/* Low-level bidirectional buffer-scanning functions for GNU Emacs.
- Copyright (C) 2000, 2001, 2004, 2005, 2009, 2010
+ Copyright (C) 2000-2001, 2004-2005, 2009-2011
Free Software Foundation, Inc.
This file is part of GNU Emacs.
#include <config.h>
#include <stdio.h>
-#include <string.h>
#include <setjmp.h>
#include "lisp.h"
#define RLO_CHAR 0x202E
#define BIDI_EOB -1
-#define BIDI_BOB -2 /* FIXME: Is this needed? */
/* Local data structures. (Look in dispextern.h for the rest.) */
/* What we need to know about the current paragraph. */
struct bidi_paragraph_info {
- int start_bytepos; /* byte position where it begins */
- int end_bytepos; /* byte position where it ends */
- int embedding_level; /* its basic embedding level */
- bidi_dir_t base_dir; /* its base direction */
+ EMACS_INT start_bytepos; /* byte position where it begins */
+ EMACS_INT end_bytepos; /* byte position where it ends */
+ int embedding_level; /* its basic embedding level */
+ bidi_dir_t base_dir; /* its base direction */
};
/* Data type for describing the bidirectional character categories. */
STRONG
} bidi_category_t;
+extern int bidi_ignore_explicit_marks_for_paragraph_level EXTERNALLY_VISIBLE;
int bidi_ignore_explicit_marks_for_paragraph_level = 1;
static Lisp_Object paragraph_start_re, paragraph_separate_re;
}
}
-void
+static void
bidi_check_type (bidi_type_t type)
{
if (type < UNKNOWN_BT || type > NEUTRAL_ON)
resolved levels in cached states. DIR, if non-zero, means search
in that direction from the last cache hit. */
static INLINE int
-bidi_cache_search (int charpos, int level, int dir)
+bidi_cache_search (EMACS_INT charpos, int level, int dir)
{
int i, i_start;
}
static INLINE bidi_type_t
-bidi_cache_find (int charpos, int level, struct bidi_it *bidi_it)
+bidi_cache_find (EMACS_INT charpos, int level, struct bidi_it *bidi_it)
{
int i = bidi_cache_search (charpos, level, bidi_it->scan_dir);
static EMACS_INT
bidi_at_paragraph_end (EMACS_INT charpos, EMACS_INT bytepos)
{
- /* FIXME: Why Fbuffer_local_value rather than just Fsymbol_value? */
Lisp_Object sep_re;
Lisp_Object start_re;
EMACS_INT val;
return pos_byte;
}
-/* Determine the direction, a.k.a. base embedding level, of the
+/* Determine the base direction, a.k.a. base embedding level, of the
paragraph we are about to iterate through. If DIR is either L2R or
R2L, just use that. Otherwise, determine the paragraph direction
- from the first strong character of the paragraph.
-
- Note that this gives the paragraph separator the same direction as
- the preceding paragraph, even though Emacs generally views the
- separartor as not belonging to any paragraph. */
+ from the first strong directional character of the paragraph.
+
+ NO_DEFAULT_P non-nil means don't default to L2R if the paragraph
+ has no strong directional characters and both DIR and
+ bidi_it->paragraph_dir are NEUTRAL_DIR. In that case, search back
+ in the buffer until a paragraph is found with a strong character,
+ or until hitting BEGV. In the latter case, fall back to L2R. This
+ flag is used in current-bidi-paragraph-direction.
+
+ Note that this function gives the paragraph separator the same
+ direction as the preceding paragraph, even though Emacs generally
+ views the separartor as not belonging to any paragraph. */
void
-bidi_paragraph_init (bidi_dir_t dir, struct bidi_it *bidi_it)
+bidi_paragraph_init (bidi_dir_t dir, struct bidi_it *bidi_it, int no_default_p)
{
EMACS_INT bytepos = bidi_it->bytepos;
+ EMACS_INT pstartbyte;
/* Special case for an empty buffer. */
if (bytepos == BEGV_BYTE && bytepos == ZV_BYTE)
/* We are either at the beginning of a paragraph or in the
middle of it. Find where this paragraph starts. */
- bytepos = bidi_find_paragraph_start (pos, bytepos);
-
+ pstartbyte = bidi_find_paragraph_start (pos, bytepos);
bidi_it->separator_limit = -1;
bidi_it->new_paragraph = 0;
- ch = FETCH_CHAR (bytepos);
- ch_len = CHAR_BYTES (ch);
- pos = BYTE_TO_CHAR (bytepos);
- type = bidi_get_type (ch, NEUTRAL_DIR);
-
- for (pos++, bytepos += ch_len;
- /* NOTE: UAX#9 says to search only for L, AL, or R types of
- characters, and ignore RLE, RLO, LRE, and LRO. However,
- I'm not sure it makes sense to omit those 4; should try
- with and without that to see the effect. */
- (bidi_get_category (type) != STRONG)
- || (bidi_ignore_explicit_marks_for_paragraph_level
- && (type == RLE || type == RLO
- || type == LRE || type == LRO));
- type = bidi_get_type (ch, NEUTRAL_DIR))
- {
- if (type == NEUTRAL_B && bidi_at_paragraph_end (pos, bytepos) >= -1)
- break;
- if (bytepos >= ZV_BYTE)
- {
- /* Pretend there's a paragraph separator at end of buffer. */
- type = NEUTRAL_B;
+
+ /* The following loop is run more than once only if NO_DEFAULT_P
+ is non-zero. */
+ do {
+ bytepos = pstartbyte;
+ ch = FETCH_CHAR (bytepos);
+ ch_len = CHAR_BYTES (ch);
+ pos = BYTE_TO_CHAR (bytepos);
+ type = bidi_get_type (ch, NEUTRAL_DIR);
+
+ for (pos++, bytepos += ch_len;
+ /* NOTE: UAX#9 says to search only for L, AL, or R types
+ of characters, and ignore RLE, RLO, LRE, and LRO.
+ However, I'm not sure it makes sense to omit those 4;
+ should try with and without that to see the effect. */
+ (bidi_get_category (type) != STRONG)
+ || (bidi_ignore_explicit_marks_for_paragraph_level
+ && (type == RLE || type == RLO
+ || type == LRE || type == LRO));
+ type = bidi_get_type (ch, NEUTRAL_DIR))
+ {
+ if (type == NEUTRAL_B && bidi_at_paragraph_end (pos, bytepos) >= -1)
break;
- }
- FETCH_CHAR_ADVANCE (ch, pos, bytepos);
- }
- if (type == STRONG_R || type == STRONG_AL) /* P3 */
- bidi_it->paragraph_dir = R2L;
- else if (type == STRONG_L)
- bidi_it->paragraph_dir = L2R;
+ if (bytepos >= ZV_BYTE)
+ {
+ /* Pretend there's a paragraph separator at end of
+ buffer. */
+ type = NEUTRAL_B;
+ break;
+ }
+ FETCH_CHAR_ADVANCE (ch, pos, bytepos);
+ }
+ if (type == STRONG_R || type == STRONG_AL) /* P3 */
+ bidi_it->paragraph_dir = R2L;
+ else if (type == STRONG_L)
+ bidi_it->paragraph_dir = L2R;
+ if (no_default_p && bidi_it->paragraph_dir == NEUTRAL_DIR)
+ {
+ /* If this paragraph is at BEGV, default to L2R. */
+ if (pstartbyte == BEGV_BYTE)
+ bidi_it->paragraph_dir = L2R; /* P3 and HL1 */
+ else
+ {
+ EMACS_INT prevpbyte = pstartbyte;
+ EMACS_INT p = BYTE_TO_CHAR (pstartbyte), pbyte = pstartbyte;
+
+ /* Find the beginning of the previous paragraph, if any. */
+ while (pbyte > BEGV_BYTE && prevpbyte >= pstartbyte)
+ {
+ p--;
+ pbyte = CHAR_TO_BYTE (p);
+ prevpbyte = bidi_find_paragraph_start (p, pbyte);
+ }
+ pstartbyte = prevpbyte;
+ }
+ }
+ } while (no_default_p && bidi_it->paragraph_dir == NEUTRAL_DIR);
}
else
abort ();
/* Contrary to UAX#9 clause P3, we only default the paragraph
direction to L2R if we have no previous usable paragraph
- direction. */
+ direction. This is allowed by the HL1 clause. */
if (bidi_it->paragraph_dir != L2R && bidi_it->paragraph_dir != R2L)
- bidi_it->paragraph_dir = L2R; /* P3 and ``higher protocols'' */
+ bidi_it->paragraph_dir = L2R; /* P3 and HL1 ``higher-level protocols'' */
if (bidi_it->paragraph_dir == R2L)
bidi_it->level_stack[0].level = 1;
else
bidi_check_type (bidi_it->type);
/* For L1 below, we need to know, for each WS character, whether
- it belongs to a sequence of WS characters preceeding a newline
+ it belongs to a sequence of WS characters preceding a newline
or a TAB or a paragraph separator. */
if (bidi_it->orig_type == NEUTRAL_WS
&& bidi_it->next_for_ws.type == UNKNOWN_BT)
we want it to be displayed as
- {RLO}STet{PDF}
+ {PDF}STet{RLO}
not as
/* This is meant to be called from within the debugger, whenever you
wish to examine the cache contents. */
+void bidi_dump_cached_states (void) EXTERNALLY_VISIBLE;
void
bidi_dump_cached_states (void)
{
fputs ("\n", stderr);
fputs ("pos ", stderr);
for (i = 0; i < bidi_cache_idx; i++)
- fprintf (stderr, "%*ld", ndigits, (long)bidi_cache[i].charpos);
+ fprintf (stderr, "%*"pI"d", ndigits, bidi_cache[i].charpos);
fputs ("\n", stderr);
}
HCoop / bpt / emacs.git / blobdiff