slime-devel April 2004

slime-devel@common-lisp.net

40 participants
103 discussions

Re: [slime-devel] compiler note buffers
by Helmut Eller 20 Apr '04

20 Apr '04

Zach Beane <xach(a)xach.com> writes: > What should I do? I tried using a typeout frame, but it doesn't look > like the compiler notes tree uses it. We could try to insert the notes tree in the output buffer. Here's a hackish way to achieve that: (setq slime-compilation-finished-hook (defun slime-insert-errors-in-output-buffer (&optional notes) (interactive) (let ((notes (or notes (slime-compiler-notes)))) (with-current-buffer (slime-output-buffer) (slime-with-output-end-mark (let ((start (point))) (insert "\n") (when (null notes) (insert "[no notes]")) (dolist (tree (slime-compiler-notes-to-tree notes)) (slime-tree-insert tree "") (insert "\n")) (insert "\n") (let ((overlay (make-overlay start (point))) (local-map (make-sparse-keymap))) (define-key local-map (kbd "RET") 'slime-compiler-notes-show-details) (overlay-put overlay 'local-map local-map)))))))) This is only marginally tested, but should be good enough to demonstrate the idea. Do you like this better? Helmut.

1 0

[slime-devel] Closing *Completions* buffer
by Tiago Maduro-Dias 20 Apr '04

20 Apr '04

Heya! I've always been a bit annoyed at how the *Completions* buffer sticks around even when it's not needed anymore. So I just added a bit of code to slime.el which closes this buffer (if one exists) after pressing space, making use of the slime-space hook. Everything else works as expected and you can still middle-click on a possible completion, etc., so I don't think this patch will get in the way of anyone's development habits. I didn't really spend much time figuring out exactly where the code should go and ended up defining a new function "slime-close-buffer" in the "Common utility functions and macros" section (as good a guess as any). Hope this helps! Tiago. -- "I conjecture that there is no good algorithm for the traveling salesman problem. My reasons are the same as for any mathematical cojecture: (1) It is a legitimate mathematical possiblity, and (2) I do not know." --Jack Edmonds, 1966

2 1

Re: [slime-devel] Some remarks
by Bill_Clementson＠peoplesoft.com 20 Apr '04

20 Apr '04

Edi Weitz <edi(a)agharta.de> writes: > On Tue, 20 Apr 2004 12:41:51 -0600, Bill_Clementson(a)peoplesoft.com wrote: > > > The function appears to be in GNU Emacs 21.3.5 but not in 21.3.1 > > Ah, OK. Is that a version from CVS? I actually wanted to try a CVS > Emacs anyway because it looks like they made some improvements to > lisp-mode after the 21.3 release. > > <nitpick> > But of course SLIME shouldn't use this function as long as it is > advertised as being usable on GNU Emacs 20... :) > </nitpick> Yes, 21.3.5 is a CVS version. The SLIME maintainers will need to decide whether they want to add the read-directory-name function to slime or re-write it in order to support earlier emacs versions. FWIW, the function is in files.el - here's the source: (defun read-directory-name (prompt &optional dir default-dirname mustmatch initial) "Read directory name, prompting with PROMPT and completing in directory DIR. Value is not expanded---you must call `expand-file-name' yourself. Default name to DEFAULT-DIRNAME if user enters a null string. (If DEFAULT-DIRNAME is omitted, the current buffer's directory is used, except that if INITIAL is specified, that combined with DIR is used.) Fourth arg MUSTMATCH non-nil means require existing directory's name. Non-nil and non-t means also require confirmation after completion. Fifth arg INITIAL specifies text to start with. DIR defaults to current buffer's directory default." (unless dir (setq dir default-directory)) (unless default-dirname (setq default-dirname (if initial (concat dir initial) default-directory))) (read-file-name prompt dir default-dirname mustmatch initial 'file-directory-p)) Cheers, Bill

1 0

Re: [slime-devel] Some remarks
by Bill_Clementson＠peoplesoft.com 20 Apr '04

20 Apr '04

Hi Edi! Just a few comments: > 2. The shortcut ,change-directory uses a function read-directory-name > which I don't seem to have. I'm using GNU Emacs 21.3. The function appears to be in GNU Emacs 21.3.5 but not in 21.3.1 > 3. Is there a specific reason that RET isn't bound to > newline-and-indent (or something similar) as in ILISP but only to > C-j? This bugged me too. I manually put the following in my .emacs file: (add-hook 'lisp-mode-hook (lambda () (slime-mode t) (define-key lisp-mode-map [(control j)] 'newline) (define-key lisp-mode-map [(control m)] 'newline-and-indent) etc....)) However, I agree that it should probably be the default. > 7. The fact that SPC (slime-space) still works and does the right > thing after you've started entering arguments (contrary to ILISP) > is very cool and a real time-saver for me!! If you like slime-space, you should try adding (slime-autodoc-mode) to your .emacs file. It shows the same info as slime-space does for whatever function your cursor is currently over. In a separate post, Edi said: > [I'm offline while I write this but I seem to remember that I've > somewhere (in CLOCC?) seen a CL package for Emacs Lisp. Does my > memory serve me right? Does this package provide Emacs Lisp regular > expressions for CL?] Yes, you are correct - CLOCC has an elisp.lisp file in the cllib package; however, it doesn't have regex support. Cheers, Bill

2 1

[slime-devel] slime-regex-apropos?
by Edi Weitz 20 Apr '04

20 Apr '04

Attached is a rather hacky patch for slime-apropos. If string is recognized as an Emacs regular expression then the result of calling SWANK:APROPOS-LIST-FOR-EMACS with "" instead of string is subsequently filtered by Emacs' string-match. Otherwise slime-apropos works as usual. This is more a proof of concept than a patch intended to be applied by the SLIME maintainers. I'd like slime-apropos to be regex-aware but I'm not sure what would be the best way. Some thoughts: 1. It'd probably be best if this regex behaviour of slime-apropos were optional. As the prefix arg is already used there are a couple of options: - Make the regex behaviour another optional y-or-n-p argument like only-external-p. - Let the user customize this behaviour via a variable like, er, slime-apropos-uses-regex-syntax-p. - Define another function slime-regex-apropos and let the user bind it to C-c C-a if he so wishes. 2. My childish hack will be inaccetably slow on many machines and Lisps because in the regex case the Lisp always has to return all symbols and Emacs subsequently has to match them all. It'd be Lispier and much faster to do the regex matching on the Lisp side. CL-PPCRE, for instance, has a facility to do that but it would certainly be overkill to include such a big package with SLIME. (Plus, I think slime-apropos should use Emacs regex syntax as it is an Emacs package.) [I'm offline while I write this but I seem to remember that I've somewhere (in CLOCC?) seen a CL package for Emacs Lisp. Does my memory serve me right? Does this package provide Emacs Lisp regular expressions for CL?] 3. My hack matches on the returned "designator" and thus a regex like "ke.*ns" will match a symbol like KERNEL:CONS-TYPE, i.e. package prefixes aren't currently treated consistently. 4. The hack pops up an empty *SLIME Apropos* buffer if there's no match - no good. Anyway, what do you think about this? Cheers, Edi.

2 1

[slime-devel] Some remarks
by Edi Weitz 20 Apr '04

20 Apr '04

Hi! I've finally found some spare hours to play with SLIME. I used the current CVS version (updated this morning) with CMUCL 18e and, wow, I'm really amazed! Great work, guys!!! While I came across a few minor bugs and lost the connection once or twice I still get the feeling that even CVS HEAD seems to be very stable and grown-up. I think I'll drop ILISP and work with SLIME exclusively from now on. Below are a couple of minor remarks which I hope are helpful: 1. The function slime-remove-notes is bound to C-c M-c while the manual says M-c. 2. The shortcut ,change-directory uses a function read-directory-name which I don't seem to have. I'm using GNU Emacs 21.3. 3. Is there a specific reason that RET isn't bound to newline-and-indent (or something similar) as in ILISP but only to C-j? 4. Load CL-PPCRE, than invoke slime-apropos-package (a cool function by the way). Now position the cursor above PPCRE-SYNTAX-ERROR and invoke slime-edit-definition. I get: No matching method for the generic-function #<Standard-Generic-Function PCL:COMPUTE-APPLICABLE-METHODS-USING-CLASSES (1) {281DF8F1}>, when called with arguments (#<Closure Over Function "DEFUN %DEFINE-CONDITION" {4879E009}> (#<CONDITIONS::CONDITION-CLASS PPCRE-SYNTAX-ERROR {486B867D}>)). [Condition of type NO-APPLICABLE-METHOD] 5. In the same *SLIME Apropos* buffer position the cursor above CREATE-SCANNER and invoke slime-edit-definition. Now I get: Error in function COMMON-LISP::%%TYPEP: Unknown type specifier: C::FILE-SOURCE-LOCATION [Condition of type SIMPLE-ERROR] 6. I think that slime-interactive-eval is a very useful function and IMHO the C-c : keybinding should also be available in the Slime REPL. Yeah, I know that you actually don't need to use the minibuffer while you're in the REPL but in the long run it'd be easier for my muscle memory to always be sure that C-c : is there without thinking... :) 7. The fact that SPC (slime-space) still works and does the right thing after you've started entering arguments (contrary to ILISP) is very cool and a real time-saver for me!! Thanks to all for this great package! Cheers, Edi.

1 0

[slime-devel] errors on startup
by Artem Baguinski 20 Apr '04

20 Apr '04

hello as soon as i (require 'slime) i get several of the following lines: Warning: `error' called with 1 args to fill 0 format field(s) in *Compile-Log* plus this *Compile-Log* buffer flushed for a split second. I understand that warning is something to expect in a alpha version of the software, but i still suspect i'm doing something wrong... e.g. compiling slime every time i (require) it. could that be helped? -- gr{oe|ee}t{en|ings} artm

1 0

Re: [slime-devel] I dig the flying letters startup banner
by Julian Stecklina 20 Apr '04

20 Apr '04

David Steuber <david(a)david-steuber.com> writes: > That is a nice bit of ascii eye candy. I do not see it with XEmacs 21.5. Is this FSF Emacs specific? Would be sad. :) Regards, -- Julian Stecklina Signed and encrypted mail welcome. Key-Server: pgp.mit.edu Key-ID: 0xD65B2AB5 FA38 DCD3 00EC 97B8 6DD8 D7CC 35D8 8D0E D65B 2AB5 Any sufficiently complicated C or Fortran program contains an ad hoc informally-specified bug-ridden slow implementation of half of Common Lisp. - Greenspun's Tenth Rule of Programming

2 1

[slime-devel] Re: slime + defsystem
by Luke Gorrie 20 Apr '04

20 Apr '04

This one's from comp.lang.lisp.. Artem Baguinski <artm(a)v2.nl> writes: > hi > > if i try to load something with defsystem it'll *always* say that > compiled files are too old or absent. it refers to correct path of > new compiled 10 min ago files though. where should i dig? Don't know about that part, but.. > NB: these are the questions it asks: > > ; - Binary file /usr/src/lisp/McCLIM/patch.x86f is old or does not exist. > ; Compile (and load) source file /usr/src/lisp/McCLIM/patch.lisp instead? > ; - Should I bother you if this happens again? > ; - Should I compile and load or not? > > also, when i'm trying to do that in slime and defsystem asks if i > want stuff recompiled i can't seem to be able to answer. i guess > slime treats my pressing Return key specially and it never reaches > the defsystem. is there a solution to this? This is a bit tricky. In general SLIME works with user input. When Lisp tries to read input, a message goes to Emacs saying "I want some user input", and while Lisp is expecting input pressing return in the REPL will do the right thing. The difficulty here is that defsystem wants to read with a timeout so it is actually _polling_ for input, i.e. most of the time it is not actually trying to read, but is waiting a while before checking if input is available. That means that most of the time Emacs thinks Lisp doesn't want input and so it interprets user input as a REPL expression to be evaluated instead (since our REPL supports issuing commands in parallel). One solution could be to add some flow-control to the REPL and disable request piplining. Then we would buffer up user input until Lisp tells us whether it wants user input or the next REPL command. Maybe there's a better way. The workaround for now is to hack defsystem so that it does a blocking read instead of polling. This loses the timeout feature: (defun mk::read-char-wait (&optional (timeout 20) input-stream (eof-error-p t) eof-value) (declare (ignore timeout)) (read-char input-stream eof-error-p eof-value)) -Luke

3 2

[slime-devel] Re: [admin] Re: Not getting list mail
by David Steuber 19 Apr '04

19 Apr '04

Erik Enge <erik(a)nittin.net> writes: > On Apr 16, 2004, at 11:17 PM, David Steuber wrote: > > > I don't know what the weirdness is. slime-devel-request was able to > > send me confirmation that I joined the list. According to the > > headers, that mail came through the postini filter. This was before > > I checked with postini. I only looked there when I found out about > > gmane and wondered why I hadn't seen any of the mail. > > According to the mailserver (Exim) log files, emails are being sent to > you. > Is anyone else experiencing this? I found the problem on my end. I'm using GNUS and created a seperate group for slime-devel mail. Then I forgot to subscribe that group. Stupid user error. Sorry. -- I wouldn't mind the rat race so much if it wasn't for all the damn cats.

1 0

← Newer
1
2
3
4
5
6
7
8
9
10
11
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

slime-devel April 2004