From: Daniel Pfeiffer <occitan@esperanto.org>
Subject: 2 fixes for compile.el
Date: Sat, 16 Nov 2002 18:27:21 +0100 [thread overview]
Message-ID: <20021116182721.578db725.occitan@esperanto.org> (raw)
[-- Attachment #1: Type: text/plain, Size: 1764 bytes --]
Hi,
Sorry, this time it's definite. My post yesterday had some glitches, and while debugging them I corrected and slimmed down quite a few of the regexps. Now I match all of the messages provided as examples except for the "horrible fc" and BSD lint 2 (both of which didn't work before either) and BSD lint 3, which still matches way too much.
While doing this it struck me that compile is a horrible cludge from prehistorical times. Now that font lock can trivially set arbitrary properties -- and does it much better than the stop & go approach of compile!!! --, it would be best to let it do the whole parsing.
next-error and the like would then only have to scan the buffer for fileinfo text properties left for it by font lock. I guess this would make the library a third smaller and more efficient too.
best regards
Daniel
---------------
Two more things:
- the word warning now gets found case insensitively in my new diff
- font-lock in CVS has for quite a few versions contained the line
,(+ ,c-type-names-depth 2))
which understandably causes an error for me in c-mode. But even taking away the comma leaves the var undefined :-( (in 1.199 which is the latest that might have worked with 21.2)
---------------
Hi,
this fixes 2 things:
- warnings and errors didn't get distinguished which finally got insupportable with strict gcc 3.2
- compile buffers are hypertext but didn't have mouse hightlight
Note that the latter uses font-lock's facename as property list, which was documented in 21.2 (as delivered by SuSE 8.1), but only works when getting a newer font-lock and syntax from CVS.
coralament / best Grötens / liebe Grüße / best regards / elkorajn salutojn
Daniel Pfeiffer
--
-- http://dapfy.bei.t-online.de/sawfish/
--
[-- Attachment #2: ChangeLog --]
[-- Type: application/octet-stream, Size: 503 bytes --]
2002-11-16 Daniel Pfeiffer <occitan@esperanto.org>
* compile.el (font-lock-compile-error-face,
font-lock-compile-warning-face): new faces
* compile.el (compile-error-or-warning-face): new fun
* compile.el (compilation-mode-font-lock-keywords): add previous
* compile.el (compilation-non-error-initials): new var
* compile.el (compilation-error-regexp-alist): added TYPE-IDX
before FORMATS, all work except HPUX fc & 4.3BSD lint pass 2/3
* compile.el (compilation-parse-errors): strip TYPE-IDX
[-- Attachment #3: compile.el.diff --]
[-- Type: application/octet-stream, Size: 9289 bytes --]
222a223,227
> (defvar compilation-non-error-initials '(?w ?W ?i)
> "These initial letters indicate a warning rather than an error.
> They are compared to the first letter matched by TYPE-IDX in
> `compilation-error-regexp-alist', if TYPE-IDX was given.")
>
227,229c232
< ;; 4.3BSD grep, cc, lint pass 1:
< ;; /usr/src/foo/foo.c(8): warning: w may be used before set
< ;; or GNU utilities:
---
> ;; GNU utilities:
231,232d233
< ;; or HP-UX 7.0 fc:
< ;; foo.f :16 some horrible error message
246,247c247,248
< \\([a-zA-Z]?:?[^:( \t\n]*[^:( \t\n0-9][^:( \t\n]*\\)[:(][ \t]*\\([0-9]+\\)\
< \\([) \t]\\|:\\(\\([0-9]+:\\)\\|[0-9]*[^:0-9]\\)\\)" 2 3 6)
---
> \\([a-zA-Z]?:?[^:( \t\n]*[^:( \t\n0-9][^:( \t\n]*\\):[ \t]*\\([0-9]+\\)\
> \\([) \t]\\|:\\(\\([0-9]+\\):\\|[0-9]*[^:0-9]\\)\\) *\\(warning\\|W:\\)?" 2 3 6 7)
252,255d252
< ("\\([a-zA-Z][-a-zA-Z._0-9]+\\): ?\
< \\([0-9]+\\)\\.\\([0-9]+\\)\
< -\\([0-9]+\\)\\.\\([0-9]+\\)\
< :" 1 2 3) ;; When ending points are supported, add line = 4 and col = 5.
257,265d253
< ("\\([a-zA-Z][-a-zA-Z._0-9]+\\): ?\
< \\([0-9]+\\)\\.\\([0-9]+\\)\
< -\\([0-9]+\\)\
< :" 1 2 3) ;; When ending points are supported, add line = 2 and col = 4.
< ;; foo.c:8-45.3: error message
< ("\\([a-zA-Z][-a-zA-Z._0-9]+\\): ?\
< \\([0-9]+\\)\
< -\\([0-9]+\\)\\.\\([0-9]+\\)\
< :" 1 2 nil) ;; When ending points are supported, add line = 2 and col = 4.
267,269c255
< ("\\([a-zA-Z][-a-zA-Z._0-9]+\\): ?\
< \\([0-9]+\\)\\.\\([0-9]+\\)\
< :" 1 2 3)
---
> ;; foo.c:8-45.3: error message
272,274c258,273
< \\([0-9]+\\)\
< -\\([0-9]+\\)\
< :" 1 2 nil);; When ending points are supported, add line = 3.
---
> \\([0-9]+\\)\\(\\.\\([0-9]+\\)\\)?\
> \\(-\\([0-9]+\\)\\)?\\(\\.\\([0-9]+\\)\\)?\
> :" 1 2 4)
>
> ;; 4.3BSD grep, cc, lint pass 1:
> ;; /usr/src/foo/foo.c(8): warning: w may be used before set
> ;; or IBM C/C++ Tools 2.01:
> ;; foo.c(2:0) : informational EDC0804: Function foo is not referenced.
> ;; foo.c(3:8) : warning EDC0833: Implicit return statement encountered.
> ;; foo.c(5:5) : error EDC0350: Syntax error.
> ("\\([^( \n\t]+\\)(\\([0-9]+\\)\\(:\\([0-9]+\\)\\)?) *: \\([iwe]\\)?" 1 2 4 5)
>
> ;; The first monster rule matched this, but wrongly
> ;; HP-UX 7.0 fc:
> ;; foo.f :16 some horrible error message
>
286c285
< : \\(error\\|warning\\) C[0-9]+:" 1 3)
---
> : \\(error\\|warning\\) C[0-9]+:" 1 3 nil 4)
295c294
< \\([0-9]+\\)\\([) \t]\\|:[^0-9\n]\\)" 4 5)
---
> \\([0-9]+\\)\\([) \t]\\|:[^0-9\n]\\)" 4 5 nil 1)
301c300,301
< ("^==[0-9]+==[^(]+\(([^:]+):([0-9]+)" 1 2)
---
> ;; (too bad he tried Perl regexps, which emacs doesn't understand :-)
> ("==[0-9]+==[^(]+(\\([^:]+\\):\\([0-9]+\\))" 1 2)
305,306c305,306
< (".*[ \t:]\\([a-zA-Z]?:?[^:( \t\n]+\\)[:(](+[ \t]*\\([0-9]+\\))[:) \t]*$"
< 1 2)
---
> ;; this doesn't match the above message
> ;;(".*[ \t:]\\([a-zA-Z]?:?[^:( \t\n]+\\)[:(](+[ \t]*\\([0-9]+\\))[:) \t]*$" 1 2)
313c313,314
< (".*([ \t]*\\([a-zA-Z]?:?[^:( \t\n]+\\)[:(][ \t]*\\([0-9]+\\))" 1 2)
---
> ;; and this still does:
> ;;(".*([ \t]*\\([a-zA-Z]?:?[^:( \t\n]+\\)[:(][ \t]*\\([0-9]+\\))" 1 2)
325c326
< ("\\(cfe\\|fort\\): [^:\n]*: \\([^ \n]*\\), line \\([0-9]+\\):" 2 3)
---
> ("\\(cfe\\|fort\\): \\([^:\n]*\\): \\([^ \n]*\\), line \\([0-9]+\\):" 3 4 nil 2)
333c334
< of[ \t]+\"?\\([a-zA-Z]?:?[^\":\n]+\\)\"?:" 3 2)
---
> of[ \t]+\"?\\([a-zA-Z]?:?[^\":\n]+\\)\"?:" 3 2 nil 1)
346c347
< \\([0-9]+\\)\\([\(.]\\([0-9]+\\)\)?\\)?[:., (-]" 1 2 4)
---
> \\([0-9]+\\)\\(\\([\(.]\\|pos \\)\\([0-9]+\\)\)?\\)?[:;., (-][^a-zA-Z]*(?\\(W)\\|warning\\)?" 1 2 5 6)
350,351d350
< ("^File \"\\([^,\" \n\t]+\\)\", line \\([0-9]+\\)," 1 2)
<
354c353
< ("^File \"\\([^,\" \n\t]+\\)\", lines? \\([0-9]+\\)[-0-9]*, characters? \\([0-9]+\\)" 1 2 3)
---
> ("File \"\\([^,\" \n\t]+\\)\", lines? \\([0-9]+\\)\\([-0-9]*, characters? \\([0-9]+\\)\\)?" 1 2 4)
360c359
< ("[a-z0-9/]+: \\([eE]rror\\|[wW]arning\\): \\([^,\" \n\t]+\\)[,:] \\(line \\)?\\([0-9]+\\):" 2 4)
---
> ("[a-z0-9/]+: \\([eE]rror\\|[wW]arning\\): \\([^,\" \n\t]+\\)[,:] \\(line \\)?\\([0-9]+\\):" 2 4 nil 1)
380c379
< ("[EW], \\([^(\n]*\\)(\\([0-9]+\\),[ \t]*\\([0-9]+\\)" 1 2 3)
---
> ("\\([EW]\\), \\([^(\n]*\\)(\\([0-9]+\\),[ \t]*\\([0-9]+\\)" 2 3 4 1)
391,396d389
< ;; IBM C/C++ Tools 2.01:
< ;; foo.c(2:0) : informational EDC0804: Function foo is not referenced.
< ;; foo.c(3:8) : warning EDC0833: Implicit return statement encountered.
< ;; foo.c(5:5) : error EDC0350: Syntax error.
< ("\\([^( \n\t]+\\)(\\([0-9]+\\):\\([0-9]+\\)) : " 1 2 3)
<
400c393
< ("\"\\(.*\\)\",\\([0-9]+\\)\\s-+\\(Error\\|Warning\\)\\[[0-9]+\\]:" 1 2)
---
> ("\"\\(.*\\)\",\\([0-9]+\\)\\s-+\\(Error\\|Warning\\)\\[[0-9]+\\]:" 1 2 nil 3)
423,424c416
< (".*: ERROR File = \\(.+\\), Line = \\([0-9]+\\)" 1 2)
< (".*: WARNING File = \\(.+\\), Line = \\([0-9]+\\)" 1 2)
---
> (".*: \\(ERROR\\|WARNING\\) File = \\(.+\\), Line = \\([0-9]+\\)" 2 3 nil 1)
434,435d425
< ("Error:.*\n.* line \\([0-9]+\\) char \\([0-9]+\\) of file://\\(.+\\)"
< 3 1 2)
438,439c428,429
< ("Warning:.*\n.* line \\([0-9]+\\) char \\([0-9]+\\) of file://\\(.+\\)"
< 3 1 2)
---
> ("\\(Error\\|Warning\\):.*\n.* line \\([0-9]+\\) char \\([0-9]+\\) of file://\\(.+\\)"
> 4 2 3 1)
443c433
< Each elt has the form (REGEXP FILE-IDX LINE-IDX [COLUMN-IDX FILE-FORMAT...])
---
> Each elt has the form (REGEXP FILE-IDX LINE-IDX [COLUMN-IDX TYPE-IDX FILE-FORMAT...])
446a437
> If TYPE-IDX is given, compare TYPE-IDX'th subexpression with `compilation-non-error-initials'.
597a589,620
> (defface font-lock-compile-error-face
> '((((type tty) (class color)) (:foreground "red"))
> (((class color) (background light)) (:foreground "Red" :weight bold))
> (((class color) (background dark)) (:foreground "Pink" :weight bold))
> (t (:inverse-video t :weight bold)))
> "Font Lock mode face used to highlight compiler errors."
> :group 'font-lock-highlighting-faces)
>
> (defface font-lock-compile-warning-face
> '((((type tty) (class color)) (:foreground "cyan" :weight bold))
> (((class color) (background light)) (:foreground "Purple" :weight bold))
> (((class color) (background dark)) (:foreground "Cyan" :weight bold))
> (t (:weight bold)))
> "Font Lock mode face used to highlight compiler warnings."
> :group 'font-lock-highlighting-faces)
>
> (defun compile-error-or-warning-face (type)
> "Choose face for whether we're highlighting an error or warning."
> `(face ,(if type
> (if (and (setq type (match-string type))
> (> (length type) 0)
> (memq (aref type 0) compilation-non-error-initials))
> 'font-lock-compile-warning-face
> 'font-lock-compile-error-face)
> (save-match-data
> (save-excursion
> (beginning-of-line)
> (if (let ((case-fold-search t)) (looking-at ".*warning"))
> 'font-lock-compile-warning-face
> 'font-lock-compile-error-face))))
> mouse-face highlight))
>
600,629c623,648
< (nconc
< ;;
< ;; Compiler warning/error lines.
< (mapcar (function
< (lambda (item)
< ;; Prepend "^", adjusting FILE-IDX and LINE-IDX accordingly.
< (let ((file-idx (nth 1 item))
< (line-idx (nth 2 item))
< (col-idx (nth 3 item))
< keyword)
< (when (numberp col-idx)
< (setq keyword
< (cons (list (1+ col-idx) 'font-lock-type-face nil t)
< keyword)))
< (when (numberp line-idx)
< (setq keyword
< (cons (list (1+ line-idx) 'font-lock-variable-name-face)
< keyword)))
< (when (numberp file-idx)
< (setq keyword
< (cons (list (1+ file-idx) 'font-lock-warning-face)
< keyword)))
< (cons (concat "^\\(" (nth 0 item) "\\)") keyword))))
< compilation-error-regexp-alist)
< (list
< ;;
< ;; Compiler output lines. Recognize `make[n]:' lines too.
< '("^\\([A-Za-z_0-9/\.+-]+\\)\\(\\[\\([0-9]+\\)\\]\\)?[ \t]*:"
< (1 font-lock-function-name-face) (3 font-lock-comment-face nil t)))
< ))
---
> `(;; Compiler warning/error lines.
> ,@(mapcar (lambda (item)
> (let ((file-idx (nth 1 item))
> (line-idx (nth 2 item))
> (col-idx (nth 3 item))
> (type-idx (nth 4 item))
> keyword)
> (when (numberp col-idx)
> (setq keyword
> `((,(1+ col-idx) font-lock-type-face nil t))))
> (when (numberp line-idx)
> (setq keyword
> `((,(1+ line-idx) font-lock-variable-name-face)
> ,@keyword)))
> (when (numberp file-idx)
> (setq keyword
> `((,(1+ file-idx)
> (compile-error-or-warning-face ,(if (numberp type-idx)
> (1+ type-idx))))
> ,@keyword)))
> ;; Prepend "^", adjusting FILE-IDX and LINE-IDX accordingly.
> (cons (concat "^\\(" (nth 0 item) "\\)") keyword)))
> compilation-error-regexp-alist)
> ;; Compiler output lines. Recognize `make[n]:' lines too.
> ("^\\([A-Za-z_0-9/\.+-]+\\)\\(\\[\\([0-9]+\\)\\]\\)?[ \t]*:"
> (1 font-lock-function-name-face) (3 font-lock-comment-face nil t))))
2162c2181
< (cons filename (cons default-directory (cdr alist))))
---
> (cons filename (cons default-directory (cddr alist))))
next reply other threads:[~2002-11-16 17:27 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-11-16 17:27 Daniel Pfeiffer [this message]
-- strict thread matches above, loose matches on Subject: below --
2002-11-16 21:12 2 fixes for compile.el Daniel Pfeiffer
2002-11-16 21:51 Daniel Pfeiffer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://www.gnu.org/software/emacs/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20021116182721.578db725.occitan@esperanto.org \
--to=occitan@esperanto.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).