Re: [Matlab-emacs-discuss] move -->github: rebase, and which branches could be deleted

[John C. <ci...@ma...>]

Hi

I just double checked the fontlonkhang branch, downloaded zip


  *
matlab-emacs-src-a78b368387c85437f8de187acf0a2b1cbeeacc3f/  (master)
  *
matlab-emacs-src-ce22c697d06ece776cfd7a9a41b3704f1815823f/ (fontlockhang)

And diff'd them - all changes are in master. I also validated that the hangs we were seeing no longer exist.

I believe what happened is that Eric and I used the fontlockhang as a "staging" area and then once solved, we manually added to master.

Thanks
John


________________________________
From: Uwe Brauer
Sent: Thursday, August 01, 2024 4:46 PM
To: John Ciolfi
Cc: Eric Ludlam; Uwe Brauer; matlab-emacs-discuss
Subject: Re: [Matlab-emacs-discuss] move -->github: rebase, and which branches could be deleted

>>> "JC" == John Ciolfi <ci...@ma...> writes:

> Regarding the fontlockhang, these have been merged in to the
> main/master branch already. I just manually compared matlab.el between
> and the fontlock improvements are in the tip.

> The fontlockhang branch can be deleted.

This is odd, (and this is why I don't trust git (and prefer mercurial 😉)

If I run clone the sourceforge repository again and run

git log   --graph --color=always --all --decorate --pretty | git name-rev --annotate-stdin | less -R

I see


| * commit ce22c697d06ece776cfd7a9a41b3704f1815823f (remotes/origin/fontlockhang) (origin/fontlockhang)
| | Author: Eric Ludlam <el...@ma...>
| | Date:   Mon Sep 26 17:44:26 2022 -0400
| |
| |     mclass.m: Added AH to test a type with a long package name.
| |
| |     This used to be very very slow or hang Emacs.
| |
| * commit c0af4943b5d45fa73e6d4ed5681121d6e1983727 (remotes/origin/fontlockhang~1)
|/  Author: Eric Ludlam <el...@ma...>
|   Date:   Mon Sep 26 17:31:41 2022 -0400
|
|       (matlab-font-lock-anchor-variable-match): Fix regex to be faster.
|
|       Fixes what was percieved as a hang for properties with lots of package.class types.
|

So this branch clearly is not merged. That is confirmed by running
git branch -a --no-merged master


  remotes/origin/copyright
  remotes/origin/fill-fix
  remotes/origin/fontlockhang
  remotes/origin/matlab-cells
  remotes/origin/org-mode
  remotes/origin/shellcomplete
  remotes/origin/wisent-parser

Could you please check again and confirm before I delete, because if I
delete an unmerged branch in git, its commits are gone (well I will do
this on my experimental github repository not in sourceforge).


> Note, in MathWorks we are using the current matlab-emacs tip minus a
> few items that no longer work such as the cedet items. Once we get
> matlab-emacs into github, I'll update it with improvements such as
> good org mode support and LSP support.


Ok, should be ready in a couple of days. I need to clarify the status of
this branch and the one Eric is looking into.

Uwe


--
I strongly condemn Hamas heinous despicable pogroms/atrocities on Israel
I strongly condemn Putin's war of aggression against Ukraine.
I support to deliver weapons to Ukraine's military.
I support the EU and NATO membership of Ukraine.