Unneeded argument in get_linestyle - #3355

This is a backwards incompatible change and I am a bit distressed that none of our tests are failing.

This needs an entry in api/api_changes. See #3349 for how we are going to start doing this.

This closes #3354

It is technically a backwards incompatible change, but it seems that
nobody has been using this particular method. I did a grep through our
codebase and did not find an instance of this method being used.
GraphicsContexts don't get directly exercised (probably because they aren't
documented very well). I am a bit concerned about why this particular
getter method was never used. Does this indicate a design flaw in our
graphics stack? Are there other getters that are not used? These are all
important questions.

But, yes, at the very least, let's note it in the api_changes for
completeness. Just because someone was sloppy in the past does not mean we
should be sloppy now.

On Sat, Aug 9, 2014 at 9:08 AM, Thomas A Caswell [email protected]
wrote:

This closes #3355 #3355

—
Reply to this email directly or view it on GitHub
#3355 (comment)
.

I am 👍 on this change, just want to make sure it gets documented correctly.

In most cases in the renderer draw_ methods, the GraphicsContext is just passed straight through to the C++ renderer functions.... I don't know if the get_ methods are required for this ?

In the python code, it mostly just requires the set_ methods; the GraphicsContext objects only seem to be created in draw methods, where the artist properties are transferred across...so no 'getting' required.

I would imagine you would never really see any of the GraphicsContext get methods being used (at least in python code)...
But, with them there, it makes it easy to make properties:

property(get_linestyle, set_linestyle, set_linestyle.__doc__)

Just checking back in on this. Is there something I need to do?

It does seem to need a rebase, or something. Why is the style.py file being added here?

moving the style module is a non-starter.

Hmm...The commit I intended to pull was bd06d88, which should only have 1 changed file..

In fact, this is the original pull (see top of page).

the more recent commits shouldn't be linked to this pull request.

Closing this as moving the style module is a non-starter.

@JamesRamm If you are still interested in working on this have a look at #3424 and the discussion around MEP25.