Re: IPC: A virtual sprint/classroom?

2018-06-28 Thread Edward K. Ream
The Leo edit pane widgets use signals using >> >> https://github.com/leo-editor/leo-editor/blob/master/leo/core/signal_manager.py > > > A gem. It's a good starting point for my studies. > BTW, the following are probably equivalent: self.listeners = defaultdict(lambda: list()) self.listeners

Re: IPC: A virtual sprint/classroom?

2018-06-28 Thread Edward K. Ream
On Thursday, June 28, 2018 at 11:16:52 AM UTC-5, Terry Brown wrote: The Leo edit pane widgets use signals using > > https://github.com/leo-editor/leo-editor/blob/master/leo/core/signal_manager.py A gem. It's a good starting point for my studies. Edward -- You received this message because

Re: IPC: A virtual sprint/classroom?

2018-06-28 Thread Edward K. Ream
On Thursday, June 28, 2018 at 11:16:52 AM UTC-5, Terry Brown wrote: mod_http.py allows arbitrary code execution within a running Leo :-} > https://github.com/leo-editor/snippets/blob/master/utils/led.sh Just ran curl for the first time. It might be the missing learning tool I have been

Re: IPC: A virtual sprint/classroom?

2018-06-28 Thread Edward K. Ream
On Thursday, June 28, 2018 at 11:16:52 AM UTC-5, Terry Brown wrote: The Leo edit pane widgets use signals using > > https://github.com/leo-editor/leo-editor/blob/master/leo/core/signal_manager.py > > but maybe should use https://pythonhosted.org/blinker/ ... > mod_http.py allows arbitrary

Re: IPC: A virtual sprint/classroom?

2018-06-28 Thread Terry Brown
On Thu, 28 Jun 2018 09:06:20 -0700 (PDT) "Edward K. Ream" wrote: > Leo already supports IPC in several ways: > > 1. The cursesGui2 plugin uses a broadcast/listener pattern for > debugging. The Leo edit pane widgets use signals using

IPC: A virtual sprint/classroom?

2018-06-28 Thread Edward K. Ream
Several important projects require inter-process communication. IPC seems like the most important area of work this year. *Call for help* This post summarizes what I do and don't know about IPC. Any advice or correction will be appreciated. *Present code* Leo already supports IPC in

Re: How to retire --no-cache?

2018-06-28 Thread Edward K. Ream
On Thu, Jun 28, 2018 at 9:52 AM, Terry Brown wrote: But where would Python have gone without the fork? ​Where it was already going, using from __future__ import ... That would have allowed people a reasonable migration path. In particular, unicode support in 3 is, in fact, only marginally

Re: How to retire --no-cache?

2018-06-28 Thread Terry Brown
On Thu, 28 Jun 2018 09:18:13 -0500 "Edward K. Ream" wrote: > On Thu, Jun 28, 2018 at 6:49 AM, jkn wrote: > > I have to say that although I think GvR's steering of Python > development as > > BDFL has generally been extraordinarily skillful, the jumping > > around with command-line parsing

Re: How to retire --no-cache?

2018-06-28 Thread Edward K. Ream
On Thu, Jun 28, 2018 at 6:49 AM, jkn wrote: I have to say that although I think GvR's steering of Python development as > BDFL has generally been extraordinarily skillful, the jumping around with > command-line parsing libraries has not been one of the high points... > ​Although I enjoy 3, I am

Re: How to retire --no-cache?

2018-06-28 Thread jkn
On Thursday, June 28, 2018 at 10:51:49 AM UTC+1, Edward K. Ream wrote: > > On Thursday, June 28, 2018 at 3:50:58 AM UTC-5, jkn wrote: > > Can you not raise an Exception with an informative message: "Please remove >> the --no-cache option and update to a more recent version of Leo"? >> > > Yes,

Re: How to retire --no-cache?

2018-06-28 Thread Edward K. Ream
On Thu, Jun 28, 2018 at 4:51 AM, Edward K. Ream wrote: Ah. I see the solution. The trick will be to scan sys.argv for > '--no-cache' *before* calling optparse.OptionParser.parse_args. > ​Done at 35d80ae in fast-read. Edward -- You received this message because you are subscribed to the

Re: How to retire --no-cache?

2018-06-28 Thread Edward K. Ream
On Thursday, June 28, 2018 at 3:50:58 AM UTC-5, jkn wrote: Can you not raise an Exception with an informative message: "Please remove > the --no-cache option and update to a more recent version of Leo"? > Yes, this would be reasonable. The problem is doing it. At present, Leo uses the

d67c242bc: Cache format changed to support dump-caches command

2018-06-28 Thread Edward K. Ream
The dump-caches command now produces much clearer descriptions of the caches. This is an important debugging tool for plugin writers. To make this work, the cacher now separates key/value pairs with ':::' instead of '_'. The change will cause Leo to ignore previous cache entries, which

Re: How to retire --no-cache?

2018-06-28 Thread jkn
On Thursday, June 28, 2018 at 9:29:38 AM UTC+1, Edward K. Ream wrote: > > > > On Wed, Jun 27, 2018 at 2:42 PM, Thomas Passin > wrote: > >> Despite what you say, I suggest keeping this option as a do-nothing >> option. No "fix" need ever be issued, or at least not until it is devised >> and

Re: How to retire --no-cache?

2018-06-28 Thread Edward K. Ream
On Wed, Jun 27, 2018 at 2:42 PM, Thomas Passin wrote: > Despite what you say, I suggest keeping this option as a do-nothing > option. No "fix" need ever be issued, or at least not until it is devised > and convenient. Leo could print a deprecation message to the (error) > console or log pane,

Re: fast-read branch fixes caching

2018-06-28 Thread Edward K. Ream
On Wed, Jun 27, 2018 at 6:21 PM, lewis wrote: > Running 1ff1e57, bookmarks plugin is working well again. > ​You're welcome. Thanks for the confirmation. Edward -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group