* lisp/progmodes/compile.el (compilation-find-file): Avoid
`expand-file-name' when computing the file names, because that will
reliably give the wrong result when there's symlinks and ".."
involved (bug#8035).
* lisp/progmodes/compile.el (compilation-error-properties):
When the end-column parameter of a compilation message rule
(in compilation-error-regexp-alist[-alist]) is a function, treat its
return value as if it were matched by the regexp, which is how it is
documented to work, and how all other parameters work.
* lisp/progmodes/compile.el (compilation-parse-errors):
Don't omit messages from the error count when FILE is a function
rather than a regexp match number.
* lisp/progmodes/compile.el (compilation-goto-locus): Pop to the
source buffer if the match buffer is the selected window (bug#47414).
Copyright-paperwork-exempt: yes
* lisp/net/tramp-integration.el
(tramp-compile-disable-ssh-controlmaster-options): New defun. Add
it to `compilation-mode-hook'. (Bug#45518)
* lisp/progmodes/compile.el (compilation-get-file-structure):
Revert commit 4ce5646d59, it isn't needed.
To scroll up/down the current displayed occurrence/error
without abandon the Occur/Grep buffer.
Add also a command 'recenter-other-window' to recenter
the other window from any kind of buffer.
* lisp/window.el (recenter-other-window): New command.
Bind recenter-other-window to S-M-C-l (Bug#46119).
* lisp/simple.el (recenter-current-error): New command.
* lisp/progmodes/grep.el (grep-mode-map):
Delete bidings for n and p.
* lisp/progmodes/compile.el (compilation-minor-mode-map):
Move here the n and p bindings.
Bind `recenter-current-error' to l.
* lisp/replace.el (occur-mode-map):
Same.
* doc/emacs/windows.texi (Other Window):
* doc/emacs/display.texi (Recentering):
Document recenter-other-window.
* etc/NEWS (Changes in Specialized Modes and Packages in Emacs 28.1):
Announce the changes.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Rule 'gradle-kotlin': don't be more forgiving than necessary; we know
exactly what the output looks like (see
https://github.com/JetBrains/kotlin/commit/\
ffe8ae3840d7b9bdc82170c8181031f05ced68bd) and there is no reason to
risk mismatches or expensive backtracking (bug#18109). Recognise
'info' level messages. Convert to rx.
When 'omake' is included in compilation-error-regexp-alist, which it
still is by default, then all other rules are modified to match with
an extra leading 6 spaces as well. The 'cucumber' pattern relied on
this in order to work as intended.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Extend the 'cucumber' pattern so that it works even when 'omake'
is not included. Move it below the 'gnu' rule so that it doesn't
match anything else.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
The comments above the regexp for the 'gnu' rule contained
references to the previous string regexp, which has been difficult
to follow ever since the translation to rx. Move the comments
to their proper places, and add some guiding notes.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
'javac': Column numbers are 1-based by default; remove subtraction and
η-reduce. Translate regexp to rx (mechanised).
* test/lisp/progmodes/compile-tests.el (compile-tests--test-regexps-data):
Adapt 'javac' test cases to the change.
* lisp/progmodes/compile.el (compilation-error-properties): There
may not be a file name (bug#40111). In that case, don't do the
`compilation-transform-file-match-alist' thing.
When matching messages in compilation-mode, keep track of the rule
employed for each match. This facilitates debugging and allows us to
verify that each test case really exercises the rule that we expect it
to.
Naturally this uncovered several test cases that didn't check what the
author thought they did; the rules affixed to
compile-tests--test-regexps-data are those actually used, so that the
tests still pass.
* lisp/progmodes/compile.el (compilation--message): Add 'rule' slot.
(compilation-directory-properties, compilation-error-properties)
(compilation-internal-error-properties, compilation-parse-errors)
(compilation--compat-parse-errors): Set the rule slot.
* test/lisp/progmodes/compile-tests.el (compile-tests--test-regexps-data)
(compile-tests--grep-regexp-testcases)
(compile-tests--grep-regexp-tricky-testcases): Add rules to test cases.
(compile--test-error-line): Check that the rule matches what we expect.
(compile-test-grep-regexps): Adapt to test case format.
Remove now superfluous ert-info.
* lisp/progmodes/compile.el (compilation-parse-errors):
Hoist the binding of case-fold-search and a memq call out of
the loop, eliminating a minor but unnecessary quadratic term.
* lisp/progmodes/compile.el (compilation-mode-map): Remove mention of
compilation-minor-mode-map (bug#43382).
As the comment in the body says:
;; Don't inherit from compilation-minor-mode-map,
;; because that introduces a menu bar item we don't want.
;; That confuses C-down-mouse-3.
and the map actually inherits from special-mode-map.
Copyright-paperwork-exempt: yes
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Extend regular expression to match optional column numbers.
*
test/lisp/progmodes/compile-tests.el (compile-tests--test-regexps-data):
Add a test.
* test/lisp/progmodes/compile-tests.el (compile-test-error-regexps):
Update the total number of compilation errors in a test.
* etc/compilation.txt: Update compilation.txt with the newly supported
message format.
* etc/NEWS: Advertise the feature.
* lisp/progmodes/compile.el (compilation-next-single-property-change):
Parse whole buffer at once (bug#42806).
Also remove the comment that mentioned that it is an option to do it
in one go as we now actually start doing. As the existence of that
comment suggested, there is not really a reason to process the buffer
in small chunks. On the contrary, processing the output in arbitrary
units can result in certain constructs not being recognized because
they begin in one arbitrary chunk, while ending in another.
* lisp/progmodes/compile.el
(compilation-error-regexp-alist-alist): Prior to this patch the
line that contains "from a.h:1:0," was seen as INFO and the line
that contains "from a.c:1:" was seen as a WARNING. This patch
makes them both INFO (bug#17826).
These comments are unnecessary because the doc-strings that follow
already cover the same ground, while being more concise. These
comments were also prefixed with too many semicolons, causing them
to be treated as outline headings.
The number of regexps is large, they are written independently of one
another, and they frequently intersect. Using case-sensitive matching
improves separation and performance, and is probably what everyone
have being assuming was used by compilation-mode all along.
* lisp/progmodes/compile.el (compilation-error-case-fold-search): New.
(compilation-parse-errors): Bind case-fold-search to
compilation-error-case-fold-search during matching.
* etc/NEWS: Announce.
When run with -p or -P, OMake regurgitates error messages that
prevented further progress, indented by 6 spaces. Use that fact
to ameliorate the modification done to other error message regexps.
* lisp/progmodes/compile.el (compilation-parse-errors):
When 'omake' is enabled, allow error messages to be indented by 0 or 6
spaces instead of any number of spaces, to avoid pathological
behaviour.
(compilation-error-regexp-alist-alist): Anchor the 'omake' pattern to
bol for performance. Repair the 'ruby-Test::Unit' pattern, which
relied on the previously over-generous 'omake' hack.
* etc/compilation.txt (OMake): Add examples.
* test/lisp/progmodes/compile-tests.el (compile-tests--test-regexps-data)
(compile-test-error-regexps): Add test for OMake (indented error).
They have similar structure, and both suffer from being able to
match leading spaces in multiple ways which leads to bad performance
when backtracking (bug#39595).
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Improved 'msft' and 'watcom' regexps.
Anchor the regexp at line-start to prevent quadratic behaviour when
it doesn't match (bug#39595). It's unclear whether the type tag, like
[ERROR], is always present; we keep it optional just in case.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Rewrite 'maven' regexp, using rx for clarity.
* etc/compilation.txt (maven): More examples.
* test/lisp/progmodes/compile-tests.el
(compile-tests--test-regexps-data): No leading spaces; they seems to
stem from a misunderstanding in bug#11517.
When the 'msft' rule was moved and modified, the old copy was left
in place by mistake.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Remove old rule.
In particular, fix some exception occurrences, fix handling of a Compilation
Mode buffer being displayed in several windows, and fix the margin when
temporarily displaying a different buffer in a window, then returning to the
compilation mode buffer. The fix is relevant for frames without fringes,
e.g. tty frames.
* lisp/progmodes/compile.el: (compilation-set-window): Always set point to
(parameter) MK.
(compilation--set-up-margin, compilation--tear-down-margin): New functions.
(compilation--set-up-arrow-spec-in-margins)
(compilation--tear-down-arrow-spec-in-margins): Renamed by introducing -- and
pluralising margin to margins. Handle the margins in _all_ windows displaying
the pertinent buffer by using get-buffer-window-list. In ...--set-up-... add
compilation--set-up-margin to window-buffer-change-functions. In
...--tear-down-... remove the hook functions added in ...--set-up-....
* lisp/progmodes/compile.el: Make ant regexp accept filenames that may
occur on Cygwin (like c:/test) and optional additional severity level
after task name.
* etc/compilation.txt: Add sample.
* lisp/progmodes/compile.el (compilation-next-error-function):
Don't bug out with an incomprehensible error message on "make[2]:"
texts (bug#5316).
(compilation-find-file): Protect against there being no file name.
* lisp/progmodes/compile.el (compilation-tear-down-arrow-spec-in-margin): Make
sure compilation-arrow-overlay is an overlay before performing overlay
operations on it. This fixes bug #37733.
