Re: [Texmacs-dev] Texmacs-dev Digest, Vol 205, Issue 10

[TeXmacs]

Hi Owen,

It should not be difficult to modify the completion algorithms
along the lines that you suggest.  We might also use the improved
tooltips to show a list of possibilities.  However, these tooltips
are "read-only" for the moment, so more work would be required to
make it possible to select a completion using the mouse.
At any rate, even easy changes would require some time, and we might
want to carefully think it over first.  For instance, one thing that
would be possible inside TeXmacs (in contrast to Emacs) is
the completion of mathematical formulas.  This sounds like a very
interesting topic, both scientifically and for practical purposes.

Best wishes, --Joris



On Sat, Jan 16, 2021 at 08:57:56PM +0100, Owen Lynch wrote:
> Hello,
> 
> Thank you for your considered and lengthy response! Essentially, Ivy/Helm 
> shows a list of completion possibilities, that you can narrow down by typing. 
> It is also possible to do "fuzzy" matching, so that what you type does not 
> have to match a prefix of an option, but instead just some portion.
> 
> If I could register a keystroke handler for the input widget, then I could 
> use the texmacs-output widget to display the list of completion 
> possibilities. Is this possible?
> 
> -Owen
> 
> Excerpts from texmacs-dev-request@gnu.org's message of januari 16, 2021 6:01 
> pm:
> > Send Texmacs-dev mailing list submissions to
> >     texmacs-dev@gnu.org
> > 
> > To subscribe or unsubscribe via the World Wide Web, visit
> >     https://lists.gnu.org/mailman/listinfo/texmacs-dev
> > or, via email, send a message with subject or body 'help' to
> >     texmacs-dev-request@gnu.org
> > 
> > You can reach the person managing the list at
> >     texmacs-dev-owner@gnu.org
> > 
> > When replying, please edit your Subject line so it is more specific
> > than "Re: Contents of Texmacs-dev digest..."
> > 
> > 
> > Today's Topics:
> > 
> >    1. Completion Framework for TeXmacs (Owen Lynch)
> >    2. Re: Completion Framework for TeXmacs (Owen Lynch)
> >    3. Re: Completion Framework for TeXmacs (Owen Lynch)
> >    4. Re: Completion Framework for TeXmacs (Massimiliano Gubinelli)
> > 
> > 
> > ----------------------------------------------------------------------
> > 
> > Message: 1
> > Date: Fri, 15 Jan 2021 21:37:01 +0100
> > From: Owen Lynch <oclynch888@gmail.com>
> > To: texmacs-dev@gnu.org
> > Subject: [Texmacs-dev] Completion Framework for TeXmacs
> > Message-ID: <1610742883.99vyfox4l2.astroid@swantrumpet.none>
> > Content-Type: text/plain; charset=utf-8
> > 
> > Hello all,
> > 
> > Couple questions.
> > 
> > Is there anything for TeXmacs that works in an analogous way to Ivy/Helm 
> > from emacs?
> > 
> > Also, when editing a scheme file, is it possible to evaluate parts of it, 
> > like one would do with C-x C-e from emacs?
> > 
> > Thanks!
> > -Owen
> > 
> > 
> > 
> > ------------------------------
> > 
> > Message: 2
> > Date: Fri, 15 Jan 2021 23:15:12 +0100
> > From: Owen Lynch <oclynch888@gmail.com>
> > To: texmacs-dev@gnu.org
> > Subject: Re: [Texmacs-dev] Completion Framework for TeXmacs
> > Message-ID: <1610748731.iuj7t0jnmj.astroid@swantrumpet.none>
> > Content-Type: text/plain; charset=utf-8
> > 
> > After more digging, it seems like this would have to be done in a similar 
> > manner to the "smart-file" input method, which is hard-coded into the 
> > TeXmacs input system, in Plugins/Widkit/Input/input_widget.cpp.
> > 
> > Is there a more flexible way of doing this? I.e., in pure scheme rather 
> > than having to rewrite the C++, just like Ivy/Helm is in pure elisp?
> > 
> > -Owen
> > 
> > Excerpts from Owen Lynch's message of januari 15, 2021 9:37 pm:
> >> Hello all,
> >> 
> >> Couple questions.
> >> 
> >> Is there anything for TeXmacs that works in an analogous way to Ivy/Helm 
> >> from emacs?
> >> 
> >> Also, when editing a scheme file, is it possible to evaluate parts of it, 
> >> like one would do with C-x C-e from emacs?
> >> 
> >> Thanks!
> >> -Owen
> >> 
> > 
> > 
> > 
> > ------------------------------
> > 
> > Message: 3
> > Date: Fri, 15 Jan 2021 23:21:10 +0100
> > From: Owen Lynch <oclynch888@gmail.com>
> > To: texmacs-dev@gnu.org
> > Subject: Re: [Texmacs-dev] Completion Framework for TeXmacs
> > Message-ID: <1610749184.a3xa3zfz0y.astroid@swantrumpet.none>
> > Content-Type: text/plain; charset=utf-8
> > 
> > I guess the real thing to have here is a fully-scriptable, multiline 
> > minibuffer widget. Is this already somewhere in the code?
> > 
> > -Owen
> > 
> > Excerpts from Owen Lynch's message of januari 15, 2021 11:15 pm:
> >> After more digging, it seems like this would have to be done in a similar 
> >> manner to the "smart-file" input method, which is hard-coded into the 
> >> TeXmacs input system, in Plugins/Widkit/Input/input_widget.cpp.
> >> 
> >> Is there a more flexible way of doing this? I.e., in pure scheme rather 
> >> than having to rewrite the C++, just like Ivy/Helm is in pure elisp?
> >> 
> >> -Owen
> >> 
> >> Excerpts from Owen Lynch's message of januari 15, 2021 9:37 pm:
> >>> Hello all,
> >>> 
> >>> Couple questions.
> >>> 
> >>> Is there anything for TeXmacs that works in an analogous way to Ivy/Helm 
> >>> from emacs?
> >>> 
> >>> Also, when editing a scheme file, is it possible to evaluate parts of it, 
> >>> like one would do with C-x C-e from emacs?
> >>> 
> >>> Thanks!
> >>> -Owen
> >>> 
> >> 
> > 
> > 
> > 
> > ------------------------------
> > 
> > Message: 4
> > Date: Sat, 16 Jan 2021 00:23:56 +0100
> > From: Massimiliano Gubinelli <m.gubinelli@gmail.com>
> > To: TeXmacs developer mailing list <texmacs-dev@gnu.org>
> > Subject: Re: [Texmacs-dev] Completion Framework for TeXmacs
> > Message-ID: <10705057-FEED-4893-B292-DCBD41E97AF6@gmail.com>
> > Content-Type: text/plain;   charset=us-ascii
> > 
> > Dear Owen, 
> > 
> > I'm not familiar with Ivy/Helm. Can you briefly describe what you would 
> > like to obtain? On one side TeXmacs provides already scheme sessions 
> > embedded in a document. So you could simply either have a document with 
> > code which can be sequentially (re)evaluated (a la jupyter notebook or 
> > mathematica files). If you want to have, on one side a scheme file and on 
> > another a buffer where to evalute, I would try to create a new document 
> > with a session and then come glue code in scheme which allows to select 
> > part of a scheme file and send them to the other document for evaluation. 
> > This should not be too difficult. 
> > 
> > Note that with scheme you can also programmatically create dialogs (but not 
> > full fledged editor windows yet). However there are a couple of widgets you 
> > could be interested in : one is texmacs-input and the other texmacs-output. 
> > The second allows to render arbitrary documents, while the first is a full 
> > editor, but you do not get the toolbars nor the menus...
> > You can see some examples in TeXmacs/progs/kernel/gui/menu-test.scm
> > 
> > 
> > for example, open a scheme session and then do
> > 
> > (show (widget6))
> > 
> > to see the two widgets I mentioned above in action.
> > 
> > 
> > Best
> > Max
> > 
> > 
> >> On 15. Jan 2021, at 23:21, Owen Lynch <oclynch888@gmail.com> wrote:
> >> 
> >> I guess the real thing to have here is a fully-scriptable, multiline 
> >> minibuffer widget. Is this already somewhere in the code?
> >> 
> >> -Owen
> >> 
> >> Excerpts from Owen Lynch's message of januari 15, 2021 11:15 pm:
> >>> After more digging, it seems like this would have to be done in a similar 
> >>> manner to the "smart-file" input method, which is hard-coded into the 
> >>> TeXmacs input system, in Plugins/Widkit/Input/input_widget.cpp.
> >>> 
> >>> Is there a more flexible way of doing this? I.e., in pure scheme rather 
> >>> than having to rewrite the C++, just like Ivy/Helm is in pure elisp?
> >>> 
> >>> -Owen
> >>> 
> >>> Excerpts from Owen Lynch's message of januari 15, 2021 9:37 pm:
> >>>> Hello all,
> >>>> 
> >>>> Couple questions.
> >>>> 
> >>>> Is there anything for TeXmacs that works in an analogous way to Ivy/Helm 
> >>>> from emacs?
> >>>> 
> >>>> Also, when editing a scheme file, is it possible to evaluate parts of 
> >>>> it, like one would do with C-x C-e from emacs?
> >>>> 
> >>>> Thanks!
> >>>> -Owen
> >>>> 
> >>> 
> >> 
> >> _______________________________________________
> >> Texmacs-dev mailing list
> >> Texmacs-dev@gnu.org
> >> https://lists.gnu.org/mailman/listinfo/texmacs-dev
> > 
> > 
> > 
> > 
> > ------------------------------
> > 
> > Subject: Digest Footer
> > 
> > _______________________________________________
> > Texmacs-dev mailing list
> > Texmacs-dev@gnu.org
> > https://lists.gnu.org/mailman/listinfo/texmacs-dev
> > 
> > 
> > ------------------------------
> > 
> > End of Texmacs-dev Digest, Vol 205, Issue 10
> > ********************************************
> > 
> 
> _______________________________________________
> Texmacs-dev mailing list
> Texmacs-dev@gnu.org
> https://lists.gnu.org/mailman/listinfo/texmacs-dev