Mercurial > hg
annotate mercurial/base85.c @ 12957:9f2ac318b92e
util: clarify purpose of MBTextWrapper class
It's easy to get confused and scared of an Unicode monster when
skimming through this code: document that this is really just
about column-counting.
author | Nicolas Dumazet <nicdumz.commits@gmail.com> |
---|---|
date | Tue, 09 Nov 2010 13:43:35 +0900 |
parents | f42ef9493fa9 |
children | 613b8bd2284e |
rev | line source |
---|---|
3283 | 1 /* |
2 base85 codec | |
3 | |
4 Copyright 2006 Brendan Cully <brendan@kublai.com> | |
5 | |
6 This software may be used and distributed according to the terms of | |
7 the GNU General Public License, incorporated herein by reference. | |
8 | |
9 Largely based on git's implementation | |
10 */ | |
11 | |
12 #include <Python.h> | |
13 | |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
14 #include "util.h" |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
15 |
3283 | 16 static const char b85chars[] = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ" |
17 "abcdefghijklmnopqrstuvwxyz!#$%&()*+-;<=>?@^_`{|}~"; | |
18 static char b85dec[256]; | |
19 | |
20 static void | |
21 b85prep(void) | |
22 { | |
23 int i; | |
24 | |
25 memset(b85dec, 0, sizeof(b85dec)); | |
26 for (i = 0; i < sizeof(b85chars); i++) | |
27 b85dec[(int)(b85chars[i])] = i + 1; | |
28 } | |
29 | |
30 static PyObject * | |
31 b85encode(PyObject *self, PyObject *args) | |
32 { | |
33 const unsigned char *text; | |
34 PyObject *out; | |
35 char *dst; | |
36 int len, olen, i; | |
37 unsigned int acc, val, ch; | |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
38 int pad = 0; |
3283 | 39 |
3369
4bad632913d8
python2.5 PyArg_ParseTuple fix
Alexis S. L. Carvalho <alexis@cecm.usp.br>
parents:
3332
diff
changeset
|
40 if (!PyArg_ParseTuple(args, "s#|i", &text, &len, &pad)) |
3283 | 41 return NULL; |
42 | |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
43 if (pad) |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
44 olen = ((len + 3) / 4 * 5) - 3; |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
45 else { |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
46 olen = len % 4; |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
47 if (olen) |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
48 olen++; |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
49 olen += len / 4 * 5; |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
50 } |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
51 if (!(out = PyBytes_FromStringAndSize(NULL, olen + 3))) |
3283 | 52 return NULL; |
53 | |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
54 dst = PyBytes_AsString(out); |
3283 | 55 |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
56 while (len) { |
3283 | 57 acc = 0; |
58 for (i = 24; i >= 0; i -= 8) { | |
59 ch = *text++; | |
60 acc |= ch << i; | |
61 if (--len == 0) | |
62 break; | |
63 } | |
64 for (i = 4; i >= 0; i--) { | |
65 val = acc % 85; | |
66 acc /= 85; | |
67 dst[i] = b85chars[val]; | |
68 } | |
69 dst += 5; | |
70 } | |
71 | |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
72 if (!pad) |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
73 _PyBytes_Resize(&out, olen); |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
74 |
3283 | 75 return out; |
76 } | |
77 | |
78 static PyObject * | |
79 b85decode(PyObject *self, PyObject *args) | |
80 { | |
81 PyObject *out; | |
82 const char *text; | |
83 char *dst; | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
84 int len, i, j, olen, c, cap; |
3283 | 85 unsigned int acc; |
86 | |
3369
4bad632913d8
python2.5 PyArg_ParseTuple fix
Alexis S. L. Carvalho <alexis@cecm.usp.br>
parents:
3332
diff
changeset
|
87 if (!PyArg_ParseTuple(args, "s#", &text, &len)) |
3283 | 88 return NULL; |
89 | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
90 olen = len / 5 * 4; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
91 i = len % 5; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
92 if (i) |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
93 olen += i - 1; |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
94 if (!(out = PyBytes_FromStringAndSize(NULL, olen))) |
3283 | 95 return NULL; |
96 | |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
97 dst = PyBytes_AsString(out); |
3283 | 98 |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
99 i = 0; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
100 while (i < len) |
3283 | 101 { |
102 acc = 0; | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
103 cap = len - i - 1; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
104 if (cap > 4) |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
105 cap = 4; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
106 for (j = 0; j < cap; i++, j++) |
3283 | 107 { |
108 c = b85dec[(int)*text++] - 1; | |
109 if (c < 0) | |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
110 return PyErr_Format( |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
111 PyExc_ValueError, |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
112 "Bad base85 character at position %d", i); |
3283 | 113 acc = acc * 85 + c; |
114 } | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
115 if (i++ < len) |
3283 | 116 { |
117 c = b85dec[(int)*text++] - 1; | |
118 if (c < 0) | |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
119 return PyErr_Format( |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
120 PyExc_ValueError, |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
121 "Bad base85 character at position %d", i); |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
122 /* overflow detection: 0xffffffff == "|NsC0", |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
123 * "|NsC" == 0x03030303 */ |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
124 if (acc > 0x03030303 || (acc *= 85) > 0xffffffff - c) |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
125 return PyErr_Format( |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
126 PyExc_ValueError, |
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
7190
diff
changeset
|
127 "Bad base85 sequence at position %d", i); |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
128 acc += c; |
3283 | 129 } |
130 | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
131 cap = olen < 4 ? olen : 4; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
132 olen -= cap; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
133 for (j = 0; j < 4 - cap; j++) |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
134 acc *= 85; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
135 if (cap && cap < 4) |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
136 acc += 0xffffff >> (cap - 1) * 8; |
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
137 for (j = 0; j < cap; j++) |
3283 | 138 { |
139 acc = (acc << 8) | (acc >> 24); | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
140 *dst++ = acc; |
3283 | 141 } |
142 } | |
143 | |
144 return out; | |
145 } | |
146 | |
147 static char base85_doc[] = "Base85 Data Encoding"; | |
148 | |
149 static PyMethodDef methods[] = { | |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
150 {"b85encode", b85encode, METH_VARARGS, |
7190
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
151 "Encode text in base85.\n\n" |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
152 "If the second parameter is true, pad the result to a multiple of " |
aecea6934fdd
Some additional space/tab cleanups
Thomas Arendsen Hein <thomas@intevation.de>
parents:
3369
diff
changeset
|
153 "five characters.\n"}, |
3288
e93c926e069e
Handle odd-sized base85 input and output
Brendan Cully <brendan@kublai.com>
parents:
3283
diff
changeset
|
154 {"b85decode", b85decode, METH_VARARGS, "Decode base85 text.\n"}, |
3283 | 155 {NULL, NULL} |
156 }; | |
157 | |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
158 #ifdef IS_PY3K |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
159 static struct PyModuleDef base85_module = { |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
160 PyModuleDef_HEAD_INIT, |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
161 "base85", |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
162 base85_doc, |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
163 -1, |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
164 methods |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
165 }; |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
166 |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
167 PyMODINIT_FUNC PyInit_base85(void) |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
168 { |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
169 b85prep(); |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
170 |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
171 return PyModule_Create(&base85_module); |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
172 } |
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
173 #else |
3283 | 174 PyMODINIT_FUNC initbase85(void) |
175 { | |
176 Py_InitModule3("base85", methods, base85_doc); | |
177 | |
178 b85prep(); | |
179 } | |
11362
f42ef9493fa9
base85.c: Added support for py3k.
Renato Cunha <renatoc@gmail.com>
parents:
10282
diff
changeset
|
180 #endif |