[slime-devel] C-c C-k with external-format latin
Hello, I have a file containing german "umlaute", which I can load with SBCL using (load "name.lisp" :external-format :latin-1) However C-c C-k yields decoding error on stream #<SB-SYS:FD-STREAM for "file /home/neuss/Programming/lisp/stundenplan.lisp" {1002304EE1}> (:EXTERNAL-FORMAT :UTF-8): the octet sequence (246 114 115 97) cannot be decoded. [Condition of type SB-INT:STREAM-DECODING-ERROR] How can I make C-c C-k (slime-compile-and-load-file) use an external format of :latin-1? Thank you, Nicolas
* Nicolas Neuss [2006-10-31 13:53+0100] writes:
How can I make C-c C-k (slime-compile-and-load-file) use an external format of :latin-1?
You can add a file variable: -*- slime-coding: iso-latin-1-unix -*- and Slime should send it to the Lisp process. It would be nicer if we could just use "coding" but that wasn't easy to do. Even nicer would be, if SBCL would detect the encoding like Emacs does. Helmut.
Helmut Eller <heller@common-lisp.net> writes:
* Nicolas Neuss [2006-10-31 13:53+0100] writes:
How can I make C-c C-k (slime-compile-and-load-file) use an external format of :latin-1?
You can add a file variable: -*- slime-coding: iso-latin-1-unix -*- and Slime should send it to the Lisp process.
It would be nicer if we could just use "coding" but that wasn't easy to do. Even nicer would be, if SBCL would detect the encoding like Emacs does.
any reason not to use buffer-file-coding-system? or at least infer a default value from it? -- -Marco Ring the bells that still can ring. Forget your perfect offering. There is a crack in everything. That's how the light gets in. -Leonard Cohen
* Marco Baringer [2006-10-31 15:58+0100] writes:
any reason not to use buffer-file-coding-system? or at least infer a default value from it?
Can't remember the details. Write a test case, try to make it work in Emacs and XEmacs, and see yourself. Helmut.
+ Helmut Eller <heller@common-lisp.net>: | Can't remember the details. Write a test case, try to make it work in | Emacs and XEmacs, and see yourself. Um. Yeah, I tend to forget that xemacs exists. Sorry 'bout the noise. - Harald
+ Helmut Eller <heller@common-lisp.net>: | * Nicolas Neuss [2006-10-31 13:53+0100] writes: | | > How can I make C-c C-k (slime-compile-and-load-file) use an external format | > of :latin-1? | | You can add a file variable: -*- slime-coding: iso-latin-1-unix -*- | and Slime should send it to the Lisp process. | | It would be nicer if we could just use "coding" but that wasn't easy | to do. Why not? What happens if you try using buffer-file-coding-system, if set, and default-buffer-file-coding-system otherwise? (Not that I doubt your word, but I get curious.) -- * Harald Hanche-Olsen <URL:http://www.math.ntnu.no/~hanche/> - It is undesirable to believe a proposition when there is no ground whatsoever for supposing it is true. -- Bertrand Russell
Helmut Eller <heller@common-lisp.net> writes:
* Nicolas Neuss [2006-10-31 13:53+0100] writes:
How can I make C-c C-k (slime-compile-and-load-file) use an external format of :latin-1?
You can add a file variable: -*- slime-coding: iso-latin-1-unix -*- and Slime should send it to the Lisp process.
Unfortunately, this does not work. Trying C-c C-k on a file "test.lisp" with the content ;;; -*- slime-coding: iso-latin-1-unix -*- (in-package :cl-user) (defun test () "Prüfung" nil) I obtain the error decoding error on stream #<SB-SYS:FD-STREAM for "file /home/neuss/CL-HOME/mathematikum/test.lisp" {1002CA2E81}> (:EXTERNAL-FORMAT :UTF-8): the octet sequence (252 102 117 110) cannot be decoded. [Condition of type SB-INT:STREAM-DECODING-ERROR] Does C-c C-k on such a file work for you? Thanks, Nicolas
* Nicolas Neuss [2006-11-03 11:58+0100] writes:
Unfortunately, this does not work. Trying C-c C-k on a file "test.lisp" with the content
;;; -*- slime-coding: iso-latin-1-unix -*- (in-package :cl-user) (defun test () "Prüfung" nil)
I obtain the error
decoding error on stream #<SB-SYS:FD-STREAM for "file /home/neuss/CL-HOME/mathematikum/test.lisp" {1002CA2E81}> (:EXTERNAL-FORMAT :UTF-8): the octet sequence (252 102 117 110) cannot be decoded. [Condition of type SB-INT:STREAM-DECODING-ERROR]
Does C-c C-k on such a file work for you?
Yes, it does. Note that file local variables are only set when you open the file, not when you insert the -*- ... -*-. Verify that it is set properly, by evaluating something like M-: slime-coding in the buffer. Helmut.
Helmut Eller <heller@common-lisp.net> writes:
Does C-c C-k on such a file work for you?
Yes, it does. Note that file local variables are only set when you open the file, not when you insert the -*- ... -*-. Verify that it is set properly, by evaluating something like M-: slime-coding in the buffer.
Thanks for your answer, but this is not the error. It looks as if somehow the flag does not get propagated to SWANK-BACKEND::SOURCE-CACHE-GET. Can you maybe read something from this backtrace? 7: (SWANK-BACKEND::READ-FILE "/home/neuss/CL-HOME/mathematikum/test.lisp") 8: (SWANK-BACKEND::SOURCE-CACHE-GET "/home/neuss/CL-HOME/mathematikum/test.lisp" 3371540165) 9: ((LAMBDA (SWANK-BACKEND::FILENAME SWANK-BACKEND::LOAD-P #1="#<...>" . #1#)) "/home/neuss/CL-HOME/mathematikum/test.lisp" T :ISO-LATIN-1-UNIX) 10: (SWANK::MEASURE-TIME-INTERVAL #<CLOSURE (LAMBDA NIL) {100352EB79}>) 11: (SWANK::SWANK-COMPILER #<CLOSURE (LAMBDA NIL) {100352EB79}>) 12: ((LAMBDA NIL)) 13: ((LAMBDA (SWANK-BACKEND::FN)) #<CLOSURE (LAMBDA NIL) {100352EAD9}>) 14: (SWANK::CALL-WITH-BUFFER-SYNTAX #<CLOSURE (LAMBDA NIL) {100352EAD9}>) 15: (SB-INT:SIMPLE-EVAL-IN-LEXENV (SWANK:COMPILE-FILE-FOR-EMACS "/home/neuss/CL-HOME/mathematikum/test.lisp" T :ISO-LATIN-1-UNIX) #<NULL-LEXENV>) I am using CVS sbcl and CVS slime. Thank you, Nicolas
participants (4)
-
Harald Hanche-Olsen -
Helmut Eller -
Marco Baringer -
Nicolas Neuss