Mark Roseman added the comment:
Screenshot extdlg.png shows result
--
Added file: http://bugs.python.org/file40119/extdlg.png
___
Python tracker
<http://bugs.python.org/issue24
Mark Roseman added the comment:
Yup. Revised extdlg-sorted.patch attached. Thanks for noticing that!
--
Added file: http://bugs.python.org/file40131/extdlg-sorted.patch
___
Python tracker
<http://bugs.python.org/issue24
Mark Roseman added the comment:
I'd like to revisit this, particularly in the context of some of the
discussions regarding future improvements of IDLE.
Assume we have a barebones Tcl/Tk framework build of e.g. 8.6.4 as part of the
installer (core Tcl/Tk, not like e.g. ActiveTcl).
New submission from Mark Roseman:
For popup menus, control-click works, but right-click on mouse buttons that
support it, doesn't work.
This is a followup to #10404, last addressed in 2010.
As noted there, right click behaviour should be supported.
The right click Tk text bindings o
Changes by Mark Roseman :
--
components: +IDLE
___
Python tracker
<http://bugs.python.org/issue24801>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Mark Roseman:
To facilitate using IDLE to learn Python, and perhaps as a first experience
with programming altogether, and taking advantage of the fact that IDLE comes
with Python out of the box, I propose the following:
1. A new "learning" mode be introduced to
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue17535>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue24225>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue13504>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue17942>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
Just as a point of information and possible future reference, there is a
'tklib' (collection of Tk-related modules written in Tcl) module called 'ctext'
which does syntax highlighting as well as implements the line numbering thing
(also
New submission from Mark Roseman:
Most of the dialogs in IDLE on OS X do respond to 'Return' key as equivalent to
hitting OK, and Escape to hitting Cancel.
Guidelines also suggest that the Enter key (on numeric keypad) should work like
'Return', and Cmd-. (period) sho
New submission from Mark Roseman:
No reason for it to be modal. Especially on OS X (where it really isn't...)
--
components: IDLE
messages: 248167
nosy: kbk, markroseman, roger.serwy, terry.reedy
priority: normal
severity: normal
status: open
title: About IDLE dialog shouldn'
New submission from Mark Roseman:
Undo/Redo in Edit menu should be disabled when there is nothing to Undo or Redo
--
components: IDLE
messages: 248168
nosy: kbk, markroseman, roger.serwy, terry.reedy
priority: normal
severity: normal
status: open
title: Disable Undo/Redo menu items when
New submission from Mark Roseman:
It's fairly easy to get IDLE to revert back to an empty menubar, i.e. just a
Python menu.
For example, open a shell, debugger, and editor window. Click on debugger
window, then editor window, then close editor window. Focus goes back to
debugger, but do
New submission from Mark Roseman:
Right now when you're running a program you can still select the 'debugger'
item in the menu... you just get an error dialog "you can only toggle the
debugger when idle" (with a title "don't debug now").
While I got a
New submission from Mark Roseman:
In an editor window, with no selection, most of the items in the format menu
(indent, tabify, etc.) aren't applicable, so the corresponding menu items
should be disabled.
--
components: IDLE
messages: 248171
nosy: kbk, markroseman, roger.
New submission from Mark Roseman:
If I have just an edit window open with my program, there's no way to run the
program with the debugger visible. Should be a way to do so without going
through the extra steps of opening up a shell window first
--
components: IDLE
messages: 2
New submission from Mark Roseman:
Rather than including the window width/height in the config dialog, just
remember the last window size and use that next time
--
components: IDLE
messages: 248176
nosy: kbk, markroseman, roger.serwy, terry.reedy
priority: normal
severity: normal
status
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue17642>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Mark Roseman:
If there are going to be highlighting themes in IDLE, and the ability to
customize them, why not the background color of the window? Light on dark is
easier for some people to read - adding one that did that would be a good
candidate for another built-in
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue1528593>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Mark Roseman:
Right now breakpoints can only be set/cleared by using a context menu on a line
in the editor. I discovered this entirely by reading through the bug database,
as right-click doesn't work on OS X (#24801).
Some other tools use an indicator (e.g. stop sig
New submission from Mark Roseman:
This builds on things like the tabbed editor suggestion, but essentially I'm
talking about a scenario where you'd have your one (editor) window open working
on your program, you click 'run...' or 'debug...' from the menu, and a sh
Mark Roseman added the comment:
While the thing about asking for a favourite colour before switching to
advanced mode was a joke, I do sincerely believe that extensive progressive
disclosure techniques should be used to keep much of IDLE's features and
options hidden out of the box.
I
Mark Roseman added the comment:
If you have your mouse set up to have two buttons, right-clicking in the editor
window doesn't bring up the context menu it's supposed to.
--
___
Python tracker
<http://bugs.python.o
Mark Roseman added the comment:
I believe the ttk::themes call is older, part of when it was still Tile. There
are a bunch of those kind of API's that were kept around (for compatibility I
presume) when the official API (style theme names) was created and documented.
As explained o
Mark Roseman added the comment:
I read #7949 as saying the text widget picks up the background color from the
system-wide GTk theme.
This one is saying that the background color of the text widget should be
changeable as part of an IDLE highlighting theme
Mark Roseman added the comment:
oh agree, definitely don't want a symbol on every line. i was already thinking
implementation where you'd probably have an object matching the background for
hit detection! :-)
but yes, visually nothing there but empty space when there isn't a b
Mark Roseman added the comment:
I'm attaching mainwin3.patch, which is a subset of the previous patches,
modified to not use ttk. It gets rid of the highlightthickness, the sunken
line/column effect, and adds a thin separator below the text widget.
--
Added file:
Mark Roseman added the comment:
note that many of these are using 'simpledialog' which has that limitation
--
___
Python tracker
<http://bugs.python.o
Mark Roseman added the comment:
Have attached macpopup.patch which removes the incorrect Tk behaviour and makes
it so that right click on Mac will bring up the context menu as appropriate.
--
keywords: +patch
Added file: http://bugs.python.org/file40148/macpopup.patch
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue9262>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
Yes that's exactly what I was thinking. If everything is a frame rather than a
toplevel it'll be much easier to reconfigure things.
--
___
Python tracker
<http://bugs.python.o
Mark Roseman added the comment:
I've attached decouple_config.patch, which removes some internal knowledge
about EditorWindow from configDialog. This is a step towards making the
preferences dialog non-modal (and also to launching either the current dialog,
or a new ttk-dependent one).
Mark Roseman added the comment:
The attached demodalize.patch (which includes the changes from the previously
posted decouple_config.patch) changes both the settings dialog and the about
dialog to be non-modal.
There's a new class UIFactory which is responsible for launching these kin
New submission from Mark Roseman:
This is a placeholder issue for adding automated functional/integration tests
to complement the existing unit tests.
--
messages: 248428
nosy: kbk, markroseman, roger.serwy, terry.reedy
priority: normal
severity: normal
status: open
title: IDLE
Changes by Mark Roseman :
--
components: +IDLE
keywords: +patch
Added file: http://bugs.python.org/file40164/functionaltests.patch
___
Python tracker
<http://bugs.python.org/issue24
Mark Roseman added the comment:
I've attached functionaltests.patch which provides a starting point, using Tk
introspection and event generation to exercise the running application. The
heart of it is the (very much in progress) TkTestCase class which provides a
bunch of Tkinter-spe
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue24039>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue2053>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
Roger's extension is an amazingly cool hack. With some of the decoupling
mentioned in #24826, the actual switching should get easier.
Regarding cosmetics, I wanted to make a suggestion. The tabs provided by
ttk::notebook aren't ideally suited for
Mark Roseman added the comment:
I did some followup on this today, and could reproduce it with a few lines of
Tcl/Tk code. As Ned noted, it seems particular to the ActiveTcl build, as when
I built my own 8.5.18 it also worked fine.
(If you're curious, the thing that is failing i
Mark Roseman added the comment:
Awesome, thanks Kevin. Have attached calltip.patch. The extra lift() call
doesn't seem to hurt on Windows or X11, so didn't make it conditional.
--
keywords: +patch
Added file: http://bugs.python.org/file40173/cal
New submission from Mark Roseman:
In EditorWindow.open_module... once switch to querydialog, display errors (e.g.
module not found) in askstring dialog itself, not open up subsequent
'showerror' dialog
--
components: IDLE
messages: 248539
nosy: kbk, markroseman, r
New submission from Mark Roseman:
One of the concerns with making significant structural changes to the IDLE
codebase is breakage of external that might import a piece of idlelib (so not
just 'import idlelib' but a particular submodule).
PEP 434 already makes the case that this be
Mark Roseman added the comment:
Exactly. The "querydialog" code (which will replace the simpledialog
askstring/askinteger calls) displays errors as shown in querydialog.png, with
the error messages disappearing as soon as you hit another key.
You can also pass in a 'validator&#
Mark Roseman added the comment:
Work in progress, have a few more tweaks to make, but here's a snapshot...
--
Added file: http://bugs.python.org/file40176/querydialog.py
___
Python tracker
<http://bugs.python.org/is
Mark Roseman added the comment:
There is no notion of middle button on Mac, hence right button is B2. It's
documented. Current Tk 8.6 has even removed the <> binding from
B2 on Aqua because of this (I've filed a Tk bug report to backport this to
8.5). 8.6 also defines a
Mark Roseman added the comment:
Yes that should be good to go
--
___
Python tracker
<http://bugs.python.org/issue24782>
___
___
Python-bugs-list mailing list
Unsub
Mark Roseman added the comment:
I am admittedly not a fan of skinnable user interfaces, especially for
non-entertainment applications. It doesn't add anything to the usability, and
makes support harder. It always says to me "hello, 2002 called and wants it's
user interface bac
Mark Roseman added the comment:
Agree about the font resizing menu items/shortcuts... your original #17642
remains open for this
--
___
Python tracker
<http://bugs.python.org/issue24
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue24790>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
Looking to get some feedback on a large chunk of new code (not yet complete),
for the 'new' preferences dialog. Not so much UI things as if I've made any
glaring structural errors which aren't obvious to this Python sorta-newbie.
For la
Mark Roseman added the comment:
Have attached macpopup-revised.patch, incorporating Serhiy's very helpful
suggestions. Unless there are any other thoughts, this one is probably ready to
go.
--
Added file: http://bugs.python.org/file40208/macpopup-revised.
Mark Roseman added the comment:
Sounds like this one should be ready to review and incorporate.
As a reminder, mainwin3.patch, which should work on Tk 8.4+, gets rid of the
highlightthickness around the text widget in the editor, the sunken line/column
effect in the status bar, and adds a
Mark Roseman added the comment:
Alessandro, would you have a chance to test this one line patch, which makes
calltips work again on OS X?
--
___
Python tracker
<http://bugs.python.org/issue24
Mark Roseman added the comment:
Terry, when you get a chance, it would be great if you could have a look at
demodalize.patch (or if you can suggest someone else who would be good to take
a peek at it).
This is sort of the baseline for the uifactory, and touches a lot of things in
small ways
Mark Roseman added the comment:
I reproduced your problem on Windows, and verified that it wasn't an issue on
either Mac or Linux.
Your fix however, didn't change anything for me (on a Windows 7 VM). I tried a
few other things (raising the window, playing with various wm attri
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue24893>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
Watched that video clip - yikes, that is bad.
I tried playing around to reproduce it, but haven't had any success yet, though
did find another weird thing. If you have a bunch of short lines without any
blanks, sometimes when you click well to the right o
Mark Roseman added the comment:
Regarding the setting background for the multiple elements.. one possibility is
that the first time in a session they change the background of an element that
was same as the background, we ask if they'd like to apply that change to the
other program ele
Mark Roseman added the comment:
Just came across this one (Terry I'm sure can figure out why).
I'm sympathetic with Guilherme's wish that Tk had done this one differently, as
an option to specify an existing text widget peer when creating a text widget,
rather than having the o
Mark Roseman added the comment:
Did a quick check... this approach seems to work. Only change was in step 4
should set both ._w and ._name for the widget object
--
___
Python tracker
<http://bugs.python.org/issue17
Mark Roseman added the comment:
I've put together a standalone tabs widget (mostly done) based on Tk canvas
widget, that emulates the behaviour of TextMate's tabs. I was able to modify
Roger's extension to use this widget. See attached screenshot newtabs.png (tabs
area is fu
Mark Roseman added the comment:
Have attached uitabs.py. From the header comments:
Standalone 'tabs' widget to switch between multiple different views.
Unlike the Tkinter ttk.Notebook widget, this widget is suitable for
displaying a potentially large number of tabs, as might be
Mark Roseman added the comment:
Have attached newTabExtension.py, which is Roger's tab extension, hacked to use
uitabs.py instead of the UI previously built into that extension.
--
Added file: http://bugs.python.org/file40265/newTabExtensi
Mark Roseman added the comment:
Have attached debugpanel.py, implementing a ttk-based mockup of one possible
redesign of the debugger UI, as seen in
http://bugs.python.org/file40234/newtabs.png
--
Added file: http://bugs.python.org/file40266/debugpanel.py
Mark Roseman added the comment:
Attached uipreferences.py, which is a newer and more complete version of the
one in prefs-wip.patch
--
Added file: http://bugs.python.org/file40276/uipreferences.py
___
Python tracker
<http://bugs.python.
Mark Roseman added the comment:
Config files affecting tests (and tests affecting config files!) is an issue.
Would like to see config files written into a different directory when running
tests. Any concerns with doing that?
--
___
Python tracker
Mark Roseman added the comment:
I'm not against generality, though don't think there are many places this
actually comes up (the cloned menus shouldn't be affecting more than a couple
people these days, and aren't the result of an explicit API call).
For this particular
New submission from Mark Roseman:
This is a followup to #14146
I don't think this is actually a Tk bug. The text widget supports an
"inactiveselectbackground" option which controls the background color of the
selection when the window is not active. On Win32, this is defined
New submission from Mark Roseman:
Right menu button to invoke context menu (as well as alternative control-click)
not set up correctly for Mac in the debugger. Patch attached.
--
components: IDLE
files: debug-mac-context-menu.patch
keywords: patch
messages: 249582
nosy: kbk
Mark Roseman added the comment:
Terry, the new code I've proposed includes a "tkextras" module that is a good
place for these sort of little convenience functions.
--
___
Python tracker
<http://bugs.pyt
New submission from Mark Roseman:
In the debugger, the values for the variables shown in the locals/globals panes
are editable (i.e. using Entry widget) but I don't see any mechanism to have
those changes affect anything. If I'm not missing something, why Entry and
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue15347>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue15348>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
Thanks Terry, sorry I missed your previous mention of this. From what I recall
the treeview doesn't allow editable items, but what I've done in similar
circumstances before (editing text items on a canvas in drawing type programs)
was 'place'
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue23551>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue22179>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue22460>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue13586>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue18590>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Mark Roseman :
--
nosy: +markroseman
___
Python tracker
<http://bugs.python.org/issue16893>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
See inactiveselection.patch. This addresses the original bug of the debugger
not highlighting the correct line, and also various issues with the find
windows not highlighting the selection they have found, e.g. #22179, #18590
--
keywords: +patch
Added
Mark Roseman added the comment:
If it's useful, I quickly threw together a routine that takes the HTML
generated by Sphinx, and using html.parser, strips out the navigation stuff,
and stuffs it into a Tk text widget, doing appropriate things with the subset
of the formatting it uses (
New submission from Mark Roseman:
Certain policy decisions should be in one place in the code (DRY).
These could be based on preferences or environment (should we be using ttk?),
which operating system we're using (what hand cursor looks best?) or even the
version of the operating s
Mark Roseman added the comment:
Ok. In case you're curious or might find it useful elsewhere, I've attached
sphinxview.py, which renders the HTML into a text widget (with formatting, it's
not converted to plain text).
--
Added file: http://bugs.python.org/file4040
New submission from Mark Roseman:
This is a placeholder for what will be a series of patches to the FileList to
generalize and enhance it.
This is part of generalizing pieces of IDLE to be more flexible, including
things like editors not necessarily being in toplevel windows, dialogs not
Mark Roseman added the comment:
This first patch is a straight refactoring so that other modules do not touch
the FileList internals.
This is done in two steps. First, we remove references to inversedict etc. from
outside FileList.
Second, we run all notifications through FileList. So for
Mark Roseman added the comment:
Have attached about.patch, which changes the about dialog to no longer be
modal, as well as does a substantial redesign, where the various help texts are
displayed within the dialog itself, rather than launching (further modal) new
windows.
Note: this new
Mark Roseman added the comment:
See #24972 for fix
--
___
Python tracker
<http://bugs.python.org/issue22179>
___
___
Python-bugs-list mailing list
Unsubscribe:
Mark Roseman added the comment:
See #24972 for fix
--
___
Python tracker
<http://bugs.python.org/issue18590>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Mark Roseman:
Right now the menubar is slightly different for the shell (has Debug) and the
editor windows (has Format and Run). I'd like to suggest the same menubar be
used for all windows.
Rationale:
1. Easier to use, especially for systems that share a single me
Mark Roseman added the comment:
Note: about dialog part of the 'demodalize' patch split off and now in #24813;
the bulk of the rest of it, which is really providing a cleaner FileList API
rather than direct access to internals from EditorWindow, is in #25031.
After the latte
New submission from Mark Roseman:
A necessary prerequisite of tabbed windows, editor and shell in same window
etc. as per #24826, is that editors stop thinking they are in their own
toplevel window.
The attached component.patch is unfortunately long, but this was necessary due
to the highly
Mark Roseman added the comment:
Debugger.py has a nested call to mainloop() ... which is almost always a bad
idea. I'm betting the close is being handled in this mainloop, and there are no
more events being generated, meaning the real mainloop towards the bottom of
PyShell (which chec
Mark Roseman added the comment:
FYI, I've gone past this point in my own playground, to the point of having a
working tabbed editor implementation. It's not that much further actually.
Here are the next steps:
1. Change so that Container is passed to Components on creation, r
Mark Roseman added the comment:
The new 'component' infrastructure provides the mechanism for passing these
kinds of notifications around. Allowing extensions (in whatever form they'll
exist with the new stuff) to take part in this notification mechanism would be
a smal
New submission from Mark Roseman:
Proposal for an alternative or perhaps replacement of both the code context
extension, as well as the class browser (rooted at a particular file, not the
path browser).
I'll direct your attention to the attached browser.png. On the right is IDLE
101 - 200 of 278 matches
Mail list logo
