And this patch makes sure that we apply the same logic also when a
line break was seen in the input string.
2024-09-15 Bruno Haible <br...@clisp.org>
unilbrk: Some more tweaks.
* lib/unilbrk/u8-possible-linebreaks.c (u8_possible_linebreaks_loop):
Assign to preceding_prop at each loop round. Assign to prev2_ea,
prev_ea, prev_nus also when a line break was seen.
* lib/unilbrk/u16-possible-linebreaks.c (u16_possible_linebreaks_loop):
Likewise.
* lib/unilbrk/u32-possible-linebreaks.c (u32_possible_linebreaks_loop):
Likewise.
diff --git a/lib/unilbrk/u16-possible-linebreaks.c
b/lib/unilbrk/u16-possible-linebreaks.c
index 9678041606..ac655f8c82 100644
--- a/lib/unilbrk/u16-possible-linebreaks.c
+++ b/lib/unilbrk/u16-possible-linebreaks.c
@@ -122,7 +122,7 @@ u16_possible_linebreaks_loop (const uint16_t *s, size_t n,
const char *encoding,
lookahead2_prop_ea = PROP_EA (LBP_BK, 0);
}
- bool nus = /* ending at the last character, there was a character
+ bool nus = /* ending at the previous character, there was a character
with line break property LBP_NU and since then only
characters with line break property LBP_SY or LBP_IS */
(prev_prop == LBP_NU
@@ -136,10 +136,8 @@ u16_possible_linebreaks_loop (const uint16_t *s, size_t n,
const char *encoding,
a CR-LF sequence. */
if (prev_prop == cr && prop == LBP_LF)
p[-1] = UC_BREAK_CR_BEFORE_LF;
- prev_prop = prop;
last_prop = LBP_BK;
seen_space = NULL;
- prev_initial_hyphen = false;
}
else
{
@@ -385,33 +383,33 @@ u16_possible_linebreaks_loop (const uint16_t *s, size_t
n, const char *encoding,
last_prop = prop;
seen_space = NULL;
}
+ }
- /* (LB9) Treat X (CM | ZWJ)* as if it were X, where X is any line
- break class except BK, CR, LF, NL, SP, or ZW. */
- if (!((prop == LBP_CM || prop == LBP_ZWJ)
- && !(prev_prop == LBP_BK || prev_prop == LBP_LF ||
prev_prop == LBP_CR
- || prev_prop == LBP_SP || prev_prop == LBP_ZW)))
- {
- prev_initial_hyphen =
- (prop == LBP_HY || uc == 0x2010)
- && (prev_prop == LBP_BK || prev_prop == LBP_CR ||
prev_prop == LBP_LF
- || prev_prop == LBP_SP || prev_prop == LBP_ZW
- || prev_prop == LBP_CB || prev_prop == LBP_GL);
- prev_prop = (prop == LBP_VI && (prev_prop == LBP_AK
- || prev_prop == LBP_AL2
- || prev_prop == LBP_AS)
- ? LBP_AKLS_VI :
- prev_prop == LBP_HL && (prop == LBP_HY
- || (prop == LBP_BA &&
!ea))
- ? LBP_HL_BA :
- prop);
- prev2_ea = prev_ea;
- prev_ea = ea;
- }
- preceding_prop = prop;
+ /* (LB9) Treat X (CM | ZWJ)* as if it were X, where X is any line
+ break class except BK, CR, LF, NL, SP, or ZW. */
+ if (!((prop == LBP_CM || prop == LBP_ZWJ)
+ && !(prev_prop == LBP_BK || prev_prop == LBP_LF || prev_prop
== LBP_CR
+ || prev_prop == LBP_SP || prev_prop == LBP_ZW)))
+ {
+ prev_initial_hyphen =
+ (prop == LBP_HY || uc == 0x2010)
+ && (prev_prop == LBP_BK || prev_prop == LBP_CR || prev_prop ==
LBP_LF
+ || prev_prop == LBP_SP || prev_prop == LBP_ZW
+ || prev_prop == LBP_CB || prev_prop == LBP_GL);
+ prev_prop = (prop == LBP_VI && (prev_prop == LBP_AK
+ || prev_prop == LBP_AL2
+ || prev_prop == LBP_AS)
+ ? LBP_AKLS_VI :
+ prev_prop == LBP_HL && (prop == LBP_HY
+ || (prop == LBP_BA && !ea))
+ ? LBP_HL_BA :
+ prop);
+ prev2_ea = prev_ea;
+ prev_ea = ea;
+ prev_nus = nus;
}
- prev_nus = nus;
+ preceding_prop = prop;
if (prop == LBP_RI)
ri_count++;
diff --git a/lib/unilbrk/u32-possible-linebreaks.c
b/lib/unilbrk/u32-possible-linebreaks.c
index c12ca3abbb..f242c9a8ee 100644
--- a/lib/unilbrk/u32-possible-linebreaks.c
+++ b/lib/unilbrk/u32-possible-linebreaks.c
@@ -115,7 +115,7 @@ u32_possible_linebreaks_loop (const uint32_t *s, size_t n,
const char *encoding,
lookahead2_prop_ea = PROP_EA (LBP_BK, 0);
}
- bool nus = /* ending at the last character, there was a character
+ bool nus = /* ending at the previous character, there was a character
with line break property LBP_NU and since then only
characters with line break property LBP_SY or LBP_IS */
(prev_prop == LBP_NU
@@ -129,10 +129,8 @@ u32_possible_linebreaks_loop (const uint32_t *s, size_t n,
const char *encoding,
a CR-LF sequence. */
if (prev_prop == cr && prop == LBP_LF)
p[-1] = UC_BREAK_CR_BEFORE_LF;
- prev_prop = prop;
last_prop = LBP_BK;
seen_space = NULL;
- prev_initial_hyphen = false;
}
else
{
@@ -378,33 +376,33 @@ u32_possible_linebreaks_loop (const uint32_t *s, size_t
n, const char *encoding,
last_prop = prop;
seen_space = NULL;
}
+ }
- /* (LB9) Treat X (CM | ZWJ)* as if it were X, where X is any line
- break class except BK, CR, LF, NL, SP, or ZW. */
- if (!((prop == LBP_CM || prop == LBP_ZWJ)
- && !(prev_prop == LBP_BK || prev_prop == LBP_LF ||
prev_prop == LBP_CR
- || prev_prop == LBP_SP || prev_prop == LBP_ZW)))
- {
- prev_initial_hyphen =
- (prop == LBP_HY || uc == 0x2010)
- && (prev_prop == LBP_BK || prev_prop == LBP_CR ||
prev_prop == LBP_LF
- || prev_prop == LBP_SP || prev_prop == LBP_ZW
- || prev_prop == LBP_CB || prev_prop == LBP_GL);
- prev_prop = (prop == LBP_VI && (prev_prop == LBP_AK
- || prev_prop == LBP_AL2
- || prev_prop == LBP_AS)
- ? LBP_AKLS_VI :
- prev_prop == LBP_HL && (prop == LBP_HY
- || (prop == LBP_BA &&
!ea))
- ? LBP_HL_BA :
- prop);
- prev2_ea = prev_ea;
- prev_ea = ea;
- }
- preceding_prop = prop;
+ /* (LB9) Treat X (CM | ZWJ)* as if it were X, where X is any line
+ break class except BK, CR, LF, NL, SP, or ZW. */
+ if (!((prop == LBP_CM || prop == LBP_ZWJ)
+ && !(prev_prop == LBP_BK || prev_prop == LBP_LF || prev_prop
== LBP_CR
+ || prev_prop == LBP_SP || prev_prop == LBP_ZW)))
+ {
+ prev_initial_hyphen =
+ (prop == LBP_HY || uc == 0x2010)
+ && (prev_prop == LBP_BK || prev_prop == LBP_CR || prev_prop ==
LBP_LF
+ || prev_prop == LBP_SP || prev_prop == LBP_ZW
+ || prev_prop == LBP_CB || prev_prop == LBP_GL);
+ prev_prop = (prop == LBP_VI && (prev_prop == LBP_AK
+ || prev_prop == LBP_AL2
+ || prev_prop == LBP_AS)
+ ? LBP_AKLS_VI :
+ prev_prop == LBP_HL && (prop == LBP_HY
+ || (prop == LBP_BA && !ea))
+ ? LBP_HL_BA :
+ prop);
+ prev2_ea = prev_ea;
+ prev_ea = ea;
+ prev_nus = nus;
}
- prev_nus = nus;
+ preceding_prop = prop;
if (prop == LBP_RI)
ri_count++;
diff --git a/lib/unilbrk/u8-possible-linebreaks.c
b/lib/unilbrk/u8-possible-linebreaks.c
index 3f32ee6289..90d64c51b8 100644
--- a/lib/unilbrk/u8-possible-linebreaks.c
+++ b/lib/unilbrk/u8-possible-linebreaks.c
@@ -122,7 +122,7 @@ u8_possible_linebreaks_loop (const uint8_t *s, size_t n,
const char *encoding,
lookahead2_prop_ea = PROP_EA (LBP_BK, 0);
}
- bool nus = /* ending at the last character, there was a character
+ bool nus = /* ending at the previous character, there was a character
with line break property LBP_NU and since then only
characters with line break property LBP_SY or LBP_IS */
(prev_prop == LBP_NU
@@ -136,10 +136,8 @@ u8_possible_linebreaks_loop (const uint8_t *s, size_t n,
const char *encoding,
a CR-LF sequence. */
if (prev_prop == cr && prop == LBP_LF)
p[-1] = UC_BREAK_CR_BEFORE_LF;
- prev_prop = prop;
last_prop = LBP_BK;
seen_space = NULL;
- prev_initial_hyphen = false;
}
else
{
@@ -385,33 +383,33 @@ u8_possible_linebreaks_loop (const uint8_t *s, size_t n,
const char *encoding,
last_prop = prop;
seen_space = NULL;
}
+ }
- /* (LB9) Treat X (CM | ZWJ)* as if it were X, where X is any line
- break class except BK, CR, LF, NL, SP, or ZW. */
- if (!((prop == LBP_CM || prop == LBP_ZWJ)
- && !(prev_prop == LBP_BK || prev_prop == LBP_LF ||
prev_prop == LBP_CR
- || prev_prop == LBP_SP || prev_prop == LBP_ZW)))
- {
- prev_initial_hyphen =
- (prop == LBP_HY || uc == 0x2010)
- && (prev_prop == LBP_BK || prev_prop == LBP_CR ||
prev_prop == LBP_LF
- || prev_prop == LBP_SP || prev_prop == LBP_ZW
- || prev_prop == LBP_CB || prev_prop == LBP_GL);
- prev_prop = (prop == LBP_VI && (prev_prop == LBP_AK
- || prev_prop == LBP_AL2
- || prev_prop == LBP_AS)
- ? LBP_AKLS_VI :
- prev_prop == LBP_HL && (prop == LBP_HY
- || (prop == LBP_BA &&
!ea))
- ? LBP_HL_BA :
- prop);
- prev2_ea = prev_ea;
- prev_ea = ea;
- }
- preceding_prop = prop;
+ /* (LB9) Treat X (CM | ZWJ)* as if it were X, where X is any line
+ break class except BK, CR, LF, NL, SP, or ZW. */
+ if (!((prop == LBP_CM || prop == LBP_ZWJ)
+ && !(prev_prop == LBP_BK || prev_prop == LBP_LF || prev_prop
== LBP_CR
+ || prev_prop == LBP_SP || prev_prop == LBP_ZW)))
+ {
+ prev_initial_hyphen =
+ (prop == LBP_HY || uc == 0x2010)
+ && (prev_prop == LBP_BK || prev_prop == LBP_CR || prev_prop ==
LBP_LF
+ || prev_prop == LBP_SP || prev_prop == LBP_ZW
+ || prev_prop == LBP_CB || prev_prop == LBP_GL);
+ prev_prop = (prop == LBP_VI && (prev_prop == LBP_AK
+ || prev_prop == LBP_AL2
+ || prev_prop == LBP_AS)
+ ? LBP_AKLS_VI :
+ prev_prop == LBP_HL && (prop == LBP_HY
+ || (prop == LBP_BA && !ea))
+ ? LBP_HL_BA :
+ prop);
+ prev2_ea = prev_ea;
+ prev_ea = ea;
+ prev_nus = nus;
}
- prev_nus = nus;
+ preceding_prop = prop;
if (prop == LBP_RI)
ri_count++;
