Mercurial > hg > xemacs-beta
view src/coding-system-slots.h @ 1330:4542b72c005e
[xemacs-hg @ 2003-03-01 07:25:26 by ben]
build patch
Makefile.in.in: Move src deletions to src/Makefile.in.in.
dump-paths.el, dumped-lisp.el: Delete. Combine stuff into setup-paths.el.
find-paths.el: Removed.
Make this file contain generic routines only. Move stuff to
compute Emacs roots to setup-paths.el.
startup.el: Removed.
Move these variables into setup-paths.el.
setup-paths.el, startup.el: Removed.
Combine all high-level code for computing the paths into
setup-paths.el. Create new function startup-find-load-path to
encapsulate all logic for computing `load-path'. Eliminate
invocation-directory and invocation-name parameters since
there is no point (false generality) -- the code references
other globals, which cannot be specified. Eliminate some code
duplicated between setup-paths.el and startup.el. Clean up
the debug-paths code and output load-path in addition.
Add logic to paths-emacs-root-p to support separated source
and build trees.
loadup.el, make-docfile.el, update-elc-2.el, update-elc.el: Rewrite to allow for separated source and build trees, as may occur
in MS Windows.
NOTE TO BUILD HACKERS:
loadup.el, make-docfile.el, update-elc.el and update-elc-2.el made two
assumptions that are no longer correct:
(1) The source and build trees are in the same place.
(2) They can make assumptions about where `.' is.
These files now compute the locations of the source and build
roots at the top of the file. *ALL* constant file names or path
snippets must now be made absolute using expand-file-name and one
of these roots.
dumped-lisp.el, packages.el: Removed.
Remove some unused lists of Lisp files. packages-hardcoded-lisp
(empty, in any case) moved to dumped-lisp.el.
startup.el: When a compiled init file is out-of-date wrt the uncompiled
version, load the uncompiled version and issue a nasty warning.
update-elc-2.el: Force touching of auto-autoloads files when REBUILD_AUTOLOADS
was set.
update-elc.el: Fix code that checks whether dumping is necessary to check against
xemacs.dmp, not xemacs.exe, when Unix and pdump.
lwlib-Xm.c: Fix compile warning.
README, config.inc.samp, xemacs.mak: -- Major reorganization and cleanup.
-- Add support for separated build tree and source tree.
-- Delete all support for X Windows building, since it's
totally bit-rotten and will never be fixed up. Instruct
people to use Cygwin if they want such support.
make-build-dir: New script to create a skeleton build tree for use with
separated build and source tree compilation.
m/acorn.h, m/alliant-2800.h, m/alliant.h, m/altos.h, m/amdahl.h, m/arm.h, m/att3b.h, m/aviion.h, m/clipper.h, m/cnvrgnt.h, m/convex.h, m/cydra5.h, m/delta.h, m/delta88k.h, m/dpx2.h, m/elxsi.h, m/ews4800r.h, m/gould.h, m/hp800.h, m/hp9000s300.h, m/i860.h, m/ibmps2-aix.h, m/ibmrs6000.h, m/ibmrt-aix.h, m/ibmrt.h, m/intel386.h, m/iris4d.h, m/iris5d.h, m/iris6d.h, m/irist.h, m/m68k.h, m/masscomp.h, m/mg1.h, m/mips-nec.h, m/mips-siemens.h, m/mips.h, m/nh3000.h, m/nh4000.h, m/ns32000.h, m/plexus.h, m/powerpc.h, m/sequent-ptx.h, m/sequent.h, m/sgi-challenge.h, m/stride.h, m/tad68k.h, m/targon31.h, m/tekxd88.h, m/template.h, m/tower32.h, m/tower32v3.h, m/ustation.h, m/wicat.h, m/xps100.h, data.c, doc.c, editfns.c, emacs.c, lrecord.h, ntheap.c, process-unix.c, sysdep.c, unexec.c: Delete all support for bit-rotten CANNOT_DUMP. Just use pdump.
Makefile.in.in: Lots o' cleanup. Use names like LISP, SRC instead of
lispdir, srcdir, for consistency with xemacs.mak and the
conventions in the rest of the file. Eliminate use of ${...}
in favor of $(...), to make it easier to move code between
this file and xemacs.mak. Fix dependency handling wrt
NEEDTODUMP to eliminate problems some people (e.g. Vin) have
been seeing with non-GNU makes. Write a long section about
the subtle but oh-so-important differences in dependency
processing between nmake, make, and GNU make. Add
unicode-encapsulate target, from xemacs.mak.
chartab.c, lrecord.h: Fix crash due to attempt to free objects across dump/undump.
author | ben |
---|---|
date | Sat, 01 Mar 2003 07:25:56 +0000 |
parents | e22b0213b713 |
children | 1d74a1d115ee |
line wrap: on
line source
/* Definitions of marked slots in coding systems Copyright (C) 1991, 1995 Free Software Foundation, Inc. Copyright (C) 1995 Sun Microsystems, Inc. Copyright (C) 2000, 2001, 2002 Ben Wing. This file is part of XEmacs. XEmacs is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version. XEmacs is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with XEmacs; see the file COPYING. If not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ /* Synched up with: ????. Split out of file-coding.h. */ /* We define the Lisp_Objects in the coding system structure in a separate file because there are numerous places we want to iterate over them, such as when defining them in the structure, initializing them, or marking them. To use, define MARKED_SLOT before including this file. In the structure definition, you also need to define CODING_SYSTEM_SLOT_DECLARATION. No need to undefine either value; that happens automatically. */ #ifndef MARKED_SLOT_ARRAY #ifdef CODING_SYSTEM_SLOT_DECLARATION #define MARKED_SLOT_ARRAY(slot, size) MARKED_SLOT(slot[size]) #else #define MARKED_SLOT_ARRAY(slot, size) do { \ int mslotidx; \ for (mslotidx = 0; mslotidx < size; mslotidx++) \ { \ MARKED_SLOT (slot[mslotidx]) \ } \ } while (0); #endif #endif /* not MARKED_SLOT_ARRAY */ /* Name and description of this coding system. The description should be suitable for a menu entry. */ MARKED_SLOT (name) MARKED_SLOT (description) /* Mnemonic string displayed in the modeline when this coding system is active for a particular buffer. */ MARKED_SLOT (mnemonic) /* Long documentation on the coding system. */ MARKED_SLOT (documentation) /* Functions to handle additional conversion after reading or before writing. #### This mechanism should be replaced by the ability to simply create new coding system types. */ MARKED_SLOT (post_read_conversion) MARKED_SLOT (pre_write_conversion) /* If this coding system is not of the correct type for text file conversion (i.e. decodes byte->char), we wrap it with appropriate char<->byte converters. This is created dynamically, when it's needed, and cached here. */ MARKED_SLOT (text_file_wrapper) /* ------------------------ junk to handle EOL ------------------------- I had hoped that we could handle this without lots of special-case code, but it appears not to be the case if we want to maintain compatibility with the existing way. However, at least with the way we do things now, we avoid EOL junk in most of the coding system methods themselves, or in the decode/encode functions. The EOL special-case code is limited to coding-system creation and to the convert-eol and undecided coding system types. */ /* If this coding system wants autodetection of the EOL type, then at the appropriate time we wrap this coding system with convert-eol-autodetect. (We do NOT do this at creation time because then we end up with multiple convert-eols wrapped into the final result -- esp. with autodetection using `undecided' -- leading to a big mess.) We cache the wrapped coding system here. */ MARKED_SLOT (auto_eol_wrapper) /* Subsidiary coding systems that specify a particular type of EOL marking, rather than autodetecting it. These will only be non-nil if (eol_type == EOL_AUTODETECT). These are chains. */ MARKED_SLOT_ARRAY (eol, 3) /* If this coding system is a subsidiary, this element points back to its parent. */ MARKED_SLOT (subsidiary_parent) /* At decoding or encoding time, we use the following coding system, if it exists, in place of the coding system object. This is how we handle coding systems with EOL types of CRLF or CR. Formerly, we did the canonicalization at creation time, returning a chain in place of the original coding system; but that interferes with `coding-system-property' and causes other complications. CANONICAL is used when determining the end types of a coding system. canonicalize-after-coding also consults CANONICAL (it has to, because the data in the lstream is based on CANONICAL, not on the original coding system). */ MARKED_SLOT (canonical) #undef MARKED_SLOT #undef MARKED_SLOT_ARRAY #undef CODING_SYSTEM_SLOT_DECLARATION