January 2006 - slime-devel - mailman3.common-lisp.net

[slime-devel] [patch] Make it possible to keep FASL files in source directory
by Maciek Pasternacki 18 Jan '06

18 Jan '06

Hello. I use SLIME with SBCL; I have two instances of SBCL in my system (sometimes more); one is system-wide installation, second is project-local, both are of the same version, but with possibly different features. Both use SLIME (project-local version loads swank by hand). Problem is that system-wide SBCL is threaded, and project-local one is not; thus, .fasl files generated by one are unreadable to other. I don't want project-local SBCL and SLIME to mess with my system-wide settings and I'd prefer to have it leave any files outside project dir alone. I made a small patch against current CVS to swank to make it possible to keep FASL files in source directory; now it is turned on by pushing :SWANK-FASL-IN-PLACE to *FEATURES*; I couldn't think of better way of signalling swank-loader.lisp that we want to have FASL with sources before loading it (i.e. before (DEFPACKAGE)s and (DEFVAR)s). If this way is ok, please apply this patch; I don't insist on this particular way of communicationg with loader (it seems kludgy), if there is better one, please treat this post as yet another feature request. -- __ Maciek Pasternacki <maciekp(a)japhy.fnord.org> [ http://japhy.fnord.org/ ] `| _ |_\ / { ...I choose to live and to grow, take and give and to ,|{-}|}| }\/ move, learn and love and to be paranoid and to lie, hate and fear \/ |____/ and to do what it takes to move through.. } ( M. J. Keenan ) -><-

2 2

[slime-devel] Beginning of defun should work for compile defun?
by Liam M. Healy 17 Jan '06

17 Jan '06

If I do C-M-a to go to beginning of defun, then do C-c C-c to compile the defun, it compiles the previous defun. Shouldn't the "beginning of defun" be the current defun? See also discussion: http://meme.b9.com/cview.html?channel=lisp&start=3346261896 Liam

2 1

[slime-devel] slime-selector r should start slime with default args
by Andreas Fuchs 17 Jan '06

17 Jan '06

Hi all, With the inferior-lisp-program variable bound to any value but a non-nil slime-lisp-implementations variable, the slime-selector r shortcut will start slime with inferior-lisp-program. This is a bit inconsistent with the behavior of M-x slime. which starts the first entry in slime-lisp-implementations (or the default entry etc.). The following patch fixes this: Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.575 diff -u -r1.575 slime.el --- slime.el 31 Dec 2005 15:06:09 -0000 1.575 +++ slime.el 8 Jan 2006 10:18:12 -0000 @@ -8573,7 +8573,7 @@ (cond ((slime-current-connection) (slime-output-buffer)) ((y-or-n-p "No connection: start Slime? ") - (slime-start)))) + (apply #'slime-start (slime-read-interactive-args))))) (def-slime-selector-method ?s "*slime-scratch* buffer." Alternatively, I could see it invoke `slime' interactively. I'm not sure what the preferred coding style is here. Have fun, -- Andreas Fuchs, (http://|im:asf@|mailto:asf@)boinkor.net, antifuchs

2 1

[slime-devel] slime-edit-definition error
by Jonathon McKitrick 17 Jan '06

17 Jan '06

I am trying to jump to the definition of a function, and got this running SBCL. Any idea what this could be? #<SB-SYS:FD-STREAM for "file /home/jonathon/.sbcl/site/cl-pdf/pdf-geom.lisp" {B799D51}> is closed. [Condition of type SIMPLE-ERROR] Restarts: 0: [ABORT-REQUEST] Abort handling SLIME request. 1: [TERMINATE-THREAD] Terminate this thread (#<THREAD "worker" {B78D109}>) Backtrace: 0: (SB-KERNEL:CLOSED-FLAME #<SB-SYS:FD-STREAM for "file /home/jonathon/.sbcl/site/cl-pdf/pdf-geom.lisp" {B799D51}>) 1: ((LAMBDA (CONDITION STREAM)) #<SB-INT:STREAM-DECODING-ERROR {B79A7A9}> #<SB-PRETTY:PRETTY-STREAM {B79AB79}>) 2: (SB-PRETTY:OUTPUT-PRETTY-OBJECT #<SB-INT:STREAM-DECODING-ERROR {B79A7A9}> #<SB-IMPL::STRING-OUTPUT-STREAM {B79AAB9}>) 3: (PRINC #<SB-INT:STREAM-DECODING-ERROR {B79A7A9}> #<SB-IMPL::STRING-OUTPUT-STREAM {B79AAB9}>) 4: (SB-FORMAT::A-FORMAT-DIRECTIVE-INTERPRETER #<SB-IMPL::STRING-OUTPUT-STREAM {B79AAB9}> #<~A> NIL #<unavailable argument> (#<SB-INT:STREAM-DECODING-ERROR {B79A7A9}>)) 5: (SB-FORMAT::INTERPRET-DIRECTIVE-LIST #<SB-IMPL::STRING-OUTPUT-STREAM {B79AAB9}> (#<~A>) (#<SB-INT:STREAM-DECODING-ERROR {B79A7A9}>) (#<SB-INT:STREAM-DECODING-ERROR {B79A7A9}>)) 6: (SB-FORMAT::%FORMAT #<SB-IMPL::STRING-OUTPUT-STREAM {B79AAB9}> "Error: ~A" (#<SB-INT:STREAM-DECODING-ERROR {B79A7A9}>) (#<SB-INT:STREAM-DECODING-ERROR {B79A7A9}>)) 7: (FORMAT NIL "Error: ~A") -Jonathon -- "I am sure that Jesus would use a command prompt. Hello? Ten Commandments??"

2 2

[slime-devel] SBCL restart-frame
by Mészáros Levente 17 Jan '06

17 Jan '06

Hi, I am relatively new to SBCL and SLIME, so maybe I'm wrong about this. I did not find restart-frame support for SBCL, so I did this patch and it seems to work for me. RCS file: /project/slime/cvsroot/slime/swank-sbcl.lisp,v retrieving revision 1.151 diff -r1.151 swank-sbcl.lisp 929c929,933 < --- > > (defimplementation restart-frame (index) > (let ((frame (nth-frame index))) > (return-from-frame index (sb-debug::frame-call-as-list frame)))) > levy ________________________________________________________________________ Nézze meg az Üvegtigrist INGYEN! T-Online Téka kódolatlan hétvége január 13-15. www.t-online.hu

3 2

[slime-devel] Debugging without stepping
by Jonathon McKitrick 14 Jan '06

14 Jan '06

I'm new to SLIME, and trying to figure out some practical debugging strategies since I'm so used to single-stepping and breakpoints, and we don't have those facilities (IIRC) under Lisp/SLIME. What are some ways to accomplish the same thing? Or do we just use C-c C-t and debug that way, reading return values from traced functions? -Jonathon -- "I am sure that Jesus would use a command prompt. Hello? Ten Commandments??"

8 10

RE: [slime-devel] invoke-debugger
by Mészáros Levente 13 Jan '06

13 Jan '06

Meanwhile I figured out that I was wrong about this, sorry. > -----Original Message----- > From: slime-devel-bounces(a)common-lisp.net [mailto:slime-devel- > bounces(a)common-lisp.net] On Behalf Of Mészáros Levente > Sent: Tuesday, January 10, 2006 10:09 AM > To: slime-devel(a)common-lisp.net > Subject: [slime-devel] invoke-debugger > > Hi, > > I'm using SBCL 0.9.7 with SLIME and have the following issue: > > (defun foo () > (error "Foo error")) > > (defun bar () > (handler-case > (foo) > (error (e) > (print "Got error") > (invoke-debugger e)))) > > (bar) > > Calling bar directly in an SBCL repl causes the "Got error" message to > appear and the debugger shows up having bar in the stack trace > correctly. > > Doing the same thing from a SLIME repl also displays the message but > the debugger shows a weird stack trace without bar. > > It's not very important to me because I have a workaround for it. I just > wanted to note it (maybe someone wants to fix this). > > Levy ________________________________________________________________________ Hideg van. Licitáljon otthonról! HunBook online könyvárverés január 20. 18:30. www.hunbook.hu

1 0

[slime-devel] Connecting to running image versus starting from emacs
by Jonathon McKitrick 12 Jan '06

12 Jan '06

Hi all, I have a question about running a Lisp app. Right now, I'm starting a repl, loading an asdf-system, and starting a swank server. Then I connect from emacs/slime. Is there a way to start emacs and load/run the app from there? -Jonathon -- "I am sure that Jesus would use a command prompt. Hello? Ten Commandments??"

3 3

[slime-devel] 2 slime inspector buffers around at the same time?
by Chisheng Huang 12 Jan '06

12 Jan '06

Hi, Is it possible to have 2 slime inspector buffers around at the same time? This can be very handy for side-by-side comparison. Best wishes, -cph

2 1

[slime-devel] unexpected behaviour of C-c C-c
by Chisheng Huang 12 Jan '06

12 Jan '06

Hi, Save the following into a file: (progn (print :aaaaaaaaaa) (finish-output)) (progn ;; C-c C-c while your cursor is at the left parenthesis on this line (print :bbbbbbbbbb) (finish-output)) Fire up Lisp and Slime. Move your cursor to the position specified in the above comment and you'll see :AAAAAAAAAA printed out in Slime's REPL. I believe doing the same thing a few weeks ago would print out :BBBBBBBBBB. I have to say I like the old way much more than the new one. A bit more info: Slime: the latest from CVS Gnu Emacs 21.4.1 CMUCL: 2005-09 snapshot Best wishes, -cph

1 0