author | Nicolas Dumazet <nicdumz.commits@gmail.com> |
Mon, 13 Jul 2009 09:50:26 +0900 | |
changeset 9136 | 31177742f54a |
parent 7190 | aecea6934fdd |
child 9427 | d37c0f4e8f48 |
permissions | -rw-r--r-- |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
1 |
/* |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
2 |
parsers.c - efficient content parsing |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
3 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
4 |
Copyright 2008 Matt Mackall <mpm@selenic.com> and others |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
5 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
6 |
This software may be used and distributed according to the terms of |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
7 |
the GNU General Public License, incorporated herein by reference. |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
8 |
*/ |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
9 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
10 |
#include <Python.h> |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
11 |
#include <ctype.h> |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
12 |
#include <string.h> |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
13 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
14 |
static int hexdigit(char c) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
15 |
{ |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
16 |
if (c >= '0' && c <= '9') |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
17 |
return c - '0'; |
7092
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
18 |
if (c >= 'a' && c <= 'f') |
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
19 |
return c - 'a' + 10; |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
20 |
if (c >= 'A' && c <= 'F') |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
21 |
return c - 'A' + 10; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
22 |
|
7092
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
23 |
PyErr_SetString(PyExc_ValueError, "input contains non-hex character"); |
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
24 |
return 0; |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
25 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
26 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
27 |
/* |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
28 |
* Turn a hex-encoded string into binary. |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
29 |
*/ |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
30 |
static PyObject *unhexlify(const char *str, int len) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
31 |
{ |
7092
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
32 |
PyObject *ret; |
6395
3f0294536b24
fix const annotation warning
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
6389
diff
changeset
|
33 |
const char *c; |
3f0294536b24
fix const annotation warning
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
6389
diff
changeset
|
34 |
char *d; |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
35 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
36 |
ret = PyString_FromStringAndSize(NULL, len / 2); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
37 |
if (!ret) |
7092
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
38 |
return NULL; |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
39 |
|
7092
fb3fc27617a2
parsers: speed up hex decoding for manifests
Matt Mackall <mpm@selenic.com>
parents:
7091
diff
changeset
|
40 |
d = PyString_AS_STRING(ret); |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
41 |
for (c = str; c < str + len;) { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
42 |
int hi = hexdigit(*c++); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
43 |
int lo = hexdigit(*c++); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
44 |
*d++ = (hi << 4) | lo; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
45 |
} |
7091
12b35ae03365
parsers: clean up whitespace
Matt Mackall <mpm@selenic.com>
parents:
6395
diff
changeset
|
46 |
|
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
47 |
return ret; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
48 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
49 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
50 |
/* |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
51 |
* This code assumes that a manifest is stitched together with newline |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
52 |
* ('\n') characters. |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
53 |
*/ |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
54 |
static PyObject *parse_manifest(PyObject *self, PyObject *args) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
55 |
{ |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
56 |
PyObject *mfdict, *fdict; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
57 |
char *str, *cur, *start, *zero; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
58 |
int len; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
59 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
60 |
if (!PyArg_ParseTuple(args, "O!O!s#:parse_manifest", |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
61 |
&PyDict_Type, &mfdict, |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
62 |
&PyDict_Type, &fdict, |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
63 |
&str, &len)) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
64 |
goto quit; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
65 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
66 |
for (start = cur = str, zero = NULL; cur < str + len; cur++) { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
67 |
PyObject *file = NULL, *node = NULL; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
68 |
PyObject *flags = NULL; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
69 |
int nlen; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
70 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
71 |
if (!*cur) { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
72 |
zero = cur; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
73 |
continue; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
74 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
75 |
else if (*cur != '\n') |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
76 |
continue; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
77 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
78 |
if (!zero) { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
79 |
PyErr_SetString(PyExc_ValueError, |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
80 |
"manifest entry has no separator"); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
81 |
goto quit; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
82 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
83 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
84 |
file = PyString_FromStringAndSize(start, zero - start); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
85 |
if (!file) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
86 |
goto bail; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
87 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
88 |
nlen = cur - zero - 1; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
89 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
90 |
node = unhexlify(zero + 1, nlen > 40 ? 40 : nlen); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
91 |
if (!node) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
92 |
goto bail; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
93 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
94 |
if (nlen > 40) { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
95 |
PyObject *flags; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
96 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
97 |
flags = PyString_FromStringAndSize(zero + 41, |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
98 |
nlen - 40); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
99 |
if (!flags) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
100 |
goto bail; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
101 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
102 |
if (PyDict_SetItem(fdict, file, flags) == -1) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
103 |
goto bail; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
104 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
105 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
106 |
if (PyDict_SetItem(mfdict, file, node) == -1) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
107 |
goto bail; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
108 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
109 |
start = cur + 1; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
110 |
zero = NULL; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
111 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
112 |
Py_XDECREF(flags); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
113 |
Py_XDECREF(node); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
114 |
Py_XDECREF(file); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
115 |
continue; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
116 |
bail: |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
117 |
Py_XDECREF(flags); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
118 |
Py_XDECREF(node); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
119 |
Py_XDECREF(file); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
120 |
goto quit; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
121 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
122 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
123 |
if (len > 0 && *(cur - 1) != '\n') { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
124 |
PyErr_SetString(PyExc_ValueError, |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
125 |
"manifest contains trailing garbage"); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
126 |
goto quit; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
127 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
128 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
129 |
Py_INCREF(Py_None); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
130 |
return Py_None; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
131 |
quit: |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
132 |
return NULL; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
133 |
} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
134 |
|
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
135 |
#ifdef _WIN32 |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
136 |
# ifdef _MSC_VER |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
137 |
/* msvc 6.0 has problems */ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
138 |
# define inline __inline |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
139 |
typedef unsigned long uint32_t; |
7122
3cf699e89e48
Fix missing uint64_t definition in parsers.c under Windows
Dhruva Krishnamurthy <dhruvakm@gmail.com>
parents:
7108
diff
changeset
|
140 |
typedef unsigned __int64 uint64_t; |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
141 |
# else |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
142 |
# include <stdint.h> |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
143 |
# endif |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
144 |
static uint32_t ntohl(uint32_t x) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
145 |
{ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
146 |
return ((x & 0x000000ffUL) << 24) | |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
147 |
((x & 0x0000ff00UL) << 8) | |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
148 |
((x & 0x00ff0000UL) >> 8) | |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
149 |
((x & 0xff000000UL) >> 24); |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
150 |
} |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
151 |
#else |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
152 |
/* not windows */ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
153 |
# include <sys/types.h> |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
154 |
# if defined __BEOS__ && !defined __HAIKU__ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
155 |
# include <ByteOrder.h> |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
156 |
# else |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
157 |
# include <arpa/inet.h> |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
158 |
# endif |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
159 |
# include <inttypes.h> |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
160 |
#endif |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
161 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
162 |
static PyObject *parse_dirstate(PyObject *self, PyObject *args) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
163 |
{ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
164 |
PyObject *dmap, *cmap, *parents = NULL, *ret = NULL; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
165 |
PyObject *fname = NULL, *cname = NULL, *entry = NULL; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
166 |
char *str, *cur, *end, *cpos; |
7174
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
167 |
int state, mode, size, mtime; |
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
168 |
unsigned int flen; |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
169 |
int len; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
170 |
char decode[16]; /* for alignment */ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
171 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
172 |
if (!PyArg_ParseTuple(args, "O!O!s#:parse_dirstate", |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
173 |
&PyDict_Type, &dmap, |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
174 |
&PyDict_Type, &cmap, |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
175 |
&str, &len)) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
176 |
goto quit; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
177 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
178 |
/* read parents */ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
179 |
if (len < 40) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
180 |
goto quit; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
181 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
182 |
parents = Py_BuildValue("s#s#", str, 20, str + 20, 20); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
183 |
if (!parents) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
184 |
goto quit; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
185 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
186 |
/* read filenames */ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
187 |
cur = str + 40; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
188 |
end = str + len; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
189 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
190 |
while (cur < end - 17) { |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
191 |
/* unpack header */ |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
192 |
state = *cur; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
193 |
memcpy(decode, cur + 1, 16); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
194 |
mode = ntohl(*(uint32_t *)(decode)); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
195 |
size = ntohl(*(uint32_t *)(decode + 4)); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
196 |
mtime = ntohl(*(uint32_t *)(decode + 8)); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
197 |
flen = ntohl(*(uint32_t *)(decode + 12)); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
198 |
cur += 17; |
7174
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
199 |
if (flen > end - cur) { |
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
200 |
PyErr_SetString(PyExc_ValueError, "overflow in dirstate"); |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
201 |
goto quit; |
7174
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
202 |
} |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
203 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
204 |
entry = Py_BuildValue("ciii", state, mode, size, mtime); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
205 |
if (!entry) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
206 |
goto quit; |
7175
5d8626b2c1db
parsers.c: do not try to untrack after a failure
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7174
diff
changeset
|
207 |
PyObject_GC_UnTrack(entry); /* don't waste time with this */ |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
208 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
209 |
cpos = memchr(cur, 0, flen); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
210 |
if (cpos) { |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
211 |
fname = PyString_FromStringAndSize(cur, cpos - cur); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
212 |
cname = PyString_FromStringAndSize(cpos + 1, |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
213 |
flen - (cpos - cur) - 1); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
214 |
if (!fname || !cname || |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
215 |
PyDict_SetItem(cmap, fname, cname) == -1 || |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
216 |
PyDict_SetItem(dmap, fname, entry) == -1) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
217 |
goto quit; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
218 |
Py_DECREF(cname); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
219 |
} else { |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
220 |
fname = PyString_FromStringAndSize(cur, flen); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
221 |
if (!fname || |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
222 |
PyDict_SetItem(dmap, fname, entry) == -1) |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
223 |
goto quit; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
224 |
} |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
225 |
cur += flen; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
226 |
Py_DECREF(fname); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
227 |
Py_DECREF(entry); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
228 |
fname = cname = entry = NULL; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
229 |
} |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
230 |
|
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
231 |
ret = parents; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
232 |
Py_INCREF(ret); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
233 |
quit: |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
234 |
Py_XDECREF(fname); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
235 |
Py_XDECREF(cname); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
236 |
Py_XDECREF(entry); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
237 |
Py_XDECREF(parents); |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
238 |
return ret; |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
239 |
} |
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
240 |
|
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
241 |
const char nullid[20]; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
242 |
const int nullrev = -1; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
243 |
|
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
244 |
/* create an index tuple, insert into the nodemap */ |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
245 |
static PyObject * _build_idx_entry(PyObject *nodemap, int n, uint64_t offset_flags, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
246 |
int comp_len, int uncomp_len, int base_rev, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
247 |
int link_rev, int parent_1, int parent_2, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
248 |
const char *c_node_id) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
249 |
{ |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
250 |
int err; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
251 |
PyObject *entry, *node_id, *n_obj; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
252 |
|
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
253 |
node_id = PyString_FromStringAndSize(c_node_id, 20); |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
254 |
n_obj = PyInt_FromLong(n); |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
255 |
if (!node_id || !n_obj) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
256 |
err = -1; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
257 |
else |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
258 |
err = PyDict_SetItem(nodemap, node_id, n_obj); |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
259 |
|
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
260 |
Py_XDECREF(n_obj); |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
261 |
if (err) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
262 |
goto error_dealloc; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
263 |
|
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
264 |
entry = Py_BuildValue("LiiiiiiN", offset_flags, comp_len, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
265 |
uncomp_len, base_rev, link_rev, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
266 |
parent_1, parent_2, node_id); |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
267 |
if (!entry) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
268 |
goto error_dealloc; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
269 |
PyObject_GC_UnTrack(entry); /* don't waste time with this */ |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
270 |
|
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
271 |
return entry; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
272 |
|
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
273 |
error_dealloc: |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
274 |
Py_XDECREF(node_id); |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
275 |
return NULL; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
276 |
} |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
277 |
|
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
278 |
/* RevlogNG format (all in big endian, data may be inlined): |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
279 |
* 6 bytes: offset |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
280 |
* 2 bytes: flags |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
281 |
* 4 bytes: compressed length |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
282 |
* 4 bytes: uncompressed length |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
283 |
* 4 bytes: base revision |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
284 |
* 4 bytes: link revision |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
285 |
* 4 bytes: parent 1 revision |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
286 |
* 4 bytes: parent 2 revision |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
287 |
* 32 bytes: nodeid (only 20 bytes used) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
288 |
*/ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
289 |
static int _parse_index_ng (const char *data, int size, int inlined, |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
290 |
PyObject *index, PyObject *nodemap) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
291 |
{ |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
292 |
PyObject *entry; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
293 |
int n = 0, err; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
294 |
uint64_t offset_flags; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
295 |
int comp_len, uncomp_len, base_rev, link_rev, parent_1, parent_2; |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
296 |
const char *c_node_id; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
297 |
const char *end = data + size; |
7169
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
298 |
char decode[64]; /* to enforce alignment with inline data */ |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
299 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
300 |
while (data < end) { |
7174
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
301 |
unsigned int step; |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
302 |
|
7169
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
303 |
memcpy(decode, data, 64); |
7186
f77c8d8331ca
clean up trailing spaces, leading spaces in C
Dirkjan Ochtman <dirkjan@ochtman.nl>
parents:
7176
diff
changeset
|
304 |
offset_flags = ntohl(*((uint32_t *) (decode + 4))); |
f77c8d8331ca
clean up trailing spaces, leading spaces in C
Dirkjan Ochtman <dirkjan@ochtman.nl>
parents:
7176
diff
changeset
|
305 |
if (n == 0) /* mask out version number for the first entry */ |
f77c8d8331ca
clean up trailing spaces, leading spaces in C
Dirkjan Ochtman <dirkjan@ochtman.nl>
parents:
7176
diff
changeset
|
306 |
offset_flags &= 0xFFFF; |
f77c8d8331ca
clean up trailing spaces, leading spaces in C
Dirkjan Ochtman <dirkjan@ochtman.nl>
parents:
7176
diff
changeset
|
307 |
else { |
7169
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
308 |
uint32_t offset_high = ntohl(*((uint32_t *) decode)); |
7186
f77c8d8331ca
clean up trailing spaces, leading spaces in C
Dirkjan Ochtman <dirkjan@ochtman.nl>
parents:
7176
diff
changeset
|
309 |
offset_flags |= ((uint64_t) offset_high) << 32; |
7133
42db22108d85
revlog parser: use ntohl() instead of ntohll() (fix endianness issues)
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7108
diff
changeset
|
310 |
} |
42db22108d85
revlog parser: use ntohl() instead of ntohll() (fix endianness issues)
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7108
diff
changeset
|
311 |
|
7169
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
312 |
comp_len = ntohl(*((uint32_t *) (decode + 8))); |
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
313 |
uncomp_len = ntohl(*((uint32_t *) (decode + 12))); |
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
314 |
base_rev = ntohl(*((uint32_t *) (decode + 16))); |
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
315 |
link_rev = ntohl(*((uint32_t *) (decode + 20))); |
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
316 |
parent_1 = ntohl(*((uint32_t *) (decode + 24))); |
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
317 |
parent_2 = ntohl(*((uint32_t *) (decode + 28))); |
6d28a399293e
Copy index before parsing to enforce alignment with inline data present.
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7154
diff
changeset
|
318 |
c_node_id = decode + 32; |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
319 |
|
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
320 |
entry = _build_idx_entry(nodemap, n, offset_flags, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
321 |
comp_len, uncomp_len, base_rev, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
322 |
link_rev, parent_1, parent_2, |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
323 |
c_node_id); |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
324 |
if (!entry) |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
325 |
return 0; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
326 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
327 |
if (inlined) { |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
328 |
err = PyList_Append(index, entry); |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
329 |
Py_DECREF(entry); |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
330 |
if (err) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
331 |
return 0; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
332 |
} else |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
333 |
PyList_SET_ITEM(index, n, entry); /* steals reference */ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
334 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
335 |
n++; |
7174
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
336 |
step = 64 + (inlined ? comp_len : 0); |
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
337 |
if (end - data < step) |
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
338 |
break; |
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
339 |
data += step; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
340 |
} |
7174
4da87407b845
parsers.c: fix integer overflows
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7168
diff
changeset
|
341 |
if (data != end) { |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
342 |
if (!PyErr_Occurred()) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
343 |
PyErr_SetString(PyExc_ValueError, "corrupt index file"); |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
344 |
return 0; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
345 |
} |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
346 |
|
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
347 |
/* create the nullid/nullrev entry in the nodemap and the |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
348 |
* magic nullid entry in the index at [-1] */ |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
349 |
entry = _build_idx_entry(nodemap, |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
350 |
nullrev, 0, 0, 0, -1, -1, -1, -1, nullid); |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
351 |
if (!entry) |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
352 |
return 0; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
353 |
if (inlined) { |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
354 |
err = PyList_Append(index, entry); |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
355 |
Py_DECREF(entry); |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
356 |
if (err) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
357 |
return 0; |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
358 |
} else |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
359 |
PyList_SET_ITEM(index, n, entry); /* steals reference */ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
360 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
361 |
return 1; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
362 |
} |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
363 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
364 |
/* This function parses a index file and returns a Python tuple of the |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
365 |
* following format: (index, nodemap, cache) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
366 |
* |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
367 |
* index: a list of tuples containing the RevlogNG records |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
368 |
* nodemap: a dict mapping node ids to indices in the index list |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
369 |
* cache: if data is inlined, a tuple (index_file_content, 0) else None |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
370 |
*/ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
371 |
static PyObject *parse_index(PyObject *self, PyObject *args) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
372 |
{ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
373 |
const char *data; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
374 |
int size, inlined; |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
375 |
PyObject *rval = NULL, *index = NULL, *nodemap = NULL, *cache = NULL; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
376 |
PyObject *data_obj = NULL, *inlined_obj; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
377 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
378 |
if (!PyArg_ParseTuple(args, "s#O", &data, &size, &inlined_obj)) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
379 |
return NULL; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
380 |
inlined = inlined_obj && PyObject_IsTrue(inlined_obj); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
381 |
|
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
382 |
/* If no data is inlined, we know the size of the index list in |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
383 |
* advance: size divided by size of one one revlog record (64 bytes) |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
7186
diff
changeset
|
384 |
* plus one for the nullid */ |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
385 |
index = inlined ? PyList_New(0) : PyList_New(size / 64 + 1); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
386 |
if (!index) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
387 |
goto quit; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
388 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
389 |
nodemap = PyDict_New(); |
7154
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
390 |
if (!nodemap) |
7fdf7a0a41b7
index parser: fix refcounting in case of errors, refactor
Benoit Boissinot <benoit.boissinot@ens-lyon.org>
parents:
7135
diff
changeset
|
391 |
goto quit; |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
392 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
393 |
/* set up the cache return value */ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
394 |
if (inlined) { |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
395 |
/* Note that the reference to data_obj is only borrowed */ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
396 |
data_obj = PyTuple_GET_ITEM(args, 0); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
397 |
cache = Py_BuildValue("iO", 0, data_obj); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
398 |
if (!cache) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
399 |
goto quit; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
400 |
} else { |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
401 |
cache = Py_None; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
402 |
Py_INCREF(Py_None); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
403 |
} |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
404 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
405 |
/* actually populate the index and the nodemap with data */ |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
406 |
if (!_parse_index_ng (data, size, inlined, index, nodemap)) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
407 |
goto quit; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
408 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
409 |
rval = Py_BuildValue("NNN", index, nodemap, cache); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
410 |
if (!rval) |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
411 |
goto quit; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
412 |
return rval; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
413 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
414 |
quit: |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
415 |
Py_XDECREF(index); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
416 |
Py_XDECREF(nodemap); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
417 |
Py_XDECREF(cache); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
418 |
Py_XDECREF(rval); |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
419 |
return NULL; |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
420 |
} |
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
421 |
|
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
422 |
|
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
423 |
static char parsers_doc[] = "Efficient content parsing."; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
424 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
425 |
static PyMethodDef methods[] = { |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
426 |
{"parse_manifest", parse_manifest, METH_VARARGS, "parse a manifest\n"}, |
7093
16bafcebd3d1
dirstate: C parsing extension
Matt Mackall <mpm@selenic.com>
parents:
7092
diff
changeset
|
427 |
{"parse_dirstate", parse_dirstate, METH_VARARGS, "parse a dirstate\n"}, |
7108
1ca878d7b849
C implementation of revlog index parsing
Bernhard Leiner <bleiner@gmail.com>
parents:
7093
diff
changeset
|
428 |
{"parse_index", parse_index, METH_VARARGS, "parse a revlog index\n"}, |
6389
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
429 |
{NULL, NULL} |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
430 |
}; |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
431 |
|
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
432 |
PyMODINIT_FUNC initparsers(void) |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
433 |
{ |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
434 |
Py_InitModule3("parsers", methods, parsers_doc); |
0231f763ebc8
manifest: improve parsing performance by 8x via a new C extension
Bryan O'Sullivan <bos@serpentine.com>
parents:
diff
changeset
|
435 |
} |