Skip to content

C
cpython
Commit 5ededdd6 authored by Andrew M. Kuchling's avatar Andrew M. Kuchling
Browse files
Options

Remove pcre module

parent f7e836e6
Hide whitespace changes
Inline Side-by-side
Showing with 0 additions and 5771 deletions
Modules/pcre-int.h deleted 100644 → 0
View file @ f7e836e6
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
#define PCRE_VERSION "1.09 28-Apr-1998"
/* This is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language. See
the file Tech.Notes for some information on the internals.
Written by: Philip Hazel <ph10@cam.ac.uk>
Copyright (c) 1998 University of Cambridge
-----------------------------------------------------------------------------
Permission is granted to anyone to use this software for any purpose on any
computer system, and to redistribute it freely, subject to the following
restrictions:
1. This software 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.
2. The origin of this software must not be misrepresented, either by
explicit claim or by omission.
3. Altered versions must be plainly marked as such, and must not be
misrepresented as being the original software.
-----------------------------------------------------------------------------
*/
/* This header contains definitions that are shared between the different
modules, but which are not relevant to the outside. */
/* Standard C headers plus the external interface definition */
#include <ctype.h>
#include <limits.h>
#include <setjmp.h>
#include <stddef.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "pcre.h"
/* In case there is no definition of offsetof() provided - though any proper
Standard C system should have one. */
#ifndef offsetof
#define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
#endif
/* Private options flags start at the most significant end of the two bytes.
The public options defined in pcre.h start at the least significant end. Make
sure they don't overlap! */
#define PCRE_FIRSTSET 0x8000 /* first_char is set */
#define PCRE_STARTLINE 0x4000 /* start after \n for multiline */
#define PCRE_COMPILED_CASELESS 0x2000 /* like it says */
/* Options for the "extra" block produced by pcre_study(). */
#define PCRE_STUDY_CASELESS 0x01 /* study was caseless */
#define PCRE_STUDY_MAPPED 0x02 /* a map of starting chars exists */
/* Masks for identifying the public options: all permitted at compile time,
only some permitted at run or study time. */
#ifdef FOR_PYTHON
#define PUBLIC_OPTIONS \
(PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY| \
PCRE_LOCALE)
#else
#define PUBLIC_OPTIONS \
(PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY)
#endif
#define PUBLIC_EXEC_OPTIONS \
(PCRE_CASELESS|PCRE_ANCHORED|PCRE_MULTILINE|PCRE_NOTBOL|PCRE_NOTEOL| \
PCRE_DOTALL|PCRE_DOLLAR_ENDONLY)
#define PUBLIC_STUDY_OPTIONS (PCRE_CASELESS)
/* Magic number to provide a small check against being handed junk. */
#define MAGIC_NUMBER 0x50435245 /* 'PCRE' */
/* Miscellaneous definitions */
typedef int BOOL;
#define FALSE 0
#define TRUE 1
/* These are escaped items that aren't just an encoding of a particular data
value such as \n. They must have non-zero values, as check_escape() returns
their negation. Also, they must appear in the same order as in the opcode
definitions below, up to ESC_Z. The final one must be ESC_REF as subsequent
values are used for \1, \2, \3, etc. There is a test in the code for an escape
greater than ESC_b and less than ESC_X to detect the types that may be
repeated. If any new escapes are put in-between that don't consume a character,
that code will have to change. */
enum { ESC_A = 1, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, ESC_W, ESC_w,
/* These are not Perl escapes, so can't appear in the */
ESC_X, /* simple table-lookup because they must be conditional */
/* on PCRE_EXTRA. */
ESC_Z,
ESC_REF };
/* Opcode table: OP_BRA must be last, as all values >= it are used for brackets
that extract substrings. Starting from 1 (i.e. after OP_END), the values up to
OP_EOD must correspond in order to the list of escapes immediately above. */
enum {
OP_END, /* End of pattern */
/* Values corresponding to backslashed metacharacters */
OP_SOD, /* Start of data: \A */
OP_NOT_WORD_BOUNDARY, /* \B */
OP_WORD_BOUNDARY, /* \b */
OP_NOT_DIGIT, /* \D */
OP_DIGIT, /* \d */
OP_NOT_WHITESPACE, /* \S */
OP_WHITESPACE, /* \s */
OP_NOT_WORDCHAR, /* \W */
OP_WORDCHAR, /* \w */
OP_CUT, /* The analogue of Prolog's "cut" operation (extension) */
OP_EOD, /* End of data: \Z. */
OP_NOT_WORD_BOUNDARY_L, /* localized \B */
OP_WORD_BOUNDARY_L, /* localized \b */
OP_NOT_WORDCHAR_L, /* localized \W */
OP_WORDCHAR_L, /* localized \w */
OP_CIRC, /* Start of line - varies with multiline switch */
OP_DOLL, /* End of line - varies with multiline switch */
OP_ANY, /* Match any character */
OP_CHARS, /* Match string of characters */
OP_NOT, /* Match anything but the following char */
OP_STAR, /* The maximizing and minimizing versions of */
OP_MINSTAR, /* all these opcodes must come in pairs, with */
OP_PLUS, /* the minimizing one second. */
OP_MINPLUS, /* This first set applies to single characters */
OP_QUERY,
OP_MINQUERY,
OP_UPTO, /* From 0 to n matches */
OP_MINUPTO,
OP_EXACT, /* Exactly n matches */
OP_NOTSTAR, /* The maximizing and minimizing versions of */
OP_NOTMINSTAR, /* all these opcodes must come in pairs, with */
OP_NOTPLUS, /* the minimizing one second. */
OP_NOTMINPLUS, /* This first set applies to "not" single characters */
OP_NOTQUERY,
OP_NOTMINQUERY,
OP_NOTUPTO, /* From 0 to n matches */
OP_NOTMINUPTO,
OP_NOTEXACT, /* Exactly n matches */
OP_TYPESTAR, /* The maximizing and minimizing versions of */
OP_TYPEMINSTAR, /* all these opcodes must come in pairs, with */
OP_TYPEPLUS, /* the minimizing one second. These codes must */
OP_TYPEMINPLUS, /* be in exactly the same order as those above. */
OP_TYPEQUERY, /* This set applies to character types such as \d */
OP_TYPEMINQUERY,
OP_TYPEUPTO, /* From 0 to n matches */
OP_TYPEMINUPTO,
OP_TYPEEXACT, /* Exactly n matches */
OP_CRSTAR, /* The maximizing and minimizing versions of */
OP_CRMINSTAR, /* all these opcodes must come in pairs, with */
OP_CRPLUS, /* the minimizing one second. These codes must */
OP_CRMINPLUS, /* be in exactly the same order as those above. */
OP_CRQUERY, /* These are for character classes and back refs */
OP_CRMINQUERY,
OP_CRRANGE, /* These are different to the three seta above. */
OP_CRMINRANGE,
OP_CLASS, /* Match a character class */
OP_NEGCLASS, /* Match a character class, specified negatively */
OP_CLASS_L, /* Match a character class */
OP_REF, /* Match a back reference */
OP_ALT, /* Start of alternation */
OP_KET, /* End of group that doesn't have an unbounded repeat */
OP_KETRMAX, /* These two must remain together and in this */
OP_KETRMIN, /* order. They are for groups the repeat for ever. */
OP_ASSERT,
OP_ASSERT_NOT,
OP_ONCE, /* Once matched, don't back up into the subpattern */
OP_BRAZERO, /* These two must remain together and in this */
OP_BRAMINZERO, /* order. */
OP_BRA /* This and greater values are used for brackets that
extract substrings. */
};
/* The highest extraction number. This is limited by the number of opcodes
left after OP_BRA, i.e. 255 - OP_BRA. We actually set it somewhat lower. */
#define EXTRACT_MAX 99
/* The texts of compile-time error messages are defined as macros here so that
they can be accessed by the POSIX wrapper and converted into error codes. Yes,
I could have used error codes in the first place, but didn't feel like changing
just to accommodate the POSIX wrapper. */
#define ERR1 "\\ at end of pattern"
#define ERR2 "\\c at end of pattern"
#define ERR3 "unrecognized character follows \\"
#define ERR4 "numbers out of order in {} quantifier"
#define ERR5 "number too big in {} quantifier"
#define ERR6 "missing terminating ] for character class"
#define ERR7 "invalid escape sequence in character class"
#define ERR8 "range out of order in character class"
#define ERR9 "nothing to repeat"
#define ERR10 "operand of unlimited repeat could match the empty string"
#define ERR11 "internal error: unexpected repeat"
#define ERR12 "unrecognized character after (?"
#define ERR13 "too many capturing parenthesized sub-patterns"
#define ERR14 "missing )"
#define ERR15 "back reference to non-existent subpattern"
#define ERR16 "erroffset passed as NULL"
#define ERR17 "unknown option bit(s) set"
#define ERR18 "missing ) after comment"
#define ERR19 "too many sets of parentheses"
#define ERR20 "regular expression too large"
#define ERR21 "failed to get memory"
#define ERR22 "unmatched brackets"
#define ERR23 "internal error: code overflow"
/* All character handling must be done as unsigned characters. Otherwise there
are problems with top-bit-set characters and functions such as isspace().
However, we leave the interface to the outside world as char *, because that
should make things easier for callers. We define a short type for unsigned char
to save lots of typing. I tried "uchar", but it causes problems on Digital
Unix, where it is defined in sys/types, so use "uschar" instead. */
typedef unsigned char uschar;
/* The real format of the start of the pcre block; the actual code vector
runs on as long as necessary after the end. */
typedef struct real_pcre {
unsigned int magic_number;
unsigned short int options;
unsigned char top_bracket;
unsigned char top_backref;
unsigned char first_char;
unsigned char code[1];
} real_pcre;
/* The real format of the extra block returned by pcre_study(). */
typedef struct real_pcre_extra {
unsigned char options;
unsigned char start_bits[32];
} real_pcre_extra;
/* Global tables from chartables.c */
extern uschar pcre_lcc[];
extern uschar pcre_fcc[];
extern uschar pcre_cbits[];
extern uschar pcre_ctypes[];
/* Bit definitions for entries in pcre_ctypes[]. */
#define ctype_space 0x01
#define ctype_letter 0x02
#define ctype_digit 0x04
#define ctype_xdigit 0x08
#define ctype_word 0x10 /* alphameric or '_' */
#define ctype_odigit 0x20 /* octal digit */
#define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
/* Offsets for the bitmap tables */
#define cbit_digit 0
#define cbit_letter 32
#define cbit_word 64
#define cbit_space 96
#define cbit_length 128 /* Length of the cbits table */
/* End of internal.h */
Modules/pcre.h deleted 100644 → 0
View file @ f7e836e6
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
/* Copyright (c) 1998 University of Cambridge */
#ifndef _PCRE_H
#define _PCRE_H
#ifdef FOR_PYTHON
#include "Python.h"
#endif
/* Have to include stdlib.h in order to ensure that size_t is defined;
it is needed here for malloc. */
#ifndef DONT_HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif
#include <stdlib.h>
/* Allow for C++ users */
#ifdef __cplusplus
extern "C" {
#endif
/* Options */
#define PCRE_CASELESS 0x0001
#define PCRE_EXTENDED 0x0002
#define PCRE_ANCHORED 0x0004
#define PCRE_MULTILINE 0x0008
#define PCRE_DOTALL 0x0010
#define PCRE_DOLLAR_ENDONLY 0x0020
#define PCRE_EXTRA 0x0040
#define PCRE_NOTBOL 0x0080
#define PCRE_NOTEOL 0x0100
#define PCRE_UNGREEDY 0x0400
#ifdef FOR_PYTHON
#define PCRE_LOCALE 0x0200
#endif
/* Exec-time error codes */
#define PCRE_ERROR_NOMATCH (-1)
#define PCRE_ERROR_BADREF (-2)
#define PCRE_ERROR_NULL (-3)
#define PCRE_ERROR_BADOPTION (-4)
#define PCRE_ERROR_BADMAGIC (-5)
#define PCRE_ERROR_UNKNOWN_NODE (-6)
#define PCRE_ERROR_NOMEMORY (-7)
/* Types */
typedef void pcre;
typedef void pcre_extra;
/* Store get and free functions. These can be set to alternative malloc/free
functions if required. */
extern void *(*pcre_malloc)(size_t);
extern void (*pcre_free)(void *);
/* Functions */
#ifdef FOR_PYTHON
extern pcre *pcre_compile(const char *, int, const char **, int *, PyObject *);
extern int pcre_exec(const pcre *, const pcre_extra *, const char *,
int, int, int, int *, int);
#else
extern pcre *pcre_compile(const char *, int, const char **, int *);
extern int pcre_exec(const pcre *, const pcre_extra *, const char *,
int, int, int *, int);
#endif
extern int pcre_info(const pcre *, int *, int *);
extern pcre_extra *pcre_study(const pcre *, int, const char **);
extern const char *pcre_version(void);
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* End of pcre.h */
Modules/pcremodule.c deleted 100644 → 0
View file @ f7e836e6
/* Pcre objects */
#include "Python.h"
#ifndef Py_eval_input
/* For Python 1.4, graminit.h has to be explicitly included */
#include "graminit.h"
#define Py_eval_input eval_input
#endif
#ifndef FOR_PYTHON
#define FOR_PYTHON
#endif
#include "pcre.h"
#include "pcre-int.h"
static PyObject *ErrorObject;
typedef struct {
PyObject_HEAD
pcre *regex;
pcre_extra *regex_extra;
int num_groups;
} PcreObject;
static PyTypeObject Pcre_Type;
#define PcreObject_Check(v) ((v)->ob_type == &Pcre_Type)
#define NORMAL 0
#define CHARCLASS 1
#define REPLACEMENT 2
#define CHAR 0
#define MEMORY_REFERENCE 1
#define SYNTAX 2
#define NOT_SYNTAX 3
#define SET 4
#define WORD_BOUNDARY 5
#define NOT_WORD_BOUNDARY 6
#define BEGINNING_OF_BUFFER 7
#define END_OF_BUFFER 8
#define STRING 9
static PcreObject *
newPcreObject(PyObject *args)
{
PcreObject *self;
self = PyObject_New(PcreObject, &Pcre_Type);
if (self == NULL)
return NULL;
self->regex = NULL;
self->regex_extra = NULL;
return self;
}
/* Pcre methods */
static void
PyPcre_dealloc(PcreObject *self)
{
if (self->regex) (pcre_free)(self->regex);
if (self->regex_extra) (pcre_free)(self->regex_extra);
PyObject_Del(self);
}
static PyObject *
PyPcre_exec(PcreObject *self, PyObject *args)
{
char *string;
int stringlen, pos = 0, options=0, endpos = -1, i, count;
int offsets[100*2];
PyObject *list;
if (!PyArg_ParseTuple(args, "t#|iii:match", &string, &stringlen,
&pos, &endpos, &options))
return NULL;
if (endpos == -1) {endpos = stringlen;}
count = pcre_exec(self->regex, self->regex_extra,
string, endpos, pos, options,
offsets, sizeof(offsets)/sizeof(int) );
/* If an error occurred during the match, and an exception was raised,
just return NULL and leave the exception alone. The most likely
problem to cause this would be running out of memory for
the failure stack. */
if (PyErr_Occurred())
{
return NULL;
}
if (count==PCRE_ERROR_NOMATCH) {Py_INCREF(Py_None); return Py_None;}
if (count<0)
{
PyObject *errval = Py_BuildValue("si", "Regex execution error", count);
PyErr_SetObject(ErrorObject, errval);
Py_XDECREF(errval);
return NULL;
}
list=PyList_New(self->num_groups+1);
if (list==NULL) return NULL;
for(i=0; i<=self->num_groups; i++)
{
PyObject *v;
int start=offsets[i*2], end=offsets[i*2+1];
/* If the group wasn't affected by the match, return -1, -1 */
if (start<0 || count<=i)
{start=end=-1;}
v=Py_BuildValue("ii", start, end);
if (v==NULL) {Py_DECREF(list); return NULL;}
PyList_SetItem(list, i, v);
}
return list;
}
static PyMethodDef Pcre_methods[] = {
{"match", (PyCFunction)PyPcre_exec, METH_VARARGS},
{NULL, NULL} /* sentinel */
};
static PyObject *
PyPcre_getattr(PcreObject *self, char *name)
{
return Py_FindMethod(Pcre_methods, (PyObject *)self, name);
}
static PyTypeObject Pcre_Type = {
PyObject_HEAD_INIT(NULL)
0, /*ob_size*/
"pcre.Pcre", /*tp_name*/
sizeof(PcreObject), /*tp_basicsize*/
0, /*tp_itemsize*/
/* methods */
(destructor)PyPcre_dealloc, /*tp_dealloc*/
0, /*tp_print*/
(getattrfunc)PyPcre_getattr, /*tp_getattr*/
0, /*tp_setattr*/
0, /*tp_compare*/
0, /*tp_repr*/
0, /*tp_as_number*/
0, /*tp_as_sequence*/
0, /*tp_as_mapping*/
0, /*tp_hash*/
};
/* --------------------------------------------------------------------- */
static PyObject *
PyPcre_compile(PyObject *self, PyObject *args)
{
PcreObject *rv;
PyObject *dictionary;
char *pattern;
const char *error;
int options, erroroffset;
if (!PyArg_ParseTuple(args, "siO!:pcre_compile", &pattern, &options,
&PyDict_Type, &dictionary))
return NULL;
rv = newPcreObject(args);
if ( rv == NULL )
return NULL;
rv->regex = pcre_compile((char*)pattern, options,
&error, &erroroffset, dictionary);
if (rv->regex==NULL)
{
Py_DECREF(rv);
if (!PyErr_Occurred())
{
PyObject *errval = Py_BuildValue("si", error, erroroffset);
PyErr_SetObject(ErrorObject, errval);
Py_XDECREF(errval);
}
return NULL;
}
rv->regex_extra=pcre_study(rv->regex, 0, &error);
if (rv->regex_extra==NULL && error!=NULL)
{
PyObject *errval = Py_BuildValue("si", error, 0);
Py_DECREF(rv);
PyErr_SetObject(ErrorObject, errval);
Py_XDECREF(errval);
return NULL;
}
rv->num_groups = pcre_info(rv->regex, NULL, NULL);
if (rv->num_groups<0)
{
PyObject *errval = Py_BuildValue("si", error, rv->num_groups);
PyErr_SetObject(ErrorObject, errval);
Py_XDECREF(errval);
Py_DECREF(rv);
return NULL;
}
return (PyObject *)rv;
}
static PyObject *
PyPcre_expand_escape(unsigned char *pattern, int pattern_len,
int *indexptr, int *typeptr)
{
unsigned char c;
int index = *indexptr;
if (pattern_len<=index)
{
PyErr_SetString(ErrorObject, "escape ends too soon");
return NULL;
}
c=pattern[index]; index++;
*typeptr=CHAR;
switch (c)
{
case('t'):
*indexptr=index;
return Py_BuildValue("c", (char)9);
case('n'):
*indexptr = index;
return Py_BuildValue("c", (char)10);
case('v'):
*indexptr = index;
return Py_BuildValue("c", (char)11);
case('r'):
*indexptr = index;
return Py_BuildValue("c", (char)13);
case('f'):
*indexptr = index;
return Py_BuildValue("c", (char)12);
case('a'):
*indexptr = index;
return Py_BuildValue("c", (char)7);
case('b'):
*indexptr=index;
return Py_BuildValue("c", (char)8);
case('\\'):
*indexptr=index;
return Py_BuildValue("c", '\\');
case('x'):
{
int x, ch, end;
x = 0; end = index;
while ( (end<pattern_len && pcre_ctypes[ pattern[end] ] & ctype_xdigit) != 0)
{
ch = pattern[end];
x = x * 16 + pcre_lcc[ch] -
(((pcre_ctypes[ch] & ctype_digit) != 0)? '0' : 'W');
x &= 255;
end++;
}
if (end==index)
{
PyErr_SetString(ErrorObject, "\\x must be followed by hex digits");
return NULL;
}
*indexptr = end;
return Py_BuildValue("c", (char)x);
}
case('E'): case('G'): case('L'): case('Q'):
case('U'): case('l'): case('u'):
{
char message[50];
PyOS_snprintf(message, sizeof(message),
"\\%c is not allowed", c);
PyErr_SetString(ErrorObject, message);
return NULL;
}
case('g'):
{
int end, i;
int group_num = 0, is_number=0;
if (pattern_len<=index)
{
PyErr_SetString(ErrorObject, "unfinished symbolic reference");
return NULL;
}
if (pattern[index]!='<')
{
PyErr_SetString(ErrorObject, "missing < in symbolic reference");
return NULL;
}
index++;
end=index;
while (end<pattern_len && pattern[end]!='>')
end++;
if (end==pattern_len)
{
PyErr_SetString(ErrorObject, "unfinished symbolic reference");
return NULL;
}
if (index==end) /* Zero-length name */
{
/* XXX should include the text of the reference */
PyErr_SetString(ErrorObject, "zero-length symbolic reference");
return NULL;
}
if ((pcre_ctypes[pattern[index]] & ctype_digit)) /* First char. a digit */
{
is_number = 1;
group_num = pattern[index] - '0';
}
for(i=index+1; i<end; i++)
{
if (is_number &&
!(pcre_ctypes[pattern[i]] & ctype_digit) )
{
/* XXX should include the text of the reference */
PyErr_SetString(ErrorObject, "illegal non-digit character in \\g<...> starting with digit");
return NULL;
}
else {group_num = group_num * 10 + pattern[i] - '0';}
if (!(pcre_ctypes[pattern[i]] & ctype_word) )
{
/* XXX should include the text of the reference */
PyErr_SetString(ErrorObject, "illegal symbolic reference");
return NULL;
}
}
*typeptr = MEMORY_REFERENCE;
*indexptr = end+1;
/* If it's a number, return the integer value of the group */
if (is_number) return Py_BuildValue("i", group_num);
/* Otherwise, return a string containing the group name */
return Py_BuildValue("s#", pattern+index, end-index);
}
case('0'):
{
/* \0 always indicates an octal escape, so we consume up to 3
characters, as long as they're all octal digits */
int octval=0, i;
index--;
for(i=index;
i<=index+2 && i<pattern_len
&& (pcre_ctypes[ pattern[i] ] & ctype_odigit );
i++)
{
octval = octval * 8 + pattern[i] - '0';
}
if (octval>255)
{
PyErr_SetString(ErrorObject, "octal value out of range");
return NULL;
}
*indexptr = i;
return Py_BuildValue("c", (unsigned char)octval);
}
case('1'): case('2'): case('3'): case('4'):
case('5'): case('6'): case('7'): case('8'):
case('9'):
{
/* Handle \?, where ? is from 1 through 9 */
int value=0;
index--;
/* If it's at least a two-digit reference, like \34, it might
either be a 3-digit octal escape (\123) or a 2-digit
decimal memory reference (\34) */
if ( (index+1) <pattern_len &&
(pcre_ctypes[ pattern[index+1] ] & ctype_digit) )
{
if ( (index+2) <pattern_len &&
(pcre_ctypes[ pattern[index+2] ] & ctype_odigit) &&
(pcre_ctypes[ pattern[index+1] ] & ctype_odigit) &&
(pcre_ctypes[ pattern[index ] ] & ctype_odigit)
)
{
/* 3 octal digits */
value= 8*8*(pattern[index ]-'0') +
8*(pattern[index+1]-'0') +
(pattern[index+2]-'0');
if (value>255)
{
PyErr_SetString(ErrorObject, "octal value out of range");
return NULL;
}
*indexptr = index+3;
return Py_BuildValue("c", (unsigned char)value);
}
else
{
/* 2-digit form, so it's a memory reference */
value= 10*(pattern[index ]-'0') +
(pattern[index+1]-'0');
if (value<1 || EXTRACT_MAX<=value)
{
PyErr_SetString(ErrorObject, "memory reference out of range");
return NULL;
}
*typeptr = MEMORY_REFERENCE;
*indexptr = index+2;
return Py_BuildValue("i", value);
}
}
else
{
/* Single-digit form, like \2, so it's a memory reference */
*typeptr = MEMORY_REFERENCE;
*indexptr = index+1;
return Py_BuildValue("i", pattern[index]-'0');
}
}
default:
/* It's some unknown escape like \s, so return a string containing
\s */
*typeptr = STRING;
*indexptr = index;
return Py_BuildValue("s#", pattern+index-2, 2);
}
}
static PyObject *
PyPcre_expand(PyObject *self, PyObject *args)
{
PyObject *results, *match_obj;
PyObject *repl_obj, *newstring;
unsigned char *repl;
int size, total_len, i, start, pos;
if (!PyArg_ParseTuple(args, "OS:pcre_expand", &match_obj, &repl_obj))
return NULL;
repl=(unsigned char *)PyString_AsString(repl_obj);
size=PyString_Size(repl_obj);
results=PyList_New(0);
if (results==NULL) return NULL;
for(start=total_len=i=0; i<size; i++)
{
if (repl[i]=='\\')
{
PyObject *value;
int escape_type;
if (start!=i)
{
int status;
PyObject *s = PyString_FromStringAndSize(
(char *)repl+start, i-start);
if (s == NULL) {
Py_DECREF(results);
return NULL;
}
status = PyList_Append(results, s);
Py_DECREF(s);
if (status < 0) {
Py_DECREF(results);
return NULL;
}
total_len += i-start;
}
i++;
value=PyPcre_expand_escape(repl, size, &i, &escape_type);
if (value==NULL)
{
/* PyPcre_expand_escape triggered an exception of some sort,
so just return */
Py_DECREF(results);
return NULL;
}
switch (escape_type)
{
case (CHAR):
PyList_Append(results, value);
total_len += PyString_Size(value);
break;
case(MEMORY_REFERENCE):
{
PyObject *r, *tuple, *result;
r=PyObject_GetAttrString(match_obj, "group");
if (r == NULL) {
Py_DECREF(results);
return NULL;
}
tuple=PyTuple_New(1);
Py_INCREF(value);
PyTuple_SetItem(tuple, 0, value);
result=PyEval_CallObject(r, tuple);
Py_DECREF(r); Py_DECREF(tuple);
if (result==NULL)
{
/* The group() method triggered an exception of some sort */
Py_DECREF(results);
Py_DECREF(value);
return NULL;
}
if (result==Py_None)
{
char message[50];
PyOS_snprintf(message, sizeof(message),
"group did not contribute to the match");
PyErr_SetString(ErrorObject,
message);
Py_DECREF(result);
Py_DECREF(value);
Py_DECREF(results);
return NULL;
}
/* typecheck that it's a string! */
if (!PyString_Check(result))
{
Py_DECREF(results);
Py_DECREF(result);
PyErr_SetString(ErrorObject,
"group() must return a string value for replacement");
return NULL;
}
PyList_Append(results, result);
total_len += PyString_Size(result);
Py_DECREF(result);
}
break;
case(STRING):
{
PyList_Append(results, value);
total_len += PyString_Size(value);
break;
}
default:
Py_DECREF(results);
PyErr_SetString(ErrorObject,
"bad escape in replacement");
return NULL;
}
Py_DECREF(value);
start=i;
i--; /* Decrement now, because the 'for' loop will increment it */
}
} /* endif repl[i]!='\\' */
if (start!=i)
{
int status;
PyObject *s = PyString_FromStringAndSize((char *)repl+start,
i-start);
if (s == NULL) {
Py_DECREF(results);
return NULL;
}
status = PyList_Append(results, s);
Py_DECREF(s);
if (status < 0) {
Py_DECREF(results);
return NULL;
}
total_len += i-start;
}
/* Whew! Now we've constructed a list containing various pieces of
strings that will make up our final result. So, iterate over
the list concatenating them. A new string measuring total_len
bytes is allocated and filled in. */
newstring=PyString_FromStringAndSize(NULL, total_len);
if (newstring==NULL)
{
Py_DECREF(results);
return NULL;
}
repl=(unsigned char *)PyString_AsString(newstring);
for (pos=i=0; i<PyList_Size(results); i++)
{
PyObject *item=PyList_GetItem(results, i);
memcpy(repl+pos, PyString_AsString(item), PyString_Size(item) );
pos += PyString_Size(item);
}
Py_DECREF(results);
return newstring;
}
/* List of functions defined in the module */
static PyMethodDef pcre_methods[] = {
{"pcre_compile", PyPcre_compile, METH_VARARGS},
{"pcre_expand", PyPcre_expand, METH_VARARGS},
{NULL, NULL} /* sentinel */
};
/*
* Convenience routine to export an integer value.
* For simplicity, errors (which are unlikely anyway) are ignored.
*/
static void
insint(PyObject *d, char *name, int value)
{
PyObject *v = PyInt_FromLong((long) value);
if (v == NULL) {
/* Don't bother reporting this error */
PyErr_Clear();
}
else {
PyDict_SetItemString(d, name, v);
Py_DECREF(v);
}
}
/* Initialization function for the module (*must* be called initpcre) */
PyMODINIT_FUNC
initpcre(void)
{
PyObject *m, *d;
Pcre_Type.ob_type = &PyType_Type;
/* Create the module and add the functions */
m = Py_InitModule("pcre", pcre_methods);
/* Add some symbolic constants to the module */
d = PyModule_GetDict(m);
ErrorObject = PyErr_NewException("pcre.error", NULL, NULL);
PyDict_SetItemString(d, "error", ErrorObject);
/* Insert the flags */
insint(d, "IGNORECASE", PCRE_CASELESS);
insint(d, "ANCHORED", PCRE_ANCHORED);
insint(d, "MULTILINE", PCRE_MULTILINE);
insint(d, "DOTALL", PCRE_DOTALL);
insint(d, "VERBOSE", PCRE_EXTENDED);
insint(d, "LOCALE", PCRE_LOCALE);
}
Modules/pypcre.c deleted 100644 → 0
View file @ f7e836e6
This source diff could not be displayed because it is too large. You can view the blob instead.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7