summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* sync win/rules.vc with Tcljan.nijtmans2022-10-131-2/+2
|
* Add one more call to update in unixWm-50.2. The CI runner says this works.culler2022-10-091-0/+1
|\
| * Tinker with unixWm-50.2.culler2022-10-091-0/+1
|/
* Add delays and updates to make tests pass on Ventura.culler2022-10-083-12/+45
|
* Fix [58e4a50642]: man.macros is neither tracked nor ignoredjan.nijtmans2022-10-061-0/+1
|
* Merge unstable_tests branch. This commit stabilizes a lot of tests that ↵fvogel2022-10-0212-1583/+1545
|\ | | | | | | could sporadically fail, mainly on macOS. No test results were changed, except for entry-5.7 and spinbox-5.7 which expected results were found to be wrong (these tests passes due to a race condition that is now fixed).
| * Temporarily restore previously existing constraints on unixEmbed-10.1 and ↵fvogel2022-10-021-2/+2
| | | | | | | | unixEmbed-10.2, so that the branch can be merged.
| * Revert previous commit, it does not seem to help in stabilizing the last ↵fvogel2022-10-021-1/+1
| | | | | | | | non-reproducible tests failures on macOS with aqua (xcode or clang), i.e. unixEmbed-10.1, unixEmbed-10.2, and (font-47.2, font-47.3, frame-12.3, textImage-4.2, these four always happening simultaneously).
| * Remove TCL_TIMER_EVENTS and TCL_DONT_WAIT flags introduced in [e0e4d873] ↵fvogel2022-09-251-1/+1
| | | | | | | | (see ticket [bfea11a916] discussing this on 2022-09-25 19:00:10).
| * Remove proc updateWidgets since it now simply runs 'update' on all platforms.fvogel2022-09-251-14/+0
| |
| * Remove proc updateWidgets since it now simply runs 'update' on all platforms.fvogel2022-09-251-427/+427
| |
| * Remove proc updateWidgets since it now simply runs 'update' on all platforms.fvogel2022-09-257-155/+155
| |
| * Restore unixEmbed-10.* to their state when they originally appeared in ↵fvogel2022-09-251-6/+4
| | | | | | | | [d28306f679].
| * Restore unixEmbed-10.1 and -10.2 as they are in core-8-6-branch. Now they ↵fvogel2022-09-251-6/+6
| | | | | | | | fail reliably on Linux, and on macOS at Github Actions (only - and not fully reliably).
| * Since [64999025af], frame-3.10 now fails at Github Actions on macOS with ↵fvogel2022-09-251-4/+10
| | | | | | | | aqua (xcode or clang), quite repeatably. Restore frame-3.10 as it is in core-8-6-branch.
| * Merge 8.6 and fix conflicts triggered by [a48b9196] (Fix [bfea11a916]: ↵fvogel2022-09-2511-88/+97
| |\ | | | | | | | | | update idletasks runs timer events on Aqua). We are on our way out of platform-specific update of widgets, yay!
| * \ merge 8.6fvogel2022-09-2042-197/+566
| |\ \
| * \ \ merge 8.6fvogel2022-08-3117-342/+2223
| |\ \ \
| * | | | Remove hardcoded values from unixEmbed-11.2fvogel2022-07-261-3/+4
| | | | |
| * | | | merge 8.6fvogel2022-07-255-46/+36
| |\ \ \ \
| * | | | | Further stabilize unixEmbed-11.2 by returning to update idletasks on the mac ↵fvogel2022-07-231-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | as it was originally.
| * | | | | Completely eradicate proc waitForMap and waitForUnmap in the test suite. The ↵fvogel2022-07-203-36/+2
| | | | | | | | | | | | | | | | | | | | | | | | tests stability problem was due to the TkMacOSXHandleMapOrUnmap() approach, now removed.
| * | | | | Remove stabilization lines in panedwindow.test. The problem was due to the ↵fvogel2022-07-201-30/+3
| | | | | | | | | | | | | | | | | | | | | | | | TkMacOSXHandleMapOrUnmap() approach, now removed.
| * | | | | merge core-8-6-branchfvogel2022-07-201-14/+12
| |\ \ \ \ \
| * | | | | | Restore comment line removed by error in the previous commit whil resolving ↵fvogel2022-07-191-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | conflicts.
| * | | | | | merge core-8-6-branchfvogel2022-07-197-99/+7
| |\ \ \ \ \ \
| * \ \ \ \ \ \ merge core-8-6-branchfvogel2022-07-1618-91/+122
| |\ \ \ \ \ \ \
| * | | | | | | | Fix my stupid mistake in previous commitfvogel2022-07-131-1/+1
| | | | | | | | |
| * | | | | | | | Attempt at fixing sporadic failures of panedwindow-23.30 at GitHub on macOS.fvogel2022-07-121-1/+4
| | | | | | | | |
| * | | | | | | | Instrument panedwindow-23.30 a bit more.fvogel2022-07-071-7/+15
| | | | | | | | |
| * | | | | | | | Further stabilize unixEmbed-11.2 because it still failed once at GitHub ↵fvogel2022-07-031-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua). [ef554a85] revealed that the .embed toplevel does not exist (even after waiting until the timeout), which may be due to an unmapped .main.f window being used in the -use option. Wait for .main.f to be mapped before creating the .embed toplevel.
| * | | | | | | | Stabilize ttk-2.8. This test was seen failing once at GitHub Actions, in ↵fvogel2022-07-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | branch unstable_tests, on macOS with xcode. Careful study of how this test works shows there is a race condition built in. This commit doesn't remove this race condition but should be enough to let the test always pass.
| * | | | | | | | Further ruggedize unixEmbed-11.2 because it still failed once at GitHub ↵fvogel2022-07-021-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua). Error message was 'bad window path name .embed' while executing 'button .embed.b -text 'Emb Button' -command {lappend result 'pushed .embed.b'}'
| * | | | | | | | Oops.fvogel2022-06-301-2/+1
| | | | | | | | |
| * | | | | | | | Instrument to understand why panedwindow-23.30 is still failing ↵fvogel2022-06-302-1/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (unstable_tests branch, macOS, xcode).
| * | | | | | | | resolve unintended forkfvogel2022-06-292-1/+1
| |\ \ \ \ \ \ \ \
| | * | | | | | | | Further stabilize panedwindow-23.30 since after [5952fd21] it failed once at ↵fvogel2022-06-292-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GitHub Actions in branch unstable_tests on macOS with clang (no, --enable-aqua).
| * | | | | | | | | In unixEmbed-11.2, make a difference in the output between the result of the ↵fvogel2022-06-281-3/+3
| |/ / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | winfo containing commands and the -command triggered.
| * | | | | | | | Stabilize textImage-3.2 since after [ce12f9ac] it failed once at GitHub ↵fvogel2022-06-281-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua).
| * | | | | | | | Further ruggedize unixEmbed-11.2 because it still failed once at GitHub ↵fvogel2022-06-282-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua). Even with [645a3198] the time out is hit (1 second in total), after which [winfo exists .embed.b] still is O. Try to update in more steps when creating toplevels or packs windows.
| * | | | | | | | Add more debug output in order to understand why panedwindow-23.30 still ↵fvogel2022-06-281-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | fails (seen once at GitHub on macOS with xcode in the unstable_tests branch).
| * | | | | | | | Further ruggedize panedwindow-23.30 since [20a906bf] was not enough. The ↵fvogel2022-06-211-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | test till failed once at GitHub Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua). [930cc27b] allowed to confirm the timeout (500 ms, in 10 steps of 50 ms) was hit ('waitForUnmap .f4: timeout!'). Double this time! And add more debug output when a timeout is hit.
| * | | | | | | | Further ruggedize unixEmbed-11.2 because it still failed once at GitHub ↵fvogel2022-06-211-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua). [930cc27b] allowed to confirm the timeout (500 ms, in 10 steps of 50 ms) was hit ('waitForMap .embed.b: timeout!'). Double this time!
| * | | | | | | | Stabilize notebook-5.1. This test once failed at least, at GitHub Actions in ↵fvogel2022-06-211-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | branch unstable_tests on macOS with clang (mem, --enable-aqua).
| * | | | | | | | Further ruggedize panedwindow-23.30 since [fe28ce65] was just plain wrong ↵fvogel2022-06-201-5/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | and the test failed once at GitHub Actions in branch unstable_tests on macOS with clang (no, --enable-aqua).
| * | | | | | | | Add debug output since unixEmbed-11.2 still failed once at GitHub Actions on ↵fvogel2022-06-191-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | macOS with clang (mem, --enable-aqua).
| * | | | | | | | Stabilize textImage-3.2 since after [ecc939fd] it failed once at GitHub ↵fvogel2022-06-181-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Actions in branch unstable_tests on macOS with xcode. What we really need is that the returned bbox be up-to-date at each of the loop steps, i.e. the line heights must have been recomputed, and this can be obtained by calling '.t sync' rather than calling 'update idletasks' after some delay.
| * | | | | | | | Further ruggedize panedwindow-23.30 since after [27e2d762] it failed once at ↵fvogel2022-06-181-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GitHub Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua).
| * | | | | | | | Further ruggedize unixEmbed-11.2 since after [768fe936] it failed once at ↵fvogel2022-06-181-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GitHub Actions in branch unstable_tests on macOS with clang (mem, --enable-aqua). Error was: ---- errorInfo: bad window path name ".embed.b" while executing "winfo ismapped $w" (procedure "waitForMap" line 3) invoked from within "waitForMap .embed.b" ("uplevel" body line 11) invoked from within "uplevel 1 $script" ---- errorCode: TK LOOKUP WINDOW .embed.b
| * | | | | | | | Restore what was working before previous commit.fvogel2022-06-151-1/+3
| | | | | | | | |
generated by cgit v0.12 at 2024-09-20 19:24:58 (GMT)