For the first time in...a few decades at least. Maybe ever. Emacs is regularly and unceremoniously segfaulting on me. Pretty clearly something to do with tree sitter, which is a bit frustrating. (#Emacs 29.3 on MacOS from brew)
One of the issues with the #REPL (and #emacs which is just a great big text-oriented repl) is that it is additive in nature; it usually takes major effort or a restart to REMOVE things once they've been added (thinking on plugins which modify app state). #Clojure
I'm finding some really interesting blogs out there, and am reminded of the early days of the internet, when it wasn't all monetized and people just wrote about stuff they were interested in for fun. And I had an RSS feeder and got updates when people posted stuff. Plus blogrolls.
So I'm sure there's a way to do this now in #emacs, right? Can someone point me in the right direction? I've heard about elfeed but don't know anything about it.
Productivity increases with the level of customization you are making in the tools you are using most often.
The downside is that the more used to these customizations you get, the more lost you'll feel when
using a system that is not configured as yours.
Simple example: create a new binding in #Vim or #Emacs. This is not only very common but
also very encouraged. After getting used to that, connect to a remote server.
So I was a bit off my org-roam habits and decided to get back on track. Since my daily driver is macOS, I am using Railwaycat's emacs-mac formulae on Homebrew. And boom! #emacs doesn't launch even before upgrading due to some wierd gcc issue.
Tried to upgrade... can't compile because of libgccjit issue (necessary for native compilation feature which makes elisp work faster). Turns out there are some breaking changes from GCC upstream and had to browse published issue to apply some workarounds.
In the process I've discovered the --HEAD option for brew which apparently allows to get the most recent-ish branch for a given formulae. So now I have Emacs 29.3 over default 29.1 for this contraption.
PS. As I like the Emacs help system which can browse its own source code to describe a function, it wasn't readily available with brew. Had to copy the code from brew cache to some static location and Emacs just asked to point for the sources there, nice!
I sometimes notice that Emacs freezes up when I press a key like d to delete (adding the tag:trashing) for a couple of seconds. It's hard to track down, but I believe that I run into a conflict with a cron job updating the notmuch database, and Emacs waits for the notmuch tag change command to finish, which waits for the cron job to finish.
Does that sound likely? How do y'all debug and improve this?
So excited that I was able to write a blog post in #orgmode#emacs and publish it to my #classicpress blog from within orgmode! The new post ain't much to look at, and there's no content to speak of. Just a proof of concept. And it proofed!
I ran into some go code that use build tags. My existing test functions didn't work. To handle build-tags, I defined go-build-tags in .dir-locals.el and used the variable in my functions.
Okay here's another Transient I've made a while back, in this case for Avy. A bit esoteric, but this has enabled me to get a lot more out of Avy than just using avy-goto-word-1 or avy-goto-char-timer. Wondering out loud if I should publish this on MELPA.
Also yes, I've read Karthik's post on using Avy wrong. I readily admit I'm not sophisticated enough to put filter-select-act into practice.
Not that anyone likely cares, but I just refactored my #emacs config away from literate org-mode, into a collection of specific elisp files. The top level init.el is now quite clean.
I got tired of having to edit source blocks in the org-mode config and having to preload org-mode just to load my config, all just so it looks prettier (?) when reading it on github.
The elisp files still have sections, rg can search them just fine. I'm happy. https://github.com/garyo/emacs-config
A question for the #emacs community: Is it possible to use the Tramp feature to connect to a remote Emacs daemon? I ask because I have a file which I always have open on a remote pseudo-tty, but sometimes it would be nice to use my desktop Emacs with its nice proportional fonts and custom sizes to edit.
I want to connect to the same remote daemon so I can see unsaved changes and not have to worry about sync.
While the LSP protocol is useful for completion or access to symbol definitions, some of its features are less appealing. In #Emacs, you can instruct Eglot to ignore any feature you dislike.
E.g. (setq eglot-ignored-server-capabilities '(:inlayHintProvider)) to remove annoying hints mixed with the code in c-mode with clangd.
Today I learned something new: don't edit "authorized_keys" file in an ssh tramp session when remote disk is full. The moment you (try to) store the new version, Emacs will backup (rename) existing file and try to write a new file. If that fails because of disk full and the file happens to be "authorized_keys" you immediately will lose ssh access to that remote machine.
Well I ran out of battery halfway through the hand install, so several hours of recovery later I can boot into #guix fsvo. Where are all my other bootable partitions though, and will I be able to tolerate #GNOME long enough to learn to live with it? These are the questions. Stay tuned, rat-fans.