Skip to content

Timezone bug in plot_date #18

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

Closed
wants to merge 227 commits into from
Closed

Conversation

faucon
Copy link
Contributor

@faucon faucon commented Feb 23, 2011

This is the patch is sent last week to the developers-list.
It fixes bug 3176823: "plot_date does not respect timezone"
(bug id from sourceforge bugtracker)

kind regards,
maximilian

mdboom and others added 30 commits July 6, 2010 15:18
…n with blitting.

svn path=/trunk/matplotlib/; revision=8515
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8514 | jdh2358 | 2010-07-06 10:48:31 -0400 (Tue, 06 Jul 2010) | 1 line
  
  update coding guide to point to 1.0 release branch
........
  r8517 | mdboom | 2010-07-06 11:30:34 -0400 (Tue, 06 Jul 2010) | 1 line
  
  Testing merging
........

svn path=/trunk/matplotlib/; revision=8518
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8519 | mdboom | 2010-07-06 11:31:53 -0400 (Tue, 06 Jul 2010) | 1 line
  
  Testing merging (removing bogus content)
........

svn path=/trunk/matplotlib/; revision=8520
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8521 | jdh2358 | 2010-07-06 10:48:16 -0500 (Tue, 06 Jul 2010) | 1 line
  
  update download site in site docs
........

svn path=/trunk/matplotlib/; revision=8522
svn path=/trunk/matplotlib/; revision=8523
… to set_major_formatter is due to an SVN merge error from the transforms branch.

svn path=/trunk/matplotlib/; revision=8538
svn path=/trunk/matplotlib/; revision=8540
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8524 | jdh2358 | 2010-07-06 17:54:21 -0400 (Tue, 06 Jul 2010) | 1 line
  
  add gridspec api to toc
........
  r8526 | mdboom | 2010-07-07 09:19:29 -0400 (Wed, 07 Jul 2010) | 2 lines
  
  Backport of r8515: Reset clipping upon exit of draw_markers to avoid negative interaction with blitting.
........
  r8539 | ryanmay | 2010-07-09 14:30:07 -0400 (Fri, 09 Jul 2010) | 1 line
  
  Fix setting of minor ticklabels. As far as I can tell, the extra call to set_major_formatter is due to an SVN merge error from the transforms branch.
........
  r8541 | leejjoon | 2010-07-13 12:39:40 -0400 (Tue, 13 Jul 2010) | 1 line
  
  Text.draw uses _set_gc_clip to set clip property
........

svn path=/trunk/matplotlib/; revision=8542
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8543 | weathergod | 2010-07-13 21:42:19 -0500 (Tue, 13 Jul 2010) | 2 lines
  
  Added documentation and examples for new, easier Axes3D use.
........

svn path=/trunk/matplotlib/; revision=8544
svn path=/trunk/matplotlib/; revision=8547
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8549 | efiring | 2010-07-14 07:37:10 -1000 (Wed, 14 Jul 2010) | 2 lines
  
  Axes.margin: bugfix--return correct values.  Thanks to Georges Schutz.
........

svn path=/trunk/matplotlib/; revision=8550
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8554 | efiring | 2010-07-15 07:34:42 -1000 (Thu, 15 Jul 2010) | 2 lines
  
  yaxis.set_ticks_position: fix bug with 'none'; thanks to Ben North
........

svn path=/trunk/matplotlib/; revision=8555
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8557 | efiring | 2010-07-15 10:45:19 -1000 (Thu, 15 Jul 2010) | 3 lines
  
  show for tkagg doesn't block in script called via ipython -pylab.
  This is the first in an expected series of changes to show().
........

svn path=/trunk/matplotlib/; revision=8558
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8559 | ianthomas23 | 2010-07-16 14:46:14 +0100 (Fri, 16 Jul 2010) | 2 lines
  
  Added tri* functions to pyplot docs.
........

svn path=/trunk/matplotlib/; revision=8560
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8559 | ianthomas23 | 2010-07-16 03:46:14 -1000 (Fri, 16 Jul 2010) | 2 lines
  
  Added tri* functions to pyplot docs.
........
  r8562 | efiring | 2010-07-16 10:44:52 -1000 (Fri, 16 Jul 2010) | 3 lines
  
  backends: factored out most of the show() code into ShowBase class.
  Also fixed various fltkagg problems.
........

svn path=/trunk/matplotlib/; revision=8563
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8562 | efiring | 2010-07-16 16:44:52 -0400 (Fri, 16 Jul 2010) | 3 lines
  
  backends: factored out most of the show() code into ShowBase class.
  Also fixed various fltkagg problems.
........
  r8564 | leejjoon | 2010-07-17 03:06:06 -0400 (Sat, 17 Jul 2010) | 1 line
  
  fix wrong baseline for multiple line legend
........

svn path=/trunk/matplotlib/; revision=8565
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8566 | dsdale | 2010-07-20 10:00:55 -0400 (Tue, 20 Jul 2010) | 2 lines
  
  Return Qt4's default cursor when leaving the canvas
........

svn path=/trunk/matplotlib/; revision=8567
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8568 | mdboom | 2010-07-23 09:17:15 -0400 (Fri, 23 Jul 2010) | 2 lines
  
  Prevent traceback when window icon can not be loaded in Gtk backend.
........

svn path=/trunk/matplotlib/; revision=8569
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8570 | mdboom | 2010-07-23 12:45:24 -0400 (Fri, 23 Jul 2010) | 2 lines
  
  Fix image clipping to a path and add a test.
........
  r8571 | mdboom | 2010-07-23 12:46:40 -0400 (Fri, 23 Jul 2010) | 2 lines
  
  Oops in last commit.
........

svn path=/trunk/matplotlib/; revision=8572
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8573 | efiring | 2010-07-24 11:59:55 -1000 (Sat, 24 Jul 2010) | 3 lines
  
  rcParams: don't include deprecated keys.
  Some other rc cleanups are included in this changeset.
........

svn path=/trunk/matplotlib/; revision=8574
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8575 | weathergod | 2010-07-25 17:24:05 -0500 (Sun, 25 Jul 2010) | 2 lines
  
  Fix "graph jumping" issue when rubber-banding a selection in GTK backends.
........

svn path=/trunk/matplotlib/; revision=8576
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8577 | efiring | 2010-07-26 08:17:18 -1000 (Mon, 26 Jul 2010) | 2 lines
  
  [3026430] delete duplicate test file with space in name
........

svn path=/trunk/matplotlib/; revision=8578
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8579 | efiring | 2010-07-26 09:14:16 -1000 (Mon, 26 Jul 2010) | 2 lines
  
  [3032390] subplot: more robust argument handling, consistent with 0.99.3
........

svn path=/trunk/matplotlib/; revision=8580
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8581 | leejjoon | 2010-07-27 11:52:47 -0400 (Tue, 27 Jul 2010) | 1 line
  
  fix pyplot.subplot2grid to support the projection keyword
........

svn path=/trunk/matplotlib/; revision=8582
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8583 | mdboom | 2010-07-27 13:26:50 -0400 (Tue, 27 Jul 2010) | 3 lines
  
  [3034778] line width arguments don't work in pcolormesh
  Also, support clipping paths on images in SVG backend.
........

svn path=/trunk/matplotlib/; revision=8584
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8585 | mdboom | 2010-07-28 14:33:11 -0400 (Wed, 28 Jul 2010) | 2 lines
  
  Fix problems displaying images with zero (logical) width.
........

svn path=/trunk/matplotlib/; revision=8586
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8588 | mdboom | 2010-07-28 14:48:14 -0400 (Wed, 28 Jul 2010) | 2 lines
  
  [3032853] Hist autorange bug using log and histtype
........

svn path=/trunk/matplotlib/; revision=8589
https://matplotlib.svn.sf.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8590 | mdboom | 2010-07-28 15:19:27 -0400 (Wed, 28 Jul 2010) | 3 lines
  
  [3031954] polar plot axis labeling problem
  Radial axis labels should now be placed correctly, even for small values of rmax
........

svn path=/trunk/matplotlib/; revision=8591
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8593 | weathergod | 2010-07-29 11:16:17 -0500 (Thu, 29 Jul 2010) | 4 lines
  
  Fix documentation for set_linestyle() and set_marker() and a few other
  functions properly display their available values on the web docs.
  Also changed pentagram to pentagon in the docstrings.
........

svn path=/trunk/matplotlib/; revision=8594
efiring and others added 19 commits February 6, 2011 17:53
https://matplotlib.svn.sourceforge.net/svnroot/matplotlib/branches/v1_0_maint

........
  r8953 | efiring | 2011-02-06 07:49:43 -1000 (Sun, 06 Feb 2011) | 2 lines
  
  bug: xlim, ylim with no arguments should not affect autoscaling
........

svn path=/trunk/matplotlib/; revision=8954
svn path=/trunk/matplotlib/; revision=8956
svn path=/trunk/matplotlib/; revision=8957
…rsions

svn path=/trunk/matplotlib/; revision=8958
svn path=/trunk/matplotlib/; revision=8965
Non-contiguous arrays, such as those resulting from pan and zoom,
are handled efficiently without additional copying.  This
noticeably speeds up zoom and pan on large images.

svn path=/trunk/matplotlib/; revision=8966
svn path=/trunk/matplotlib/; revision=8968
svn path=/trunk/matplotlib/; revision=8969
svn path=/trunk/matplotlib/; revision=8984
svn path=/trunk/matplotlib/; revision=8985
…est on narrow Python builds.)

svn path=/trunk/matplotlib/; revision=8986
…s now valid).

Enabling this required fixes in both scatter() and in collections.py.
Collections now support empty versions of themselves more correctly due to these fixes.

svn path=/trunk/matplotlib/; revision=8988
This marks all maintenance branches as merged.
majfmt = AutoDateFormatter(majloc, tz=unit)
tz = None
if getattr(unit, "tzinfo", None):
tz = unit.tzinfo
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These three lines can be concisely written as:

tz = getattr(unit, "tzinfo", None)

"""
import datetime
# should be enough to inform the unit conversion interface
# dates are comng in
self.update_units(datetime.date(2009,1,1))
if tz is not None:
if isinstance(tz, str):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This won't work in general; for example, the developing 0.11 version of ipython seems to be turning all strings into unicode (via turning on some future feature?). You could use cbook.is_string_like(), but using the first part of its logic should be adequate here:

if isinstance(tz, (str, unicode)):

@efiring efiring closed this May 16, 2011
@efiring
Copy link
Member

efiring commented May 16, 2011

Github seems to have gotten confused, claiming this involves 227 commits, and it seemed that some substantial change was needed anyway, so I am closing it; the successor is #116.

@faucon
Copy link
Contributor Author

faucon commented May 19, 2011

Hello Eric,

thanks for your help. As far as is understand it seems not to be my
fault that there were so much commits? But maybe i merged somehow from
master to my bug-fixing-branch.

And in addition, thanks for showing how to "do it right", i will have a
closer look at your changes this weekend, to understand some things better.

regards
maximilian

Am 16.05.2011 02:38, schrieb efiring:

Github seems to have gotten confused, claiming this involves 227
commits, and it seemed that some substantial change was needed
anyway, so I am closing it; the successor is
#116.

@efiring
Copy link
Member

efiring commented May 19, 2011

On 05/18/2011 08:52 PM, faucon wrote:

Hello Eric,

thanks for your help. As far as is understand it seems not to be my
fault that there were so much commits? But maybe i merged somehow from
master to my bug-fixing-branch.

Maximilian,

Not your fault in the least; it was a combination of an earlier
less-than-ideal commit or commits, plus what appears to be a bug in
github. No problem, really.

And in addition, thanks for showing how to "do it right", i will have a
closer look at your changes this weekend, to understand some things better.

You were trying to fix things in one of the more obscure corners of mpl;
units support has always been hard to understand, often troublesome, and
sometimes a moving target. I think I came up with a reasonable
modification of your changes, but whether it is "right" remains to be
seen. You were close.

Eric

regards
maximilian

Am 16.05.2011 02:38, schrieb efiring:

Github seems to have gotten confused, claiming this involves 227
commits, and it seemed that some substantial change was needed
anyway, so I am closing it; the successor is
#116.

@richbwood richbwood mentioned this pull request Dec 19, 2012
fariza added a commit to fariza/matplotlib that referenced this pull request Sep 22, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.