Summary: | dev-python/matplotlib-1.3.1 - src_compile(): virtualmake: the wrap_setup distutils-r1_python_compile failed. | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Dennis Schridde <dschridde+gentoobugs> |
Component: | [OLD] Development | Assignee: | Python Gentoo Team <python> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | jlec |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | build.log |
Description
Dennis Schridde
2014-10-26 09:03:21 UTC
Created attachment 387422 [details]
build.log
Not related to bug #504202, as I do not even have dev-python/wxpython installed. Maybe related: Xlib: extension "RANDR" missing on display ":1". (message early in the compile phase) The issue persists. Any ideas or suggestions how to fix or work around it? $ emerge =dev-python/matplotlib-1.3.1 -pv These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] x11-base/xorg-server-1.16.1:0/1.16.1 USE="glamor ipv6 nptl suid udev xorg xvfb* -dmx -doc -kdrive -minimal (-selinux) -static-libs -systemd -tslib -unwind -wayland -xnest" 0 KiB [ebuild R ] dev-python/matplotlib-1.3.1 USE="cairo gtk qt4 wxwidgets -doc -examples -excel -fltk -gtk3 -latex -pyside {-test} -tk" PYTHON_TARGETS="python2_7 python3_3 python3_4 (-python3_2)" 0 KiB Perhaps you can report the state of your emerged x11-base/xorg-server re USE flags. The key one is xvfb Also portage has 1.4.0 and 1.4.2, what happens with those? (In reply to Ian Delaney from comment #5) > Perhaps you can report the state of your emerged x11-base/xorg-server re USE > flags. The key one is xvfb $ emerge =dev-python/matplotlib-1.3.1 xorg-server -pv […] [ebuild R ] x11-base/xorg-server-1.16.1:0/1.16.1 USE="glamor ipv6 nptl suid systemd udev wayland xorg xvfb -dmx -doc -kdrive -minimal (-selinux) -static-libs -tslib -unwind -xnest" 0 KiB [ebuild R ] dev-python/matplotlib-1.3.1 USE="cairo gtk latex qt4 -doc -examples -excel -fltk -gtk3 -pyside {-test} -tk -wxwidgets" PYTHON_TARGETS="python2_7 python3_3 python3_4%* (-python3_2) (-python2_6%)" 0 KiB […] (In reply to Ian Delaney from comment #6) > Also portage has 1.4.0 and 1.4.2, what happens with those? $ equery depends matplotlib […] sci-mathematics/sage-6.3-r1 (>=dev-python/matplotlib-1.3.1[python_targets_python2_7(-)?,-python_single_target_python2_7(-)]) (<dev-python/matplotlib-1.4.0[python_targets_python2_7(-)?,-python_single_target_python2_7(-)]) This does occur for me with DISPLAY set to :1. Also fails without the output of
Xlib: extension "RANDR" missing ...
with DISPLAY set to ":0".
Also, by "portage has 1.4.0 and 1.4.2, what happens with those?" I meant can you build these versions, like so;
~/cvsPortage/gentoo-x86/dev-python/matplotlib $ PYTHON_TARGETS=python2_7 ebuild matplotlib-1.4.0.ebuild clean compile
* python2_7: running distutils-r1_run_phase python_compile_all
>>> Source compiled.
Normally new version that works trumps a broken past version however
(<dev-python/matplotlib-1.4.0[python_targets_python2_7(-)?,-python_single_target_python2_7(-)])
kinf of over-rides this.
jlec: Why are we wrapping the src_compile phase with virtualmake? Do we actually need and X server to build this? Looking at the ebuilds in the tree, it looks like this was removed in later ebuilds. Please try >=matplotlib-1.4.0. This works: # emerge -1a --nodeps \>matplotlib-1.4 These are the packages that would be merged, in order: [ebuild U ] dev-python/matplotlib-1.4.2 [1.3.1] PYTHON_TARGETS="python3_4%*" Can you please backport the change to matplotlib-1.3.1, as I need that version for sage? Please reopen if this problem still exists with version 1.4 (In reply to Justin Lecher from comment #12) > Please reopen if this problem still exists with version 1.4 As mentioned in comment #11, version 1.4.2 worked. |