2019 was supposedly going to be the year of playful prototypes. Instead, it 
was consumed by the following ponderous projects :-)

*1. Support for Qt docks*

A ponderous success. Qt docks are not easy to use, but some people 
appreciate having them, despite the difficulties.

*2. The pyzo_in_leo plugin*

At present, this plugin is mostly a failure. There is no real connection 
between pyzo running within Leo and Leo itself.

*3. The beautify and fstringify commands*

These commands are complete successes. They work well within Leo, and they 
use the new token-order classes in leoAst.py effectively.

There was a big surprise, however. Leo's beautify commands must needs be 
completely separate from the external black program. This means that the 
Orange class in leoAst.py can't be considered a replacement for black. Leo 
6.1 added the black command. Leo 6.2 will remove it. It's a bad idea.

*4. Coverage testing using traditional unit tests*

This got a thorough vetting while testing new code in leoAst.py. This is a 
marvelous new pattern for me. It made the last four month's work worthwhile.


*Plans for 2020*

Three items aim to complete projects started in 2019:

1. I am considering making Leo's legacy operation (no Qt docks) the 
default. This will simplify life for newbies. I'll ask for comments in 
another thread.

2. It may be possible to use pyzo's yoton framework to communicate between 
Leo and the embedded copy of pyzo. I few hours of work will show whether 
this is feasible. As a bonus, it may be possible to use *Leo's* plugin 
manager within pyzo. I do not intend this to turn into a major project.

3. I will likely announce the new token order generators (in leoAst.py) to 
the wider world. This announcement will be strictly a FYI. There will be no 
claims that Leo's beautify command is superior to black.

After these items are complete I'll focus on continual incremental 
improvements to Leo.

*Schedule*

Leo 6.2 will go out the door when all bugs scheduled for 6.2 have been 
fixed. This will likely take a few more weeks.

It's too late to fix #1437 
<https://github.com/leo-editor/leo-editor/issues/1437>. That must wait 
until early in the 6.3 release cycle.

*Summary*

2019 involved heavy work on projects with minimal impact on most Leo users. 
This is disappointing. I would like to focus on more consequential 
projects. 

On a happier note, except for VS Code, there are no more plans to embed Leo 
in other editors. This removes a long term distraction for me.

Instead of embedding, Leo's bridge module allows IPC (Inter-Process 
Communication) between Leo and other editors. IPC may offer a *low-cost* 
way of connecting Leo with pyzo, thereby rescuing the pyzo_in_leo plugin. 
No way am I going to spend months on this project. It will fail if it 
doesn't succeed quickly.

And that's it. As always, fixing bugs will be near the top of the list.

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 view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/691b6c42-0f00-47af-9483-1bbde1ab96f6%40googlegroups.com.

Reply via email to