Mercurial > hg > xemacs-beta
annotate etc/TERMS @ 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 | 376386a54a3c |
children |
rev | line source |
---|---|
0 | 1 This file describes what you must or might want to do to termcap entries |
2 to make terminals work properly and efficiently with Emacs. Information | |
3 on likely problems with specific types of terminals appears at the end | |
4 of the file. | |
5 | |
6 *** What you want in a terminal *** | |
7 | |
8 Vital | |
9 1. Easy to compute suitable padding for. | |
10 2. Never ever sends ^S/^Q unless you type them, at least in one mode. | |
11 | |
12 Nice for speed | |
13 1. Supports insert/delete of multiple lines in one command. | |
14 2. Same for multiple characters, though doing them one by | |
15 one is usually fast enough except on emulators running on | |
16 machines with bitmap screens. | |
17 | |
18 Nice for usability | |
19 1. Considerably more than 24 lines. | |
20 2. Meta key (shift-like key that controls the 0200 bit | |
21 in every character you type). | |
22 | |
23 *** New termcap strings *** | |
24 | |
25 Emacs supports certain termcap strings that are not described in the | |
26 4.2 manual but appear to be standard in system V. The one exception | |
27 is `cS', which I invented. | |
28 | |
29 `AL' insert several lines. Takes one parameter, the number of | |
30 lines to be inserted. You specify how to send this parameter | |
31 using a %-construct, just like the cursor positions in the `cm' | |
32 string. | |
33 | |
34 `DL' delete several lines. One parameter. | |
35 | |
36 `IC' insert several characters. One parameter. | |
37 | |
38 `DC' delete several characters. One parameter. | |
39 | |
40 `rp' repeat a character. Takes two parameters, the character | |
41 to be repeated and the number of times to repeat it. | |
42 Most likely you will use `%.' for sending the character | |
43 to be repeated. Emacs interprets a padding spec with a * | |
44 as giving the amount of padding per repetition. | |
45 | |
46 WARNING: Many terminals have a command to repeat the | |
47 *last character output* N times. This means that the character | |
48 will appear N+1 times in a row when the command argument is N. | |
49 However, the `rp' string's parameter is the total number of | |
50 times wanted, not one less. Therefore, such repeat commands | |
51 may be used in an `rp' string only if you use Emacs's special | |
52 termcap operator `%a-c\001' to subtract 1 from the repeat count | |
53 before substituting it into the string. It is probably safe | |
54 to use this even though the Unix termcap does not accept it | |
55 because programs other than Emacs probably won't look for `rp' | |
56 anyway. | |
57 | |
58 `cs' set scroll region. Takes two parameters, the vertical | |
59 positions of the first line to include in the scroll region | |
60 and the last line to include in the scroll region. | |
61 Both parameters are origin-zero. The effect of this | |
62 should be to cause a following insert-line or delete-line | |
63 not to move lines below the bottom of the scroll region. | |
64 | |
65 This is not the same convention that Emacs version 16 used. | |
66 That is because I was led astray by unclear documentation | |
67 of the meaning of %i in termcap strings. Since the termcap | |
68 documentation for `cs' is also unclear, I had to deduce the | |
69 correct parameter conventions from what would make the VT-100's | |
70 `cs' string work properly. From an incorrect assumption about | |
71 %i, I reached an incorrect conclusion about `cs', but the result | |
72 worked correctly on the VT100 and ANSII terminals. In Emacs | |
73 version 17, both `cs' and %i work correctly. | |
74 | |
75 The version 16 convention was to pass, for the second parameter, | |
76 the line number of the first line beyond the end of the | |
77 scroll region. | |
78 | |
79 `cS' set scroll region. Differs from `cs' in taking parameters | |
80 differently. There are four parameters: | |
81 1. Total number of lines on the screen. | |
82 2. Number of lines above desired scroll region. | |
83 3. Number of lines below (outside of) desired scroll region. | |
84 4. Total number of lines on the screen, like #1. | |
85 This is because an Ambassador needs the parameters like this. | |
86 | |
87 `cr', `do', `le' | |
88 Emacs will not attempt to use ^M, ^J or ^H for cursor motion | |
89 unless these capabilities are present and say to use those | |
90 characters. | |
91 | |
92 `km' Says the terminal has a Meta key. | |
93 | |
94 Defining these strings is important for getting maximum performance | |
95 from your terminal. | |
96 | |
97 Make sure that the `ti' string sets all modes needed for editing | |
98 in Emacs. For example, if your terminal has a mode that controls | |
99 wrap at the end of the line, you must decide whether to specify | |
100 the `am' flag in the termcap entry; whichever you decide, the `ti' | |
101 string should contain commands to set the mode that way. | |
102 (Emacs also sends the `vs' string after the `ti' string. | |
103 You can put the mode-setting commands in either one of them.) | |
104 | |
105 *** Specific Terminal Types *** | |
106 | |
107 Watch out for termcap entries for Ann Arbor Ambassadors that | |
108 give too little padding for clear-screen. 7.2 msec per line is right. | |
109 These are the strings whose padding you probably should change: | |
110 :al=1*\E[L:dl=1*\E[M:cd=7.2*\E[J:cl=7.2*\E[H\E[J: | |
111 I have sometimes seen `\E[2J' at the front of the `ti' string; | |
112 this is a clear-screen, very slow, and it can cause you to get | |
113 Control-s sent by the terminal at startup. I recommend removing | |
114 the `\E[2J' from the `ti' string. | |
115 The `ti' or `vs' strings also usually need stuff added to them, such as | |
116 \E[>33;52;54h\E[>30;37;38;39l | |
117 You might want to add the following to the `te' or `ve' strings: | |
118 \E[>52l\E[>37h | |
119 The following additional capabilities will improve performance: | |
120 :AL=1*\E[%dL:DL=1*\E[%dM:IC=4\E[%d@:DC=4\E[%dP:rp=1*%.\E[%a-c\001%db: | |
121 If you find that the Meta key does not work, make sure that | |
122 :km: | |
123 is present in the termcap entry. | |
124 | |
125 Watch out for termcap entries for VT100's that fail to specify | |
126 the `sf' string, or that omit the padding needed for the `sf' and `sr' | |
127 strings (2msec per line affected). What you need is | |
128 :sf=2*^J:sr=2*\EM:cs=\E[%i%d;%dr: | |
129 | |
130 The Concept-100 and Concept-108 have many modes that `ti' strings | |
131 often fail to initialize. If you have problems on one of these | |
132 terminals, that is probably the place to fix them. These terminals | |
133 can support an `rp' string. | |
134 | |
135 Watch out on HP terminals for problems with standout disappearing on | |
136 part of the mode line. These problems are due to the absence of | |
137 :sg#0: which some HP terminals need. | |
138 | |
139 The vi55 is said to require `ip=2'. | |
140 | |
141 The Sun console should have these capabilities for good performance. | |
142 :AL=\E[%dL:DL=\E[%dM:IC=\E[%d@:DC=\E[%dP: | |
143 | |
144 The vt220 needs to be set to vt220 mode, 7 bit, space parity | |
145 in order to work fully with TERM=vt220. | |
146 | |
147 If you are using a LAT terminal concentrator, you need to issue these | |
148 commands to turn off flow control: | |
149 | |
150 set port flow control disable | |
151 define port flow control disable | |
152 | |
153 On System V, in the terminfo database, various terminals may have | |
154 the `xt' flag that should not have it. `xt' should be present only | |
155 for the Teleray 1061 or equivalent terminal. | |
156 | |
157 In particular, System V for the 386 often has `xt' for terminal type | |
158 AT386 or AT386-M, which is used for the console. You should delete | |
159 this flag. Here is how: | |
160 | |
161 You can get a copy of the terminfo "source" for at386 using the | |
162 command: `infocmp at386 >at386.tic'. Edit the file at386.tic and remove | |
163 the `xt' flag. Then compile the new entry with: `tic at386.tic'. | |
164 | |
165 It is also reported that these terminal types sometimes have the wrong | |
166 reverse-scroll string. It should be \E[T, but sometimes is given as \E[S. | |
167 | |
168 Here is what watserv1!maytag!focsys!larry recommends for these terminals: | |
169 | |
170 # This copy of the terminfo description has been fixed. | |
171 # The suggestions came from a number of usenet postings. | |
172 # | |
173 # Intel AT/386 for color card with monochrome display | |
174 # | |
175 AT386-M|at386-m|386AT-M|386at-m|at/386 console, | |
176 am, bw, eo, xon, | |
177 cols#80, lines#25, | |
178 acsc=``a1fxgqh0jYk?lZm@nEooppqDrrsstCu4vAwBx3yyzz{{||}}~~, | |
179 bel=^G, blink=\E[5m, bold=\E[1m, cbt=\E[Z, | |
180 clear=\E[2J\E[H, | |
181 cr=\r, cub=\E[%p1%dD, cub1=\E[D, cud=\E[%p1%dB, | |
182 cud1=\E[B, cuf=\E[%p1%dC, cuf1=\E[C, | |
183 cup=\E[%i%p1%02d;%p2%02dH, cuu=\E[%p1%dA, cuu1=\E[A, | |
184 dch=\E[%p1%dP, dch1=\E[P, dl=\E[%p1%dM, dl1=\E[1M, | |
185 ech=\E[%p1%dX,ed=\E[J, el=\E[K, el1=\E[1K\E[X, flash=^G, home=\E[H, | |
186 hpa=\E[%i%p1%dG, ich=\E[%p1%d@, ich1=\E[1@, il=\E[%p1%dL, il1=\E[1L, | |
187 ind=\E[S, indn=\E[%p1%dS, invis=\E[9m, | |
188 is2=\E[0;10;38m, kbs=\b, kcbt=^], kclr=\E[2J, | |
189 kcub1=\E[D, kcud1=\E[B, kcuf1=\E[C, kcuu1=\E[A, | |
190 kdch1=\E[P, kend=\E[Y, kf1=\EOP, kf10=\EOY, kf11=\EOZ, | |
191 kf12=\EOA, kf2=\EOQ, kf3=\EOR, kf4=\EOS, kf5=\EOT, | |
192 kf6=\EOU, kf7=\EOV, kf8=\EOW, kf9=\EOX, khome=\E[H, | |
193 kich1=\E[@, knp=\E[U, kpp=\E[V, krmir=\E0, rev=\E[7m, ri=\E[T, | |
194 rin=\E[%p1%dT, rmacs=\E[10m, rmso=\E[m, rmul=\E[m, | |
195 sgr=\E[10m\E[0%?%p1%p3%|%t;7%;%?%p2%t;4%;%?%p4%t;5%;%?%p6%t;1%;%?%p9%t;12%;%?%p7%t;9%;m, | |
196 sgr0=\E[0;10m, smacs=\E[12m, smso=\E[7m, smul=\E[4m, | |
197 | |
198 # | |
199 # AT&T 386 color console | |
200 # | |
201 AT386|at386|386AT|386at|at/386 console, | |
202 colors#8, ncv#3, pairs#64, | |
203 is2=\E[0;10;39m, | |
204 op=\E[0m, | |
205 setb=\E[%?%p1%{0}%=%t40m | |
206 %e%p1%{1}%=%t44m | |
207 %e%p1%{2}%=%t42m | |
208 %e%p1%{3}%=%t46m | |
209 %e%p1%{4}%=%t41m | |
210 %e%p1%{5}%=%t45m | |
211 %e%p1%{6}%=%t43m | |
212 %e%p1%{7}%=%t47m%;, | |
213 setf=\E[%?%p1%{0}%=%t30m | |
214 %e%p1%{1}%=%t34m | |
215 %e%p1%{2}%=%t32m | |
216 %e%p1%{3}%=%t36m | |
217 %e%p1%{4}%=%t31m | |
218 %e%p1%{5}%=%t35m | |
219 %e%p1%{6}%=%t33m | |
220 %e%p1%{6}%=%t33m | |
221 %e%p1%{7}%=%t37m%;, | |
222 use=at386-m, | |
223 # | |
224 # Color console version that supports underline but maps blue | |
225 # foreground color to cyan. | |
226 # | |
227 AT386-UL|at386-ul|386AT-UL|386at-ul|at/386 console, | |
228 is2=\E[0;10;38m, | |
229 use=at386, |