* configure.ac (comma_version, comma_space_version) [mingw32]:
New output variables.
(nt/emacs.rc, nt/emacsclient.rc) [mingw32]: New output files.
* make-dist: Update nt/ for *.rc -> *.rc.in changes.
* nt/emacs.rc.in, nt/emacsclient.rc.in: Rename from nt/emacs.rc, emacsclient.rc.
Let configure generate the real files, and set the version numbers.
* lib-src/Makefile.in (emacsclient.res): Update deps for nt/emacsclient.rc
now being in the build directory, not the source directory.
* nt/Makefile.in (distclean): Delete *.rc.
(emacs.res): Update deps for nt/emacsclient.rc now being in the
build directory, not the source directory.
* admin/admin.el (set-version): No more need to update nt/*.rc.
* admin/authors.el (authors-renamed-files-alist): Add .rc.in files.
This is in preparation for generating the former automatically
from the latter.
* admin/notes/bugtracker, admin/notes/copyright, admin/notes/newfile:
ChangeLog -> change log
* admin/notes/changelogs: Remove, merging old contents to ...
* admin/notes/repo: ... here.
* doc/emacs/maintaining.texi (Change Log): Mention that ChangeLog files may
be copied to or from a version control system.
* doc/emacs/trouble.texi (Sending Patches): Point to the commit messages.
* doc/lispref/intro.texi (Acknowledgments): ChangeLog file -> change log entries.
* doc/lispref/tips.texi (Library Headers): Emacs uses a version control system.
* etc/CONTRIBUTE: Give advice about git commit messages and how
to generate proposed patches containing them.
... for detecting MinGW-w64. _W64 is not specific of MinGW-w64, it is
defined for compatibility with MS VC++.
* nt/inc/ms-w32.h: Define MINGW_W64.
* admin/CPP-DEFINES: Mention MINGW_W64.
* nt/addpm.c: Use it.
* nt/addsection.c: Use it.
* nt/preprep.c: Use it.
* src/w32.c: Use MINGW_W64 instead of _W64
* src/w32term.c: Likewise.
All bzr revision IDS, and all CVS revision IDs for which a commit
could be identified, were changed to time-date!committer version
stamps. All .cvsignore files in the history became .gitignore files.
Fixes-bug annotations from bzr were copied into the corresponding
commit comments.
(The first .cvsignore commit was 1999-09-30T14:07:54Z!fx@gnu.org>.
The last CVS commit was <2009-12-27T08:11:12Z!cyd@stupidchicken.com>)
Committer/author email addresses are generally correct for the
transition day, not necessarily when the commit was originally made.
All bzr revision IDS, and all CVS revision IDs for which a commit
could be identified, were changed to time-date!committer version
stamps. All .cvsignore files in the history became .gitignore files.
Fixes-bug annotations from bzr were copied into the corresponding
commit comments.
(The first .cvsignore commit was 1999-09-30T14:07:54Z!fx@gnu.org>.
The last CVS commit was <2009-12-27T08:11:12Z!cyd@stupidchicken.com>)
Committer/author email addresses are generally correct for the
transition day, not necessarily when the commit was originally made.
* configure.ac (doc/man/emacs.1): Generate it.
* Makefile.in (top_bootclean): Remove doc/man/emacs.1.
* make-dist: Do not distribute doc/man/emacs.1.
* admin/admin.el (set-version): No need to update doc/man/emacs.1.
* doc/man/emacs.1.in: Rename from emacs.1.
* .bzrignore: Add doc/man/emacs.1.
* grammars/c.by (template-type): Add :template-specifier and
:typevar to capture extra details about the template.
(opt-post-fcn-modifiers): Splice in the found symbol into the
return value correctly.
(QUESTION): New punctuation.
(expression): Add ternary conditional support.
* grammars/scheme.by (MODULE): New token.
(scheme): Handle expanding the MODULE tag.
(scheme-list): Remove closeparen required match.
(scheme-in-list): Remove extraneous matches for DEFINE. Add
support for MODULE Simplify matching for code & make work.
(name-args, name-arg-list, name-arg-expand): Make it work.
* semantic.el (semantic-mode): Add/remove 3
completion-at-point-functions.
(semantic-completion-at-point-function): Removed.
(semantic-analyze-completion-at-point-function)
(semantic-analyze-notc-completion-at-point-function)
(semantic-analyze-nolongprefix-completion-at-point-function): New
completion at point functions.
* semantic/doc.el (semantic-doc-snarf-comment-for-tag): Fix case
when comment-end is empty string.
* semantic/debug.el
(semantic-debug-parser-debugger-source): New buffer local
variable.
(semantic-debug-interface): Add 'nil' initform to overlays.
(semantic-debug-mode): Remove read-only tags from buffers on exit.
(semantic-debug): Add autoload cookie. Force the language
specific debugger to load.
* semantic/db.el (generic::semanticdb-full-filename): New generic
method to allow this method to be used on buffer names via an
associated database.
* semantic/symref.el
(semantic-symref-cleanup-recent-buffers-fcn): When cleaning up
buffers, don't clean up buffers that are being used (i.e., in a
window) when the hook fires.
(semantic-symref-recently-opened-buffers): New tracking variable.
(semantic-symref-cleanup-recent-buffers-fcn): New hook fcn.
(semantic-symref-result-get-tags): Move logic into
`semantic-symref-hit-to-tag-via-buffer', and cleanup buffers via
the symref cleanup function in post-command-hook.
(semantic-symref-hit-to-tag-via-buffer): Logic that used to be
from above.
(semantic-symref-hit-to-tag-via-db): New.
* semantic/analyze.el:
(semantic-analyze-find-tag-sequence-default): If first entry in
sequence is the only one, apply tagclass filter.
(semantic-analyze-princ-sequence): Show slot even if empty.
(semantic-analyze-find-tag-sequence)
(semantic-analyze-find-tag-sequence-default): Add flags argument.
Add support for forcing the final entry of the sequence to be of
class variable.
(semantic-analyze-find-tag): Fix bug where input class filter was
ignored if there was a typecache match.
(semantic-analyze-current-context-default): For assignments, the
assignee now must be of class variable.
* semantic/analyze/complete.el
(semantic-analyze-possible-completions-default): Add
'no-longprefix' flag. When used, the prefix and prefixtypes are
shortened to just the last symbol.
* semantic/bovine/c.el (semantic-c-do-lex-if): Catch errors from
'hideif', and push to the parser warning stack.
(semantic-lex-cpp-define): When a comment is at the end of a
macro, do not subtract an extra 1 from the found position. Fixes
bug with: #define foo (a)/**/ adding an extra ')' to the stream.
* semantic/bovine/scm.el (semantic-lex-scheme-symbol): Allow
symbols to be one char long.
* semantic/bovine/grammar.el
(bovine-grammar-calculate-source-on-path): New.
(bovine-grammar-setupcode-builder): Use it.
* ede.el (ede/detect): New require.
(ede-version): Bump version
(ede-initialize-state-current-buffer): Use new
`ede-detect-directory-for-project' to detect projects first
instead of depending on currente dir only.
(ede-delete-project-from-global-list): New.
(ede-flush-deleted-projects): Use above.
(ede-check-project-query-fcn): New variable
(ede-check-project-directory): Use above when querying the user.
Added to support unit testing of this security measure.
(ede-initialize-state-current-buffer): Use
`ede-directory-project-cons' instead of the -detect- fcn to take
advantage of the cache. Pass found project into
`ede-load-project-file'.
(ede-load-project-file): Add new input DETECTIN.
(ede-rescan-toplevel): Get the proj root a better way.
(ede-load-project-file): Return the loaded object. When asking
for existing project, ask for an exact match.
(ede-initialize-state-current-buffer): Simplify some conditional
logic.
(ede-load-project-file): Simplify conditional logic.
(ede-global-list-sanity-check): New Testing fcn.
(ede-parent-project): Replace old code with call to faster
`ede-find-subproject-for-directory'.
(ede-load-project-file): Use
`ede-directory-get-toplevel-open-project' instead of above
deleted. Rename "pfc" to "autoloader". Use
`ede-directory-project-cons' to detect a project. Delete no
project found case where we search up the tree.
* ede/auto.el (ede-project-autoload): Fix doc typo. Add
`:root-only' slot.
(ede-auto-load-project): Doc update: warn to not use.
(ede-dir-to-projectfile): Deleted.
(ede-project-autoload-dirmatch): Add subdir-only slot. Make
configdatastash unbound by default.
(ede-do-dirmatch): If subdir-only is true, then don't allow exact
matches. Account for configdatastash as unbound. Assume value of
nil means no tool installed. Make sure loaded path matches from
beginning. Stash the regexp, not the raw string.
(ede-project-class-files): Note that makefile and automake are not
root only.
(ede-auto-detect-in-dir): New (for use with `ede/detect.el').
(ede-project-dirmatch-p): Deleted.
(ede-project-root-directory): Remove body, return nil.
(ede-project-autoload): :proj-root-dirmatch can be null & doc fix.
(ede-auto-detect-in-dir): If there is no :proj-file, check for a
dirmatch.
* ede/generic.el (ede/config): Replace require of ede.
(ede-generic-new-autoloader): Generic projects are now safe by
default. Note this is NOT a root only project.
(project-rescan, ede-project-root, ede-generic-target-java)
(ede-java-classpath, ede-find-subproject-for-directory): New.
(ede-enable-generic-projects): Add new autoloaders for git, bzr,
hg, sv, CVS.
(ede-generic-vc-project)
(ede-generic-vc-project::ede-generic-setup-configuration): New.
(ede-generic-config): Remove slots: c-include-path,
c-preprocessor-table, c-preprocessor-files, classpath,
build-command, debug-command, run command. Inherit from
ede-extra-config-build, ede-extra-config-program. Make
run-command :value match :custom so only strings are accepted.
Add some more :group slot specifiers.
(ede-generic-project): Add mixins `ede-project-with-config-c' and
`ede-project-with-config-java'. Inherit from
`ede-project-with-config-build',
`ede-project-with-config-program'. Subclass
`ede-project-with-config'. Remove duplication from new baseclass.
(ede-generic-target): Inherit from `ede-target-with-config-build',
`ede-target-with-config-program'. Subclass `ede-target-with-config'.
(ede-generic-target-c-cpp): Add mixin `ede-target-with-config-c'.
(ede-generic-target-java): Add mixin `ede-target-with-config-java'.
(ede-preprocessor-map, ede-system-include-path)
(edejava-classpath): Deleted, moved to config.el.
(project-compile-project, project-compile-target)
(project-debug-target, project-run-target): Deleted.
(ede-generic-get-configuration, ede-generic-setup-configuration)
(ede-commit-project, project-rescan)
(ede-generic-project::ede-customize)
(ede-generic-target::ede-customize)
(ede-generic-config::eieio-done-customizing)
(ede-generic-config::ede-commit): Deleted. Subsumed by new
baseclass.
(ede-preprocessor-map, ede-system-include-path)
(project-debug-target, project-run-target): Call new
`ede-config-get-configuration' instead of old version.
(ede-generic-load): Do not add to global list here.
* ede/files.el (ede-find-project-root)
(ede-files-find-existing)
(ede-directory-get-toplevel-open-project-new): Deleted.
(ede-project-root-directory): Use `ede-project-root' first.
(ede-project-directory-remove-hash)
(ede--directory-project-from-hash)
(ede--directory-project-add-description-to-hash): Rename to make
internal symbols (via --). Expand input dir first.
(ede-directory-project-p): Doc fix (note obsoleted.)
(ede-toplevel-project-or-nil): Alias to `ede-toplevel-project'.
(ede-toplevel-project): Doc Fix. Delete commented out old code.
Simplify returning result from ede-detect-directory-for-project.
(ede-directory-get-open-project): Support when
inodes are disabled. If disabled to str compare on root project.
(ede-directory-get-toplevel-open-project): Enabled nested
projects. When doing directory name matching, save the 'short'
version of an answer (non-exact match) and eventually select the
shortest answer at the end. Expand the filename of tested
projects. Better support for when inodes are disabled. Add
'exact' option so that it will return a project that is an exact
match.
(ede-find-subproject-for-directory): Small optimization to run
`file-truename' less often.
(ede-directory-project-p): Move content, then use
`ede-directory-project-cons'. Use
`ede-detect-directory-for-project', replacing old detection loop.
(ede-directory-project-cons): New, from above.
(ede-toplevel-project): Toss old scanning code. Use
`ede-detect-directory-for-project' instead.
(ede-directory-get-toplevel-open-project-new): New.
* ede/linux.el (ede-linux-project-root): Deleted.
(ede-project-autoload): Remove dirmatch entry - it is no longer
needed.
* lisp/cedet/ede/proj.el (project-rescan): Replace direct
manipulation of `ede-projects' with equivalent and better
functions.
(ede-proj-load): Replace call to test if dir has project to
explicity ask filesystem if Project.ede is there.
* ede/config.el:
* ede/detect.el: New files.
* ede/project-am.el (project-run-target): Add "./" to program to
run for systems where '.' isn't in PATH.
(project-am-load): Remove old code regarding `ede-constructing'.
Just read in the makefiles.
* ede/linux.el (ede-linux-load): Do not add to global list here.
Don't check for existing anymore.
(project-rescan): New.
(ede-linux-project-list, ede-linux-file-existing): Deleted.
(ede-linux-project-root): Delete body. Need symbol for autoloads
for now.
(ede-linux-project): No longer instance tracker.
(ede-project-autoload): Don't provide :proj-root
* ede/emacs.el (ede-emacs-load): Do not add project to global list
here. Don't look for existing first.
(ede-project-autoload): Remove dirmatch entry - it is no longer
needed. Don't provide proj-root anymore.
(ede-emacs-project-list, ede-emacs-file-existing): Delete.
(ede-emacs-project-root): Remove body (need symbol for loaddefs
still).
(ede-emacs-project): Do not instance track anymore.
* ede/cpp-root.el (initialize-instance): Remove commented code.
Add note about why we are adding the project to the master list.
Make sure if we are replacing a prev version, remove from global
list.
(ede-cpp-root-file-existing)
(ede-cpp-root-project-file-for-dir)
(ede-cpp-root-count, ede-cpp-root-project-root, ede-cpp-root-load)
(ede-project-autoload cpp-root): Deleted.
(ede-project-root-directory): Return :directory instead of
calculating from :file.
(project-rescan): New.
* ede/base.el (ede-toplevel): Only use buffer cached value if
subproj not passed in.
* srecode/java.el (srecode-semantic-handle-:java): Fix case when
an EDE project didn't support java paths.
* lisp/cedet/ede/proj-elisp.el
(ede-proj-target-elisp::ede-proj-tweak-autoconf): Kill buffer
after saving modified elisp-comp script, as to avoid "file has
changed on disk; really edit the buffer" questions when script
gets rewritten.
* emacs-lisp/eieio-custom.el (eieio-customize-object): Set
eieio-cog (current group) to g, which is an improved form of input
group.
* srecode/doc-cpp.srt (mode): Set to c mode so this works with
both C and C++.
* admin/notes/repo: Avoid "DVCS" acronym without first explaining it.
Mention using the first line of a ChangeLog as the topic line, and
that commit messages should use UTF-8.
* lisp/mouse.el (mouse-drag-line): Fix misspelling of "right-fringe".
src/Makefile.in (repo-update): Fix a typo and slightly reword
commentary.
admin/notes/repo (Notes): Reword the stylistic guidance for commit log
messages so that they are in line with Emacs development practices
and style guidance.
admin/notes/bzr: Renamed to notes/repo, reorganixed to separate
VCS-dependent from VCS-independent stuff. Added guidance about
commit-comment format under DVCSes.
admin/notes/BRANCH: Merged into notes/repo.
lisp/version.el: Fix some fallback values to conform to the actual
release number.
admin: Changes in several documents.
autogen.sh: Neutralize language specific to a repository type.
doc/misc/efaq-w32.texi: Neutralized language specific to a repository type.
doc/misc/gnus-coding.txt: Neutralized language specific to a repository type.
lisp/Makefile.in: Change some production names so they're neutral
about the repository type.
1. A significant portion of the advice is independent of any
specific version-control sytem.
2. Doing the rename now means it is easier to compose a multi-file
patch that can be applied to fix the tree on the day of the git move.
