Summary: | dev-tcltk/blt-2.4z-r13: shipped demos segfault with tcl/tk 8.6 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Fabio Rossi <rossi.f> |
Component: | Current packages | Assignee: | TCL/TK Project <tcltk> |
Status: | RESOLVED FIXED | ||
Severity: | normal | ||
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
See Also: |
https://bugs.debian.org/753476 https://bugs.gentoo.org/show_bug.cgi?id=783714 |
||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Fabio Rossi
2019-11-05 10:17:59 UTC
I'm retiring this package. Unless some one provide a working ebuild, it is gone in 30 days # Alfredo Tupone <tupone@gentoo.org> (2019-11-08) # Does not work with tcl/tk 8.6 without upstream # Removal in 30 days. dev-tcltk/blt Looking here https://wiki.tcl-lang.org/page/BLT it seems there are three unofficial ports to tcl-8.6, I should see what is available in the debian repositories and maybe align on that. What do you think? I am not a direct user of the package but is needed for other packages like ngspice[tcl] The last rite has been removed. However I tested BLT-2.5. It is not 8.6 compatible. I made it work and some demo actually work, but other segfault. e.g bltwish barchart1.tcl work but barchart2.tcl segfault I saw the freebsd port but patching is quite heavy. Other I have not tested. I would suggest to follow the patches of other distribution where there is someone working actively to keep alive the package. Otherwise we can decide to follow a fork, it seems tkblt is quite updated (last release is 23 Oct 2019 as available here https://sourceforge.net/projects/tkblt/files/) and they claim it is "a derived version of the BLT Toolkit, Tcl/Tk 8.5/8.6 TEA compatible". This last option requires of course a compatibility check of all packages in portage that need blt, here is a list: $ equery d -a dev-tcltk/blt * These packages depend on dev-tcltk/blt: dev-lang/python-2.7.9-r1 (tk ? dev-tcltk/blt) dev-lang/python-2.7.9-r2 (tk ? dev-tcltk/blt) dev-lang/python-2.7.10 (tk ? dev-tcltk/blt) dev-lang/python-2.7.10-r1 (tk ? dev-tcltk/blt) dev-lang/python-2.7.15 (tk ? dev-tcltk/blt:0) dev-lang/python-2.7.16 (tk ? dev-tcltk/blt:0) dev-lang/python-3.3.5-r1 (tk ? dev-tcltk/blt) dev-lang/python-3.4.0 (tk ? dev-tcltk/blt) dev-lang/python-3.4.1 (tk ? dev-tcltk/blt) dev-lang/python-3.4.2 (tk ? dev-tcltk/blt) dev-lang/python-3.4.3 (tk ? dev-tcltk/blt) dev-lang/python-3.4.3-r1 (tk ? dev-tcltk/blt) dev-lang/python-3.5.5 (tk ? dev-tcltk/blt:0) dev-lang/python-3.5.7 (tk ? dev-tcltk/blt:0) dev-lang/python-3.6.5 (tk ? dev-tcltk/blt:0) dev-lang/python-3.6.8 (tk ? dev-tcltk/blt:0) dev-lang/python-3.6.9 (tk ? dev-tcltk/blt:0) dev-lang/python-3.7.2 (tk ? dev-tcltk/blt:0) dev-lang/python-3.7.3 (tk ? dev-tcltk/blt:0) dev-lang/python-3.7.4-r1 (tk ? dev-tcltk/blt:0) dev-lang/python-3.7.5 (tk ? dev-tcltk/blt:0) dev-lang/python-3.7.5-r1 (tk ? dev-tcltk/blt:0) sci-astronomy/skycat-3.1.2-r2 (>=dev-tcltk/blt-2.4) sci-astronomy/skycat-3.1.3 (dev-tcltk/blt) sci-chemistry/nmrpipe-8.7.2015.320.15.26 (dev-tcltk/blt) sci-electronics/magic-7.5.202-r1 (dev-tcltk/blt) sci-electronics/magic-7.5.231 (dev-tcltk/blt) sci-electronics/magic-8.0.138 (dev-tcltk/blt) sci-electronics/magic-8.0.207 (dev-tcltk/blt) sci-electronics/ngspice-26 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-27 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-27 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-27-r1 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-27-r1 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-29 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-30 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-31 (tcl ? dev-tcltk/blt) sci-electronics/ngspice-31 (tcl ? dev-tcltk/blt) I have added the r14 with some fix. Could you please check if there are some regression (hope no) and some fixes TIA (In reply to Tupone Alfredo from comment #5) > I have added the r14 with some fix. > > Could you please check if there are some regression (hope no) and some fixes > > TIA I have tried running all the demos in /usr/lib64/blt2.4/demos Most demos now run but they exit with SEGFAULT after interacting with the mouse to zoom an area (i.e. clicking once, dragging the mouse to draw a rectangle and clicking again to finish the operation) or after pressing a button. Here is a list of examples that doesn't run at all: $ ./barchart2.tcl Error in startup script: invalid command name ".header.stacked" while executing ".header.stacked invoke" (file "./barchart2.tcl" line 184) $ ./barchart5.tcl Error in startup script: can't read "graph": no such variable while executing "$graph element create sin -relief raised -bd 1 -x x -y y -barwidth $barWidth" (file "./barchart5.tcl" line 104) $ ./graph2.tcl Segmentation fault $ ./graph5.tcl Segmentation fault $ ./graph7.tcl Segmentation fault $ ./spline.tcl Segmentation fault $ ./stripchart1.tcl Segmentation fault $ ./tabnotebook2.tcl Segmentation fault $ ./tabnotebook3.tcl Segmentation fault $ ./tabset2.tcl Segmentation fault $ ./tabset3.tcl Segmentation fault Here is a list of problems with other demos: $ ./bgexec1.tcl Segmentation fault when clicking on the quit button $ ./bgexec2.tcl ... it runs but when closing the window ... NULL main window while executing "table forget .l1" ("after" script) invalid command name ".l5" while executing ".l$name2 configure -text "$var($name2)"" (procedure "ShowResult" line 3) invoked from within "ShowResult var 5 w" (write trace on "var(5)") Error in startup script: can't invoke "label" command: application has been destroyed while executing "label .l$i " ("for" body line 2) invoked from within "for { set i 1 } { $i <= 20 } { incr i } { label .l$i table . .l$i $i,0 set pid [bgexec var($i) du /usr/include &] .l$i configure -tex..." (file "./bgexec2.tcl" line 37) $ ./bgexec3.tcl Segmentation fault when clicking on the quit button $ ./bgexec4.tcl Segmentation fault when clicking on the quit button $ ./bitmap.tcl Segmentation fault when clicking one of the buttons $ ./busy1.tcl Segmentation fault when clicking one of the buttons $ ./busy2.tcl Segmentation fault when clicking on the Hold button $ ./dnd2.tcl Segmentation fault when clicking on the Dismiss button $ ./eps.tcl Segmentation fault when clicking on the buttons $ ./graph1.tcl Segmentation fault when drawing rectangle for the zoom $ ./graph3.tcl Segmentation fault when drawing rectangle for the zoom $ ./graph4.tcl Segmentation fault when drawing rectangle for the zoom $ ./graph6.tcl Segmentation fault when drawing rectangle for the zoom $ ./hierbox2.tcl Segmentation fault when clicking on the buttons $ ./hierbox4.tcl Segmentation fault when clicking on the Test button $ ./hiertable1.tcl 160 entries Aborted $ ./hiertable2.tcl 160 entries Aborted $ ./treeview1.tcl 159 entries Error in startup script: can't find tag or id "c_files" in ".t" while executing ".t entry configure "c_files" -foreground green4" (file "./treeview1.tcl" line 181) My primary concern was about "other packages like ngspice[tcl]", if at least the one you use works or not I have added 2.5.3. I have tested some demo. The situation should be better now. Still some segfault but not so many (In reply to Tupone Alfredo from comment #8) > I have added 2.5.3. Thanks! There is a bug in the ebuild which creates /usr/$(get_libdir)/blt2.4/demos/bitmaps instead of *blt2.5* > I have tested some demo. The situation should be better now. > > Still some segfault but not so many I confirm the situation with the demos is definitely much better now, no more segfaults but other issues, e.g. some scripts return: X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 138 (RENDER) Minor opcode of failed request: 4 (RenderCreatePicture) Serial number of failed request: 130 Current serial number in output stream: 135 which, in according to what is written in https://bugzilla.redhat.com/show_bug.cgi?id=504388, seems related to xft fonts management of blt. I have fixed the bug with demos installed under blt2.4 blt-2.4z version is removed from the tree |