Add explanatory text for rasterization demo #18012
Conversation
examples/misc/rasterization_demo.py
Outdated
| @@ -4,50 +4,57 @@ | |||
| ================== | |||
|
|
|||
| """ | |||
|
|
|||
| # Rasterization is a method where an image described in a vector graphics format is being converted into a raster image (pixels). | |||
There was a problem hiding this comment.
Can you add a sentence about how we can flag particular parts of a Figure to be rasterized and some comments about why you would want to do this? Something like:
This demo is showing how you may want to balance most things (text, lines, ...) being vectorized (where you get all of the benefits of vector graphics), but turning some of the artists into rasterized images (because embedding a 2k by 2k mesh in an SVG results in a huge file that takes a long time to both save and load).
?
|
Thanks for working on this @marychrisgo ! We enforce a sub-set of the flake8 code style checks on all of the Python source in our repo, hopefully the linter's comments are clear? |
|
Hello @tacaswell ! I've read all the comments and resolved all the issues: Flake8 guide and your suggested paragraph. Thanks! |
| ax4.text(0.5, 0.5, "Text", alpha=0.2, | ||
| zorder=-15, | ||
| va="center", ha="center", size=50, transform=ax4.transAxes) | ||
|
|
||
| # Set zorder value below which artists will be rasterized |
There was a problem hiding this comment.
What? I don't even think this should be a thing, and we definitely should not advertise it. The artists we want rasterized should explicitly get a rasterized=True kwarg, not depend on zorder, which should not have anything to do with rasterization.
There was a problem hiding this comment.
set_rasterization_zorder was meant to be used for the ps backend family which does not support transparency, i.e., rasterizing an artist may affect the appearance of other artists with lower zorder. So the idea was rasterizing all artists below certain zorder and then add non-rasterized artists on top of it (like texts).
| ax2.set_aspect(1) | ||
| ax2.set_title("Rasterization") | ||
|
|
||
| m = ax2.pcolormesh(xx, yy, d) |
There was a problem hiding this comment.
I think this is better spelled as ax2.pcolormesh(xx, yy, d, rasterized=True)
| # ax2.title.set_rasterized(True) # should display a warning | ||
|
|
||
| # Save files in pdf and eps format | ||
| plt.savefig("test_rasterization.pdf", dpi=150) |
There was a problem hiding this comment.
Why are we saving these?
Can you also compare how big they are w/o the rasterization to with the rasterization?
|
Hello @jklymak ! Thank you for the comments. I've edited the paragraph you suggested. I won't edit the coding part as this PR is mainly for the documentation. |
In the end, the whole example is documentation (the code is for illustration, it's not part of the library). We encourage to also make the example code more concise where possible. |
|
Thank you @timhoffm ! Changing some based on your comments, pushing a new request now. |
|
Thinks for this work @marychrisgo ! If I recall correctly this was your first Matplotlib pr 🎉 ! Hopefully we will hear from you again. |
PR Summary
This is solving a request in issue##17708 by @timhoffm. I added an explanation of the rasterization_demo.py.
PR Checklist