(byte-compile-constants, byte-compile-variables): Fix docstring.
(byte-compile-initial-macro-environment): Use `byte-compile-eval' to
execute `eval-whenc-compile's body.
(byte-compile-unresolved-functions): Fix docstring.
(byte-compile-eval): New function.
(byte-compile-callargs-warn): Check if the function will be available
at runtime (via property `byte-compile-noruntime').
(byte-compile-print-syms): New function.
(byte-compile-warn-about-unresolved-functions): Also warn about
`noruntime' functions (and use `byte-compile-print-syms').
(byte-compile-file): Capitalize the message.
function defined by `quickurl-format-function'. Before only the
text of the URL was passed. Now the whole URL structure is passed
and the function is responsible for extracting the parts it
requires. Changed the default of `quickurl-format-function'
accordingly.
(quickurl-insert): Changed the `funcall' of
`quickurl-format-function' to match the above change.
(quickurl-list-insert): Changed the `url' case so that it makes
use of `quickurl-format-function', previous to this the format was
hard wired.
(w32_menu_show): Set help string correctly.
(add-menu-item): Set help string in MIIM_DATA for menu item.
Load SetMenuItemInfoA explicitly.
(w32_menu_display_help): New function.
(QCdata): Moved to xdisp.c.
Include fontset.h after dispextern.h.
(Fx_create_frame): Do not create fontset.
(w32_load_system_font): Doc fix.
(Fx_close_connection): Free full_name if it is not shared.
regarding the "parent" handle. These just separate out the different
usages based on the type of parent (interval vs lisp object); later
changes will do type checking and enforcement.
* intervals.h (NULL_INTERVAL): Cast to INTERVAL type.
(INT_LISPLIKE): New macro.
(NULL_INTERVAL_P): Use it.
(INTERVAL_HAS_PARENT, INTERVAL_HAS_OBJECT, SET_INTERVAL_PARENT,
SET_INTERVAL_OBJECT, INTERVAL_PARENT, COPY_INTERVAL_PARENT,
GET_INTERVAL_OBJECT, INTERVAL_PARENT_OR_NULL): New macros.
* alloc.c (make_interval, gc_sweep): Use new macros; eliminate all
explicit references to "parent" field of struct interval and
associated unclean type conversions.
* intervals.c (create_root_interval, root_interval, rotate_right,
rotate_left, balance_possible_root_interval, split_interval_right,
split_interval_left, interval_start_pos, find_interval,
next_interval, previous_interval, update_interval,
adjust_intervals_for_insertion, delete_node, delete_interval,
adjust_intervals_for_deletion, merge_interval_right,
merge_interval_left, reproduce_tree, graft_intervals_into_buffer,
copy_intervals_to_string): Likewise.
* intervals.h (AM_LEFT_CHILD, AM_RIGHT_CHILD, RESET_INTERVAL):
Likewise.
* syntax.c (update_syntax_table): Likewise.
* intervals.c (reproduce_tree_obj): New function, like
reproduce_tree but takes a Lisp_Object for the parent. Declare
with prototype.
(graft_intervals_into_buffer): Use it when appropriate.
(reproduce_tree): Declare with prototype.
(balance_possible_root_interval): Check that the parent is a lisp
object before trying to examine its type.
definitions for non-Emacs compilation.
(enum re_opcode_t): Remove (not)wordchar and move (not)syntaxspec
outside of `#ifdef emacs'.
(print_partial_compiled_pattern): Update.
(regex_compile): Use (not)syntaxspec(Sword) instead of (not)wordchar.
(re_compile_fastmap): Merge handling of charset and charset_not (for
emacs and non-emacs compilation as well).
Similarly for (not)categoryspec and (not)syntaxspec.
Don't use the fastmap when reaching `anychar' since the added
complexity is not justified.
(re_match_2_internal): Merge (not)wordchar (emacs and non-emacs)
and (not)syntaxspec. Merge (not)categoryspec.
(cvs-make-cvs-buffer): Change the header part by removing the startup
message and adding a `Module' entry. Also replace the FOOTER and
HEADER special fileinfos with the new support in ewoc for updating
its own footer and header.
(cvs-update-header): Update to use the header/footer of the ewoc.
(cvs-mode): Use define-derived-mode and set truncate-lines to t.
(cvs-is-within-p): New function.
(cvs-mode-run): Take advantage of `save-some-buffers's new ability
to only examine some subset of the buffers.
* pcvs-info.el (cvs-fileinfo-pp): Use the new property-preserving
`format' instead of our own ad-hoc functions.
Remove HEADER and FOOTER cases, now handled in the EWOC.
(cvs-fileinfo<): Remove HEADER and FOOTER cases.
* pcvs-parse.el (cvs-parse-run-table): Change message for unknown
output to avoid scaring the user.
(cvs-parse-table): Catch message for non-up-to-date commits.
* pcvs-defs.el (cvs-startup-message): Remove.
(cvs-global-menu): New autoloaded menu.
* pcvs-util.el (cvs-string-fill): Remove.
* emacs-lisp/ewoc.el (ewoc--create-special-node): Remove.
(ewoc--refresh-node): Don't take the whole EWOC but only the relevant
PP part of it and also make it work for footers and headers.
(ewoc-create): Drop POS and BUFFER arguments.
Use the DLL's dummy node to store the end-of-footer position.
(ewoc-map, ewoc-invalidate): Update call to ewoc--refresh-node.
(ewoc-refresh): Remove unused `header' variable.
(ewoc-(get|set)-hf): New functions.
log-view-*-(message|file) and use easy-mmode-define-navigation.
(log-view-message-re): Match SCCS format as well.
And match the revision line rather than the dashed separator line.
(log-view-mode): Use the new define-derived-mode.
(log-view-current-tag): Fill in with an actual implementation.
cvs-status-(prev|next)-entry and use easy-mmode-define-navigation.
(cvs-tree-dstr-*): Rename from cvstree-dstr-* and use two ascii chars
to let the output "breathe" a little more (more readable).
(smerge-basic-map): Rename from smerge-basic-keymap.
Change the bindings for smerge-diff-*.
(smerge-*-map): Use easy-mmode-defmap.
(smerge-(next|prev)): Use easy-mmode-define-navigation.
(smerge-keep-*): Use smerge-auto-leave.
(tags-loop-eval): New function. Bind case-fold-search around eval
depending on the value of tags-case-fold-search.
(tags-loop-continue): Use tags-loop-eval.
(find-tag-in-order): Bind case-fold-search depending on the value
of tags-case-fold-search.
