Mercurial > emacs
view src/doprnt.c @ 13728:4b7903cfa7ee
(bibtex-auto-fill-function): Adapted for use with
changed autofill policy of emacs-19.30 (uses now fill-prefix
instead of indent-line-function).
(bibtex-indent-line-function): Removed (not used any more).
(bibtex-make-field): Was broken when called
non-interactively.
(bibtex-make-field): Point is now placed on closing
brace or quote (suggested by Karl Eichwalder <ke@ke.Central.DE>).
(bibtex-clean-entry): Comma after last field isn't
deleted anymore (new standard in BibTeX 0.99 and 1.xx).
(bibtex-enclosing-reference-maybe-empty-head): Works with entries
with comma after last field.
(bibtex-reference): Permits entries with comma after last field.
(bibtex-font-lock-keywords): Enhanced to support new field-name
characters (suggested by Martin Maechler
<maechler@stat.math.ethz.ch>).
(bibtex-field-name): Now numbers (not as the first sign), dashes,
and underscores are allowed (suggested by Martin Maechler
<maechler@stat.math.ethz.ch> and Oren Patashnik
<opbibtex@labrea.Stanford.EDU>).
(bibtex-make-field): Was broken on lines containing
non-parenthesized entries (reported by Karl Eichwalder
<ke@ke.Central.DE>).
(bibtex-validate-buffer): Changed so that preamble
references are ignored (same as string entries) (reported by
Martin Maechler <maechler@stat.math.ethz.ch>).
(bibtex-enclosing-reference-maybe-empty-head):
New function to be used in case reference head may be empty.
(bibtex-clean-entry, bibtex-pop-previous, bibtex-pop-next): Uses
now bibtex-enclosing-reference-maybe-empty-head.
(bibtex-mode): Added support for font-lock mode.
(bibtex-font-lock-keywords): New variable with font-lock keywords
for BibTeX mode.
(bibtex-make-optional-field): Not longer interactive
(suggested by Karl Eichwalder <karl@pertron.central.de>).
(bibtex-maintain-sorted-entries): Set to nil, since it
requires more user attention and more restricted files to have
this set to t.
(bibtex-maintain-sorted-entries,
bibtex-sort-ignore-string-entries): Made buffer local, since it
may depend on the buffer which preferences to use.
(bibtex-validate-buffer): Looking for correct sort order only when
bibtex-maintain-sorted-entries is non-nil.
Put a comment in the `KNOWN BUGS' section about the
quote-inside-quotes problem.
(whole file): Changed string `true' in some documentation strings
to `non-nil' (e.g. `if variable has a true/non-nil value').
(bibtex-mode-map): Changed `move/edit' to `bibtex-edit'.
(bibtex-sort-entries): Now works correctly with
`@String' entries inside BibTeX files (i.e. after the occurence of
other references).
(bibtex-validate-buffer): Inserted code which looks if entries are
balanced (a single non-escaped quote inside braces was not
detected till now, but bibtex-sort-entries stumbles about it).
(bibtex-entry): bibtex-move-outside-of-entry is only
called when bibtex-maintain-sorted-entries is nil (otherwise
bibtex-find-entry-location determines the correct location).
(bibtex-find-entry-location): Now uses binary search. As before,
it assumes that the buffer is sorted without duplicates (but as
before it is only called when bibtex-maintain-sorted-entries is
t). Ignores `@String' entries if told so via variable
bibtex-sort-ignore-string-entries.
(bibtex-clean-entry): Respect
bibtex-maintain-sorted-entries when inserting autokey.
(bibtex-validate-buffer): Searching whole buffer for duplicates
and correct order is now done directly instead of calling
bibtex-find-entry-location (since this is to be reprogrammed to
use a binary search instead a sequential one).
(bibtex-parse-keys): May now be called with an
optional parameter which (if t) tells bibtex-parse-keys that it
should abort if input is pending.
(bibtex-mode): The instance of bibtex-parse-keys called in
auto-save-mode-hook is now called with this new parameter set to
t, so an auto-save caused by exceeding auto-save-interval is now
aborted immediately if user is still typing.
(bibtex-print-help-message, bibtex-clean-entry): Use
now constant strings instead of custom ones.
(bibtex-clean-entry): Changed the call of
bibtex-enclosing-reference to a more specific call so entries
without a key (here allowed) can be handled.
(bibtex-reference-key): Cleared off parentheses (caused string
entries enclosed by parentheses instead of braces to be not added
to bibtex-completion-candidates).
(bibtex-complete-string): Made it use bibtex-string.
(bibtex-keys,
bibtex-buffer-last-parsed-for-keys-tick): New buffer-local
variables to make parsing of BibTeX buffer for reference keys
(needed by TAB completion in minibuffer when entering key) more
occasional.
(bibtex-parse-keys): New function to parse for keys (functionality
was partially included in bibtex-entry).
(bibtex-entry): Changed to use bibtex-parse-keys.
(bibtex-mode): Installs bibtex-parse-keys as an
auto-save-mode-hook, so whole buffer is parsed at most when it is
autosaved.
(bibtex-clean-entry): Calls bibtex-parse-keys on the new entry, so
bibtex-keys remains consistent for new entries that are finished
by calling this function (most should).
(bibtex-inside-field): Be independent on current
setting of bibtex-field-right-delimiter (allows more intermixing
between quotes and braces).
(bibtex-make-field): Last change didn't make it work correctly
when called non-interactively by bibtex-entry (fixed).
author | Karl Heuer <kwzh@gnu.org> |
---|---|
date | Wed, 13 Dec 1995 20:26:13 +0000 |
parents | 626f0886b434 |
children | ee40177f6c68 |
line wrap: on
line source
/* Output like sprintf to a buffer of specified size. Also takes args differently: pass one pointer to an array of strings in addition to the format string which is separate. Copyright (C) 1985 Free Software Foundation, Inc. This file is part of GNU Emacs. GNU Emacs 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. GNU Emacs 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 GNU Emacs; see the file COPYING. If not, write to the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */ #include <config.h> #include <stdio.h> #include <ctype.h> #include "lisp.h" extern long *xmalloc (), *xrealloc (); static int doprnt1 (); /* Generate output from a format-spec FORMAT, terminated at position FORMAT_END. Output goes in BUFFER, which has room for BUFSIZE chars. If the output does not fit, truncate it to fit. Returns the number of characters stored into BUFFER. ARGS points to the vector of arguments, and NARGS says how many. A double counts as two arguments. String arguments are passed as C strings. Integers are passed as C integers. */ doprnt (buffer, bufsize, format, format_end, nargs, args) char *buffer; register int bufsize; char *format; char *format_end; int nargs; char **args; { return doprnt1 (0, buffer, bufsize, format, format_end, nargs, args); } /* Like doprnt except that strings in ARGS are passed as Lisp_Object. */ doprnt_lisp (buffer, bufsize, format, format_end, nargs, args) char *buffer; register int bufsize; char *format; char *format_end; int nargs; char **args; { return doprnt1 (1, buffer, bufsize, format, format_end, nargs, args); } static int doprnt1 (lispstrings, buffer, bufsize, format, format_end, nargs, args) int lispstrings; char *buffer; register int bufsize; char *format; char *format_end; int nargs; char **args; { int cnt = 0; /* Number of arg to gobble next */ register char *fmt = format; /* Pointer into format string */ register char *bufptr = buffer; /* Pointer into output buffer.. */ /* Use this for sprintf unless we need something really big. */ char tembuf[100]; /* Size of sprintf_buffer. */ int size_allocated = 100; /* Buffer to use for sprintf. Either tembuf or same as BIG_BUFFER. */ char *sprintf_buffer = tembuf; /* Buffer we have got with malloc. */ char *big_buffer = 0; register int tem; char *string; char fixed_buffer[20]; /* Default buffer for small formatting. */ char *fmtcpy; int minlen; int size; /* Field width factor; e.g., %90d */ char charbuf[2]; /* Used for %c. */ if (format_end == 0) format_end = format + strlen (format); if ((format_end - format + 1) < sizeof (fixed_buffer)) fmtcpy = fixed_buffer; else fmtcpy = (char *) alloca (format_end - format + 1); bufsize--; /* Loop until end of format string or buffer full. */ while (fmt != format_end && bufsize > 0) { if (*fmt == '%') /* Check for a '%' character */ { int size_bound; fmt++; /* Copy this one %-spec into fmtcpy. */ string = fmtcpy; *string++ = '%'; while (1) { *string++ = *fmt; if (! (*fmt >= '0' && *fmt <= '9') && *fmt != '-' && *fmt != ' '&& *fmt != '.') break; fmt++; } *string = 0; /* Get an idea of how much space we might need. */ size_bound = atoi (&fmtcpy[1]); /* Avoid pitfall of negative "size" parameter ("%-200d"). */ if (size_bound < 0) size_bound = -size_bound; size_bound += 50; if (size_bound > (unsigned) (1 << (BITS_PER_INT - 1))) error ("Format padding too large"); /* Make sure we have that much. */ if (size_bound > size_allocated) { if (big_buffer) big_buffer = (char *) xrealloc (big_buffer, size_bound); else big_buffer = (char *) xmalloc (size_bound); sprintf_buffer = big_buffer; size_allocated = size_bound; } minlen = 0; switch (*fmt++) { default: error ("Invalid format operation %%%c", fmt[-1]); /* case 'b': */ case 'd': case 'o': case 'x': if (cnt == nargs) error ("Not enough arguments for format string"); if (sizeof (int) == sizeof (EMACS_INT)) ; else if (sizeof (long) == sizeof (EMACS_INT)) /* Insert an `l' the right place. */ string[1] = string[0], string[0] = string[-1], string[-1] = 'l', string++; else abort (); sprintf (sprintf_buffer, fmtcpy, args[cnt++]); /* Now copy into final output, truncating as nec. */ string = sprintf_buffer; goto doit; case 'f': case 'e': case 'g': { union { double d; char *half[2]; } u; if (cnt + 1 == nargs) error ("not enough arguments for format string"); u.half[0] = args[cnt++]; u.half[1] = args[cnt++]; sprintf (sprintf_buffer, fmtcpy, u.d); /* Now copy into final output, truncating as nec. */ string = sprintf_buffer; goto doit; } case 'S': string[-1] = 's'; case 's': if (cnt == nargs) error ("not enough arguments for format string"); if (fmtcpy[1] != 's') minlen = atoi (&fmtcpy[1]); if (lispstrings) { string = (char *) XSTRING (((Lisp_Object *) args)[cnt])->data; tem = XSTRING (((Lisp_Object *) args)[cnt])->size; cnt++; } else { string = args[cnt++]; tem = strlen (string); } goto doit1; /* Copy string into final output, truncating if no room. */ doit: tem = strlen (string); doit1: if (minlen > 0) { while (minlen > tem && bufsize > 0) { *bufptr++ = ' '; bufsize--; minlen--; } minlen = 0; } if (tem > bufsize) tem = bufsize; bcopy (string, bufptr, tem); bufptr += tem; bufsize -= tem; if (minlen < 0) { while (minlen < - tem && bufsize > 0) { *bufptr++ = ' '; bufsize--; minlen++; } minlen = 0; } continue; case 'c': if (cnt == nargs) error ("not enough arguments for format string"); *charbuf = (EMACS_INT) args[cnt++]; string = charbuf; tem = 1; if (fmtcpy[1] != 'c') minlen = atoi (&fmtcpy[1]); goto doit1; case '%': fmt--; /* Drop thru and this % will be treated as normal */ } } *bufptr++ = *fmt++; /* Just some characters; Copy 'em */ bufsize--; }; /* If we had to malloc something, free it. */ if (big_buffer) xfree (big_buffer); *bufptr = 0; /* Make sure our string end with a '\0' */ return bufptr - buffer; }