38 |
*/ |
*/ |
39 |
|
|
40 |
|
|
41 |
/* This module contains internal functions for testing newlines when more than |
/* This module contains internal functions for testing newlines when more than |
42 |
one kind of newline is to be recognized. When a newline is found, its length is |
one kind of newline is to be recognized. When a newline is found, its length is |
43 |
returned. In principle, we could implement several newline "types", each |
returned. In principle, we could implement several newline "types", each |
44 |
referring to a different set of newline characters. At present, PCRE supports |
referring to a different set of newline characters. At present, PCRE supports |
45 |
only NLTYPE_FIXED, which gets handled without these functions, and NLTYPE_ALL, |
only NLTYPE_FIXED, which gets handled without these functions, NLTYPE_ANYCRLF, |
46 |
so for now the type isn't passed into the functions. It can easily be added |
and NLTYPE_ANY. The full list of Unicode newline characters is taken from |
|
later if required. The full list of Unicode newline characters is taken from |
|
47 |
http://unicode.org/unicode/reports/tr18/. */ |
http://unicode.org/unicode/reports/tr18/. */ |
48 |
|
|
49 |
|
|
60 |
|
|
61 |
Arguments: |
Arguments: |
62 |
ptr pointer to possible newline |
ptr pointer to possible newline |
63 |
|
type the newline type |
64 |
endptr pointer to the end of the string |
endptr pointer to the end of the string |
65 |
lenptr where to return the length |
lenptr where to return the length |
66 |
utf8 TRUE if in utf8 mode |
utf8 TRUE if in utf8 mode |
69 |
*/ |
*/ |
70 |
|
|
71 |
BOOL |
BOOL |
72 |
_pcre_is_newline(const uschar *ptr, const uschar *endptr, int *lenptr, |
_pcre_is_newline(const uschar *ptr, int type, const uschar *endptr, |
73 |
BOOL utf8) |
int *lenptr, BOOL utf8) |
74 |
{ |
{ |
75 |
int c; |
int c; |
76 |
if (utf8) { GETCHAR(c, ptr); } else c = *ptr; |
if (utf8) { GETCHAR(c, ptr); } else c = *ptr; |
77 |
switch(c) |
|
78 |
|
if (type == NLTYPE_ANYCRLF) switch(c) |
79 |
|
{ |
80 |
|
case 0x000a: *lenptr = 1; return TRUE; /* LF */ |
81 |
|
case 0x000d: *lenptr = (ptr < endptr - 1 && ptr[1] == 0x0a)? 2 : 1; |
82 |
|
return TRUE; /* CR */ |
83 |
|
default: return FALSE; |
84 |
|
} |
85 |
|
|
86 |
|
/* NLTYPE_ANY */ |
87 |
|
|
88 |
|
else switch(c) |
89 |
{ |
{ |
90 |
case 0x000a: /* LF */ |
case 0x000a: /* LF */ |
91 |
case 0x000b: /* VT */ |
case 0x000b: /* VT */ |
110 |
|
|
111 |
Arguments: |
Arguments: |
112 |
ptr pointer to possible newline |
ptr pointer to possible newline |
113 |
|
type the newline type |
114 |
startptr pointer to the start of the string |
startptr pointer to the start of the string |
115 |
lenptr where to return the length |
lenptr where to return the length |
116 |
utf8 TRUE if in utf8 mode |
utf8 TRUE if in utf8 mode |
119 |
*/ |
*/ |
120 |
|
|
121 |
BOOL |
BOOL |
122 |
_pcre_was_newline(const uschar *ptr, const uschar *startptr, int *lenptr, |
_pcre_was_newline(const uschar *ptr, int type, const uschar *startptr, |
123 |
BOOL utf8) |
int *lenptr, BOOL utf8) |
124 |
{ |
{ |
125 |
int c; |
int c; |
126 |
ptr--; |
ptr--; |
130 |
GETCHAR(c, ptr); |
GETCHAR(c, ptr); |
131 |
} |
} |
132 |
else c = *ptr; |
else c = *ptr; |
133 |
switch(c) |
|
134 |
|
if (type == NLTYPE_ANYCRLF) switch(c) |
135 |
|
{ |
136 |
|
case 0x000a: *lenptr = (ptr > startptr && ptr[-1] == 0x0d)? 2 : 1; |
137 |
|
return TRUE; /* LF */ |
138 |
|
case 0x000d: *lenptr = 1; return TRUE; /* CR */ |
139 |
|
default: return FALSE; |
140 |
|
} |
141 |
|
|
142 |
|
else switch(c) |
143 |
{ |
{ |
144 |
case 0x000a: *lenptr = (ptr > startptr && ptr[-1] == 0x0d)? 2 : 1; |
case 0x000a: *lenptr = (ptr > startptr && ptr[-1] == 0x0d)? 2 : 1; |
145 |
return TRUE; /* LF */ |
return TRUE; /* LF */ |