http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c1 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- URL| |http://bugs.ghostscript.com | |/show_bug.cgi?id=696317 CC| |jsmeix@suse.com, | |vuntz@suse.com Found By|--- |Development Flags| |needinfo?(vuntz@suse.com) --- Comment #1 from Johannes Meixner --- For example libspectre build fails currently with --------------------------------------------------------------------------- spectre-gs.c: In function 'critic_error_code': spectre-gs.c:46:9: error: 'e_Fatal' undeclared (first use in this function) case e_Fatal: ^ spectre-gs.c:46:9: note: each undeclared identifier is reported only once for each function it appears in spectre-gs.c:51:9: error: 'e_ExecStackUnderflow' undeclared (first use in this function) case e_ExecStackUnderflow: ^ spectre-gs.c: In function 'spectre_gs_process': spectre-gs.c:112:20: error: 'e_NeedInput' undeclared (first use in this function) error = error == e_NeedInput ? 0 : error; ^ --------------------------------------------------------------------------- I would like to assign this issue for now to the libspectre maintainers so that they could adapt libspectre for Ghostscript 9.18 but what the libspectre maintainers are seems contradicting: ----------------------------------------------------------------------- $ osc maintainer libspectre Defined in package: X11:common:Factory/libspectre bugowner of libspectre : KDEMaintainers maintainer of libspectre : group:gnome-maintainers ----------------------------------------------------------------------- Vincent Untz, because you are listed several times in libspectre.changes: Do you know who the right libspectre maintainer is? -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c3 Dominique Leuenberger changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |dimstar@opensuse.org Assignee|os.gnome.maintainers@gmail. |dimstar@opensuse.org |com | --- Comment #3 from Dominique Leuenberger --- (In reply to Johannes Meixner from comment #1)

$ osc maintainer libspectre Defined in package: X11:common:Factory/libspectre bugowner of libspectre : KDEMaintainers

maintainer of libspectre : group:gnome-maintainers -----------------------------------------------------------------------

That's what I call a joint effort: bugs owned by the KDE Team, package maintained by the GNOME Team :) -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c5 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Flags| |needinfo?(dimstar@opensuse. | |org) --- Comment #5 from Johannes Meixner --- Dominique Leuenberger, it seems under certain circumstances your testing code in the %prep section fails. To verify whether or not it works with ghostscript-9.18 I copied X11:common:Factory/libspectre to home:jsmeix/libspectre and Printing/ghostscript to home:jsmeix/ghostscript Currently I have this state: ----------------------------------------------------------------------------- $ osc results -v home:jsmeix libspectre | cut -b-140 \ | egrep 'Factory|Tumbleweed|13.2' openSUSE_Tumbleweed i586 succeeded openSUSE_Tumbleweed x86_64 succeeded(unpublished) openSUSE_Factory_ARM armv7l blocked: ghostscript-devel, ghostscript openSUSE_Factory i586 outdated (was: succeeded) openSUSE_Factory x86_64 succeeded openSUSE_13.2 i586 failed openSUSE_13.2 x86_64 failed ----------------------------------------------------------------------------- The interesting point is the build log on openSUSE_13.2 ----------------------------------------------------------------------------- $ osc rbl home:jsmeix libspectre openSUSE_13.2 x86_64 | less ... [ 42s] [114/115] cumulate ghostscript-9.18-1.1 [ 42s] [115/115] cumulate ghostscript-devel-9.18-1.1 ... [ 89s] Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.zliGto ... [ 89s] + IFS=. [ 89s] + read GS_MAJ GS_MIN [ 89s] ++ rpm -q --qf '%{version}' ghostscript-devel [ 89s] + IFS=' [ 89s] ' [ 89s] + '[' 9 18 -gt 9 -o 9 18 -eq 9 -a -gt 17 ']' [ 89s] /var/tmp/rpm-tmp.zliGto: line 40: [: too many arguments [ 89s] + exit 0 ... [ 101s] spectre-gs.c: In function 'critic_error_code': [ 101s] spectre-gs.c:46:9: error: 'e_Fatal' undeclared (first use in this function) [ 101s] case e_Fatal: [ 101s] ^ [ 101s] spectre-gs.c:46:9: note: each undeclared identifier is reported only once for each function it appears in [ 101s] spectre-gs.c:51:9: error: 'e_ExecStackUnderflow' undeclared (first use in this function) [ 101s] case e_ExecStackUnderflow: [ 101s] ^ [ 101s] spectre-gs.c: In function 'spectre_gs_process': [ 101s] spectre-gs.c:112:20: error: 'e_NeedInput' undeclared (first use in this function) [ 101s] error = error == e_NeedInput ? 0 : error; [ 101s] ^ ----------------------------------------------------------------------------- versus the build log on openSUSE_Factory ----------------------------------------------------------------------------- $ osc rbl home:jsmeix libspectre openSUSE_Factory x86_64 | less ... [ 59s] [127/128] cumulate ghostscript-9.18-1.1 [ 59s] [128/128] cumulate ghostscript-devel-9.18-1.1 ... [ 94s] Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.Tf0sPX ... [ 94s] + IFS=. [ 94s] + read GS_MAJ GS_MIN [ 94s] ++ rpm -q --qf '%{version}' ghostscript-devel [ 94s] + IFS=' [ 94s] ' [ 94s] + '[' 9 -gt 9 -o 9 -eq 9 -a 18 -gt 17 ']' [ 94s] + echo 'Patch #0 (libspectre-gs-9.18.patch):' [ 94s] Patch #0 (libspectre-gs-9.18.patch): [ 94s] + /usr/bin/cat /home/abuild/rpmbuild/SOURCES/libspectre-gs-9.18.patch [ 94s] + /usr/bin/patch -p1 --fuzz=0 [ 94s] patching file libspectre/spectre-gs.c [ 94s] + exit 0 ----------------------------------------------------------------------------- -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c7 --- Comment #7 from Dominique Leuenberger --- For arguments sake: The test works on Factory (see build log, still against 9.16) https://build.opensuse.org/build/X11:common:Factory/openSUSE_Factory/x86_64/... [ 66s] + IFS=. [ 66s] + read GS_MAJ GS_MIN [ 66s] ++ rpm -q --qf '%{version}' ghostscript-devel [ 66s] + IFS=' [ 66s] ' [ 66s] + '[' 9 -gt 9 -o 9 -eq 9 -a 16 -gt 17 ']' [ 66s] + exit 0 hence, if 13.2 support which does not like the read GS_MAJ GS_MIN is needed, resorting to zypper vcmp (or anything else that manages to properly compare versions) is needed... (we can't guarantee that GS won't ever do a strange thing like a hot fix 9.20a; so pure mathematical conversion will always be flawed) -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c9 --- Comment #9 from Johannes Meixner --- Even on my SLE11 system read X Y Z < <(echo foo bar baz) works: ------------------------------------------------------------------------ # read X Y Z < <(echo foo bar baz) # echo X:$X Y:$Y Z:$Z X:foo Y:bar Z:baz ------------------------------------------------------------------------ Accordingly on my SLE11 system the following works: ------------------------------------------------------------------------------- # rpm -q --qf "%{version}" cups 1.3.9 # ( IFS=. read X Y Z < <(rpm -q --qf "%{version}" cups) ; echo X:$X Y:$Y Z:$Z ) X:1 Y:3 Z:9 ------------------------------------------------------------------------------- and same on my SLE12 system: ------------------------------------------------------------------------------- # rpm -q --qf "%{version}" cups 1.7.5 # ( IFS=. read X Y Z < <(rpm -q --qf "%{version}" cups) ; echo X:$X Y:$Y Z:$Z ) X:1 Y:7 Z:5 ------------------------------------------------------------------------------- -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c11 --- Comment #11 from Johannes Meixner --- A general recommendation when doing read X Y Z < <(output of a command) it is fail-safe programming to always add a dummy variable where possibly trailing junk gets stored. On my SLE11 system (with bash-3.2): bad (probably "3.9" is unexpected): ----------------------------------------------------------------------------- # ( IFS=. read maj min < <(rpm -q --qf "%{version}" cups) ; \ echo maj:$maj min:$min ) maj:1 min:3.9 ----------------------------------------------------------------------------- good: ----------------------------------------------------------------------------- # ( IFS=. read maj min junk < <(rpm -q --qf "%{version}" cups) ; \ echo maj:$maj min:$min ) maj:1 min:3 ----------------------------------------------------------------------------- -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c13 --- Comment #13 from Johannes Meixner --- Dominique Leuenberger, what about testing for the actual requirement which is whether or not e_Fatal e_ExecStackUnderflow e_NeedInput are defined in a Ghostscript header file? Something like: for error in e_Fatal e_ExecStackUnderflow e_NeedInput ; \ do grep -q "$error" /usr/include/ghostscript/*.h || need_patch="yes" ; \ done if test -n "need_patch" ; then ... -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c15 --- Comment #15 from Johannes Meixner --- Typo in comment#13 - I meant if test -n "$need_patch" -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=953149 http://bugzilla.opensuse.org/show_bug.cgi?id=953149#c17 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(jsmeix@suse.com) | --- Comment #17 from Johannes Meixner ---

From my point of view building home:dimstar:Factory/libspectre works well.

-- You are receiving this mail because: You are on the CC list for the bug.