Uh oh!
There was an error while loading.Please reload this page.
- Notifications
You must be signed in to change notification settings - Fork7.9k
Improve formatting of imshow() cursor data independently of colorbar.#20949
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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.
Already on GitHub?Sign in to your account
Uh oh!
There was an error while loading.Please reload this page.
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
This makes sense to me.
d280424 tob00423fComparelib/matplotlib/_api/__init__.py Outdated
| @@ -297,3 +298,20 @@ def warn_external(message, category=None): | |||
| break | |||
| frame = frame.f_back | |||
| warnings.warn(message, category, stacklevel) | |||
| def g_sig_digits(value, delta): | |||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
This function is not API related. I think it should becbook._g_sig_digits.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
Sure.
Currently, when a colorbar is present, the cursor data under imshow() isformatted using the colorbar's cursor formatter (the idea being thatthat formatter should be able to "smartly" take normalization limitsinto account); if no colorbar is present a fixed format string ("%0.3g")is used.In fact, there is a better scale that defines the number of significantdigits one should display in an imshow cursor data: it arises becausecolormaps are discrete (usually with 256 colors, but in any case thevalue is available as `cmap.N`). This quantization tells us, for agiven value, by how much one needs to move before the underlying colorchanges (at all); that step size can be used to to determine a numberof significant digits to display. (Even if that's not necessarilyalways the best number, it should at least be reasonable *given theuser's choice of normalization*.)Also, note that because ScalarFormatter has now changed to take pixelsize into account when determining *its* number of significant digits,the previous approach of relying on the colorbar formatter has becomea less good approximation, as that means that the number of digitsdisplayed for an imshow() cursor could depend on the physical size ofthe associated colorbar (if present).Also factor out and reuse some logic to compute the number ofsignificant digits to use in format strings for given value/error pairs,already used by the linear and polar tickers.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
Anybody can merge after CI pass.
@meeseeksdev backport to 3.5.x |
Something went wrong ... Please have a look at my logs. It seem that the branch you are trying to backport to does not exists. |
@meeseeksdev backport to v3.5.x |
…ata independently of colorbar.
…949-on-v3.5.xBackport PR#20949 on branch v3.5.x (Improve formatting of imshow() cursor data independently of colorbar.)
Improve formatting of imshow() cursor data independently of colorbar.
Improve formatting of imshow() cursor data independently of colorbar.
Uh oh!
There was an error while loading.Please reload this page.
Currently (since#12459), when a colorbar is present, the cursor data under imshow() is
formatted using the colorbar's cursor formatter (the idea being that
that formatter should be able to "smartly" take normalization limits
into account); if no colorbar is present a fixed format string ("%0.3g")
is used (see#12473 for a stalled attempt to work around that).
In fact, there is a better scale that defines the number of significant
digits one should display in an imshow cursor data: it arises because
colormaps are discrete (usually with 256 colors, but in any case the
value is available as
cmap.N). This quantization tells us, for agiven value, by how much one needs to move before the underlying color
changes (at all); that step size can be used to to determine a number
of significant digits to display. (Even if that's not necessarily
always the best number, it should at least be reasonablegiven the
user's choice of normalization.)
Also, note that because ScalarFormatter has now changed to take pixel
size into account when determiningits number of significant digits (#16776),
the previous approach of relying on the colorbar formatter has become
a less good approximation, as that means that the number of digits
displayed for an imshow() cursor could depend on the physical size of
the associated colorbar (if present).
Also factor out and reuse some logic to compute the number of
significant digits to use in format strings for given value/error pairs,
already used by the linear and polar tickers.
PR Summary
PR Checklist
pytestpasses).flake8on changed files to check).flake8-docstringsand runflake8 --docstring-convention=all).doc/users/next_whats_new/(follow instructions in README.rst there).doc/api/next_api_changes/(follow instructions in README.rst there).