New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
unable to build docs locally #2529
Comments
I can't reproduce this myself, but I have Sphinx 1.2b3, not 1.2b1 (don't know if that makes a difference). The docstring in question is most likely this one: https://github.com/matplotlib/matplotlib/blob/master/lib/matplotlib/patches.py#L1285 But I'm not sure why it's failing for you and not me -- that's a Unicode string, and Sphinx should be generating everything in utf-8. If we could coax a traceback out of Sphinx to see where that encoding is happening, that might help. |
This is the identical error I reported back during the release candidate |
Can one of you apply this patch and see if that gives us a better traceback out of Sphinx? diff --git a/doc/make.py b/doc/make.py
index ab48fef..7cda626 100755
--- a/doc/make.py
+++ b/doc/make.py
@@ -142,7 +142,7 @@ def html():
options = "-D plot_formats=\"[('png', 80)]\""
else:
options = ''
- if os.system('sphinx-build %s -b html -d build/doctrees . build/html' % options):
+ if os.system('sphinx-build %s -T -b html -d build/doctrees . build/html' % options):
raise SystemExit("Building HTML failed.")
figures_dest_path = 'build/html/pyplots' |
This seems to be a windows vs linux problem: see pandas-dev/pandas#5245 for a similar error in pandas and windows |
I am a Linux user... never tried building docs on Windows. |
I was also on linux. I will try the patch on Monday when I'm back on my machine. I currently have intermittent internet access hence my need for local docs build. |
If you want a local copy of the docs, you can also check out the Michael Droettboom
|
The |
Maybe a more useful traceback from 1.2b3
|
When you say "-T" flag did not work, does that mean it did not give you an extended traceback? I was not expecting it to change behavior, only to provide a more detailed traceback. Perhaps that's what we're getting from 1.2b3 above, anyway. In any case, it's not very helpful -- the traceback leads us to a Around line 617 in
Could you try replacing it with
and sending me the output? |
Ah, that's what the T does. I thought maybe it was an encoding thing and didn't check. Helpful to know. Looks like a problem with numpydoc.
|
Numpydoc version.
|
Heh |
@mdboom Your patch works for this problem, so feel free to close this as you see fit. |
Heh. Sorry I didn't remember my own bug that I had reported. Obviously, I have the local patch, which is why it "Works for me[TM]". Anyway, thanks for reporting this and +1'ing the numpydoc bug. I'm going to keep this open and NOTE TO SELF to harass (in a friendly way) the numpydoc folks to get this done before the next matplotlib release. |
Closing this because the patch has been merged into |
Is this a local error? Was having the problem, so I updated to master and I still see it. haven't attempted to debug.
The text was updated successfully, but these errors were encountered: