- slime-devel - mailman3.common-lisp.net

[slime-devel] Additional source location information for SBCL
by Nikodemus Siivola 01 Jun '05

01 Jun '05

Development environments such as Slime benefit from being able attach additional source location information (eg. buffer associations) to functions. Currently Slime does this by patching SBCL in order to be able to attach its own plists to SB-C::DEBUG-SOURCE structures. I propose we add :SOURCE-PLIST option to WITH-COMPILATION-UNIT, with the following tentative spec: :SOURCE-PLIST Plist-Form Attaches the value returned by the Plist-Form to internal source location information of functions and definitions compiled in within the dynamic contour. Primarily for use by development environments, in order to eg. associate function definitions with editor-buffers. Can be accessed as SB-INTROSPECT:DEFINITION-SOURCE-PLIST. If multiple, nested WITH-COMPILATION-UNITs provide :SOURCE-PLISTs, they are appended togather, innermost left, unless :OVERRIDE is true, in which case previous :SOURCE-PLISTs are ignored. The space-costs of this are nominal: size of DEBUG-SOURCE structures is increased by one slot/word, but there is only one of those per compiled file -- though one per function loaded from source or entered at the toplevel. I think calling it a :SOURCE-PLIST is a reasonable compromise between more abstract names such as :INFO, the documentary nature of the -PLIST vrt. the appending behaviour, by YMMW. To illustrate the usage, the significant part of SWANK-COMPILE-STRING[*] could eg. become (with-compilation-unit (:source-plist (list 'emacs-buffer buffer 'emacs-string string 'emacs-position position)) (with-input-from-string (s string) (load s)))))) I've the SBCL parts of this implemented in my local tree, and most of swank-sbcl adapted for this. The questions to SBCL and Slime folks respectively: 1. Does the interface seem sane and something we would be happy to support? Are the non-slime use cases this doesn't reasonably cover? 2. Does this fit the needs of Slime? If this existed on other platforms as well, would you be tempted to use this instead of whatever you currently use? [*] To nitpick, I venture the opinion that SWANK-COMPILE-STRING is a misnomer, as as far as I can tell the intention is something like "evaluate the forms contained in this string as if they had been entered at the toplevel, but making sure all functions are compiled". Cheers, -- Nikodemus Schemer: "Buddha is small, clean, and serious." Lispnik: "Buddha is big, has hairy armpits, and laughs."

2 3

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 31 May '05

31 May '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.699 slime/ChangeLog:1.700 --- slime/ChangeLog:1.699 Tue May 24 21:10:08 2005 +++ slime/ChangeLog Tue May 31 20:38:58 2005 @@ -1,3 +1,14 @@ +2005-05-27 Espen Wiborg <espenhw(a)grumblesmurf.org> + + * swank-corman.lisp: New file, swank for Corman Lisp. + + * swank.lisp (simple-announce-function): force-output after + announcing. + (symbol-external-p): Be extra paranoid about the symbol's package; + find-symbol barfs on a nil package in Corman Lisp. + + * swank-loader.lisp: Add Corman Lisp support. + 2005-05-24 Alan Ruttenberg <alanr-l(a)mumble.net> * slime.el text-property-default-nonsticky not defined in @@ -178,6 +189,7 @@ slime-complete-symbol* and slime-simple-complete-symbol. 2005-05-06 Alan Ruttenberg <alanr-l(a)mumble.net> + * swank-openmcl.lisp specializer-name didn't handle structure-class which caused meta-. of methods specialized on defstruct arguments to fail.

1 0

[slime-devel] updating slime during idle time
by Chris Baker 29 May '05

29 May '05

Hi Slimers, I've recently started running slime with clisp on windows. This works great for most things, but now I'm starting to do windows gui programming, and windows wants each program to have it's own event loop, which interferes with slime, which has its own server update loop. Unfortunately, clisp is single threaded, so I can only update windows events or update slime requests at a given time. Is there a function I can call periodically to process slime requests at idle time? Or alternately, can I add a hook to swank so I can pump windows events when slime isn't processing anything? Many thanks for this great piece of software, and thanks in advance for any help anyone can provide. cbb -- 19:45:04 up 9 days, 21:54, 4 users, load average: 0.55, 0.27, 0.10

2 2

[slime-devel] Way to jump into file to error location?
by Jeffrey Cunningham 27 May '05

27 May '05

Hi, When I compile a buffer I often get an error message like this: -+ Errors (1) `-- Read error in form starting at 3192: ";; (list `(:p :class "small"" End-of-File on #<Stream for file "/home/jcunningham/code/lisp/html/test.lisp"> [+] Notes (2) My question is: is there a way (key-code maybe?) to find that start point in the file? Is there some way to get at least a line number? I've looked through some of the documents and googled a bit but haven't come up with anything yet. That number must be useful for something beyond counting characters from the beginning. ;-) --jeff cunningham

2 2

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 24 May '05

24 May '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.696 slime/ChangeLog:1.699 --- slime/ChangeLog:1.696 Mon May 23 04:31:12 2005 +++ slime/ChangeLog Tue May 24 21:10:08 2005 @@ -1,4 +1,25 @@ -2005-05-20 Alan Ruttenberg <alanr-l(a)mumble.net> +2005-05-24 Alan Ruttenberg <alanr-l(a)mumble.net> + + * slime.el text-property-default-nonsticky not defined in + xemacs. oops. + +2005-05-24 Alan Ruttenberg <alanr-l(a)mumble.net> + + * slime.el meta-w now removes properties before insertion if you + cut just a portion of the presentation. Added xemacs + support. Enabled in xemacs. + +2005-05-23 Alan Ruttenberg <alanr-l(a)mumble.net> + + * slime.el slime-presentation-menu - use with-current-buffer, so + that menus work even if you are not in the buffer with the + presentation. + + * present.lisp More menu items for pathnames. Remember last + slime-stream-p value. *can-print-presentation* t during + swank-compiler and during presentation menu action. + +2005-05-22 Alan Ruttenberg <alanr-l(a)mumble.net> * present.lisp. (slime-stream-p) check if a stream is destined for output in a slime listener. (checks *connections* looks into pretty-print @@ -15,7 +36,7 @@ In cmucl, use fwrappers to modify behaviour rather than redefinition. -2005-05-20 Alan Ruttenberg <alanr-l(a)mumble.net> +2005-05-22 Alan Ruttenberg <alanr-l(a)mumble.net> * present.lisp. mouse-3 now gives a menu for actions on the presentation. See documentation in file for information about how

1 0

Re: [slime-devel] For presentation fans
by Alan Ruttenberg 24 May '05

24 May '05

Couple more developments: Now check that we are writing to a slime stream (or an pretty-printer stream that directs to a slime stream) before presenting. This should ensure that you don't land up with the in-band messages in places you didn't intend to, at the expense of possibly missing presentations if you direct your io to slime via some wrapping stream. (but you can specialize slime-stream-p for that stream, if you are motivated) First readable object : pathnames are now presented. E.g. load file (when *load-verbose* t) log message has sensitive pathname. Check out the menu (right click). Turn that off by setting *enable-presenting-readable-objects* nil You still have to explicitly load present.lisp. Please do. Unless there are objections I'll probably make this the default in a few days. Couple of more notes in the ChangeLog To do: - Make it work for non-dedicated-stream - When you copy the text of just a portion of a presentation with meta-w, remove the text properties. - Properly factor for swank.lisp, swank-backend, and lisp specific portions. - boxed text for multiple lines is ugly (lispms boxed the outside of all the text - emacs boxes each line). So see if there is a way to get nicer behavior, or consider changing default mouseover style. - Maybe replace x-popup-menu with menu-keymaps (volunteers welcome) - Assuming presenting readable objects doesn't cause problems, decide what other objects should so be printed - next candidate: structures. - Try to implement nested presentations. Strategy: slime-repl-old-output becomes a list of ids of object point is part of. Copy presentation and mouseover highlighting need extra work to properly gather all the parts. -Alan On May 22, 2005, at 2:58 AM, Alan Ruttenberg wrote: > There is now support for menus on the presentations. Assuming you have > present.lisp loaded, mouse-3 will bring it up. > Documentation on how to define menus is in present.lisp. > > Note that this uses x-popup-menu. Don't know if that works in xemacs. > > -Alan >

3 5

[slime-devel] Backups and Slime
by GP lisper 24 May '05

24 May '05

Hi, I have a weird problem with Emacs, and it apparently started around the same time I started using Slime. I hadn't been coding for a few months and didn't notice that the simple filename.~N~ emacs backups had stopped working....sometimes. After disabling various features in .emacs (or default.el), it doesn't seem to be slime related, but the time history matches up. After a week of guessing, I thought that maybe the emacs wizards here might have some better guesses. Some facts: With the root user, ~N~ backups are fine, regardless of the file ext. With the lisp user (I have one username exclusively for lisp work), ~N~ backups work on everything but .lisp files. So the version control settings would seem to be active, and testing in a *scratch* emacs buffer shows expected values. .default.el: ; version control settings (setq backup-by-copying t backup-directory-alist '(("." . "~/.saves")) delete-old-versions t kept-new-versions 9 kept-old-versions 1 version-control t ) ; SLIME (add-to-list 'load-path "/usr/share/emacs/site-lisp/slime") (require 'slime) (setq inferior-lisp-program "/usr/local/bin/lisp") (slime-setup :autodoc t) ;(add-hook 'slime-connected-hook 'slime-ensure-typeout-frame) (setq common-lisp-hyperspec-root "file:///usr/share/doc/HyperSpec/") (setq common-lisp-hyperspec-symbol-table "file:///usr/share/doc/HyperSpec/Data") (eval-after-load "hyperspec" '(load "hyperspec-addon")) Root and lisp users have identical .emacs, except for username in email (setq-default mail-default-reply-to "spambait(a)clouddancer.com") Grepping thru either .../slime/ or the emacs /lisp show that changing version-control is rarely done, and doesn't seem to be in slime at all. This is GNU emacs 21.4, I tried dropping back to 21.3 with the same outcome. Losing ~N~ backups is a problem, I'm used to being able to drop back to the code that had been working last hour, etc. Doing a lot of CVS submits just isn't a habit. Any guesses?? Any replacement mechanisms? -- With sufficient thrust, pigs fly fine.

2 3

[slime-devel] Typo or intentional ? (Slime class browser)
by R. Mattes 24 May '05

24 May '05

Hello list, While hacking on the class browser i stumbled across the following. in file slime.el, line 7834: (defun slime-browse-classes (name) "Read the name of a class and show its subclasses." (interactive (list (slime-read-symbol-name "Class Name: "))) (slime-call-with-browser-setup "*slime class browser*" (slime-current-package) "Class Browser" (lambda () (widget-create 'tree-widget :tag name :dynargs 'slime-expand-class-node :has-echildren t)))) AFAIK there's no option has-echildren - most likely a typo. Cheers Ralf Mattes

1 0

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 23 May '05

23 May '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.695 slime/ChangeLog:1.696 --- slime/ChangeLog:1.695 Sun May 22 08:52:13 2005 +++ slime/ChangeLog Mon May 23 04:31:12 2005 @@ -1,5 +1,22 @@ 2005-05-20 Alan Ruttenberg <alanr-l(a)mumble.net> + * present.lisp. (slime-stream-p) check if a stream is destined for + output in a slime listener. (checks *connections* looks into pretty-print + streams in openmcl and cmucl) + Don't present unless (slime-stream-p stream). + + Variable *enable-presenting-readable-objects* The only readable object + which is presented are pathnames (e.g. pathnames printed when loading + and *load-verbose* is t). Try the useful menu :) + More to come if this doesn't cause problems. (nil this if it does) + + *can-print-presentation* t around compile-string-for-emacs, + load-file, interactive-eval. + + In cmucl, use fwrappers to modify behaviour rather than redefinition. + +2005-05-20 Alan Ruttenberg <alanr-l(a)mumble.net> + * present.lisp. mouse-3 now gives a menu for actions on the presentation. See documentation in file for information about how to define menus. Also, disable presentations in inspector. Initial bits

1 0

[slime-devel] Bugs?
by Lars Brinkhoff 23 May '05

23 May '05

I tried FAIRLY-STABLE this morning, and had some problems: In swank-sbcl.lisp preferred-communication-style there were complaints that the linux_no_threads_p symbol was undefined. I'm using SBCL 0.8.16, Linux 2.6.11.6, and glibc 2.3.2. (Unconditionally returning :spawn works.) In swank.lisp clear-repl-results, there seems to be one superflous closing parenthesis.

2 2