There is a reasonable way of making leo faster - switching to native code for hot spots (e.g. through cython or sip).
If we switched to pygments to do highlighting, we would probably be disappointed on performance. What we could do instead is use this: http://srchiliteqt.sourceforge.net/source-highlight-qt.html This would shove the whole syntax highlighting to native side. I used to think about Leo performance a lot, but these days I don't see any problems. The only thing that needs work IMO is saving an outline, which takes a while with large outlines. On Sun, Oct 9, 2011 at 4:26 PM, Edward K. Ream <[email protected]> wrote: > Yesterday I used SciTe, the Scintilla text editor, to study some > IPython code. > > The contrast between Leo and SciTe is instructive. Sure, SciTe lacks > many (all?) of Leo's fundamental features, but it has some important > advantages too, namely: > > 1. I can "just open" the file without importing it or creating @auto > nodes. > > 2. It feels faster and zippier. > > In some vague way, studying IPython with SciTe is/was a "dummy > whammy": both SciTe and IPython are different enough from "the Leo > way" as to issue a challenge and provoke a response and new ideas. > > As I think about point 1, several options come to mind: > > - using the read_only_nodes plugin (perhaps with some improvements). > > - using @edit. > > About IPython > =========== > > I greatly admire how IPython uses a wide range of tools, including > Python itself, Qt, pygments, readline, MatPlotLib and recently zmq/ > zeromq. > > There is sophisticated adapter code in IPython that integrates these > tools, but it does seem the the Fernando et. al. have gone out of > their way to use the best tools possible. > > I'm going to be studying pygments and zmq soon. > > Will it be possible to replace Leo's colorizer with pygments? The > essential question involves performance. You may remember the effort > that went into improving Leo's colorizer's speed. I'm thinking that > if pygments works for large pieces of text, it may have some very > important lessons to teach me about speed. > > zmq is a framework for parallel processing. This could be useful in > integrating Leo with other programs. > > Finally(?), the IPython notebook creates a *high quality* (good > looking) web interface for IPython. I want to know, in detail, how > that was done. I've studied the Firebug code, and it's a maze of > javascript hacks. I suspect the IPython way uses a higher-level web > tool. > > Finally (again), I have the nagging suspicion that something important > is brewing. Leo seems too "heavy" in comparison with IPython or > SciTe. In the game of go, "heavy" is a pejorative term (GvR used the > term fancy-smancy in the same way to describe Leo). Furthermore, I > have the suspicion (nothing more than that) that somehow IPython and > Leo can help each other in new ways. > > There are this morning's thoughts. Your comments please, Amigos. > > Edward > > -- > You received this message because you are subscribed to the Google Groups > "leo-editor" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/leo-editor?hl=en. > > -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/leo-editor?hl=en.
