summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* patch 8.0.0395: testing the + register fails with Motifv8.0.0395Bram Moolenaar2017-03-014-7/+13
| | | | | | Problem: Testing the + register fails with Motif. Solution: Also ignore the "failed to create input context" error in the second gvim. Don't use msg() when it would result in a dialog.
* patch 8.0.0394: tabs are not aligned when scrolling horizontallyv8.0.0394Bram Moolenaar2017-03-016-5/+60
| | | | | | | | Problem: Tabs are not aligned when scrolling horizontally and a Tab doesn't fit. (Axel Bender) Solution: Handle a Tab as a not fitting character. (Christian Brabandt) Also fix that ":redraw" does not scroll horizontally to show the cursor. And fix the test that depended on the old behavior.
* patch 8.0.0393: order of duplicate tags is not preservedv8.0.0393Bram Moolenaar2017-03-013-28/+58
| | | | | | | Problem: When the same tag appears more than once, the order is unpredictable. (Charles Campbell) Solution: Besides using a dict for finding duplicates, use a grow array for keeping the tags in sequence.
* patch 8.0.0392: GUI test fails with Athena and Motifv8.0.0392Bram Moolenaar2017-03-016-0/+58
| | | | | | Problem: GUI test fails with Athena and Motif. Solution: Add test_ignore_error(). Use it to ignore the "failed to create input context" error.
* patch 8.0.0391: arabic support is verbose and not well testedv8.0.0391Bram Moolenaar2017-03-013-344/+465
| | | | | Problem: Arabic support is verbose and not well tested. Solution: Simplify the code. Add more tests.
* patch 8.0.0390: when the window scrolls the popup menu may be garbledv8.0.0390Bram Moolenaar2017-03-012-16/+21
| | | | | | Problem: When the window scrolls horizontally when the popup menu is displayed part of it may not be cleared. (Neovim issue #6184) Solution: Remove the menu when the windows scrolled. (closes #1524)
* patch 8.0.0389: test for arabic does not check what is displayedv8.0.0389Bram Moolenaar2017-02-282-24/+67
| | | | | | Problem: Test for arabic does not check what is displayed. Solution: Improve what is asserted. (Dominique Pelle, closes #1523) Add a first shaping test.
* patch 8.0.0388: filtering lines changes foldsv8.0.0388Bram Moolenaar2017-02-283-11/+37
| | | | | | | Problem: filtering lines through "cat", without changing the line count, changes manual folds. Solution: Change how marks and folds are adjusted. (Matthew Malcomson, from neovim #6194.
* patch 8.0.0387: compiler warningsv8.0.0387Bram Moolenaar2017-02-283-3/+6
| | | | | Problem: compiler warnings Solution: Add type casts. (Christian Brabandt)
* patch 8.0.0386: tiny build has a problem with generating the options testv8.0.0386Bram Moolenaar2017-02-272-5/+8
| | | | | Problem: Tiny build has a problem with generating the options test. Solution: Change the "if" to skip over statements.
* patch 8.0.0385: no tests for arabicv8.0.0385Bram Moolenaar2017-02-274-1/+97
| | | | | Problem: No tests for arabic. Solution: Add a first test for arabic. (Dominique Pelle, closes #1518)
* patch 8.0.0384: timer test failed for no apparent reasonv8.0.0384Bram Moolenaar2017-02-272-0/+3
| | | | | Problem: Timer test failed for no apparent reason. Solution: Mark the test as flaky.
* patch 8.0.0383: misplaced #ifdefv8.0.0383Bram Moolenaar2017-02-272-1/+4
| | | | | Problem: Misplaced #ifdef. (Christ van Willigen) Solution: Split assignment.
* patch 8.0.0382: warning in tiny build for unused variablev8.0.0382Bram Moolenaar2017-02-272-1/+12
| | | | | Problem: Warning in tiny build for unused variable. (Tony Mechelynck) Solution: Add #ifdefs.
* patch 8.0.0381: diff mode is not sufficiently testedv8.0.0381Bram Moolenaar2017-02-262-0/+77
| | | | | Problem: Diff mode is not sufficiently tested. Solution: Add more diff mode tests. (Dominique Pelle, closes #1515)
* patch 8.0.0380: with 'linebreak' double wide char wraps badlyv8.0.0380Bram Moolenaar2017-02-263-3/+35
| | | | | | | Problem: With 'linebreak' set and 'breakat' includes ">" a double-wide character results in "<<" displayed. Solution: Check for the character not to be replaced. (Ozaki Kiichi, closes #1456)
* patch 8.0.0379: CTRL-Z and mouse click use CTRL-O unnecessaryv8.0.0379Bram Moolenaar2017-02-263-4/+8
| | | | | Problem: CTRL-Z and mouse click use CTRL-O unnecessary. Solution: Remove stuffing CTRL-O. (James McCoy, closes #1453)
* patch 8.0.0378: possible overflow when reading corrupted undo filev8.0.0378Bram Moolenaar2017-02-262-4/+5
| | | | | Problem: Another possible overflow when reading corrupted undo file. Solution: Check if allocated size is not too big. (King)
* patch 8.0.0377: possible overflow when reading corrupted undo filev8.0.0377Bram Moolenaar2017-02-262-2/+5
| | | | | Problem: Possible overflow when reading corrupted undo file. Solution: Check if allocated size is not too big. (King)
* patch 8.0.0376: size computations in spell file reading are offv8.0.0376Bram Moolenaar2017-02-262-2/+4
| | | | | Problem: Size computations in spell file reading are not exactly right. Solution: Make "len" a "long" and check with LONG_MAX.
* patch 8.0.0375: the "+ register is not testedv8.0.0375Bram Moolenaar2017-02-262-0/+43
| | | | | | Problem: The "+ register is not tested. Solution: Add a test using another Vim instance to change the "+ register. (Kazuki Kuriyama)
* patch 8.0.0374: invalid memory access when using :sc in Ex modev8.0.0374Bram Moolenaar2017-02-264-2/+21
| | | | | Problem: Invalid memory access when using :sc in Ex mode. (Dominique Pelle) Solution: Avoid the column being negative. Also fix a hang in Ex mode.
* patch 8.0.0373: build fails without +foldingv8.0.0373Bram Moolenaar2017-02-252-1/+3
| | | | | Problem: Build fails without +folding. Solution: Move misplaced #ifdef.
* patch 8.0.0372: more options are not always definedv8.0.0372Bram Moolenaar2017-02-253-39/+137
| | | | | Problem: More options are not always defined. Solution: Consistently define all possible options.
* patch 8.0.0371: leaking memory when setting v:completed_itemv8.0.0371Bram Moolenaar2017-02-252-1/+3
| | | | | Problem: Leaking memory when setting v:completed_item. Solution: Or the flags instead of setting them.
* patch 8.0.0370: invalid memory access when setting wildchar emptyv8.0.0370Bram Moolenaar2017-02-254-3/+19
| | | | | | | Problem: Invalid memory access when setting wildchar empty. Solution: Avoid going over the end of the option value. (Dominique Pelle, closes #1509) Make option test check all number options with empty value.
* patch 8.0.0369: a few options are not defined, depending on featuresv8.0.0369Bram Moolenaar2017-02-253-8/+28
| | | | | | | | Problem: The 'balloondelay', 'ballooneval' and 'balloonexpr' options are not defined without the +balloon_eval feature. Testing that an option value fails does not work for unsupported options. Solution: Make the options defined but not supported. Don't test if setting unsupported options fails.
* patch 8.0.0368: not all options are tested with a range of valuesv8.0.0368Bram Moolenaar2017-02-255-1/+216
| | | | | Problem: Not all options are tested with a range of values. Solution: Generate a test script from the source code.
* patch 8.0.0367: types in include files may be inconsistentv8.0.0367Bram Moolenaar2017-02-256-8/+12
| | | | | | Problem: If configure defines _LARGE_FILES some include files are included before it is defined. Solution: Include vim.h first. (Sam Thursfield, closes #1508)
* patch 8.0.0366: build fails with tiny featuresv8.0.0366Bram Moolenaar2017-02-252-0/+4
| | | | | Problem: Build fails with tiny features. Solution: Add #ifdef.
* patch 8.0.0365: might free a dict item that wasn't allocatedv8.0.0365Bram Moolenaar2017-02-2512-50/+40
| | | | | | Problem: Might free a dict item that wasn't allocated. Solution: Call dictitem_free(). (Nikolai Pavlov) Use this for b:changedtick.
* patch 8.0.0364: ]s does not move cursor with two spell errors in one linev8.0.0364Bram Moolenaar2017-02-255-6/+30
| | | | | | | Problem: ]s does not move cursor with two spell errors in one line. (Manuel Ortega) Solution: Don't stop search immediately when wrapped, search the line first. (Ken Takata) Add a test.
* patch 8.0.0363: Travis is too slow to keep up with patchesv8.0.0363Bram Moolenaar2017-02-232-1/+3
| | | | | Problem: Travis is too slow to keep up with patches. Solution: Increase git depth to 20
* patch 8.0.0362: tests fail on MS-Windowsv8.0.0362Bram Moolenaar2017-02-232-2/+4
| | | | | Problem: Tests fail on MS-Windows. Solution: Use $*.vim instead of $<.
* patch 8.0.0361: GUI initialisation is not sufficiently testedv8.0.0361Bram Moolenaar2017-02-2311-49/+156
| | | | | Problem: GUI initialisation is not sufficiently tested. Solution: Add the gui_init test. (Kazuki Kuriyama)
* patch 8.0.0360: sometimes VimL is used instead of "Vim script"v8.0.0360Bram Moolenaar2017-02-2315-30/+32
| | | | | Problem: Sometimes VimL is used, which is confusing. Solution: Consistently use "Vim script". (Hirohito Higashi)
* patch 8.0.0359: 'number' and 'relativenumber' are not properly testedv8.0.0359Bram Moolenaar2017-02-239-103/+288
| | | | | | Problem: 'number' and 'relativenumber' are not properly tested. Solution: Add tests, change old style to new style tests. (Ozaki Kiichi, closes #1447)
* patch 8.0.0358: invalid memory access in C-indent codev8.0.0358Bram Moolenaar2017-02-233-1/+11
| | | | | Problem: Invalid memory access in C-indent code. Solution: Don't go over end of empty line. (Dominique Pelle, closes #1492)
* patch 8.0.0357: crash when setting 'guicursor' to weird valuev8.0.0357Bram Moolenaar2017-02-233-3/+13
| | | | | Problem: Crash when setting 'guicursor' to weird value. Solution: Avoid negative size. (Dominique Pelle, closes #1465)
* patch 8.0.0356: leaking memory when setting 'ttytype'v8.0.0356Bram Moolenaar2017-02-232-0/+3
| | | | | Problem: Leaking memory when setting 'ttytype'. Solution: Get free_oldval from the right option entry.
* patch 8.0.0355: using uninitialized memory when 'isfname' is emptyv8.0.0355Bram Moolenaar2017-02-233-10/+17
| | | | | | Problem: Using uninitialized memory when 'isfname' is empty. Solution: Don't call getpwnam() without an argument. (Dominique Pelle, closes #1464)
* patch 8.0.0354: test to check that setting termcap key fails sometimesv8.0.0354Bram Moolenaar2017-02-232-1/+3
| | | | | Problem: Test to check that setting termcap key fails sometimes. Solution: Check for "t_k1" to exist. (Christian Brabandt, closes #1459)
* patch 8.0.0353: if [RO] is translated it may be truncatedv8.0.0353Bram Moolenaar2017-02-232-1/+3
| | | | | | Problem: If [RO] in the status line is translated to a longer string, it is trunctted to 4 bytes. Solution: Skip over the resulting string. (Jente Hidskes, closes #1499)
* patch 8.0.0352: not easy to see when a typval needs to be clearedv8.0.0352Bram Moolenaar2017-02-232-24/+18
| | | | | | Problem: The condition for when a typval needs to be cleared is too complicated. Solution: Init the type to VAR_UNKNOWN and clear it always.
* patch 8.0.0351: no test for concatenating an empty stringv8.0.0351Bram Moolenaar2017-02-232-0/+7
| | | | | | Problem: No test for concatenating an empty string that results from out of bounds indexing. Solution: Add a simple test.
* patch 8.0.0350: not enough test coverage for Perlv8.0.0350Bram Moolenaar2017-02-232-6/+108
| | | | | Problem: Not enough test coverage for Perl. Solution: Add more Perl tests. (Dominique Perl, closes #1500)
* patch 8.0.0349: redrawing errors with GTK 3v8.0.0349Bram Moolenaar2017-02-232-3/+20
| | | | | | Problem: Redrawing errors with GTK 3. Solution: When updating, first clear all rectangles and then draw them. (Kazunobu Kuriyama, Christian Ludwig, closes #848)
* patch 8.0.0348: using shadow dir on Mac lack +clipboardv8.0.0348Bram Moolenaar2017-02-222-1/+3
| | | | | | Problem: When building with a shadow directory on macOS lacks the +clipboard feature. Solution: Link *.m files, specifically os_macosx.m. (Kazunobu Kuriyama)
* patch 8.0.0347: when using completion comment leader wont workv8.0.0347Bram Moolenaar2017-02-213-0/+24
| | | | | | Problem: When using CTRL-X CTRL-U inside a comment, the use of the comment leader may not work. (Klement) Solution: Save and restore did_ai. (Christian Brabandt, closes #1494)
* patch 8.0.0346: Vim relies on limits.h to be included indirectlyv8.0.0346Bram Moolenaar2017-02-213-10/+6
| | | | | | Problem: Vim relies on limits.h to be included indirectly, but on Solaris 9 it may not be. (Ben Fritz) Solution: Always include limits.h.