Summary: MIME support added for e-mail processing that
skips encoded regions.  Allow user to skip saving Fcc messages with
large attachments.  Fixed region skipping bug with multi-line
comments - e.g.  tex $ regions spanning multiple lines.
Added support for postscript and uuencoded regions.
Redundant dictionary file names purged.
Dictionary definition field name changed from "Character Set"
to "Coding System".  Fixed bug in reloading dictionaries.
Modified headers to reflect new version.  XEmacs menu now adds
customize item.
(ispell-check-version): No longer an aliased function.
Returns library path if not called interactively.
Variable `temporary-file-directory' protected if not loaded.
(check-ispell-version): Now the alias for `ispell-check-version'.
(ispell-message-fcc-skip): New variable that determines if and when
to query about saving Fcc copy of message if an attachment is large.
(ispell-skip-html): Declared buffer-local.
(ispell-local-dictionary-alist): Docstring expanded.
Tag name changed from "Character Set" to "Coding System".
(ispell-dictionary-alist-1): Removed redundant command-line option to
load brasileiro, british, and castellano dictionary files.
(ispell-dictionary-alist-2): Removed redundant command-line option to
load czech dictionary file.
(ispell-dictionary-alist-3): Moved francais-tex here.
(ispell-dictionary-alist-4): Removed german and german8 dictionaries.
The deutsch ones are the correct definitions.  `nederlands'
and `nederlands8' dictionaries moved here.
(ispell-dictionary-alist-5): `polish' and `portugues' dictionaries
moved here.  Removed redundant command-line option to `norsk'
and `portugues'.
(ispell-dictionary-alist-6): Removed redundant command-line option to
load `russian' and `slovak' dictionary files.
(ispell-dictionary-alist): Tag name changed from "Character Set" to
"Coding System".
(ispell-version): Updated to 3.6.
(ispell-library-directory): Calls non-deprecated function.
(ispell-valid-dictionary-list): New function returning all valid
dictionaries on machine.
(ispell-checking-message): Documentation string improved.
(ispell-skip-region-alist): Added uuencoded and postscript region
skipping.  Improved http/e-mail/file regexp to not match `/.\w'.
(ispell-html-skip-alists): New variable for html region support.
(ispell-send-string): Removed redundant xemacs check.
(ispell-word): Fix spelling error in documentation string, added
extent information to support highlighting in ispell-minor-mode.
(ispell-command-loop): Disable horizontal scrollbar in XEmacs
choices buffer.
(ispell-show-choices): Directly select `choices-window'.
(ispell-help): Use default buffer size for electric help.
(ispell-adjusted-window-height): Correct for xemacs detection.
(ispell-start-process): Don't double specify dictionary file name.
(ispell-init-process): Set `ispell-library-path' each call.
(ispell-change-dictionary): Now only completes valid dictionaries.
(ispell-region): Add support for MIME region skipping and Fcc
message query for large attachments.
(ispell-begin-skip-region-regexp): Add documentation string.
Added message support and cleaned up code for generic and html regions.
(ispell-begin-skip-region): Function is now requires alist argument.
(ispell-begin-tex-skip-regexp): Added comments and support
improved html and message regions.
(ispell-skip-region-list): New function for MIME and region skipping.
(ispell-tex-arg-end): Add documentation string.
(ispell-ignore-fcc): New function to query saving Fcc message.
(ispell-skip-region): Calculate alist for key match dynamically,
html skipping pushed to alists.
(ispell-get-line): Add support for multi-line comment regions.
(ispell): Check that variables to continue spelling are bound.
(ispell-message-text-end): Postscript and uuencoded regions now
supported as MIME regions, rather than as end-of-message region.
(ispell-mime-multipartp): New function supporting MIME.
(ispell-mime-skip-part): New function supporting MIME.
(ispell-message): Add MIME support.
(ispell-buffer-local-parsing): Variable `ispell-skip-html' now local.
(ispell-buffer-local-dict): Fixed bug for detecting and reloading
new dictionary.

*** empty log message ***

(Creating Strings): Update split-string specification and examples.

(split-string): Implement specification that splitting on explicit separators
retains null fields.  Add new argument OMIT-NULLS.  Special-case (split-string
"a string").

*** empty log message ***

(difftime) [!HAVE_DIFFTIME]: Define.
(strerror) [!HAVE_STRERROR && !WINDOWSNT]: New.

(Indentation): Explain the concepts.
(Just Spaces): Explain why preventing tabs for indentation might
be useful.

Include lisp.h, not ../src/lisp.h.
(make_drawing_gcs): Remove unused `xswa', `mask'.

Include lisp.h, not ../src/lisp.h.
(lw_lucid_widget_p): Remove unused `mw'.
(xlw_update_one_widget, xlw_pop_instance) [PROTOTYPES]: Provide
ISO C arglists.

Remove redundant test for term.h.  Test for
difftime.

(ENCODE_UNSAFE_CHARACTER): Adjusted for the name change
of CODING_REPLACEMENT_CHARACTER.
(decode_coding_iso2022): If CODING_FLAG_ISO_SAFE, set
CODING_MODE_INHIBIT_UNENCODABLE_CHAR flag in coding->mode, and
check this flag on encoding.
(encode_coding_sjis_big5): Check
CODING_MODE_INHIBIT_UNENCODABLE_CHAR flag of coding->mode.
(Fset_terminal_coding_system_internal): Set
CODING_MODE_INHIBIT_UNENCODABLE_CHAR flag in terminal_coding.mode
instead of setting CODING_FLAG_ISO_SAFE flag in
terminal_coding.flags.

(CODING_REPLACEMENT_CHARACTER): Renamed from
CODING_INHIBIT_CHARACTER_SUBSTITUTION.
(CODING_MODE_INHIBIT_UNENCODABLE_CHAR): New macro.

Updated.

*** empty log message ***

("chinese-py-punct-b5"): Change the title
Chinese characters from GB to Big5.

(select-safe-coding-system): Try
default-buffer-file-coding-system too for automatic selection.

*** empty log message ***

utf bits.  useXIM.

useXIM.  index entries.

(x_window, x_window): Use use_xim.

(use_xim): Initialize.
(xim_open_dpy, xim_initialize, xim_close_dpy): Use use_xim.
(x_term_init): Maybe set use_xim.

(use_xim) [HAVE_X_I18N]: Declare.

(unicodedata-file): New.
(unicode-data): New (adapted from unicode branch).
(describe-char): Use it.  Print char's unicode differently.  Avoid
elements with null cadr when formatting list.  Clarify error
message when used in Help buffer.
(button): Require when compiling.
(describe-char-after): Alias for obsolete command.

(main) [WINDOWSNT]: Move it here.

(_start): Remove _fmode initialization.

(gud-gdba-command-name): Use -noasync option for Gdb
with MS windows.
(gdb-display-end): Only make buffer writeable temporarily.
Move "View" submenu up one level.

*** empty log message ***

(get_local_map): Don't get char-property of previous
point any more: get_pos_property already does it and better.

Do not instruct the user to autoload `ansi-color-for-comint-mode-on'.

(ctl-x-map): Remove the C-x U binding added recently.

*** empty log message ***

(gdb-view-source, gdb-selected-view): New variables.
(gdba): Inhibit DOS window in MS Windows.
(gdb-inferior-io-mode): Use hexl instead of cat.
(gdb-info-breakpoints-custom, gdb-assembler-custom, gdb-reset):
Use text in margin for MS Windows as there is no image support.
(gdb-restore-windows, gdb-setup-windows): Restore/start with
assembler view if appropriate.
(gdb-assembler-custom): Assembler code should display at point.
Parse address correctly.
(gdb-frame-handler): Accommodate selection of display of source
or assembler.
Add radio buttons to select display of source or assembler.

*** empty log message ***

Notes on fonts, Debian packages.

*** empty log message ***

Check HAVE_LIBLOCKFILE like HAVE_LIBMAIL.

Define LIBS_MAIL=-llockfile.

Append * to s390-*-linux-gnu case.
(LIBMAIL) <lockfile>: Don't define.

(MAIL_USE_FLOCK): Make it conditional.

*** empty log message ***

(desktop-save): Ensure parameter is expanded and ends with a slash before
assigning it to desktop-dirname and default-directory.
(desktop-read): Put buffers existing prior to evaluating the desktop (and not
reused) at the end of the buffer list.
(desktop-create-buffer): Don't update desktop-first-buffer for old desktop
files.

(xml-name-regexp): Wrap in `eval-and-compile'.

*** empty log message ***

Doc fixes.
(xml-parse-file, xml-parse-region): Autoload.
(xml-syntax-table, xml-name-regexp): New.
(xml-parse-region): Narrow to region, set syntax-table and
case-fold-search.  Reject fewer valid documents.
(xml-parse-tag): Remove arg END.  Callers changed.
(xml-parse-tag): Use skip-syntax-forward.  Use PARSE-DTD arg
properly.  Don't use buffer-substring-no-properties.  Don't bind
case-fold-search.  Fix syntax for empty elements.  Hoist consing
of end-of-tag regexp out of loop.
(xml-parse-attlist): Remove arg.  Callers changed.  Use
skip-syntax-forward, replace-regexp-in-string, forward-sexp.
Allow non-ASCII names.
(xml-skip-dtd): Remove arg.  Callers changed.  Change matching
code.
(xml-parse-dtd): Grok external DTDs.  Allow non-ASCII.  Don't use
match-string-no-properties.
(xml-ucs-to-string): Deleted.
(xml-substitute-entity): New.
(xml-substitute-special): Use it.
(xml-debug-print-internal): Simplify insertions.
(xml-parse-file): Avoid finding file in xml-mode.

* simple.el (kill-whole-line): New function.
* bindings.el (global-map): Bind it.

*** empty log message ***

Correct invariant section names.

(goto-address-fontify-maximum-size): Value t means no limit.
(goto-address-fontify): Implement that feature.

(find-function-on-key): Move the call to find-function-other-window
outside the save-excursion.

(ff-special-constructs): Delete the Ada entry.

(x-create-frame-with-faces): Don't call face-set-after-frame-default.
(face-set-after-frame-default): Copy attrs of global `default'
face to FRAME's `default' face.

Mark as obsolete.

Comment change.

(Finternal_set_lisp_face_attribute): Handle Qunspecified
as value for QCfont attribute.
(realize_default_face): lface should already exist; crash if not.
Specify 0 for FORCE_P when calling set_lface_from_font_name.

(Fignore_event): Doc fix.

(decode_coding_string): Handle post-read-conversion
even if the coding doesn't require decoding.

*** empty log message ***

*** empty log message ***

(skeleton-internal-1): Allow - as alternate interesting point marker and revert
@ to just setting skeleton-positions.

(put-arrow): Rename gdb-put-arrow and simplify.
(put-string): Rename gdb-put-string and simplify.
(remove-strings): Rename gdb-remove-strings.
(remove-arrow): Rename gdb-remove-arrow.
(gdb-assembler-custom): Try to get line marker (arrow) to display
in window (revisited).
Use with-current-buffer where possible.

*** empty log message ***

(beginning-of-defun-raw): Use shy-group.

(executable-set-magic): Remove unused vars `point' and `buffer-modified-p'.

(mail-mode): Don't bother setting paragraph-start.

Use explicit coding tag.

(sgml-xml-auto-coding-function, sgml-html-meta-auto-coding-function):
Don't assume point-min == 1.

(hz-set-msb-table): Build when compiling.
(big5-to-cns): Pre-build the alist when compiling.

(indian-regexp-of-hashtbl-keys): Don't sort the argument to regexp-opt.
(indian--puthash-char): Use dolist rather than mapc.

*** empty log message ***

Update copyright years.

(ibuffer-display-summary): New customizable variable.
(ibuffer-update-title-and-summary): Respect it.

*** empty log message ***

2003-05-18  Michael Kifer  <kifer@cs.stonybrook.edu>

* ediff-mult.el (ediff-default-filtering-regexp): new variable.

* ediff-util.el (ediff-maybe-save-and-delete-merge): change in a
message.

* ediff.el (ediff-directories,ediff-directory-revisions,
ediff-directories3,ediff-merge-directories,
ediff-merge-directories-with-ancestor,ediff-merge-directory-revisions,
ediff-merge-directory-revisions-with-ancestor): make use of the new
ediff-default-filtering-regexp variable.

*** empty log message ***

(Fcall_process_region): Doc fix.

(which-func-current, which-func-format):
Preload the risky-local-variable properties for them.

(prepare-change-group): Reinstate BUFFER arg; make it work.

*** empty log message ***

(Fload): Print a message if package is obsolete.

Don't require vc.el anymore.
(log-edit-mode-map): Fold vc-log-mode-map into it.
(vc-log-mode-map): Redefine as an alias.
(vc-previous-comment): Don't widen.
(vc-comment-to-change-log): Don't redundantly set paragraph-separate.
Don't require `add-log' since it's already required at toplevel.

(vc-log-mode-map, vc-maximum-comment-ring-size)
(vc-comment-ring, vc-comment-ring-index, vc-last-comment-match):
Move vars to log-edit.el.
(vc-new-comment-index, vc-previous-comment, vc-next-comment)
(vc-comment-search-reverse, vc-comment-search-forward)
(vc-comment-to-change-log): Move funs to log-edit.el.
(vc-clear-context): Don't empty the comment-ring.
(vc-finish-logentry): Don't add the comment onto the comment-ring.

(vc-log-mode-map, vc-maximum-comment-ring-size, vc-comment-ring)
(vc-comment-ring-index, vc-last-comment-match): Move vars from vc.el.
(vc-new-comment-index, vc-previous-comment, vc-next-comment)
(vc-comment-search-reverse, vc-comment-search-forward)
(vc-comment-to-change-log): Move funs from vc.el.
(log-edit-show-files): Remove unused var `editbuf'.

(macro-declaration-function): Avoid `dolist' and `cadr'.

New icons for debugger.

(gdb): Define gud-nexti.
(gud-menu-map): Add gud-nexti to map.
(gud-tool-bar-map): Add gud-nexti and gud-stepi to toolbar.

*** empty log message ***

(number-to-register): Use `string-to-number' instead
of `string-to-int'.

(tcl-indent-exp): Remove dead code.
(tcl-tab-always-indent): Default to tab-always-indent.

(skkdic-okuri-nasi-entries, skkdic-okuri-nasi-entries-count):
Use defvar for non-constants.

(perl-tab-always-indent): Default to tab-always-indent.
(perl-end-of-function): Remove unsed var `npos'.

(which-func-format): Make it risky-local-variable.
(which-func-table): New var.
(which-func-current): Make it into a constant modeline spec.
(which-func-previous): Remove.
(which-func-update): Only update the selected window.
(which-func-update-1): Use the new var to allow the current
function to be different for a buffer shown in two windows.

(with-selected-window): New macro.
(dolist, dotimes, with-current-buffer): Use backquotes.
(when, unless, save-match-data, combine-after-change-calls)
(with-output-to-string, with-temp-buffer): Add `declare' info.
(listify-key-sequence): Don't allocate unnecessarily.
(read-quoted-char): Allow up to base 36.
(prepare-change-group): Remove unimplemented argument.

(Fselect_window): Add optional `norecord' arg.
(select_window_1): Fold into Fselect_window.
(select_window_norecord): New function.
(temp_output_buffer_show): Use it.  Preserve current buffer.
(delete_window, Fother_window, Fset_window_configuration):
Pass nil as new arg to Fselect_window.

(Fselect_window): Add optional arg `norecord'.

(pp-to-string, Info-goto-node): Don't autoload.
(widget-choose, widget-map-buttons): Use with-current-buffer.
(widget-field-add-space): Change to nil (and to defconst).
(widget-info-link-action): Use `info'.

(read_process_output): Back out change from 2003-03-09.

*** empty log message ***

(get_pos_property): Declare.

(create_root_interval, graft_intervals_into_buffer): Use BEG.
(merge_properties, intervals_equal, merge_properties_sticky): Use XCAR, XCDR.
(adjust_for_invis_intang): Pass new arg to text_property_stickiness.
(get_local_map): Use get_pos_property (for stickiness and empty overlays).

(get_pos_property): Don't assume that `object' = nil.