<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div><br></div><blockquote type="cite"><div><span>That's what I do, also. For the installation from source I now use either "pip install ." or "pip install -e .". </span></div></blockquote><div><br></div><div>-e for development for sure. But pip will try to bring in missing dependencies for you, which is less than ideal if you don't want a conda-pip jumble.</div><div><br></div><div>I think pip -e has a --no-deps flag, or something like that, which is useful.</div><div><br></div><blockquote type="cite"><div><span>On 2017/08/05 1:27 PM, Thomas Caswell wrote:</span><br><blockquote type="cite"><span>I sometimes take a blunter approach, install Matplotlib via conda and then uninstall it which leaves behind the dependencies ;)</span><br></blockquote></div></blockquote><div><br></div><div>D-uh! I should have thought of that!</div><div><br></div><div>That may not get you all the build deps but it will be close.</div><div><br></div><div>Oh, and taking a look at the Vonda recipe would be a better way to get the deps list:</div><div><br></div><div><a href="https://github.com/conda-forge/matplotlib-feedstock/blob/master/recipe/meta.yaml">https://github.com/conda-forge/matplotlib-feedstock/blob/master/recipe/meta.yaml</a></div><div><br></div><div>-CHB</div><div><br></div><br><blockquote type="cite"><div><blockquote type="cite"><span>Tom</span><br></blockquote><blockquote type="cite"><span>On Sat, Aug 5, 2017 at 5:33 PM Klymak Jody <<a href="mailto:jklymak@uvic.ca">jklymak@uvic.ca</a> <<a href="mailto:jklymak@uvic.ca">mailto:jklymak@uvic.ca</a>>> wrote:</span><br></blockquote><blockquote type="cite"><span> Hi Chris. Thanks so much. That'll be useful</span><br></blockquote><blockquote type="cite"><span> I think I was partially tricked by the fact that %matplotlib</span><br></blockquote><blockquote type="cite"><span> notebook wasn't working for master in a Jupiter notebook. but I</span><br></blockquote><blockquote type="cite"><span> see now that it doesn't wrk for anyone so I guess I was installing</span><br></blockquote><blockquote type="cite"><span> things adequately ;-)</span><br></blockquote><blockquote type="cite"><span> Thanks again. Jody</span><br></blockquote><blockquote type="cite"><span> Sent from my iPhone</span><br></blockquote><blockquote type="cite"><span> On Aug 5, 2017, at 14:22, Chris Barker <<a href="mailto:chris.barker@noaa.gov">chris.barker@noaa.gov</a></span><br></blockquote><blockquote type="cite"><span> <<a href="mailto:chris.barker@noaa.gov">mailto:chris.barker@noaa.gov</a>>> wrote:</span><br></blockquote><blockquote type="cite"><blockquote type="cite"><span> This should be pretty straightforward, but MPL's dependencie are</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> pretty complex. IN thoery, simplyL</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 1) create an environment</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 2) install the dev dependencies</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 3) build away.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> but in fact, step (2) there is not so straightforward, so here we</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> go...</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 1) I recommend using conda-forge as a source of packages -- it is</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> more up to date than defaults, and you are developing....</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> conda config --add channels conda-forge</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 2) create an environment for doing your development:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> conda create -n mpl-dev python=3</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 3) activate that environment:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> source activate mpl-dev</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 4) install the requirements of mpl:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> OK -- this one is kinda messy for MPL. It has a lot of</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> requirements that depend someone on which sub-packages you are</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> going to build.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> but first you need pk-config, which MPL used to find/configure</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> stuff on *nix systems.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> (you may have this from Brew already)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> conda install pkg-config</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> python setup.py --requires doesn't work, as mpl does a lot of</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> ative configuration.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> but:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> python ./setup.py dry-run</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> will get you a good list:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> REQUIRED DEPENDENCIES AND EXTENSIONS</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> numpy: yes [not found. pip may install it below.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> six: yes [six was not found.pip will attempt to</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> install</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> it after matplotlib.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> dateutil: yes [dateutil was not found. It is</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> required for date</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> axis support. pip/easy_install may attempt to</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> install it after matplotlib.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> backports.functools_lru_cache: yes [Not required]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> subprocess32: yes [Not required]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> pytz: yes [pytz was not found. pip/easy_install may</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> attempt to install it after matplotlib.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> cycler: yes [cycler was not found.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> pip/easy_install may</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> attempt to install it after matplotlib.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> tornado: yes [tornado was not found. It is required</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> for the</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> WebAgg backend. pip/easy_install may</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> attempt to</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> install it after matplotlib.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> pyparsing: yes [pyparsing was not found. It is</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> required for</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> mathtext support. pip/easy_install may</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> attempt to</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> install it after matplotlib.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> libagg: yes [pkg-config information for 'libagg'</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> could not</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> be found. Using local copy.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> freetype: no [The C/C++ header for freetype2</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> (ft2build.h)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> could not be found. You may need to</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> install the</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> development package.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> png: yes [version 1.6.26]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> qhull: yes [pkg-config information for 'libqhull'</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> could not</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> be found. Using local copy.]</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> you don't want pip to try to install anything, so I created a</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> requirements file (enclosed)</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> conda install --file conda_requirements_dev.txt</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> And now you should be able to build!</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> python setup.py build</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Note that you won't get all the back-ends -- other dependencies</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> need to be installed for that.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> BTW: it would be nice to have something like that</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> conda_requirements_dev.txt in the MPL repo -- the setup-tools</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> stuff is way to tied into the build/packaging/installing system.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> HTH,</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> -CHB</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> On Tue, Jun 13, 2017 at 6:24 AM, Jody Klymak <<a href="mailto:jklymak@gmail.com">jklymak@gmail.com</a></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <<a href="mailto:jklymak@gmail.com">mailto:jklymak@gmail.com</a>>> wrote:</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Hi all,</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> I’m somewhat stymied trying to understand how to install</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> matplotlib in a development environment using conda.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> I have MacOS, and I have brew gcc etc. I know how to set up a</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> env in conda.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> I eventually want to be able to run the tests, so that I can</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> stop uploading things with PEP8 errors etc.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Sorry if this is too vague. I can start over again and send</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> specific errors if that is easier, but I would bet someone has</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> a step-by-step.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Thanks, Jody</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> _______________________________________________</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Matplotlib-devel mailing list</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <a href="mailto:Matplotlib-devel@python.org">Matplotlib-devel@python.org</a> <<a href="mailto:Matplotlib-devel@python.org">mailto:Matplotlib-devel@python.org</a>></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <a href="https://mail.python.org/mailman/listinfo/matplotlib-devel">https://mail.python.org/mailman/listinfo/matplotlib-devel</a></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> -- </span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Christopher Barker, Ph.D.</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Oceanographer</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Emergency Response Division</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> NOAA/NOS/OR&R (206) 526-6959 <<a href="tel:%28206%29%20526-6959">tel:%28206%29%20526-6959</a>> voice</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> 7600 Sand Point Way NE (206) 526-6329 <<a href="tel:%28206%29%20526-6329">tel:%28206%29%20526-6329</a>> fax</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Seattle, WA 98115 (206) 526-6317 <<a href="tel:%28206%29%20526-6317">tel:%28206%29%20526-6317</a>> main reception</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <a href="mailto:Chris.Barker@noaa.gov">Chris.Barker@noaa.gov</a> <<a href="mailto:Chris.Barker@noaa.gov">mailto:Chris.Barker@noaa.gov</a>></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <conda_requirements_dev.txt></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> _______________________________________________</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> Matplotlib-devel mailing list</span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <a href="mailto:Matplotlib-devel@python.org">Matplotlib-devel@python.org</a> <<a href="mailto:Matplotlib-devel@python.org">mailto:Matplotlib-devel@python.org</a>></span><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><span> <a href="https://mail.python.org/mailman/listinfo/matplotlib-devel">https://mail.python.org/mailman/listinfo/matplotlib-devel</a></span><br></blockquote></blockquote><blockquote type="cite"><span> _______________________________________________</span><br></blockquote><blockquote type="cite"><span> Matplotlib-devel mailing list</span><br></blockquote><blockquote type="cite"><span> <a href="mailto:Matplotlib-devel@python.org">Matplotlib-devel@python.org</a> <<a href="mailto:Matplotlib-devel@python.org">mailto:Matplotlib-devel@python.org</a>></span><br></blockquote><blockquote type="cite"><span> <a href="https://mail.python.org/mailman/listinfo/matplotlib-devel">https://mail.python.org/mailman/listinfo/matplotlib-devel</a></span><br></blockquote><blockquote type="cite"><span>_______________________________________________</span><br></blockquote><blockquote type="cite"><span>Matplotlib-devel mailing list</span><br></blockquote><blockquote type="cite"><span><a href="mailto:Matplotlib-devel@python.org">Matplotlib-devel@python.org</a></span><br></blockquote><blockquote type="cite"><span><a href="https://mail.python.org/mailman/listinfo/matplotlib-devel">https://mail.python.org/mailman/listinfo/matplotlib-devel</a></span><br></blockquote><span></span><br><span>_______________________________________________</span><br><span>Matplotlib-devel mailing list</span><br><span><a href="mailto:Matplotlib-devel@python.org">Matplotlib-devel@python.org</a></span><br><span><a href="https://mail.python.org/mailman/listinfo/matplotlib-devel">https://mail.python.org/mailman/listinfo/matplotlib-devel</a></span><br></div></blockquote></body></html>