Re: Need help installing Leo on Windows

I believe you need to type two dashes for the "editable" parameter: pip install --editable On Monday, October 3, 2022 at 7:23:41 AM UTC-4 uu86...@gmail.com wrote: > I also tried to avoid cloud storage and whitespace issues by copying Leo > to another folder, but it did not help either: >

Re: Approximate String Searching

On Monday, October 3, 2022 at 6:18:43 AM UTC-4 Edward K. Ream wrote: > On Sun, Oct 2, 2022 at 11:57 PM Thomas Passin wrote: > >> Here is a fairly easy way to search in a body of text for a string when >> there might be single-character errors in a pattern, like searching

Approximate String Searching

Here is a fairly easy way to search in a body of text for a string when there might be single-character errors in a pattern, like searching for "butterfly" but the text has "budterfly". The paper, which I found some years ago, dates back to I think the early 1990s. It has a basic method with

Re: LeoInteg 1.0.8 released! Major bug fixes and features update!

Apart from why this isn't working, it looks like the bridge should guard against receiving None here when it expected to get an actual commander; and an error message could be output as well. On Friday, September 30, 2022 at 4:26:04 PM UTC-4 Félix wrote: > Hmm.. that's weird... from your error

More Git Magic

>From time to time Git issues show up here. My understanding about using Git is very superficial, to say the least. Today's column from Raymond Chen, in The Old New Thing, discusses a complex situation. I don't really understand it yet, but the post might be worthwhile for other Git users to

Re: Discuss: require Qt 6.1+ for Leo 6.7.1?

As always, the problem is most acute for Linux users. Often, the newer packages just aren't available. In some or many Linux distros you are supposed to install things like Qt or even PyQt through the distro's package manager - for example, Debian doesn't have a Python 3.10 package yet. It

Re: Recent changes to devel

Great! Can we set up a new milestone, say 6.7.1, so that we can set a target for new issues? On Tuesday, September 20, 2022 at 12:49:25 PM UTC-4 Edward K. Ream wrote: > Absent new bug reports, I plan no further changes for Leo 6.7.0. Let the > testing continue! > -- You received this

Re: A problem with @button write-leoPyRef

PM UTC-4 Edward K. Ream wrote: > On Tue, Sep 20, 2022 at 11:29 AM Thomas Passin wrote: > > Looking through the write-leoPyRef, I infer that we are expected to create >> a leoPy.leo outline - presumably by saving LeoPyRef by that name, doing all >> our work in leoPy, then usi

Re: Recent changes to devel

I see an improvement in that on my Windows system, the freeze time when saving LeoPyRef is now only a few seconds instead of the previous 20 - 30 seconds. That's a worthwhile improvment! On Tuesday, September 20, 2022 at 12:14:39 PM UTC-4 Edward K. Ream wrote: > PR #2877

Re: A problem with @button write-leoPyRef

I'm confused about leoPy.leo vs LeoPyRef.leo. Some time ago, things were changed so that opening leoPy from the File/Open Specific Leo File menu actually opened LeoPyRef. There is no leoPy.leo in the leo/core directory any more. Looking through the write-leoPyRef, I infer that we are

Re: Typing Annotation Question

I thought something like that would work; I was hoping for a way that doesn't involve that area of the codebase. Thanks for your help. On Monday, September 19, 2022 at 4:10:55 PM UTC-4 jacc...@gmail.com wrote: > I'm not very familiar with the Leo codebase and don't have a working > version

Typing Annotation Question

In Leo's code base, there are many method calls like this: c.config = leoConfig.LocalConfigManager(c, previousSettings) However, mypy doesn't know that c (known to be a leo.core.leoCommands.Commands) has an attribute *config* (which we actually know is a leo.core.leoConfig.LocalConfigManager).

Re: Two more PRs merged into devel

Of course; I just ran those by hand to try them out. On Sunday, September 18, 2022 at 7:35:05 PM UTC-4 Edward K. Ream wrote: > On Sun, Sep 18, 2022 at 4:00 PM Thomas Passin wrote: > >> Just trying to make sure I understand correctly. >> >> py -m unittest >> Ran 94

Re: Two more PRs merged into devel

Just trying to make sure I understand correctly. py -m unittest Ran 941 tests in 12.162s OK py -m unittest leo.unittests.core.test_leoserver Ran 3 tests in 2.017s OK On Sunday, September 18, 2022 at 4:00:13 PM UTC-4 Edward K. Ream wrote: > On Sun, Sep 18, 2022 at 2:57 PM Thomas Passin wr

Re: Two more PRs merged into devel

So what's the way for devs to test after making changes, if we don't have test-all? On Sunday, September 18, 2022 at 3:53:55 PM UTC-4 Edward K. Ream wrote: > On Sun, Sep 18, 2022 at 2:47 PM Viktor Ransmayr wrote: > > > It looks like the 'test-all' command does no longer work ... > > Correct. I

Re: OT-ish: convert script from camelCase to snake_case

For me, I barely notice typing the underscore, but my hands really dislike typing a "#" for a comment. So when I need a line-oriented data format of my own, I usually allow a ";" as well as a "#" to comment out a line. On Thursday, September 15, 2022 at 9:29:13 AM UTC-4 jkn wrote: > Yeah,

Re: OT-ish: convert script from camelCase to snake_case

FWIW (maybe not much), Wikipedia's page on CamelCase includes this - 'A 2010 follow-up study, with other subjects containing mainly pre-trained programmers and using an improved measurement method with use of eye-tracking equipment, indicates: "While results indicate no difference in accuracy

Re: OT-ish: convert script from camelCase to snake_case

pylint by default will issue message C0103 for functions and methods that should be snake_case but aren't (Leo's configuration doesn't seem to pick that up). Since the pylint message will include line and position of the name, it shouldn't be hard to write a program to convert these instances.

Re: Leo is (and should be) complete

It sounds like retiring all over again. There are a lot of folks who are very happy that you've stuck to it those 30 years! Thank you. On Wednesday, September 14, 2022 at 6:51:43 AM UTC-4 Edward K. Ream wrote: > At 5 am this morning, I declared the last Leo issue to be complete. I do > not

Re: UUID gnx support

I wondered about that message. I saw it after pulling the latest changeset. I haven't set any uAs of any variety, myself. Maybe some plugin? On Wednesday, September 14, 2022 at 4:53:20 AM UTC-4 Edward K. Ream wrote: > On Wednesday, September 14, 2022 at 3:38:40 AM UTC-5 Edward K. Ream

Re: UUID gnx support

It's far from clear to me that any code that uses gnx's needs to know anything except their string value. You want to avoid inadvertent duplications of gnx's (so copies can be distinguished from clones, for one thing), that's clear. Not that I've studied the code or anything useful like

Re: where does "leo_auto_file" come from?

Always more to find out about ... On Monday, September 12, 2022 at 4:30:53 PM UTC-4 map...@gmail.com wrote: > *follow up from > https://github.com/leo-editor/leo-editor/issues/2824#issuecomment-1241276950 > > > so

Re: I wonder why...

If you want to try using PyQt5, you can pip-install it by hand. Here's how I get Leo to use PyQt5 instead of PyQt6; I think it's the easiest way to do it. I navigate to Lib/site-packages and find the PyQt5 and PyQt6 directories. I just rename the PyQt6 directory to something else, usually

Re: I wonder why...

I see that the linked page with dependencies still says that Leo can run using Python 3.6. But when I tried to run it using Python3.7, it wouldn't run and gave a message that 3.9 or higher is required. The language should be updated for the 6.7 branch, I think. On Sunday, September 11, 2022

Re: PR #2813 has been merged into devel

One good thing I've noticed - the "marked node" icons turn on and off without a noticeable delay. I've posted before about unpleasant delays with these icons, so this is a real improvement. On Sunday, September 11, 2022 at 2:33:03 AM UTC-4 Edward K. Ream wrote: > PR #2813

Re: Leo deployment on Windows - past story, current story

I don't install from my git clone, and I don't install with --editable. Instead, I pip-install Leo, whatever version that may be. This brings in the dependencies that Leo needs. I run Leo using my git clone by setting the PYTHONPATH variable to point to the clone's directory. I then run Leo

Re: Devel contains a *huge* performance boost

Marking/unmarking a node still incurs a perceptible delay on my system, though. Do icons get cached? On Saturday, September 3, 2022 at 1:40:04 PM UTC-4 Edward K. Ream wrote: > PR #2803 fix one of > the biggest performance bugs in Leo's

Re: Hewlett-Packard Calculator Emulators

/paste the calculation results to/from the clipboard, and there are MacOS and Linux versions available. On Friday, August 26, 2022 at 11:42:51 PM UTC-4 Thomas Passin wrote: > Sometimes there is just no substitute for a good calculator. Even though > your computer could do a little calcula

Re: Leo 6.6.4 released

Works in Debian 11/Python 3.92 for me too. On Thursday, September 1, 2022 at 4:56:42 PM UTC-4 Edward K. Ream wrote: > On Thu, Sep 1, 2022 at 1:31 PM Viktor Ransmayr > wrote: > > A confirmation that Leo 6.6.4 is working properly, when installed from >> PyPI into a Debian 11 - as well as - a

Re: newbie stuck

Let's try that again. Maybe this will be usable: "tbp100 dot tp" [this line intentionally left blank] "at" [this line intentionally left blank] "gmail.com" On Thursday, September 1, 2022 at 7:28:52 AM UTC-4 Thomas Passin wrote: > Sometimes I can get Groups to s

Re: Pls test --gui=console in devel and ekr-undo

I got the crash in devel, and no crash but no functionality in the ekr-undo branch, I have never used the console gui in practice, but I would imagine I would want to be able to use the minibuffer. On Saturday, November 14, 2020 at 11:54:58 AM UTC-5, Edward K. Ream wrote: > > > It looks like

Re: Google groups no longer provides pinned posts, program highlighting?

*New Google Groups to replace classic Groups on Nov. 16 - * https://its.umich.edu/communication/collaboration/google/update/new-google-groups-replace-classic-groups-nov-16 On Friday, November 13, 2020 at 8:17:15 AM UTC-5, Edward K. Ream wrote: > > > > On Thursday, November 12, 2020 at 3:00:05

Re: Google groups no longer provides pinned posts, program highlighting?

I'm still using classic groups. When they changed me over, I disliked the new skin and found some link to get back, but I forget where it was. And yes, no ability to edit, GRRR. I figured that at some point, we'd all get stuck with the new look. On Thursday, November 12, 2020 at 4:00:05 PM

Re: Themes documentation

On Monday, November 9, 2020 at 3:22:04 PM UTC-5, Rob wrote: > > Having trouble loading different themes, so checking the online > documentation here as > well as looking at the main LeoSettings.leo and other theme files, I'm > confused. > >

Re: ENB: Simplifying undo

On Sunday, November 8, 2020 at 9:08:19 AM UTC-5, Edward K. Ream wrote: > > > The first phase is nearing completion. It simplifies the logic concerning > headlines. This is tricky because editing headlines involves *ephemeral *Qt > "headline editor" widgets. > Ephemeral - do you mean you

Re: Leo Themes with VR3

That's not a message from VR3. The stylesheets for VR3 are contained in the leo/plugins/viewrendered3 directory. You can override that by the setting @string vr3-rst-stylesheet, or vr3-md-stylesheet for Markdown, so you could check to see if that setting exists in your myLeoSettings.leo

Re: Placement behaviour of VR vs. VR3 pane?

On Wednesday, November 4, 2020 at 5:44:17 PM UTC-5, Viktor Ransmayr wrote: > > Am Mi., 4. Nov. 2020 um 23:40 Uhr schrieb Thomas Passin >: > >> Odd, since for me the position is stable on the right after the default >> layout command. >> >> Does "

Re: Placement behaviour of VR vs. VR3 pane?

Odd, since for me the position is stable on the right after the default layout command. Does "PyVE" in the paths indicate you are using a virtual environment? Not that I can see why that would make a difference. Let me just confirm that you deleted the db apps directory. On Wednesday,

Leo Themes with VR3

It has been pointed out that VR3 does not pick up dark Leo themes, like the current default EKRDark. Viewrendered, on the other hand, does. I've been looking at addressing this, and could use some help on a few things. First of all, I haven't been able to see how the Leo theme gets applied to

Re: Placement behaviour of VR vs. VR3 pane?

I have just checked the 6.3 branch, and get the same results. After deleting the .leo/db/g_app_db, VR3 first opened on the left. After applying vr3-use-std-layout, VR3 opened correctly on the right for every outline I opened, and did so again after closing and restarting Leo. VR opened on

Re: Placement behaviour of VR vs. VR3 pane?

it here later. Would you be so good as to try it yourself and let me know what happens? I recall that you had said earlier that deleting the db database didn't help, but I think that was on a pre-6.3.1 version. On Tuesday, November 3, 2020 at 9:29:47 AM UTC-5, Thomas Passin wrote

Re: Placement behaviour of VR vs. VR3 pane?

On Tuesday, November 3, 2020 at 1:53:35 AM UTC-5, Viktor Ransmayr wrote: > > > Looking forward to your findings. - Let me know if I can help. - From a > platform POV its Fedora & Debian Linux, but if needed I can start a Windows > VM ... > And I have Mint and Ubuntu as VMs also. I develop on

Re: Placement behaviour of VR vs. VR3 pane?

I'm not sure that VR and VR3 have different behavior. There seems to be an element of caprice in VR3's placement - still trying to track that down - and I have seen VR also open in unplanned positions on my system. The layout code is practically identical between the two plugins, so I would

Re: Placement behaviour of VR vs. VR3 pane?

Well, I'm still working on the VR3 problem. The main questions, I think, are who would approve a PR, and who is going to keep the devel branch on GutHub up to date. Edward has written that he will still have some involvement - he won't be completely absent. On Sunday, November 1, 2020 at

Re: Placement behaviour of VR vs. VR3 pane?

On Wednesday, October 28, 2020 at 3:52:08 PM UTC-4, Viktor Ransmayr wrote: > Thanks a lot for your two messages. - Applying the command - or - the > suggested button is a good workaround. > > But IMO other Leo users will experience similar issues, when the have to > switch from the

Re: Placement behaviour of VR vs. VR3 pane?

my named layout, and then if necessary running the vr3-use-default-layout command. You could also save another layout named, say, vr3-right after you get the layout with VR3 that you like, and restoring that should always work. On Tuesday, October 27, 2020 at 4:08:07 PM UTC-4, Thomas Passin

Re: Placement behaviour of VR vs. VR3 pane?

Hi, Viktor, I don't think you need to open an issue on Github since I plan to keep working on the problem after Leo 6.3 Final is released. In the meantime, once you get a layout that you like, I suggest saving it using the splitter bar context menu. After a restart you can load that named

Re: Placement behaviour of VR vs. VR3 pane?

Hi, Viktor, In recent development releases of Leo, the VR3 pane has sometimes started to show up in unexpected places. I have also had the VR pane open in undesired locations too (VR and VR3 use almost identical code for the on/off behavior). This seems to be caused by some bug/feature in

Re: ENB: Are huge Leo outlines possible?

Remembering back to long ago when we only had 640k of RAM at the most, there were editors that kept three screens of data in memory at once - the current page, the previous page, and the next page. When the user scrolled forward (say), parts of the next page would be brought in, and when

Re: Discuss: drop support for Qt docks?

Just for historical interest and for your entertainment (since we've dropped docks) - "... offers a more deterministic dock placement API (in Qt it is very difficult to programatically generate complex dock arrangements)" - from an example of pyqtgraph's dock widget system. On Friday,

Re: I am about to start a sabbatical

How will PRs going to be handled? On Friday, October 16, 2020 at 10:49:05 AM UTC-4, Edward K. Ream wrote: > > This post will be pre-writing for a locked, pinned post. > > Don't panic: during my sabbatical I'll continue to answer questions and > fix serious bugs. > > Now seems like an excellent

Viewrendered3 How-To -- Insert An Interactive Plot Using Holoview and Bokeh

An interactive plot can be inserted into the VR3 rendering pane. This example uses Holoview and Bokeh. Copy the following code into a node, open VR3, and execute it. A screenshot is attached. This can be done using Bokeh alone, but it's a little simpler using Holoview. Matplotlib figures

Re: Undo Removed A Sequence Of Changes Instead Of Just The Last One

. Ream wrote: > > On Thu, Oct 15, 2020 at 8:22 AM Thomas Passin > wrote: > >> Thanks, Edward. BTW, I can't find any reference to it in the docs I >> looked at. I had to search LeoPyRef.leo to find it and the values it >> recognizes. Of course, I would have had no i

Re: Undo Removed A Sequence Of Changes Instead Of Just The Last One

. Ream wrote: > > On Wed, Oct 14, 2020 at 3:19 PM Thomas Passin > wrote: > > I needed to add some spaces between parts of a string. I added several, >> then wanted to undo just the last one. Instead, all of the added spaces >> were removed with a single command.

Undo Removed A Sequence Of Changes Instead Of Just The Last One

I needed to add some spaces between parts of a string. I added several, then wanted to undo just the last one. Instead, all of the added spaces were removed with a single command.. This behavior repeated several times. Leo 6.3-devel, devel branch, build b6c70a7c7e 2020-10-13 16:26:16

Re: Select Multiple Nodes With Mouse Does Not Highlight Uppermost Node

Could you tell if it's only a highlight problem or if the selection itself is affected? On Wednesday, October 14, 2020 at 11:33:48 AM UTC-4, Edward K. Ream wrote: > > > > On Wed, Oct 14, 2020 at 9:50 AM Thomas Passin > wrote: > >> Leo 6.3-devel, devel branch, build b6c

Select Multiple Nodes With Mouse Does Not Highlight Uppermost Node

Leo 6.3-devel, devel branch, build b6c70a7c7e 2020-10-13 16:26:16 -0400 Python 3.8.5, PyQt version 5.15.1 Windows 10 AMD64 (build 10.0.19041) SP0 When I click on a node in the outline pane, hold the shift key and then click on a later node - expecting all the nodes between clicks to be selected

Re: How to diff one file across branches *or* revs

In which branch have you done the renaming? On Tuesday, October 13, 2020 at 9:54:46 AM UTC-4, Edward K. Ream wrote: > > leoPy.leo contains the following script. I have just renamed it to > "script: diff-branches/revs (one file)": > > import leo.commands.editFileCommands as efc > >

Re: More protections and suggestions about branches

On Tuesday, October 13, 2020 at 9:50:10 AM UTC-4, Edward K. Ream wrote: > > I'm no expert on git workflow, but I don't see how PR's can possibly be a > substitute for cooperation based on close communication. > I wouldn't disagree, but I've seen on some other projects that a PR sometimes

Re: Thomas, did you mean to change the VR3 layout?

No. This happens to me once in a while, too. The only remedy I know is to run the vr3-use-default-layout command. Once the position is back to normal, I haven't been able to make it misbehave on purpose. Very annoying, because it's nearly impossible to debug. The only clue I know is that

Re: More protections and suggestions about branches

Do you mean that every branch that is not devel, and that person A did not create, should never be pushed to or merged by anyone else besides person A? That's an easy-enough rule to understand. Up til now, I had thought that if some feature was being worked on, say layout code, that everyone

Re: How to Make Replace-all Work?

Ah, I see. I didn't see anywhere about the part. Not obvious to me. Thanks! On Tuesday, October 13, 2020 at 3:49:25 AM UTC-4, Edward K. Ream wrote: > > On Mon, Oct 12, 2020 at 7:52 PM Thomas Passin > wrote: > >> Every time I try to use the command replace-all, I can't

How to Make Replace-all Work?

Every time I try to use the command replace-all, I can't get it to work. I enter the search string and replacement string. Then what? If I enter replace-all in the minibuffer, I see the search string highlighted. If I hit enter, the first found instance gets highlighted but not replaced. I

Re: layouts branch has been merged into devel

I made a few changes to my own version, and they eliminated an error I got when I tried to invoke the new vr3-use-default-layout command after opening VR3 using the splitter context menu Insert/viewrendered3. On Monday, October 12, 2020 at 1:32:46 PM UTC-4, Thomas Passin wrote: Note

Re: layouts branch has been merged into devel

On Monday, October 12, 2020 at 12:09:24 PM UTC-4, Edward K. Ream wrote: > > The fix is in the layout2 branch. The corresponding PR is #1705 > . I have walked back > some of the changes I made to Terry's code, but the actual cause of the >

Errors opening LeoPyRef.leo from Leo 6.3-devel, devel branch, build 486336f1e2

I got these errors opening the latest devel LeoPyRef.leo - initWriteIvars Did not create ../plugins for ../plugins/leo_cloud.py = PERFECT IMPORT FAILED = leo_cloud.py len(s1): 585 len(s2): 0 all common lines match initWriteIvars Did not create ../plugins for

Re: Leo has a new default look

On Monday, October 12, 2020 at 10:48:48 AM UTC-4, Edward K. Ream wrote: > > On Mon, Oct 12, 2020 at 8:32 AM Rob > > wrote: > > resolve_urls None ==> NOT FOUND >> > > This message is probably not relevant. > > What do you have in your myLeoSettings.leo for @string theme-name ? > > Setting it to

Re: layouts branch has been merged into devel

I do have a problem. It's with add_adjacent(), in class NestedSplitter. It has an inner function hunter(layout, id_). hunter() expects to receive a layout dictionary with a key 'content'. However , when add_adjacent() is called by VR3, what gets handed to hunter() is not a dictionary but a

Re: Viewrendered3 Updated to V3.0rc4

Speaking of that, what changes did you make to the splitter layout methods? I'm getting a number of exceptions that didn't happen before. On Sunday, October 11, 2020 at 3:20:25 PM UTC-4, Edward K. Ream wrote: > You can do whatever you like in VR3, but I'm not going to worry about the > kind

Re: Viewrendered3 Updated to V3.0rc4

I'll put in a conditional call in VR3 in case one or the other method doesn't work. On Sunday, October 11, 2020 at 2:51:11 PM UTC-4, Edward K. Ream wrote: > > > > On Sun, Oct 11, 2020 at 10:41 AM Thomas Passin > wrote: > >> Strange ... nothing has been changed about t

Re: Leo has a new default look

PM UTC-4, Edward K. Ream wrote: > > On Sunday, October 11, 2020 at 1:14:30 PM UTC-5, Thomas Passin wrote: > > > It seems to me that the .leo directory would be a good place to put > modified theme outlines. There could be a .leo/themes subdirectory. > > Leo

Re: Leo has a new default look

It seems to me that the .leo directory would be a good place to put modified theme outlines. There could be a .leo/themes subdirectory. On Sunday, October 11, 2020 at 12:46:19 PM UTC-4, Thomas Passin wrote: > > I don't want to modify the default theme file or my changes would be lost &

Re: Leo has a new default look

My problem with this is not individual settings like font size, but that I don't know how to modify them. I tried copying specific settings to myLeoSettings.leo under a new node @theme Default, but they don't take effect. When I open myLeoSettings.leo, then I seem to get some kind of strange

Re: Viewrendered3 Updated to V3.0rc4

Strange ... nothing has been changed about this in VR3. Edward, has something been changed in JEditColorizer? I have never seen this error before, and it ought to be showing up all the time.. On Sunday, October 11, 2020 at 9:43:10 AM UTC-4, zhaohe wang wrote: > > run vr3, errors displayed >

Re: Viewrendered3 Updated to V3.0rc4

One thing I see is that the VR3 version has some experimental code that I had not intended to have outside of my own clone. I'll have to revert that. It must have been some mistake on my part.But that would not be the cause of this problem, which I'm now seeing but didn't a day or two before.

Re: undoing unmarks nodes

I figured it must be VIM or emacs! On Thursday, October 8, 2020 at 8:43:55 AM UTC-4, gar wrote: > > I am VIMer, so I cannot imagine too complicated actions :-) I can see only > different levels of simplicity > > чт, 8 окт. 2020 г. в 15:35, Thomas Passin > >: > >>

Re: undoing unmarks nodes

@gar: "Actually when I started to use Leo I expected that there would be several undoes: for tree operations, for editors, for marks/clones etc." That sounds too complicated to me. I would expect undo actions to happen in the (reverse) order that I did them in the first place, whether they

Viewrendered3 Updated to V3.0rc4

I have just merged a new release candidate of VR3 into the devel branch. The main changes for users are 1. There are two new VR3-* Leo commands to zoom or unzoom the zoom size in the rendering pane. These commands can be linked to shortcut keys - typically, one would use the same keys as

Re: Leo won't start

I feel for you! Good luck. On Tuesday, October 6, 2020 at 2:12:10 PM UTC-4, Rob wrote: > > No, I get nothing, just returns to command line. I have Python installed > through Anaconda and now it's stuck trying to update. I'm afraid I'll have > to reinstall everything; not looking forward to

Re: Leo won't start

Do you get anything from python -V? Probably better just to reinstall and be done with it. On Tuesday, October 6, 2020 at 12:47:21 PM UTC-4, Rob wrote: > > Well, it's not a Leo problem as I can't even execute a simple Hello World > script. I must have a corrupted Python installation. Looks

Re: TravisCI checks no longer required for PR's

Well, I did submit the PR. Is there something else I should be doing with it? On Tuesday, October 6, 2020 at 9:56:15 AM UTC-4, Edward K. Ream wrote: > > There seems to be something wrong with the TravisCI checks. They aren't > being run. > > For this reason, I have temporarily removed the

Re: Leo won't start

Pretty strange, all right. If you open the Windows Task Manager, can you tell if it is running? There should be an entry for "Python", which should look something like the attached image. If you see it, Leo is running but somehow invisible. If not, Leo has quit. Also, is it possible that

Re: ENB: Saving restoring layouts

On Monday, October 5, 2020 at 3:30:12 PM UTC-4, Edward K. Ream wrote: > > ... > >> >> I had slightly modified this code. But when I restored it to your >> version, I still found that a new ViewRenderedController3 object gets >> instantiated each time the rendering panel gets hidden and then

Re: ENB: Saving restoring layouts

l. That call instantiates a new VR/VR3 widget. The other day, I tried caching the widget outside the layout and controllers, and returning the cached object when ns_provide() was called. But I got errors from pyqt saying that the c/c++ wrapper had already been torn down. On Monday, October 5,

Re: ENB: Saving restoring layouts

OK, I'll take a look at that. On Monday, October 5, 2020 at 11:45:31 AM UTC-4, Edward K. Ream wrote: > > > > On Sat, Oct 3, 2020 at 7:41 AM tbp1...@gmail.com > wrote: > >> There are two widgets involved here. One is the actual displayed widget, >> such as a QWebView. That's the one checked by

Re: undoing unmarks nodes

I can confirm that a marked node gets unmarked by an undo operation for a change later than the change that marked the node. I don't think that should happen. I think that marking nodes should be undoable - if only because the marks get saved when the outline is saved, so they are a real

Re: How to zoom-in or zoom-out vr3 font size?

imental. They will probably be included in the devel >> branch soon. >> >> On Saturday, October 3, 2020 at 12:08:10 PM UTC-4, Thomas Passin wrote: >>> >>> Actually, there is a setting to change the default font size: >>> >>> @int qweb-view-f

Re: mathematical formulas display in vr3

That's right; I only illustrated a math block. Mathjax does a nice math rendering, doesn't it? On Saturday, October 3, 2020 at 8:52:11 PM UTC-4, zhaohe wang wrote: > > Thanks a lot! It is excellent! > > The inline mode is \\( [(0.2-0)^2 + (0.7-0)^2 + (0.1-1)^2]*0.5=0.67 \\) > > Not > > $

Re: How to zoom-in or zoom-out vr3 font size?

> >>> I have added two Leo commands to VR3 that can be linked to the same keys >>> when the focus is not in the VR3 rendering pane: >>> >>> vr3-zoom-view, and vr3-shrink-view. >>> >>> These are still experimental. They will probably be inclu

Re: How to zoom-in or zoom-out vr3 font size?

12.1 >> darwin >> >> 在2020年10月4日星期日 UTC+8 上午6:38:18 写道： >> >>> I have added two Leo commands to VR3 that can be linked to the same keys >>> when the focus is not in the VR3 rendering pane: >>> >>> vr3-zoom-view, and vr3-shrink-view. >&

Re: How to zoom-in or zoom-out vr3 font size?

eo commands to VR3 that can be linked to the same keys >>> when the focus is not in the VR3 rendering pane: >>> >>> vr3-zoom-view, and vr3-shrink-view. >>> >>> These are still experimental. They will probably be included in the devel >>>

Re: How to zoom-in or zoom-out vr3 font size?

VR3 that can be linked to the same keys >>> when the focus is not in the VR3 rendering pane: >>> >>> vr3-zoom-view, and vr3-shrink-view. >>> >>> These are still experimental. They will probably be included in the devel >>> branch soon. >>> >>

Re: How to zoom-in or zoom-out vr3 font size?

UTC-4, Thomas Passin wrote: > > Actually, there is a setting to change the default font size: > > @int qweb-view-font-size = 16 > > (or use some other size). > > On Saturday, October 3, 2020 at 11:48:28 AM UTC-4 tbp1...@gmail.com wrote: > >> The rendering pane is b

Re: The docks3 branch is ready for testing

When I do not use docks, VR3 always opens for me to the right of the body pane (which is to the left of the outline). It's true that VR3 may not be visible when an outline opens, but I have linked vr3-toggle to , so it's easy to make it visible. (From time to time I have been getting an error

Re: Discuss: drop support for Qt docks?

I have given docks a try several times, though I have not had a need for anything they can do. Legacy works well for me, and I would not miss docks if they were to go away. On Tuesday, September 15, 2020 at 4:13:26 PM UTC-4, vitalije wrote: > > I agree that removing support for docks will

Re: The docks3 branch is ready for testing

On Tuesday, September 15, 2020 at 11:45:28 AM UTC-4, Edward K. Ream wrote: > > On Tue, Sep 15, 2020 at 10:22 AM Thomas Passin > wrote: > >> I wouldn't disagree, but then I haven't used the plugin. >> > > At present, the plugin is little more than a proof of

Re: The docks3 branch is ready for testing

I wouldn't disagree, but then I haven't used the plugin. If we drop having docks and use vitalije's suggestion, I notice that the new windows don't have a title (except for "Leo unnamed Window"). Perhaps it wouldn't be too hard to put the same name on the new window as was shown on the "Open

Re: The docks3 branch is ready for testing

of decode(). val is returned by QT5 as a C++ bye array. I don't know whether pyqt5 turns that into a Python bytearray or a Python bytes object, but either should decode as above. Not that this would solve the current issue :) On Monday, September 14, 2020 at 1:15:49 PM UTC-4, Thomas Passin

Re: The docks3 branch is ready for testing

On Monday, September 14, 2020 at 3:09:00 PM UTC-4, Edward K. Ream wrote: > > Leo is creating a new QMainWindow for every outline. (Each DynamicWindow > is a subclass of QMainWindow). That may be causing problems, maybe even > exposing a latent Qt bug. > ... > > The next step will be to see

Re: Python 3 on MacOS

This seems to be something wrong about your Python installation, not Leo. My Python 3.8 installation in Windows has the exact same line at the same line number (class RegexFlag(enum.IntFlag):) and it works for me. You could check this as follows: 1. Open a Python3 session. 2. Type the

<    4   5   6   7   8   9   10   11   12   13   >