Re: ENB: Simple design principles for style sheets

I just went through a really painful experience when I adopted a dark theme. The existing ones that are in leoSettings were unusable, because too many of the syntax colors were unreadable against the dark background. I've gotten it working more or less to my satisfaction (there's one color th

Re: ENB: Simple design principles for style sheets

On Saturday, March 10, 2018 at 1:18:38 PM UTC-5, Terry Brown wrote: > > On Sat, 10 Mar 2018 06:58:42 -0800 (PST) > Thomas Passin > wrote: > > in 5.7, LeoSettings has (at least) four dark themes: "base dark", "windows ekr dark", "leo_dark theme 1&

Re: ENB: Simple design principles for style sheets

On Sat, Mar 10, 2018 at 8:58 AM, Thomas Passin > wrote: > >> I just went through a really painful experience when I adopted a dark >> theme. >> > > ​I agree with Terry, this is excellent feedback. Here are some ideas. > > *Color samples directly in Leo* > &g

Re: The design of Leo+Ipython+Jupyter+Lit-computing

If I could step in here some months later, and move to a higher level of conversation, I think that there are several levels of engagement with Jupyter that we could contemplate. For example, we could display a notebook, graphics output and all, in a separate pane. ViewRendered does that pret

Re: Atom might be Leo's future

It seems to me that many python-based systems could be run in a Leo process, which would give access to their internal APIs. The challenge would be to get their input and output into Leo cells. For non-python systems, in some cases there could be a python wrapper that handles communication (

Using Holoview in @Jupyter Nodes

I have been playing with Holoview, which may be a good system for graphing and plotting, as a replacement fro pyplot. It uses more declarative and less procedural progamming, which is usually a good thing. It already integrates with Jupyter Notebook, and so I decided to see if the ViewRender

Re: The design of Leo+Ipython+Jupyter+Lit-computing

; On Sat, 10 Mar 2018 17:41:47 -0800 (PST) > Thomas Passin > wrote: > > > If I could step in here some months later, and move to a higher level > > of conversation, I think that there are several levels of engagement > > with Jupyter that we could contemplate. >

Re: Using Holoview in @Jupyter Nodes

s I can think of until one of them works. Just as a reminder, this is using VR, *not* VR2. On Sunday, March 11, 2018 at 8:35:06 AM UTC-4, Edward K. Ream wrote: > > On Saturday, March 10, 2018 at 8:26:21 PM UTC-6, Thomas Passin wrote: > > > Run the code in Jupyter Notebook. Let it finis

Re: ENB: Simple design principles for style sheets

I didn't have too much trouble finding color constants per se. I would do a web search for, say, "color #553399", and that would show me the color. There are also on-line color pickers. But colors look different depending on their surroundings, and I had to do a lot of trial and error to hit

Re: The design of Leo+Ipython+Jupyter+Lit-computing

I haven't worked on anything much in-browser since before html5 came out. So I didn't know anything about "~=", for example. Even then I tried to only work with the simpler constructs (both javascript and css), so I probably wouldn't have used that particular construct anyway. On Saturday, Ma

Re: Viewrendered3 Read For Beta Testing

I think that between the time that I modified qt_gui.py and the time that I pushed it to my clone of the devel branch, that it got reformatted in origin/ devel. Between my VR3 branch and my devel branch, I see many changes in the file diffs - I'm sure they are what Edward noticed - that are mi

Re: Some excellent pytest articles/presentations

On Saturday, March 14, 2020 at 7:33:28 AM UTC-4, Edward K. Ream wrote: > > > Otoh, Leo's Qt dock code depends on complex state set on program startup > and shutdown, as well as arbitrary *sequences* of user actions. I agree > that the code is hard to test and hard to use. That does not mean that

Re: Some excellent pytest articles/presentations

On Saturday, March 14, 2020 at 9:39:15 AM UTC-4, Thomas Passin wrote: > > On Saturday, March 14, 2020 at 7:33:28 AM UTC-4, Edward K. Ream wrote: >> >> >> Otoh, Leo's Qt dock code depends on complex state set on program startup >> and shutdown, as well as arbit

Re: Some excellent pytest articles/presentations

On Saturday, March 14, 2020 at 10:12:06 AM UTC-4, vitalije wrote: > > That does not mean that testing that code will ever be easy. If anyone >> knows how to test the Qt code, I would appreciate knowing how. > > [snip] > I really believe that doing decoupling piece by piece, module by module, > c

Re: Viewrendered3 Read For Beta Testing

you please just add them so we can get this done? On Saturday, March 14, 2020 at 7:49:33 AM UTC-4, Edward K. Ream wrote: > > > > On Fri, Mar 13, 2020 at 9:48 PM Thomas Passin > wrote: > >> I think that between the time that I modified qt_gui.py and the time that >&

Re: Viewrendered3 Read For Beta Testing

Thank you, Edward! On Saturday, March 14, 2020 at 4:16:39 PM UTC-4, Edward K. Ream wrote: > > > > On Sat, Mar 14, 2020 at 11:05 AM Thomas Passin > wrote: > >> Thanks, Ed, >> >> I have not been able to get my repo on GitHub straightened out so that >> y

Re: launchLeo on ubuntu: Reuse existing editor process?

I don't know how Ubuntu does this any more - I did it a few times several years ago. After some experimenting (on the Unity desktop), I can get you a Leo launch icon on the desktop, I can get Leo added to the list of applications that can be used to open a file, but I can't get the launch icon

Re: Viewrendered3 Read For Beta Testing

On Sunday, March 15, 2020 at 7:08:32 AM UTC-4, Edward K. Ream wrote: > > > Your version of qt_gui was way out of date, so I made the change by hand. > The new code is: > That's so strange, because I got it right from the devel branch and just added my little bit of code. Anyway, thanks again.

Re: launchLeo on ubuntu: Reuse existing editor process?

On Sunday, March 15, 2020 at 3:32:11 AM UTC-4, Yaakov Belch wrote: > > Thank you very much for your help, Thomas. > > What I want is simply that by pressing Window-1, the open Leo session is > brought to the foreground --- or (if there is no running Leo session) a new > Leo session is created.

Re: Some excellent pytest articles/presentations

On Sunday, March 15, 2020 at 6:39:46 AM UTC-4, Edward K. Ream wrote: > > On Sat, Mar 14, 2020 at 9:12 AM vitalije > > wrote: > > I really believe that doing decoupling piece by piece, module by module, >> can make Leo's code much easier to test and more importantly easier to >> (re)use. >> > >

Re: Viewrendered3 Read For Beta Testing

On Sunday, March 15, 2020 at 11:35:26 AM UTC-4, Edward K. Ream wrote: > > > In future, please run pylint on your code. It will catch these kinds of > things. > I thought that pyflakes was doing that job now. It didn't say anything. Well, I didn't run it on qt_gui.py, because I just copied it

Re: Viewrendered3 Read For Beta Testing

On Sunday, March 15, 2020 at 4:46:22 PM UTC-4, Thomas Passin wrote: > > > > On Sunday, March 15, 2020 at 11:35:26 AM UTC-4, Edward K. Ream wrote: >> >> >> In future, please run pylint on your code. It will catch these kinds of >> things. >> > >

Re: A button to run rust snippets

On Sunday, March 15, 2020 at 1:57:10 PM UTC-4, Edward K. Ream wrote: > > Here is the @button node that I use to run the tutorial snippets in the Rust > Book . > That's the kind of thing I was thinking of to get VR3 to run other languages. -- Y

Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

Windows 10, Python 3.8. Running Leo 6.2-b1, VR3 fails to create its toolbar (which causes VR3 to fail), but the very same file, viewrendered3.py, works as expected running Leo 6.1-final. Most of my development used Leo 6.1-final, though using Leo from the devel branch from several weeks ago wo

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

There is one specific thing that is failing: g.app.dock is returning False. When that happens, the rest of create_pane() gets skipped (and of course my plugin won't work). It is supposed to be returning True when docks are in use. This happens when I start Leo with --init-docks. When I start

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

On Monday, March 16, 2020 at 6:58:21 AM UTC-4, Edward K. Ream wrote: > > Thomas, I'm not sure why this was a mysterious problem for you. I continue > to suspect something is not right on your end. > > Please verify, using gitk, that you see the changes I made in both the 6.2 > and devel branche

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

On Monday, March 16, 2020 at 8:08:57 AM UTC-4, Thomas Passin wrote: > > There is one specific thing that is failing: > > g.app.dock is returning False. When that happens, the rest of > create_pane() gets skipped (and of course my plugin won't work). It is > supposed to

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

On Monday, March 16, 2020 at 8:32:24 AM UTC-4, Edward K. Ream wrote: > > Alright. I've fixed the problem in vr3. Hope you recover from your > problems :-) > Thanks. In addition, I found two other issues, both fixed in the attached zip file: 1) The name of the RsT stylesheet was obsolete( i.

Re: About rust and python

On Monday, March 16, 2020 at 7:17:17 AM UTC-4, Edward K. Ream wrote: > > In the "What's Next post > I > said: > > "As far as wasm [WebAssembly] goes, rust has a (temporary?) advantage over > python, because wasm doesn't yet

Re: Some excellent pytest articles/presentations

On Monday, March 16, 2020 at 6:23:50 AM UTC-4, Edward K. Ream wrote: > > As noted in the issue, it's not clear whether changing `g.app` to `self` > in the LeoApp class would make any real difference. It might well mislead > people. > I always thought that you had used this kind of idiom as a

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

On Monday, March 16, 2020 at 10:56:51 AM UTC-4, Edward K. Ream wrote: > > > Your copy of qt_gui.py was, once again, not formatted properly. Neither > file changed the name of the stylesheet. > I probably should not have included the qt_gui.py at all. I just wanted to make sure it didn't get d

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

On Monday, March 16, 2020 at 12:06:39 PM UTC-4, Edward K. Ream wrote: > > On Mon, Mar 16, 2020 at 10:57 AM Thomas Passin > wrote: > > > The correct file name is "vr3_rst.css". > > So I should change leo_vr3.css to vr3_rst.css? > Yes, please do that. I just

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

Thank you Matt. I followed your suggested steps, and they worked. I now have an all-new fork. One of the missing things for me (before you wrote) was how to delete my entire GitHub fork. I don't know how long that would have taken me to find if you had not pointed me to Settings. On Monday,

Re: Viewrendered3 Fails to Start With Leo 6.2-b1 But Works With Leo 6.1-final

One thing that made it a bit easier is that my work on VR3 took place outside of my Github clone. My development tree is in Mercurial on my own computer. I have a script hot-keyed to F5 that copies the files to the Github clone, and the only work I was doing was on VR3 - no other branches wer

Re: Initial F/B for installation of Leo 6.2b1 via pip from PyPI

Success on Linux Mint on VirtualBox- tom@tom-VirtualBox:~$ python3.6 -m pip install --upgrade leo==6.2b1 -- 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 leo-edito

Re: Initial F/B for installation of Leo 6.2b1 via pip from PyPI

It turns out that those were long-forgotten files in my workbook.leo outline. They didn't even exist on-disk in my Linux VMs. On Sunday, March 22, 2020 at 1:35:51 PM UTC-4, Matt Wilkie wrote: > > > not found: /home/tom/.leo/d:/test/jython/unicode/punct_py.py >> not found: /home/tom/.leo/d:/test/

Run Javascript Nodes With GraalJS

A few days ago, Edward (@ekr) showed how he runs Rust programs from Leo: https://groups.google.com/forum/#!topic/leo-editor/QBvmeT0zQyM It's possible to run Javascript programs from a Leo node, too. First, get GraalJS, which is available for all the major OSs. See https://github.com/graalvm/

Re: Run Javascript Nodes With GraalJS

GraalJS can use classes in jar files, and it even supports nodejs, but I haven't learned how yet. On Sunday, March 22, 2020 at 2:46:42 PM UTC-4, Thomas Passin wrote: > > A few days ago, Edward (@ekr) showed how he runs Rust programs from Leo: > > https://groups.google.com/

Re: A better way to send code to language processors

I often use mixtures of code and non-code blocks. If there are different kinds of code blocks, though, there can be problems if you want to execute them. If you execute each code block with a different language processor, you each block has to be independent of the others. They can't share c

Re: What's next, continued

On Monday, March 23, 2020 at 9:11:04 AM UTC-4, Edward K. Ream wrote: > > I have been working on this post for about a week. The way forward has > just become clear: > > The work in 2019 will be the foundation for this year's work. > > The work in 2020 will focus on bridges between programs. > >

Re: What's next, continued

On Monday, March 23, 2020 at 8:50:16 PM UTC-4, Thomas Passin wrote: > On Monday, March 23, 2020 at 9:11:04 AM UTC-4, Edward K. Ream wrote: >> >> >> The VR (and VR3?) plugins should be able to support Bokeh fairly easily. >> > > VR3 already does in this in a

Re: A better way to send code to language processors

be good at what they are good at, and let Leo be good at what it does well. On Tuesday, March 24, 2020 at 7:00:34 AM UTC-4, Marcel Franke wrote: > > > Am Dienstag, 24. März 2020 01:32:29 UTC+1 schrieb Thomas Passin: > > I often use mixtures of code and non-code blocks. If ther

Re: What's next, continued

On Tuesday, March 24, 2020 at 2:12:39 AM UTC-4, Edward K. Ream wrote: > > On Mon, Mar 23, 2020 at 7:56 PM Thomas Passin > wrote: > > > Here's a simple Bokeh program that VR3 executes: [snip] > > Thanks for this. Looks like holoviews/bokeh are just what we need. &

Re: What's next, continued

That's funny. I edited my post twice, but it seems that the original unedited version hung around. I don't know how that happened, but I'm sorry for the confusion. On Wednesday, March 25, 2020 at 7:49:47 AM UTC-4, Edward K. Ream wrote: > > On Wednesday, March 25, 2020 at 6:33:23 AM UTC-5, Edwa

Re: What's next, continued

On Wednesday, March 25, 2020 at 7:49:47 AM UTC-4, Edward K. Ream wrote: > > But I'm still having troubles. Importing matplotlib directly does work, > but the last line here throws an exception: > > import holoviews as hv > import holoviews.util > hv.extension('matplotlib') > > Here is the excep

Re: Crash in vr3

I get that too, sometimes, and I don't know why yet. On my system(s), VR3 keeps working even after that, though the same thing sometimes gets raised over and over, after it happens once. On my copy, I have changed the position of a try statement to catch this and at least be a little calmer ab

Re: What's next, continued

On Wednesday, March 25, 2020 at 10:32:23 AM UTC-4, Edward K. Ream wrote: > > On Wed, Mar 25, 2020 at 9:29 AM Thomas Passin > wrote: > > > This is one of those "I dunno, it works on my machine" things. I do not > recall installing anything besides matplotli

Re: What's next, continued

On Wednesday, March 25, 2020 at 4:56:34 PM UTC-4, Edward K. Ream wrote: > > On Wed, Mar 25, 2020 at 3:27 PM Thomas Passin > wrote: > > The exact packages that get installed probably doesn't matter. What *does* > matter is that python gets upgraded to 3.8 whenever I in

Re: What's next, continued

I installed Holoviews with pip on my Python 3.6 installation on a Linux Mint VM. After installation, the code above worked, and so did the bokeh example. Python was not upgraded to 3.8. So that must be an Anaconda thing. On Wednesday, March 25, 2020 at 6:26:31 PM UTC-4, Thomas Passin wrote

Re: Proposed change to setup.py

I still don't understand what "--editable" accomplishes, so I've never used it. What I do, when I want to run from a git clone, is to set PYTHONPATH to the top directory of the clone. Then Python will use, say, the Leo clone but it find all its other packages from the regular installed site-pa

Re: Partially solved: installing holoviews

On Thursday, March 26, 2020 at 10:09:37 AM UTC-4, Edward K. Ream wrote: > > [snip] > *Summary* > > Using pip to install matplotlib, bokeh and holoviews works on a Python 3.7 > conda environment. > > Still to do: allow holoviews to use matplotlib. > > I still don't understand how to use conda to

Re: What's next, continued

On Thursday, March 26, 2020 at 12:31:52 PM UTC-4, Edward K. Ream wrote: > > On Tuesday, March 24, 2020 at 6:20:15 PM UTC-5, Thomas Passin wrote: > > You can also embed an interactive Bokeh plot in a Markdown document using >> an iframe. The second attached screen capture fil

Re: What's next, continued

e complete. And yes, this version does work with the short form version of an element. These changes to viewrendered3.py are available in the VR3 branch of my repo at https://github.com/tbpassin/leo-editor On Thursday, March 26, 2020 at 4:26:53 PM UTC-4, Thomas Passin wrote: > > [Lat

Re: Partially solved: installing holoviews

da. But using it would leave out all those Leo users that don't run Anaconda. On Friday, March 27, 2020 at 6:01:07 AM UTC-4, Edward K. Ream wrote: > On Thursday, March 26, 2020 at 9:29:35 AM UTC-5, Thomas Passin wrote: > > This all makes me glad I've never worked with Anaconda.

Re: Partially solved: installing holoviews

> > Note that when I try to use Holoviews with Bokeh, I get this error: > > RuntimeError: Neither firefox and geckodriver nor a variant of chromium > browser and chromedriver are available on system PATH. You can install the > former with 'conda install -c conda-forge firefox geckodriver'. > > H

Re: Partially solved: installing holoviews

Ream wrote: > > On Fri, Mar 27, 2020 at 8:02 AM Thomas Passin > wrote: > > I discovered how to use Bokeh in Holoviews without needing that >> geckodriver. If you write the actual interactive output - instead of a png >> file - you don't even need a renderer. >

Re: Partially solved: installing holoviews

On Friday, March 27, 2020 at 7:44:33 PM UTC-4, Edward K. Ream wrote: > > >> RuntimeError: Neither firefox and geckodriver nor a variant of chromium >> browser and chromedriver are available on system PATH. You can install the >> former with 'conda install -c conda-forge firefox geckodriver'. >> >

Re: Partially solved: installing holoviews

for what seem to have been updates to Bokeh (or maybe my misunderstanding), so they are, I think, minimal examples that illustrate how to write the code. Edward, would you try them out and see what happens with your setup? Tom On Friday, March 27, 2020 at 10:09:39 PM UTC-4, Thomas Passin

Re: Partially solved: installing holoviews

On Saturday, March 28, 2020 at 4:47:31 AM UTC-4, Edward K. Ream wrote: > > On Fri, Mar 27, 2020 at 9:06 PM Thomas Passin > wrote: > > How do you want to work it? >> > > I'll send a zoom invite to tbp1...@gmail.com . You check > that regularly, right? >

Re: Partially solved: installing holoviews

That is strange. Oh, I forgot to write that I ran the non-VR3 examples using CTRL-B. I imagine that's how you tried them, too. On Saturday, March 28, 2020 at 4:45:07 AM UTC-4, Edward K. Ream wrote: > > On Fri, Mar 27, 2020 at 11:02 PM Thomas Passin > wrote: > >>

Re: Partially solved: installing holoviews

duced: curve1.opts(height=350, width=350) On Saturday, March 28, 2020 at 4:45:07 AM UTC-4, Edward K. Ream wrote: > > > > On Fri, Mar 27, 2020 at 11:02 PM Thomas Passin > wrote: > >> Here is a Leo outline containing >> >> 1 - Holoview examples using Bokeh, wit

Re: 6.2 final released. Public announcements later

Wait, I just upgraded om Windows using pip with install --upgrade leo==6.2, but here's what actually got installed: Leo 6.3-devel, VR3 branch, build 22239f2c7b On Saturday, March 28, 2020 at 3:39:30 AM UTC-4, Viktor Ransmayr wrote: > > Hello Edward, hello Matt, > > Am Freitag, 27. März 2020 11:2

Re: Partially solved: installing holoviews

March 28, 2020 at 7:26:58 AM UTC-4, Thomas Passin wrote: > > If you want to see something else really strange, the same > matplotlib-without-VR3 example that runs successfully with CTRL-B fails to > run from a (non-Leo) terminal. The error says that this line invokes > parameters

Re: 6.2 final released. Public announcements later

I got 6.2 final on Linux too, but not on Windows. On Saturday, March 28, 2020 at 7:59:09 AM UTC-4, Viktor Ransmayr wrote: > > Hello Thomas, > > Am Sa., 28. März 2020 um 12:37 Uhr schrieb Thomas Passin < > tbp1...@gmail.com >: > >> Wait, I just upgraded om Windows us

Re: 6.2 final released. Public announcements later

Did you use --upgrade leo==6.2 or just --upgrade leo? On Saturday, March 28, 2020 at 9:11:27 AM UTC-4, Viktor Ransmayr wrote: > > Hello Thomas, > > Am Sa., 28. März 2020 um 13:12 Uhr schrieb Thomas Passin < > tbp1...@gmail.com >: > >> I got 6.2 final on Linux too, bu

Re: 6.2 final released. Public announcements later

Passin wrote: > > I got 6.2 final on Linux too, but not on Windows. > > On Saturday, March 28, 2020 at 7:59:09 AM UTC-4, Viktor Ransmayr wrote: >> >> Hello Thomas, >> >> Am Sa., 28. März 2020 um 12:37 Uhr schrieb Thomas Passin < >> tbp1...@gmail.com>: &

Re: New installation log

Linux Mint is still on Python 3.6.9. I think that's what Debian is at too, since the Mint package is basically the Debian one, though I'm not 100% sure. Ubuntu is up to 3.7.6 On Saturday, March 28, 2020 at 10:00:14 AM UTC-4, Edward K. Ream wrote: > > Leo is supposed to run on python 3.6, but

Re: Partially solved: installing holoviews

I have updated this Leo file to add Seaborn plots, and to make a few minor adjustments in the other examples. Note that Seaborn uses matplotlib behind the scenes. On Saturday, March 28, 2020 at 12:02:15 AM UTC-4, Thomas Passin wrote: > > Here is a Leo outline containing > >

Re: Proposed change to setup.py

On Saturday, March 28, 2020 at 4:45:01 PM UTC-4, Matt Wilkie wrote: > > > I've always installed Leo using pip. I still don't understand what >> "--editable" accomplishes, so I've never used it. >> > > *Pip install* puts the package and all the files it says it needs under > PYTHONHOME/Lib/site-p

Re: Leo's legacy layout is now the default in version 6.2

I kind of get you there adding a setting @string central-dock-widget = outline I have attached a screen shot. BTW, you can add this setting just to one outline, so you can experiment with changing the setting(s) by closing the outline and reloading it, rather than needing to restart Leo each t

Re: Proposed change to setup.py

On Sunday, March 29, 2020 at 12:34:56 AM UTC-4, Matt Wilkie wrote: > > So if I had a clone at d:\leo-clone, I'd use the following? >> >> python -m pip install --editable d:\leo-clone >> > > Yes. > > I have been setting PYTHONPATH to d:\leo-clone. This has always seemed >> to work well, as lon

Re: Leo's legacy layout is now the default in version 6.2

On Sunday, March 29, 2020 at 10:42:33 AM UTC-4, Viktor Ransmayr wrote: > > Hello Thomas, > > Am So., 29. März 2020 um 15:09 Uhr schrieb Thomas Passin < > tbp1...@gmail.com >: > >> I kind of get you there adding a setting >> >> @string central-dock-widge

Re: New installation log

On Sunday, March 29, 2020 at 2:49:30 PM UTC-4, vitalije wrote: > On Sunday, March 29, 2020 at 8:21:05 PM UTC+2, Matt Wilkie wrote: > > >> Linux can add a wrinkle with system vs user installs. User is safer in my >> experience. Morale of a painful story I havn't shared: *never use `sudo >> pip

Re: Proposed change to setup.py

On Sunday, March 29, 2020 at 2:01:53 PM UTC-4, Matt Wilkie wrote: > > > >> Then you could keep your development code entirely out of the repo until >> you got it to the point it was worth adding it in. I'd like that. >> > > A fork would accomplish this and then be easier to add to the main re

Re: New installation log

On Sunday, March 29, 2020 at 10:07:37 PM UTC-4, Matt Wilkie wrote: > > > OTOH, I have installed as a user on Linux distros many times using pip >> install, and never had a problem. Never needed to consider using sudo. >> > > Did/do you need to pass `--user` flag? > > pip install --user foobaz >

Re: Proposed change to setup.py

On Sunday, March 29, 2020 at 10:29:36 PM UTC-4, Matt Wilkie wrote: > > > b) merge early, merge often: as in daily. This way any given upstream > change can (usually) be inspected and accepted/changed in small easy to > understand pieces. Use an interactice tool like WinMerge or Kmeld to do the >

Re: About VR3, holoviews and bokeh

On Monday, March 30, 2020 at 6:48:29 AM UTC-4, Edward K. Ream wrote: > > Thomas Passin and I have been discussing his vr3 plugin in private. This > is a great plugin. Our conversation deserves to be public. > > Here I am going to discuss what I think I know about vr3 and its

Re: About VR3, holoviews and bokeh

On Monday, March 30, 2020 at 6:48:29 AM UTC-4, Edward K. Ream wrote: > > Thomas Passin and I have been discussing his vr3 plugin in private. This > is a great plugin. Our conversation deserves to be public. > > *Holoview and bokeh* > > Holoviews and bokeh are both imp

Re: About VR3, holoviews and bokeh

And it turns out that it's just as easy to work with Seaborn, On Monday, March 30, 2020 at 6:48:29 AM UTC-4, Edward K. Ream wrote: > > Thomas Passin and I have been discussing his vr3 plugin in private. This > is a great plugin. Our conversation deserves to be public. > >

Re: About VR3, holoviews and bokeh

On Monday, March 30, 2020 at 8:12:25 AM UTC-4, Thomas Passin wrote: > On Monday, March 30, 2020 at 6:48:29 AM UTC-4, Edward K. Ream wrote: >> >> Thomas Passin and I have been discussing his vr3 plugin in private. This >> is a great plugin. Our conversation deserves to be pu

Leo and Jupyter

Users of Leo who also know about Jupyter and Jupyter notebooks sometimes see a resemblance. Now that the Viewrendered3 plugin is working (though still a beta version), it is possible to render code and the resulting text and plots in the same node, which looks even more like Jupyter. Leo alre

Re: Leo and Jupyter

On Monday, March 30, 2020 at 12:06:47 PM UTC-4, Thomas Passin wrote: > > > This thread is for discussion about how or whether Leo might be able to > play with Jupyter. > > I think there are basically four general ways that Leo could interact with > Leo. I'll put them

Re: Leo and Jupyter

On Monday, March 30, 2020 at 12:08:15 PM UTC-4, Thomas Passin wrote: > > On Monday, March 30, 2020 at 12:06:47 PM UTC-4, Thomas Passin wrote: >> >> >> This thread is for discussion about how or whether Leo might be able to >> play with Jupyter. >> >>

Re: Leo and Jupyter

Leo Notebooks are easy to imagine, but they would never gain traction because Jupyter is too widespread already. On Monday, March 30, 2020 at 12:13:04 PM UTC-4, Thomas Passin wrote: > On Monday, March 30, 2020 at 12:08:15 PM UTC-4, Thomas Passin wrote: >> >> On Monday, March 30

Re: Leo and Jupyter

On Monday, March 30, 2020 at 12:15:44 PM UTC-4, Thomas Passin wrote: > > > The questions are: what benefit could be gotten by interfacing Leo and > Jupyter, and how complicated would it be? > It seems to me that it would be essential to be able to embed images, interactive plot

Re: Leo and Jupyter

On Monday, March 30, 2020 at 1:04:12 PM UTC-4, Edward K. Ream wrote: > > On Mon, Mar 30, 2020 at 11:08 AM Thomas Passin wrote: > > > There are at least four ways that Leo could be used with Jupyter: > > All four look like the hard way. I tried the hard way several times las

Re: Jupyter vs Leo + VR3

On Tuesday, March 31, 2020 at 7:24:52 AM UTC-4, Edward K. Ream wrote: > > Thomas, this post is an indirect response to some of your recent remarks. > There is some chance I understand them now ;-) > It's hard to grasp some of these things until you get time to work with VR3 for a while. I hav

Re: Jupyter vs Leo + VR3

On Tuesday, March 31, 2020 at 9:28:01 AM UTC-4, Edward K. Ream wrote: > > On Tue, Mar 31, 2020 at 8:19 AM Thomas Passin > wrote: > > Thomas, seems like we agree on points 1 through 3. > > >> 4. In the (unlikely?) event that people want to use both Leo and > jupyter,

Re: Leo and Jupyter

On Tuesday, March 31, 2020 at 11:44:23 AM UTC-4, tfer wrote: > > There is another option we could consider, rather than full juypter stuff, > just add an Ipython console tab to the log pane. Though not as sexy as > Juypter, it has most of its guts, is embed-able, and has a qt-console > version.

Re: font size shrank

On Tuesday, March 31, 2020 at 2:36:38 PM UTC-4, andyjim wrote: > > All was well this morning, but just now when I came back to Leo I find the > font size in the bodies of all nodes has shrunk to unreadably small size. > Text in the node headings is ok. This is only for one of six files I have

Re: font size shrank

On Tuesday, March 31, 2020 at 2:36:38 PM UTC-4, andyjim wrote: > > All was well this morning, but just now when I came back to Leo I find the > font size in the bodies of all nodes has shrunk to unreadably small size. > Text in the node headings is ok. This is only for one of six files I have

Re: Jupyter vs Leo + VR3

On Tuesday, March 31, 2020 at 6:42:01 PM UTC-4, Edward K. Ream wrote: > As you say, at present the ipynb.do_cell method (in importers/ipynb.py) > doesn't save much. It should be easy to save more, if you would like that. > Easy, perhaps, but I would go slowly, because we need to look at how bes

Re: Tom, please fix vr3 in devel

It's always annoying when there are a lot of whitespace and style changes. I'll see about it. (For the benefit of anyone else who is wondering what this is about, this commit tried to remove a lot of complaints from pylint about whitespace and other sylistic nits that had build up over the ye

Re: Tom, please fix vr3 in devel

OK, I'll take a run at it. It can be difficult when you have a lot of changes that are mostly whitespace and style. I messed one up that way recently, too. On Wednesday, April 1, 2020 at 6:34:20 AM UTC-4, Edward K. Ream wrote: > > Somehow it contains many conflict markers from a git merge. > >

Re: FYI: pylint-leo-rc-ref.txt disables two new pylint tests

It's really a comprehension vs brevity thing, and a matter of personal style. Removing the else gives shorter code, which is *usually* more clear. But without the *else*, you have to *infer* its virtual presence, and that interferes with comprehension. I'm not consistent about this, myself.

Re: warning viewrender3.py plugin has been commited with unresolved conflicts.

Right, and Edward has already asked me to clear this up. On Wednesday, April 1, 2020 at 11:38:11 AM UTC-4, vitalije wrote: > > Title tells everything. While executing unit tests, two are failing: check > syntax of all files and pluginTestCase for viewrender3. > > Vitalije > -- You received this

Re: Thomas, the fix branch fixes vr3.py

I merged the files. For some reason I couldn't push my changes upstream to the "fix" branch. Although I was logged in to Github, it said it couldn't authenticate. So I pushed the changes to the devel branch of my own fork and sent a PR. On Wednesday, April 1, 2020 at 6:31:35 PM UTC-4, Edwar

Re: For Thomas: C++ problems in pyqt

Thanks, that looks like it could be relevant. On Wednesday, April 1, 2020 at 6:37:34 PM UTC-4, Edward K. Ream wrote: > > Here's the link. I found it while cleaning out my bookmarks :-) > > http://enki-editor.org/2014/08/23/Pyqt_mem_mgmt.html > > Maybe it will help find the memory problems in vr3.p

Re: Thomas, the fix branch fixes vr3.py

Edward, if you retrieved this file before about 9:30 PM Eastern, please get it again. I found a bug where the line h = c.hash() in store_layout had been removed during my merge. It needs to be there. On Wednesday, April 1, 2020 at 8:00:27 PM UTC-4, Thomas Passin wrote: > > I merged the

Re: For Thomas: C++ problems in pyqt

20 at 8:14:04 PM UTC-4, Thomas Passin wrote: > > Thanks, that looks like it could be relevant. > > On Wednesday, April 1, 2020 at 6:37:34 PM UTC-4, Edward K. Ream wrote: >> >> Here's the link. I found it while cleaning out my bookmarks :-) >> >> http://enki-e

  1   2   3   4   5   6   7   8   9   10   >