Mercurial > hg > xemacs-beta
view configure.usage @ 665:fdefd0186b75
[xemacs-hg @ 2001-09-20 06:28:42 by ben]
The great integral types renaming.
The purpose of this is to rationalize the names used for various
integral types, so that they match their intended uses and follow
consist conventions, and eliminate types that were not semantically
different from each other.
The conventions are:
-- All integral types that measure quantities of anything are
signed. Some people disagree vociferously with this, but their
arguments are mostly theoretical, and are vastly outweighed by
the practical headaches of mixing signed and unsigned values,
and more importantly by the far increased likelihood of
inadvertent bugs: Because of the broken "viral" nature of
unsigned quantities in C (operations involving mixed
signed/unsigned are done unsigned, when exactly the opposite is
nearly always wanted), even a single error in declaring a
quantity unsigned that should be signed, or even the even more
subtle error of comparing signed and unsigned values and
forgetting the necessary cast, can be catastrophic, as
comparisons will yield wrong results. -Wsign-compare is turned
on specifically to catch this, but this tends to result in a
great number of warnings when mixing signed and unsigned, and
the casts are annoying. More has been written on this
elsewhere.
-- All such quantity types just mentioned boil down to EMACS_INT,
which is 32 bits on 32-bit machines and 64 bits on 64-bit
machines. This is guaranteed to be the same size as Lisp
objects of type `int', and (as far as I can tell) of size_t
(unsigned!) and ssize_t. The only type below that is not an
EMACS_INT is Hashcode, which is an unsigned value of the same
size as EMACS_INT.
-- Type names should be relatively short (no more than 10
characters or so), with the first letter capitalized and no
underscores if they can at all be avoided.
-- "count" == a zero-based measurement of some quantity. Includes
sizes, offsets, and indexes.
-- "bpos" == a one-based measurement of a position in a buffer.
"Charbpos" and "Bytebpos" count text in the buffer, rather than
bytes in memory; thus Bytebpos does not directly correspond to
the memory representation. Use "Membpos" for this.
-- "Char" refers to internal-format characters, not to the C type
"char", which is really a byte.
-- For the actual name changes, see the script below.
I ran the following script to do the conversion. (NOTE: This script
is idempotent. You can safely run it multiple times and it will
not screw up previous results -- in fact, it will do nothing if
nothing has changed. Thus, it can be run repeatedly as necessary
to handle patches coming in from old workspaces, or old branches.)
There are two tags, just before and just after the change:
`pre-integral-type-rename' and `post-integral-type-rename'. When
merging code from the main trunk into a branch, the best thing to
do is first merge up to `pre-integral-type-rename', then apply the
script and associated changes, then merge from
`post-integral-type-change' to the present. (Alternatively, just do
the merging in one operation; but you may then have a lot of
conflicts needing to be resolved by hand.)
Script `fixtypes.sh' follows:
----------------------------------- cut ------------------------------------
files="*.[ch] s/*.h m/*.h config.h.in ../configure.in Makefile.in.in ../lib-src/*.[ch] ../lwlib/*.[ch]"
gr Memory_Count Bytecount $files
gr Lstream_Data_Count Bytecount $files
gr Element_Count Elemcount $files
gr Hash_Code Hashcode $files
gr extcount bytecount $files
gr bufpos charbpos $files
gr bytind bytebpos $files
gr memind membpos $files
gr bufbyte intbyte $files
gr Extcount Bytecount $files
gr Bufpos Charbpos $files
gr Bytind Bytebpos $files
gr Memind Membpos $files
gr Bufbyte Intbyte $files
gr EXTCOUNT BYTECOUNT $files
gr BUFPOS CHARBPOS $files
gr BYTIND BYTEBPOS $files
gr MEMIND MEMBPOS $files
gr BUFBYTE INTBYTE $files
gr MEMORY_COUNT BYTECOUNT $files
gr LSTREAM_DATA_COUNT BYTECOUNT $files
gr ELEMENT_COUNT ELEMCOUNT $files
gr HASH_CODE HASHCODE $files
----------------------------------- cut ------------------------------------
`fixtypes.sh' is a Bourne-shell script; it uses 'gr':
----------------------------------- cut ------------------------------------
#!/bin/sh
# Usage is like this:
# gr FROM TO FILES ...
# globally replace FROM with TO in FILES. FROM and TO are regular expressions.
# backup files are stored in the `backup' directory.
from="$1"
to="$2"
shift 2
echo ${1+"$@"} | xargs global-replace "s/$from/$to/g"
----------------------------------- cut ------------------------------------
`gr' in turn uses a Perl script to do its real work,
`global-replace', which follows:
----------------------------------- cut ------------------------------------
: #-*- Perl -*-
### global-modify --- modify the contents of a file by a Perl expression
## Copyright (C) 1999 Martin Buchholz.
## Copyright (C) 2001 Ben Wing.
## Authors: Martin Buchholz <martin@xemacs.org>, Ben Wing <ben@xemacs.org>
## Maintainer: Ben Wing <ben@xemacs.org>
## Current Version: 1.0, May 5, 2001
# This program 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.
#
# This program 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.
eval 'exec perl -w -S $0 ${1+"$@"}'
if 0;
use strict;
use FileHandle;
use Carp;
use Getopt::Long;
use File::Basename;
(my $myName = $0) =~ s@.*/@@; my $usage="
Usage: $myName [--help] [--backup-dir=DIR] [--line-mode] [--hunk-mode]
PERLEXPR FILE ...
Globally modify a file, either line by line or in one big hunk.
Typical usage is like this:
[with GNU print, GNU xargs: guaranteed to handle spaces, quotes, etc.
in file names]
find . -name '*.[ch]' -print0 | xargs -0 $0 's/\bCONST\b/const/g'\n
[with non-GNU print, xargs]
find . -name '*.[ch]' -print | xargs $0 's/\bCONST\b/const/g'\n
The file is read in, either line by line (with --line-mode specified)
or in one big hunk (with --hunk-mode specified; it's the default), and
the Perl expression is then evalled with \$_ set to the line or hunk of
text, including the terminating newline if there is one. It should
destructively modify the value there, storing the changed result in \$_.
Files in which any modifications are made are backed up to the directory
specified using --backup-dir, or to `backup' by default. To disable this,
use --backup-dir= with no argument.
Hunk mode is the default because it is MUCH MUCH faster than line-by-line.
Use line-by-line only when it matters, e.g. you want to do a replacement
only once per line (the default without the `g' argument). Conversely,
when using hunk mode, *ALWAYS* use `g'; otherwise, you will only make one
replacement in the entire file!
";
my %options = ();
$Getopt::Long::ignorecase = 0;
&GetOptions (
\%options,
'help', 'backup-dir=s', 'line-mode', 'hunk-mode',
);
die $usage if $options{"help"} or @ARGV <= 1;
my $code = shift;
die $usage if grep (-d || ! -w, @ARGV);
sub SafeOpen {
open ((my $fh = new FileHandle), $_[0]);
confess "Can't open $_[0]: $!" if ! defined $fh;
return $fh;
}
sub SafeClose {
close $_[0] or confess "Can't close $_[0]: $!";
}
sub FileContents {
my $fh = SafeOpen ("< $_[0]");
my $olddollarslash = $/;
local $/ = undef;
my $contents = <$fh>;
$/ = $olddollarslash;
return $contents;
}
sub WriteStringToFile {
my $fh = SafeOpen ("> $_[0]");
binmode $fh;
print $fh $_[1] or confess "$_[0]: $!\n";
SafeClose $fh;
}
foreach my $file (@ARGV) {
my $changed_p = 0;
my $new_contents = "";
if ($options{"line-mode"}) {
my $fh = SafeOpen $file;
while (<$fh>) {
my $save_line = $_;
eval $code;
$changed_p = 1 if $save_line ne $_;
$new_contents .= $_;
}
} else {
my $orig_contents = $_ = FileContents $file;
eval $code;
if ($_ ne $orig_contents) {
$changed_p = 1;
$new_contents = $_;
}
}
if ($changed_p) {
my $backdir = $options{"backup-dir"};
$backdir = "backup" if !defined ($backdir);
if ($backdir) {
my ($name, $path, $suffix) = fileparse ($file, "");
my $backfulldir = $path . $backdir;
my $backfile = "$backfulldir/$name";
mkdir $backfulldir, 0755 unless -d $backfulldir;
print "modifying $file (original saved in $backfile)\n";
rename $file, $backfile;
}
WriteStringToFile ($file, $new_contents);
}
}
----------------------------------- cut ------------------------------------
In addition to those programs, I needed to fix up a few other
things, particularly relating to the duplicate definitions of
types, now that some types merged with others. Specifically:
1. in lisp.h, removed duplicate declarations of Bytecount. The
changed code should now look like this: (In each code snippet
below, the first and last lines are the same as the original, as
are all lines outside of those lines. That allows you to locate
the section to be replaced, and replace the stuff in that
section, verifying that there isn't anything new added that
would need to be kept.)
--------------------------------- snip -------------------------------------
/* Counts of bytes or chars */
typedef EMACS_INT Bytecount;
typedef EMACS_INT Charcount;
/* Counts of elements */
typedef EMACS_INT Elemcount;
/* Hash codes */
typedef unsigned long Hashcode;
/* ------------------------ dynamic arrays ------------------- */
--------------------------------- snip -------------------------------------
2. in lstream.h, removed duplicate declaration of Bytecount.
Rewrote the comment about this type. The changed code should
now look like this:
--------------------------------- snip -------------------------------------
#endif
/* The have been some arguments over the what the type should be that
specifies a count of bytes in a data block to be written out or read in,
using Lstream_read(), Lstream_write(), and related functions.
Originally it was long, which worked fine; Martin "corrected" these to
size_t and ssize_t on the grounds that this is theoretically cleaner and
is in keeping with the C standards. Unfortunately, this practice is
horribly error-prone due to design flaws in the way that mixed
signed/unsigned arithmetic happens. In fact, by doing this change,
Martin introduced a subtle but fatal error that caused the operation of
sending large mail messages to the SMTP server under Windows to fail.
By putting all values back to be signed, avoiding any signed/unsigned
mixing, the bug immediately went away. The type then in use was
Lstream_Data_Count, so that it be reverted cleanly if a vote came to
that. Now it is Bytecount.
Some earlier comments about why the type must be signed: This MUST BE
SIGNED, since it also is used in functions that return the number of
bytes actually read to or written from in an operation, and these
functions can return -1 to signal error.
Note that the standard Unix read() and write() functions define the
count going in as a size_t, which is UNSIGNED, and the count going
out as an ssize_t, which is SIGNED. This is a horrible design
flaw. Not only is it highly likely to lead to logic errors when a
-1 gets interpreted as a large positive number, but operations are
bound to fail in all sorts of horrible ways when a number in the
upper-half of the size_t range is passed in -- this number is
unrepresentable as an ssize_t, so code that checks to see how many
bytes are actually written (which is mandatory if you are dealing
with certain types of devices) will get completely screwed up.
--ben
*/
typedef enum lstream_buffering
--------------------------------- snip -------------------------------------
3. in dumper.c, there are four places, all inside of switch()
statements, where XD_BYTECOUNT appears twice as a case tag. In
each case, the two case blocks contain identical code, and you
should *REMOVE THE SECOND* and leave the first.
author | ben |
---|---|
date | Thu, 20 Sep 2001 06:31:11 +0000 |
parents | 277f059a321b |
children | 512e4a478e9d |
line wrap: on
line source
Usage: configure [--OPTION[=VALUE] ...] [CONFIGURATION] Set compilation and installation parameters for XEmacs, and report. Note that for most of the following options, you can explicitly enable them using `--OPTION=yes' and explicitly disable them using `--OPTION=no'. This is especially useful for auto-detected options. The option `--without-FEATURE' is a synonym for `--with-FEATURE=no'. Options marked with a (*) are auto-detected. Many features require external packages to be installed first. Get them from ftp://ftp.xemacs.org/pub/xemacs/aux. Use colons (or quoted spaces) to separate directory names in option values which are PATHs (i.e. lists of directories). The results of configure tests are saved in config.log, which is useful for diagnosing problems. General options: --help Issue this usage message. --verbose Accepted but ignored. --extra-verbose Display more information, useful for debugging. Compilation options: --compiler=PROG C compiler to use --xemacs-compiler=PROG compiler to use to compile just the xemacs executable --with-gcc (*) Use GCC to compile XEmacs. --cflags=FLAGS Compiler flags (such as -O) --cpp=PROG C preprocessor to use (e.g. /usr/ccs/lib/cpp or cc -E) --cppflags=FLAGS C preprocessor flags (e.g. -I/foo or -Dfoo=bar) --libs=LIBS Additional libraries (e.g. -lfoo) --ldflags=FLAGS Additional linker flags (e.g. -L/foo) --site-includes=PATH List of directories to search first for header files --site-libraries=PATH List of directories to search first for libraries --site-prefixes=PATH List of directories to search for include/ and lib/ subdirectories, just after 'site-includes' and 'site-libraries' --site-runtime-libraries=PATH List of ALL directories to search for dynamically linked libraries at run time --dynamic=yes Link dynamically if supported by system. --dynamic=no Force static linking on systems where dynamic linking is the default. --srcdir=DIR Look for the XEmacs source files in DIR. Works best when using GNU Make. Installation options: --prefix=DIR Install files below DIR. Defaults to `/usr/local'. --with-prefix=no Don't compile the value of --prefix into the executable. Window-system options: --with-gtk Support GTK on the X Window System. (EXPERIMENTAL) --with-gnome Support GNOME on the X Window System. (EXPERIMENTAL) --with-x11 (*) Support the X Window System. --x-includes=DIR Search for X header files in DIR. --x-libraries=DIR Search for X libraries in DIR. --with-msw (*) Support MS Windows as a window system (only under Cygwin and MinGW). --with-toolbars=no Don't compile with any toolbar support. --with-wmcommand=no Compile without realized leader window which will keep the WM_COMMAND property. --with-athena=TYPE Use TYPE Athena widgets (xaw, 3d, next, 95, or xpm) --with-menubars=TYPE Use TYPE menubars (lucid, motif, or no). The Lucid widgets emulate Motif (mostly) but are faster. *WARNING* The Motif menubar is currently broken. --with-scrollbars=TYPE Use TYPE scrollbars (lucid, motif, athena, or no). --with-dialogs=TYPE Use TYPE dialog boxes (lucid, motif, athena, or no). Lucid menubars and scrollbars are the default. Motif dialog boxes will be used if Motif can be found. --with-widgets=TYPE Use TYPE widgets (lucid, motif, athena, or no). Motif widgets will be used if Motif can be found. Other widget types are currently unsupported. --with-dragndrop Compile in the generic drag and drop API. This is automatically added if one of the drag and drop protocols is found (currently CDE, OffiX, MSWindows, and GTK). *WARNING* The Drag'n'drop support is under development and is considered experimental. --with-cde Compile in support for CDE drag and drop. --with-offix Compile in support for OffiX drag and drop. *WARNING* If you compile in OffiX, you may not be able to use multiple X displays success- fully. If the two servers are from different vendors, the results may be unpredictable. --with-xmu=no (*) For those unfortunates whose vendors don't ship Xmu. --external-widget Compile with external widget support. TTY (character terminal) options: --with-tty=no Don't support ttys. --with-ncurses (*) Use the ncurses library for tty support. --with-gpm (*) Compile in GPM mouse support for ttys. Image options: --with-xpm (*) Compile with support for XPM images. PRACTICALLY REQUIRED. Although this library is nonstandard and a real hassle to build, many basic things (e.g. toolbars) depend on it, and you will run into many problems without it. --with-png (*) Compile with support for PNG images. Recommended because the images on the About page are not viewable without it. --with-jpeg (*) Compile with support for JPEG images. Useful if you are using a mail, news reader, or web browser in XEmacs, so that JPEG images can be displayed. --with-tiff (*) Compile with support for TIFF images. Possibly useful, for the same reason as JPEG images. --with-xface (*) Compile with support for X-Face mail headers. Requires the compface package. Of doubtful usefulness. --with-gif=no Compile without the (builtin) support for GIF images. Sound options: --with-sound=TYPE,[TYPE],... Compile with native sound support. Valid types are `native', `nas' and `esd'. Prefix a type with 'no' to disable. The first option can be `none' or `all'. `none' is a synonym for `nonative,nonas,noesd'. `all' is a synonym for native,nas,esd or `all'. The default is to autodetect all sound support. --native-sound-lib=LIB Native sound support library. Needed on Suns with --with-sound=both because both sound libraries are called libaudio. Database options: --with-database=TYPE (*) Compile with database support. Valid types are `no' or a comma-separated list of one or more of `berkdb' and either `dbm' or `gnudbm'. --with-ldap (*) Compile with support for the LDAP protocol. --with-postgresql (*) Compile with support for the PostgreSQL RDBMS. Mail options: --mail-locking=TYPE (*) Specify the locking to be used by movemail to prevent concurrent updates of mail spool files. Valid types are `lockf', `flock', `dot', `locking' or `mmdf'. --with-pop Support POP for mail retrieval. --with-kerberos Support Kerberos-authenticated POP. --with-hesiod Support Hesiod to get the POP server host. Additional features: --with-tooltalk (*) Support the ToolTalk IPC protocol. --with-workshop Support the Sun WorkShop (formerly Sparcworks) development environment. --with-socks Compile with support for SOCKS (an Internet proxy). --with-dnet (*) Compile with support for DECnet. --with-modules Compile in experimental support for dynamically loaded libraries (Dynamic Shared Objects). --with-netinstall Compile in support for installation over the internet. --with-site-lisp=yes Allow for a site-lisp directory in the XEmacs hierarchy searched before the installation packages. --with-site-modules=no Disable site-modules directory in the XEmacs hierarchy, which is searched before the installation modules. --package-path=PATH Directories to search for packages to dump with xemacs. PATH splits into three parts separated by double colons (::), an early, a late, and a last part, corresponding to their position in the various system paths: The early part is always first, the late part somewhere in the middle, and the last part at the very back. Only the late part gets seen at dump time. If PATH has only one component, that component is late. If PATH has two components, the first is early, the second is late. --infodir=DIR Directory to install XEmacs Info manuals and dir in. --infopath=PATH Directories to search for Info documents, info dir and localdir files in case run-time searching for them fails. --moduledir=DIR Directory to install dynamic modules in. --pdump New, experimental, non-working, don't-sue-me-if- your-house-collapses-and-your-wife-leaves-you, portable dumper. --with-file-coding Allows transparent use of "foreign" line break conventions in text files (such as LF-delimited text imported from a Unix system to a Windows environment), optionally including autodetection. Defaults to ON on Windows, OFF on Unix. Internationalization options: --with-mule Compile with Mule (MUlti-Lingual Emacs) support, needed to support non-Latin-1 (including Asian) languages. --with-xim=xlib Compile with support for X input methods, --with-xim=motif (*) Used in conjunction with Mule support. Use either raw Xlib to provide XIM support, or the Motif XmIm* routines (when available). NOTE: On some systems bugs in X11's XIM support will cause XEmacs to crash, so by default, no XIM support is compiled in, unless running on Solaris and the XmIm* routines are detected. --with-canna (*) Compile with support for Canna (a Japanese input method used in conjunction with Mule support). --with-wnn (*) Compile with support for WNN (a multi-language input method used in conjunction with Mule support). --with-wnn6 (*) Compile with support for the commercial package WNN6. --with-i18n3 Compile with I18N level 3 (support for message translation). This doesn't currently work. --with-xfs Compile with XFontSet support for bilingual menubar. Can't use this option with --with-xim=motif or xlib. And should have --with-menubars=lucid. Debugging options: --debug Compile with support for debugging XEmacs. (Causes code-size increase and little loss of speed.) --error-checking=TYPE[,TYPE]... Compile with internal error-checking added. Causes noticeable loss of speed. Valid types are extents, bufpos, malloc, gc, typecheck. --error-checking=none Disable all internal error-checking (the default). --error-checking=all Enable all internal error-checking. --memory-usage-stats Compile with additional code to allow you to determine what XEmacs's memory is being used for. Causes a small code increase but no loss of speed. Normally enabled when --debug is given. --no-doc-file Don't rebuild the DOC file unless it's explicitly deleted. Only use during development. (It speeds up the compile-run-test cycle.) --use-union-type Enable or disable use of a union, instead of an int, for the fundamental Lisp_Object type; this provides stricter type-checking. Only works with some systems and compilers. --with-quantify Add support for performance debugging using Quantify. --with-purify Add support for memory debugging using Purify. Other options: --rel-alloc Use the relocating allocator (default for this option is system-dependent). --with-dlmalloc Control usage of Doug Lea malloc on systems that have it in the standard C library (default is to use it if it is available). --with-system-malloc Force use of the system malloc, rather than GNU malloc. --with-debug-malloc Use the debugging malloc package. --with-clash-detection Use lock files to detect multiple edits of the same file. The default is to do clash detection. You may also specify any of the `path' variables found in Makefile.in, including --bindir, --libdir, --docdir, --lispdir, --sitelispdir, --datadir, --infodir, --mandir and so on. Note that we recommend against explicitly setting any of these variables. See the INSTALL file for a complete list plus the reasons we advise not changing them. If successful, configure leaves its status in config.status. If unsuccessful after disturbing the status quo, it removes config.status. The configure script also recognizes some environment variables, each of which is equivalent to a corresponding configure flag. Configure flags take precedence over environment variables, if both are specified. environment corresponding variable configure flag ----------- -------------- CC --compiler XEMACS_CC --xemacs-compiler CPP --cpp CFLAGS --cflags CPPFLAGS --cppflags LDFLAGS --ldflags LIBS --libs LD_RUN_PATH --site-runtime-libraries For more details on the install process, consult the INSTALL file.