(catch :abort
;; do something
(when condition
(message "A bad thing happened")
(throw :abort nil))
;; do something else
)When the functionality I really wanted was:
(progn
;; do something
(when condition
(user-error "A bad thing happened"))
;; do something else
)I knew the former felt sketchy, but I couldn't think of a better way to do it until just now.
#emacs #lisp
FOSS Dev reshared this.
#Elisp logic:
All interned symbols can be found in a lookup table. This table is bound to the obarray symbol.
Hang on a minute...
I can only assume that the underlying C code has its own pointer to this table and the obarray symbol is only provided as a convenience for elisp functions that can't see this pointer?
#emacs #lisp
No no, the obarray you see from elisp is the same one used by the reader. Elisp is an old-style Lisp here, and the obarray is a first-class thing: you can make a new one, rebind obarray, etc.
That's the sort of thing people don't do much anymore, but used to do. The documentation covers it reasonably well gnu.org/software/emacs/manual/…
I'm starting to get the impression that my doctor's office is trying to drum up business by giving me repeated heart attacks with vague voicemails about some test results I'm waiting on.
For the record, according to the information I have at this very moment, everything is (probably) fine.
I'm a little self-conscious about it as non-trivial is relative, but...
(defmacro lambdamoo-chatter-interact
(func-name to msg docstring fmtstr &rest vals)
"Define a function for interacting with another player"
(let ((proc (gensym))
(str (gensym)))
`(defun ,func-name (,proc ,str)
,docstring
(let ((,to lambdamoo-chatter)
(,msg (substring-no-properties (lambdamoo-command-text ,str))))
(if ,to
(funcall lambdamoo-send-line ,proc
(format ,fmtstr . ,vals))
(message "No chatter specified"))))))
That's awesome. I need to hear my own advice, of course, but don't be inhibited to share something that isn't finished. It's the Fediverse! We're all anarchists! The kind sort, I mean.
Lisp macros are just so powerful.
Wait, how do you get the awesome code formatting? Anybody know how to configure this on fediscience.org?
AM I GOING TO HAVE TO SPIN UP MY OWN INSTANCE AGAIN
@James Endres Howell Frendica has a markdown add-on.
I just typed:
```lisp
...code...
```Typing that however was trickier.
RIP Sterrance (my #sourdough starter) who met a tragic end in his glass jar at the hands of gravity and the kitchen floor.
Also, happy birthday to Stella who is going to take a bunch of work to make into another viable starter.
reshared this
Jonathan Lamothe likes this.
@screwlisp @Judy Anderson
I've been looking to migrate more of my workflow into emacs, in this particular case I'm looking to moo via emacs which I believe you both do?
I believe @screwlisp has mentioned using rmoo, but the only repo I found for that hasn't been updated in over a decade. Is there something more recent I'm not aware of?
> had do a little finessing to get it to installed
out of curiosity: what was the problem?
Is it that I didn't make a (M)ELPA package out of it? (nobody just drops things in their ./emacs directory anymore?)
or some other issue?
screwlisp reshared this.
M-x package-install-file, it didn't like that the file didn't end with:;;; mud-mcp.el ends hereI just had to add that and it was all good.
ok, so the answer is indeed
"nobody just drops things in their ./emacs directory anymore"
(really, that's all it's supposed to be.
well okay, that plus
M-x load-library mud-mcp
which is the old-school way of doing things)
(wRog needs to learn MELPA. Film at 11.)
screwlisp reshared this.
@Roger Crew✅❌☑🗸❎✖✓✔ @Judy Anderson @screwlisp It essentially already was a valid ELPA package with the mentioned exception.
I'm currently in the process of adding my own custonizations. I've added a rudimebtary shim that processes lines entered bu the user so that it can support commands that get processed on the client side.
Here's an excerpt:
(require 'mud-mcp)
(defun lambdamoo ()
"Connect to LambdaMOO"
(interactive)
(mud-mcp-connect "LambdaMOO" "lambda.moo.mud.org" 8888)
(setq lambdamoo-send-line comint-input-sender
comint-input-sender #'lambdamoo-process-line))
(defconst lambdamoo-commands
'(("send" . lambdamoo-send)
("test" . lambdamoo-test))
"Command functions")
(defvar lambdamoo-send-line nil
"The function that is called to send a line to the server")
(defun lambdamoo-process-line (proc str)
"Process input sent by the user"
(if (string-prefix-p "/" str)
(lambdamoo-process-command proc str)
(funcall lambdamoo-send-line proc str)))
(defun lambdamoo-process-command (proc str)
"Process a command"
(let* ((words (split-string str))
(command (string-trim-left (car words) "/"))
(found (assoc (downcase command)
lambdamoo-commands
#'string=))
(func (and found (cdr found))))
(if func
(funcall func proc str)
(message "Command '%s' not found." command))))After I wrote all this, I found comments in the file detailing how to add functionality.
Is there a more "proper" way I could've done this?
Shannon Prickett reshared this.
like this
I am dangerously close to unleashing my first #emacs package on the public. It's nothing fancy and still relatively niche, but I deem it potentially useful enough to be worth publishing.
There are a couple small features I want to add and a few things that still need some polish, but it's almost ready for a version 0.1 release.
It's not anything ground breaking or anything. I'm still pretty much an #elisp novice, but I'm proud of it anyway.
More details when it's released.
like this
@Álvaro R. At this point all I need to add is a README and two features (which will mostly reuse code I've already written just in a slightly different way).
Surprisingly enough, the hardest part of the whole project was getting it to display numbers with thousands separators. That code might exist in the bowels of the calc package, but it was easier to just roll my own.
Okay, my first #Emacs package is officially released. It was strongly inspired by @Soroban Exam Website's work, providing practice tools for the #soroban. This is the first Emacs package I've ever released. It's probably not perfect, but I welcome feedback on how it can be improved.
I wonder if there is an overlap of more than say five people who are both soroban and emacs users. 🙃
Anyhow, it can be found at: codeberg.org/jlamothe/soroban
reshared this
vim guy here. happy to see I inspire others...
May be you could post on our forum. Not sure you will get more users, though
@Soroban Exam Website Might as well. I wrote it mainly for myself, partly because I don't own a printer and this makes it easier to practice when working from a computer screen, but also just to see if I could.
Still, if someone else is going to find it useful, that's probably the place I'll find them.
May be you didn't see you that you can generate an interactive HTML output, on the site.
That was designed for people who don't want to print.
Should I make it more visible?
like this
Shannon Prickett reshared this.
Well, everything's mostly set up. Cable management needs some definite work, but at least the layout of my desk is more or less unchanged.
The new arrangement makes more logistical sense, but will require some getting used to. Just about every room in the apartment's been rearranged.
(roll m3tti)
in reply to Jonathan Lamothe • • •CLHS: Function APROPOS, APROPOS-LIST
www.lispworks.comMarce Coll
in reply to (roll m3tti) • • •dziban
dziban.netBjörn Gohla
in reply to Jonathan Lamothe • • •well, there is hoogle for that.
Although I agree, it should be possible to search by signature in GHCi for all modules in scope, without building an index first.
Jonathan Lamothe
in reply to Björn Gohla • •Björn Gohla
in reply to Jonathan Lamothe • • •Jonathan Lamothe
in reply to Björn Gohla • •Dave Tenny
in reply to Jonathan Lamothe • • •FYI in case it's useful to you.
If you use generic functions (via `defgeneric` or `defmethod`) the resulting methods specialize on classes (and other things, but it isn't the same as all types). The slime inspector will show you all the method signatures for different methods, and you could no doubt write tools though I don't know if the method type signatures are available via standard API (it might be, I just haven't tried).
Jonathan Lamothe
in reply to Dave Tenny • •Digital Mark λ ☕️ 🕹 👽
in reply to Jonathan Lamothe • • •As a Schemer (and formerly/sometimes still Objective-C), everything is pretty verbose, and my own functions even more so, so I can search by function name knowing the type and parameters. (vector-index vec searchfunc), (draw-rect-with-edge-color rect edge-width color), etc.
There's no excuse for hardcore Lisp functions like (wadsf w q) "wander down stack frames for word query" (fictional but not unlikely).
#lisp
Jonathan Lamothe
in reply to Digital Mark λ ☕️ 🕹 👽 • •