Engineering is intensely enjoyable because it involves the interplay 
between big visions and endless picky but important details. The big 
picture is now pretty clear:

1. Release PyInstaller version of Leo on all platforms asap.
2. Release an official version, 5.1.1, asap.
3. Create api's and commands for using icons, attributes and windows.
4. Decouple panes so that any kind of pane can appear anywhere.

In this post, I'm going to discuss the details that support points 1 and 2 
above.

*Anaconda distributions*

I am now basing all my windows work on Anaconda 
<https://www.continuum.io/downloads>. Install Anaconda and it's unlikely 
you will have to install anything else.  I installed the 32-bit versions 
for both Python 2 and 3. This should allow soon-to-be-created PyInstaller 
version of Leo to work on any 32-bit or 64-bit system.

*pylint*

I have spent several days removing complaints issued by the latest pylint. 
In particular, most of the code now uses isinstance(obj)  instead of tests 
against type(obj). This simplified the code, but was potentially dangerous, 
as it affected g.isString, etc. and some parts of Leo's read/write code. 
However, all appears well. Imo, simplifying fundamental code seemed 
worthwhile.

I spent many happy hours simplifying leo/test/pylint-leo-rc.txt, pylint's 
configuration file. The new file is much simpler, shorter and clearer than 
the old. Most importantly, it now uses option *names* instead of option 
*numbers*. This is something I've wanted forever. It's much easier to see 
what tests are disabled.

pylint-leo-rc.txt disables several categories of checks that are pretty 
much irrelevant in Leonine contexts. This speeds up pylint considerably.

I also added the -s option to leo-editor/pylint-leo.py, the stand-alone 
version of pylint adapted to Leo. This silent option shows only pylint 
errors.

*Schedule*

I plan to install Anaconda on my Linux and MacOS machines next. I should 
then be in a position to release completely stand-alone PyInstaller 
versions of Leo. Maybe in a day or so.

I doubt that Leo 5.1.1 will go out the door this Friday, March 4, as 
originally proposed. True, minor bugs shouldn't inhibit releases, but it 
seems wasteful not to fix as many bugs as possible before going to all the 
work of getting an official release out the door.  I'll probably give 
myself another week to fix bugs. At present, seven bugs are targeted for 
5.1.1 
<https://github.com/leo-editor/leo-editor/issues?q=is%3Aopen+is%3Aissue+label%3Abug+milestone%3A5.1.1>.
 
The hard crashes look like release blockers, but I think all 7 bugs should 
be fixed if possible.

All comments welcome.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/leo-editor.
For more options, visit https://groups.google.com/d/optout.

Reply via email to