test * tests/parseExpr.test * unix/tclUnixTest.c * unix/tclXtTest.c * win/tclWinTest.cfile: [b18e42ccdc] check-in: [e476c22fec] user: dgp branch: trunk, size: 41940 | |
|
2002-07-19
| ||
| 12:31 | Global symbols are now all either prefixed with 'tcl' (or 'Tcl' or ...) or have file-scope. file: [4126660f15] check-in: [86e27ff753] user: dkf branch: trunk, size: 44537 | |
|
2002-06-10
| ||
| 05:33 | Merging with TOT as of 06/09/2002. file: [ed60c139a6] check-in: [73b68fb238] user: wolfsuit branch: macosx-8-4-branch, size: 44540 | |
|
2002-05-30
| ||
| 03:27 |
* unix/configure: regen'ed * unix/configure.in: replaced bigendian check with autoconf standard AC_C_BIG_ENDIAN, which defined WORDS_BIGENDIAN on bigendian systems. * generic/tclUtf.c (Tcl_UniCharNcmp): * generic/tclInt.h (TclUniCharNcmp): use WORDS_BIGENDIAN instead of TCL_OPTIMIZE_UNICODE_COMPARE to enable memcmp alternative.file: [308dcb7710] check-in: [5a5c16e5a7] user: hobbs branch: trunk, size: 44533 | |
|
2002-05-29
| ||
| 10:35 | Made Tcl_UniCharNcmp faster on big-endian machines; the system memcmp()is probably optimized far in excess of anything we could do! Little-endian just use the old code... file: [384a26efe9] check-in: [b3535ea391] user: dkf branch: trunk, size: 44570 | |
| 09:09 |
* generic/tclInt.decls: * generic/tclIntDecls.h: * generic/tclStubInit.c: * generic/tclUtf.c: added TclpUtfNcmp2 private command that mirrors Tcl_UtfNcmp, but takes n in bytes, not utf-8 chars. This provides a faster alternative for comparing utf strings internally. (Tcl_UniCharNcmp, Tcl_UniCharNcasecmp): removed the explicit end of string check as it wasn't correct for the function (by doc and logic). * generic/tclCmdMZ.c (Tcl_StringObjCmd): reworked the string equal comparison code to use TclpUtfNcmp2 as well as short-circuit for equal objects or unequal length strings in the equal case. Removed the use of goto and streamlined the other parts. * generic/tclExecute.c (TclExecuteByteCode): added check for object equality in the comparison instructions. Added short-circuit for != length strings in INST_EQ, INST_NEQ and INST_STR_CMP. Reworked INST_STR_CMP to use TclpUtfNcmp2 where appropriate, and only use Tcl_UniCharNcmp when at least one of the objects is a Unicode obj with no utf bytes.file: [03dbd6b1d6] check-in: [c78da914be] user: hobbs branch: trunk, size: 44362 | |
|
2002-02-08
| ||
| 02:52 |
* Partial TIP 27 rollback. Following routines restored to return (char *): Tcl_DStringAppend, Tcl_DStringAppendElement, Tcl_JoinPath, Tcl_TranslateFileName, Tcl_ExternalToUtfDString, Tcl_UtfToExternalDString, Tcl_UniCharToUtfDString, Tcl_GetCwd, Tcl_WinTCharToUtf. Also restored Tcl_WinUtfToTChar to return (TCHAR *) and Tcl_UtfToUniCharDString to return (Tcl_UniChar *). Modified some callers. This change recognizes that Tcl_DStrings are de-facto white-box objects. * generic/tclCmdMZ.c: corrected use of C++-style comment.file: [8496aab82e] check-in: [bb1a244cde] user: dgp branch: trunk, size: 43084 | |
|
2002-02-05
| ||
| 02:21 | Merging with the current TOT. Very few conflicts, mostly in the generated files. file: [2bd859f5d7] check-in: [f469a31a06] user: wolfsuit branch: macosx-8-4-branch, size: 43105 | |
|
2002-01-26
| ||
| 01:10 |
* Sought out and eliminated instances of CONST-casting that are no longer needed after the TIP 27 effort.file: [e7d378c3b1] check-in: [4bca1d26db] user: dgp branch: trunk, size: 43096 | |
|
2002-01-17
| ||
| 03:03 |
* Updated APIs in generic/tclUtf.c and generic/tclRegexp.c according to the guidelines of TIP 27. Updated callers.file: [73d8335019] check-in: [17ade15700] user: dgp branch: trunk, size: 43105 | |
|
2002-01-02
| ||
| 13:52 | Fixed fault with case-insensitive string matching (Bug#233257) and rewrote some tests to test what they claimed to be testing. file: [608df3bfe8] check-in: [99e550c5be] user: dkf branch: trunk, size: 43103 | |
|
2001-10-16
| ||
| 05:31 | Undo of mistaken commit. Sorry! file: [c911d35dc6] check-in: [092e282e8d] user: dgp branch: trunk, size: 43048 | |
| 05:10 |
| |
|
2001-09-27
| ||
| 13:49 | Backing out unwise changes file: [35b0daf5fe] check-in: [4c0c25f627] user: dkf branch: dkf-64bit-support-branch, size: 43052 | |
|
2001-09-26
| ||
| 14:23 | Now builds on Solaris8/SPARC with both SunPro CC *and* GCC. file: [af6d32ff69] check-in: [4850711173] user: dkf branch: dkf-64bit-support-branch, size: 43066 | |
|
2001-09-13
| ||
| 19:33 |
* generic/tclUtf.c (Tcl_UtfPrev): corrected to return the proper location when the middle of a UTF-8 byte was passed in. [Bug #450504]file: [08c7f3b96f] check-in: [5b19d3b799] user: hobbs branch: core-8-3-1-branch, size: 38202 | |
| 19:31 |
* generic/tclUtf.c (Tcl_UtfPrev): corrected to return the proper location when the middle of a UTF-8 byte was passed in. [Bug #450504]file: [0411922794] check-in: [5f02e9a154] user: hobbs branch: trunk, size: 43050 | |
|
2001-07-16
| ||
| 23:14 |
2001-07-02 Jeff Hobbs <[email protected]>
* tests/util.test: added util-4.6 * generic/tclUtil.c (Tcl_ConcatObj): Corrected walking backwards over utf-8 chars. [Bug #227512] 2001-06-27 Jeff Hobbs <[email protected]> * generic/tclUtf.c (Tcl_UtfBackslash): Corrected backslash handling of multibyte utf-8 chars. [Bug #217987] * generic/tclCmdIL.c (InfoProcsCmd): fixed potential mem leak in info procs that created objects without using them. * generic/tclCompCmds.c (TclCompileStringCmd): fixed mem leak when string command failed to parse the subcommand. 2001-05-22 Jeff Hobbs <[email protected]> * generic/tclObj.c (TclAllocateFreeObjects): simplified objSizePlusPadding to use sizeof(Tcl_Obj) (max)file: [b1e3fb83a7] check-in: [bef6467977] user: hobbs branch: core-8-3-1-branch, size: 38256 | |
|
2001-06-28
| ||
| 01:10 |
* tests/subst.test: * generic/tclUtf.c (Tcl_UtfBackslash): Corrected backslash handling of multibyte utf-8 chars. [Bug #217987]file: [32045ec19e] check-in: [7eba85bf31] user: hobbs branch: trunk, size: 43023 | |
|
2001-04-06
| ||
| 10:50 | Fixed problem with [string compare \x00 \x01] and hopefully sped the command up in a few cases too (notably byte arrays and UNICODE objects.) [Bug #219201] file: [ca14319b76] check-in: [6677432d73] user: dkf branch: trunk, size: 42633 | |
|
2000-06-05
| ||
| 23:36 | Comment typo correction. file: [7d4c2af295] check-in: [68d103692c] user: ericm branch: trunk, size: 42731 | |
|
2000-05-08
| ||
| 22:06 | removed unreferenced var file: [60f04b3f80] check-in: [5482dfcb56] user: hobbs branch: trunk, size: 42731 | |
| 21:59 |
* doc/Utf.3: * generic/tclStubInit.c: * generic/tcl.decls: * generic/tclDecls.h: * generic/tclUtf.c: Added new functions Tcl_UniCharNcasecmp and Tcl_UniCharCaseMatch (unicode parallel to Tcl_StringCaseMatch) * generic/tclUtil.c: rewrote Tcl_StringCaseMatch algorithm for optimization and made Tcl_StringMatch just call Tcl_StringCaseMatchfile: [7a3ece8e99] check-in: [52c8e2d16d] user: hobbs branch: trunk, size: 42750 | |
|
2000-01-11
| ||
| 22:08 |
* generic/tclUtf.c: changed Tcl_UtfBackslash to not allow non-octal digits (8,9) in \ooo substs. [Bug: 3975] * generic/tcl.h: noted need to change win/tcl.m4 and tools/tclSplash.bmp for minor version changesfile: [072d41fd6b] check-in: [f664655d33] user: hobbs branch: trunk, size: 37858 | |
|
1999-07-22
| ||
| 01:08 | * doc/Utf.3: * generic/tcl.decls: * generic/tclInt.decls: * generic/tclDecls.h: * generic/tclIntDecls.h: * generic/tclUtf.c: * compat/strftime.c: * unix/tclUnixTime.c: Changed function declarations in non-platform-specific APIs to use "unsigned long" instead of "size_t", which may not be defined on certain compilers (rather than include sys/types.h, which may not exist). file: [1063baf081] check-in: [22b143003b] user: redman branch: trunk, size: 37736 | |
|
1999-06-24
| ||
| 03:27 |
* unix/Makefile.in: Changed install-doc to install-man.
* tools/uniParse.tcl: * tools/uniClass.tcl: * tools/README: * tests/string.test: * generic/regc_locale.c: * generic/tclUniData.c: * generic/tclUtf.c: * doc/string.n: Updated Unicode character tables to reflect latest Unicode 2.1 data. Also rationalized "regexp" and "string is" definitions of character classes. file: [c48c2fa19e] check-in: [9d26b83359] user: stanton branch: trunk, size: 37717 | |
|
1999-06-02
| ||
| 20:21 | * generic/tclUtf.c (Tcl_UtfNcasecmp): Fixed incorrect computation of relative ordering. [Bug: 2135] file: [5cabcf8d5a] check-in: [58260de12d] user: stanton branch: trunk, size: 37568 | |
|
1999-05-22
| ||
| 01:20 | Merged changes from scriptics-tclpro-1-3-b2 branch file: [f55a65074e] check-in: [f692388d07] user: stanton branch: trunk, size: 37532 | |
|
1999-05-20
| ||
| 23:40 | lint in comments file: [10c8f0fe13] check-in: [9c2ceb7303] user: hershey branch: trunk, size: 34755 | |
| 00:03 |
Merged in various changes submitted by Jeff Hobbs:
* generic/tcl.decls: * generic/tclUtf.c: Added Tcl_UniCharIs* functions for control, graph, print, and punct classes. * generic/tclUtil.c: * doc/StrMatch.3: Added Tcl_StringCaseMatch() implementation to support case-insensitive globbing. * doc/string.n: * unix/mkLinks: * tests/string.test: * generic/tclCmdMZ.c: Added additional character class tests, added -nocase switch to "string match", changed string first/last to use offsets. file: [7706b36834] check-in: [61e23396c3] user: stanton branch: scriptics-tclpro-1-3-b2-branch, size: 37579 | |
|
1999-05-06
| ||
| 18:46 | * doc/Utf.3: * generic/tclStubInit.c: * generic/tclDecls.h: * generic/tclUtf.c: * generic/tcl.decls: Added Tcl_UtfNcmp and Tcl_UtfNcasecmp. file: [fbdeeb90bf] check-in: [d9e7608a90] user: stanton branch: trunk, size: 34798 | |
|
1999-04-30
| ||
| 16:22 | Tcl_UtfToUpper and Tcl_UtfToTitle now works on badly formed Utf strings. file: [2f760fc34e] check-in: [8cc1ac5095] user: hershey branch: trunk, size: 32513 | |
|
1999-04-29
| ||
| 00:04 | fixed part of bug 1791: Tcl_UtfToUpper and Tcl_UtfToLower now work on badly formed Utf strings. file: [5737c04290] check-in: [a9c361524c] user: hershey branch: trunk, size: 32180 | |
|
1999-04-21
| ||
| 00:42 | Deleted: Added comments for 4/19 and 4/20 check-in: [6c6fc5d7b2] user: redman branch: scriptics-tclpro-1-2-old, size: 0 | |
|
1999-04-16
| ||
| 00:46 | Added: merged tcl 8.1 branch back into the main trunk file: [3f7d7e0b3e] check-in: [f3b32fb71c] user: stanton branch: trunk, size: 30544 | |
|
1999-04-02
| ||
| 23:44 | * generic/regc_locale.c: * generic/regcustom.h: * generic/tcl.decls: * generic/tclCmdIL.c: * generic/tclCmdMZ.c: * generic/tclInt.h: * generic/tclRegexp.c: * generic/tclScan.c: * generic/tclTest.c: * generic/tclUtf.c: * win/tclWinFCmd.c: * win/tclWinFile.c: Made various Unicode utility functions public. The following functions were made public and added to the stubs table: Tcl_UtfToUniCharDString, Tcl_UniCharToUtfDString, Tcl_UniCharLen, Tcl_UniCharNcmp, Tcl_UniCharIsAlnum, Tcl_UniCharIsAlpha, Tcl_UniCharIsDigit, Tcl_UniCharIsLower, Tcl_UniCharIsSpace, Tcl_UniCharIsUpper, Tcl_UniCharIsWordChar file: [f80e00437b] check-in: [de06484e63] user: stanton branch: core-8-1-branch-old, size: 30548 | |
|
1998-11-04
| ||
| 04:39 |
Merged Henry's latest changes to add support for character ranges
in cvec data type
Added support for Unicode character classes in regular expressions We now support the following character classes: alnum, alpha, blank, cntrl, digit, graph, lower, print, punct, space, upper, xdigit These all follow the example set by the GNU regular expression package for Java except that "digit" only matches the ASCII '0'-'9' characters. Renamed tclUtf.h to tclUniData.c file: [e9daa1c432] check-in: [6a4b8e2ee0] user: stanton branch: core-8-1-branch-old, size: 30526 | |
|
1998-10-21
| ||
| 20:57 | fixed typo in include file: [c75a37c3d7] check-in: [078afb3c89] user: stanton branch: core-8-1-branch-old, size: 30447 | |
|
1998-10-16
| ||
| 01:16 |
Added Unicode character table support:
added TclUniCharIsWordChar
tclCmdMZ.c (Tcl_StringObjCmd): added "totitle" subcommand, changed "wordend" and "wordstart" to properly handle Unicode word characters and connector punctuationfile: [e7c8a7e302] check-in: [a448adffae] user: stanton branch: core-8-1-branch-old, size: 30447 | |
|
1998-10-03
| ||
| 01:56 | replaced SCCS with RCS strings file: [f4f9e2686e] check-in: [c65ae5da0d] user: stanton branch: core-8-1-branch-old, size: 30133 | |
|
1998-09-21
| ||
| 23:39 | Added: Added contents of Tcl 8.1a2 file: [988d81327c] check-in: [8c56dc8868] user: stanton branch: core-8-1-branch-old, size: 30108 | |