Show More
@@ -1,60 +1,59 | |||||
1 | # Files that just need to be migrated to the formatter. |
|
1 | # Files that just need to be migrated to the formatter. | |
2 | # Do not add new files here! |
|
2 | # Do not add new files here! | |
3 | mercurial/cext/base85.c |
|
3 | mercurial/cext/base85.c | |
4 | mercurial/cext/charencode.c |
|
4 | mercurial/cext/charencode.c | |
5 | mercurial/cext/charencode.h |
|
5 | mercurial/cext/charencode.h | |
6 | mercurial/cext/diffhelpers.c |
|
6 | mercurial/cext/diffhelpers.c | |
7 | mercurial/cext/dirs.c |
|
7 | mercurial/cext/dirs.c | |
8 | mercurial/cext/manifest.c |
|
8 | mercurial/cext/manifest.c | |
9 | mercurial/cext/mpatch.c |
|
9 | mercurial/cext/mpatch.c | |
10 | mercurial/cext/osutil.c |
|
10 | mercurial/cext/osutil.c | |
11 | mercurial/cext/pathencode.c |
|
|||
12 | mercurial/cext/revlog.c |
|
11 | mercurial/cext/revlog.c | |
13 | # Vendored code that we should never format: |
|
12 | # Vendored code that we should never format: | |
14 | contrib/python-zstandard/c-ext/bufferutil.c |
|
13 | contrib/python-zstandard/c-ext/bufferutil.c | |
15 | contrib/python-zstandard/c-ext/compressiondict.c |
|
14 | contrib/python-zstandard/c-ext/compressiondict.c | |
16 | contrib/python-zstandard/c-ext/compressionparams.c |
|
15 | contrib/python-zstandard/c-ext/compressionparams.c | |
17 | contrib/python-zstandard/c-ext/compressionwriter.c |
|
16 | contrib/python-zstandard/c-ext/compressionwriter.c | |
18 | contrib/python-zstandard/c-ext/compressobj.c |
|
17 | contrib/python-zstandard/c-ext/compressobj.c | |
19 | contrib/python-zstandard/c-ext/compressor.c |
|
18 | contrib/python-zstandard/c-ext/compressor.c | |
20 | contrib/python-zstandard/c-ext/compressoriterator.c |
|
19 | contrib/python-zstandard/c-ext/compressoriterator.c | |
21 | contrib/python-zstandard/c-ext/constants.c |
|
20 | contrib/python-zstandard/c-ext/constants.c | |
22 | contrib/python-zstandard/c-ext/decompressionwriter.c |
|
21 | contrib/python-zstandard/c-ext/decompressionwriter.c | |
23 | contrib/python-zstandard/c-ext/decompressobj.c |
|
22 | contrib/python-zstandard/c-ext/decompressobj.c | |
24 | contrib/python-zstandard/c-ext/decompressor.c |
|
23 | contrib/python-zstandard/c-ext/decompressor.c | |
25 | contrib/python-zstandard/c-ext/decompressoriterator.c |
|
24 | contrib/python-zstandard/c-ext/decompressoriterator.c | |
26 | contrib/python-zstandard/c-ext/frameparams.c |
|
25 | contrib/python-zstandard/c-ext/frameparams.c | |
27 | contrib/python-zstandard/c-ext/python-zstandard.h |
|
26 | contrib/python-zstandard/c-ext/python-zstandard.h | |
28 | contrib/python-zstandard/zstd.c |
|
27 | contrib/python-zstandard/zstd.c | |
29 | contrib/python-zstandard/zstd/common/bitstream.h |
|
28 | contrib/python-zstandard/zstd/common/bitstream.h | |
30 | contrib/python-zstandard/zstd/common/entropy_common.c |
|
29 | contrib/python-zstandard/zstd/common/entropy_common.c | |
31 | contrib/python-zstandard/zstd/common/error_private.c |
|
30 | contrib/python-zstandard/zstd/common/error_private.c | |
32 | contrib/python-zstandard/zstd/common/error_private.h |
|
31 | contrib/python-zstandard/zstd/common/error_private.h | |
33 | contrib/python-zstandard/zstd/common/fse.h |
|
32 | contrib/python-zstandard/zstd/common/fse.h | |
34 | contrib/python-zstandard/zstd/common/fse_decompress.c |
|
33 | contrib/python-zstandard/zstd/common/fse_decompress.c | |
35 | contrib/python-zstandard/zstd/common/huf.h |
|
34 | contrib/python-zstandard/zstd/common/huf.h | |
36 | contrib/python-zstandard/zstd/common/mem.h |
|
35 | contrib/python-zstandard/zstd/common/mem.h | |
37 | contrib/python-zstandard/zstd/common/pool.c |
|
36 | contrib/python-zstandard/zstd/common/pool.c | |
38 | contrib/python-zstandard/zstd/common/pool.h |
|
37 | contrib/python-zstandard/zstd/common/pool.h | |
39 | contrib/python-zstandard/zstd/common/threading.c |
|
38 | contrib/python-zstandard/zstd/common/threading.c | |
40 | contrib/python-zstandard/zstd/common/threading.h |
|
39 | contrib/python-zstandard/zstd/common/threading.h | |
41 | contrib/python-zstandard/zstd/common/xxhash.c |
|
40 | contrib/python-zstandard/zstd/common/xxhash.c | |
42 | contrib/python-zstandard/zstd/common/xxhash.h |
|
41 | contrib/python-zstandard/zstd/common/xxhash.h | |
43 | contrib/python-zstandard/zstd/common/zstd_common.c |
|
42 | contrib/python-zstandard/zstd/common/zstd_common.c | |
44 | contrib/python-zstandard/zstd/common/zstd_errors.h |
|
43 | contrib/python-zstandard/zstd/common/zstd_errors.h | |
45 | contrib/python-zstandard/zstd/common/zstd_internal.h |
|
44 | contrib/python-zstandard/zstd/common/zstd_internal.h | |
46 | contrib/python-zstandard/zstd/compress/fse_compress.c |
|
45 | contrib/python-zstandard/zstd/compress/fse_compress.c | |
47 | contrib/python-zstandard/zstd/compress/huf_compress.c |
|
46 | contrib/python-zstandard/zstd/compress/huf_compress.c | |
48 | contrib/python-zstandard/zstd/compress/zstd_compress.c |
|
47 | contrib/python-zstandard/zstd/compress/zstd_compress.c | |
49 | contrib/python-zstandard/zstd/compress/zstd_opt.h |
|
48 | contrib/python-zstandard/zstd/compress/zstd_opt.h | |
50 | contrib/python-zstandard/zstd/compress/zstdmt_compress.c |
|
49 | contrib/python-zstandard/zstd/compress/zstdmt_compress.c | |
51 | contrib/python-zstandard/zstd/compress/zstdmt_compress.h |
|
50 | contrib/python-zstandard/zstd/compress/zstdmt_compress.h | |
52 | contrib/python-zstandard/zstd/decompress/huf_decompress.c |
|
51 | contrib/python-zstandard/zstd/decompress/huf_decompress.c | |
53 | contrib/python-zstandard/zstd/decompress/zstd_decompress.c |
|
52 | contrib/python-zstandard/zstd/decompress/zstd_decompress.c | |
54 | contrib/python-zstandard/zstd/dictBuilder/cover.c |
|
53 | contrib/python-zstandard/zstd/dictBuilder/cover.c | |
55 | contrib/python-zstandard/zstd/dictBuilder/divsufsort.c |
|
54 | contrib/python-zstandard/zstd/dictBuilder/divsufsort.c | |
56 | contrib/python-zstandard/zstd/dictBuilder/divsufsort.h |
|
55 | contrib/python-zstandard/zstd/dictBuilder/divsufsort.h | |
57 | contrib/python-zstandard/zstd/dictBuilder/zdict.c |
|
56 | contrib/python-zstandard/zstd/dictBuilder/zdict.c | |
58 | contrib/python-zstandard/zstd/dictBuilder/zdict.h |
|
57 | contrib/python-zstandard/zstd/dictBuilder/zdict.h | |
59 | contrib/python-zstandard/zstd/zstd.h |
|
58 | contrib/python-zstandard/zstd/zstd.h | |
60 | hgext/fsmonitor/pywatchman/bser.c |
|
59 | hgext/fsmonitor/pywatchman/bser.c |
@@ -1,765 +1,763 | |||||
1 | /* |
|
1 | /* | |
2 | pathencode.c - efficient path name encoding |
|
2 | pathencode.c - efficient path name encoding | |
3 |
|
3 | |||
4 | Copyright 2012 Facebook |
|
4 | Copyright 2012 Facebook | |
5 |
|
5 | |||
6 | This software may be used and distributed according to the terms of |
|
6 | This software may be used and distributed according to the terms of | |
7 | the GNU General Public License, incorporated herein by reference. |
|
7 | the GNU General Public License, incorporated herein by reference. | |
8 | */ |
|
8 | */ | |
9 |
|
9 | |||
10 | /* |
|
10 | /* | |
11 | * An implementation of the name encoding scheme used by the fncache |
|
11 | * An implementation of the name encoding scheme used by the fncache | |
12 | * store. The common case is of a path < 120 bytes long, which is |
|
12 | * store. The common case is of a path < 120 bytes long, which is | |
13 | * handled either in a single pass with no allocations or two passes |
|
13 | * handled either in a single pass with no allocations or two passes | |
14 | * with a single allocation. For longer paths, multiple passes are |
|
14 | * with a single allocation. For longer paths, multiple passes are | |
15 | * required. |
|
15 | * required. | |
16 | */ |
|
16 | */ | |
17 |
|
17 | |||
18 | #define PY_SSIZE_T_CLEAN |
|
18 | #define PY_SSIZE_T_CLEAN | |
19 | #include <Python.h> |
|
19 | #include <Python.h> | |
20 | #include <assert.h> |
|
20 | #include <assert.h> | |
21 | #include <ctype.h> |
|
21 | #include <ctype.h> | |
22 | #include <stdlib.h> |
|
22 | #include <stdlib.h> | |
23 | #include <string.h> |
|
23 | #include <string.h> | |
24 |
|
24 | |||
25 | #include "util.h" |
|
25 | #include "util.h" | |
26 |
|
26 | |||
27 | /* state machine for the fast path */ |
|
27 | /* state machine for the fast path */ | |
28 | enum path_state { |
|
28 | enum path_state { | |
29 |
START, |
|
29 | START, /* first byte of a path component */ | |
30 |
A, |
|
30 | A, /* "AUX" */ | |
31 | AU, |
|
31 | AU, | |
32 |
THIRD, |
|
32 | THIRD, /* third of a 3-byte sequence, e.g. "AUX", "NUL" */ | |
33 |
C, |
|
33 | C, /* "CON" or "COMn" */ | |
34 | CO, |
|
34 | CO, | |
35 |
COMLPT, |
|
35 | COMLPT, /* "COM" or "LPT" */ | |
36 | COMLPTn, |
|
36 | COMLPTn, | |
37 | L, |
|
37 | L, | |
38 | LP, |
|
38 | LP, | |
39 | N, |
|
39 | N, | |
40 | NU, |
|
40 | NU, | |
41 |
P, |
|
41 | P, /* "PRN" */ | |
42 | PR, |
|
42 | PR, | |
43 |
LDOT, |
|
43 | LDOT, /* leading '.' */ | |
44 |
DOT, |
|
44 | DOT, /* '.' in a non-leading position */ | |
45 |
H, |
|
45 | H, /* ".h" */ | |
46 |
HGDI, |
|
46 | HGDI, /* ".hg", ".d", or ".i" */ | |
47 | SPACE, |
|
47 | SPACE, | |
48 |
DEFAULT |
|
48 | DEFAULT, /* byte of a path component after the first */ | |
49 | }; |
|
49 | }; | |
50 |
|
50 | |||
51 | /* state machine for dir-encoding */ |
|
51 | /* state machine for dir-encoding */ | |
52 | enum dir_state { |
|
52 | enum dir_state { | |
53 | DDOT, |
|
53 | DDOT, | |
54 | DH, |
|
54 | DH, | |
55 | DHGDI, |
|
55 | DHGDI, | |
56 | DDEFAULT |
|
56 | DDEFAULT, | |
57 | }; |
|
57 | }; | |
58 |
|
58 | |||
59 | static inline int inset(const uint32_t bitset[], char c) |
|
59 | static inline int inset(const uint32_t bitset[], char c) | |
60 | { |
|
60 | { | |
61 | return bitset[((uint8_t)c) >> 5] & (1 << (((uint8_t)c) & 31)); |
|
61 | return bitset[((uint8_t)c) >> 5] & (1 << (((uint8_t)c) & 31)); | |
62 | } |
|
62 | } | |
63 |
|
63 | |||
64 | static inline void charcopy(char *dest, Py_ssize_t *destlen, size_t destsize, |
|
64 | static inline void charcopy(char *dest, Py_ssize_t *destlen, size_t destsize, | |
65 | char c) |
|
65 | char c) | |
66 | { |
|
66 | { | |
67 | if (dest) { |
|
67 | if (dest) { | |
68 | assert(*destlen < destsize); |
|
68 | assert(*destlen < destsize); | |
69 | dest[*destlen] = c; |
|
69 | dest[*destlen] = c; | |
70 | } |
|
70 | } | |
71 | (*destlen)++; |
|
71 | (*destlen)++; | |
72 | } |
|
72 | } | |
73 |
|
73 | |||
74 | static inline void memcopy(char *dest, Py_ssize_t *destlen, size_t destsize, |
|
74 | static inline void memcopy(char *dest, Py_ssize_t *destlen, size_t destsize, | |
75 | const void *src, Py_ssize_t len) |
|
75 | const void *src, Py_ssize_t len) | |
76 | { |
|
76 | { | |
77 | if (dest) { |
|
77 | if (dest) { | |
78 | assert(*destlen + len < destsize); |
|
78 | assert(*destlen + len < destsize); | |
79 | memcpy((void *)&dest[*destlen], src, len); |
|
79 | memcpy((void *)&dest[*destlen], src, len); | |
80 | } |
|
80 | } | |
81 | *destlen += len; |
|
81 | *destlen += len; | |
82 | } |
|
82 | } | |
83 |
|
83 | |||
84 | static inline void hexencode(char *dest, Py_ssize_t *destlen, size_t destsize, |
|
84 | static inline void hexencode(char *dest, Py_ssize_t *destlen, size_t destsize, | |
85 | uint8_t c) |
|
85 | uint8_t c) | |
86 | { |
|
86 | { | |
87 | static const char hexdigit[] = "0123456789abcdef"; |
|
87 | static const char hexdigit[] = "0123456789abcdef"; | |
88 |
|
88 | |||
89 | charcopy(dest, destlen, destsize, hexdigit[c >> 4]); |
|
89 | charcopy(dest, destlen, destsize, hexdigit[c >> 4]); | |
90 | charcopy(dest, destlen, destsize, hexdigit[c & 15]); |
|
90 | charcopy(dest, destlen, destsize, hexdigit[c & 15]); | |
91 | } |
|
91 | } | |
92 |
|
92 | |||
93 | /* 3-byte escape: tilde followed by two hex digits */ |
|
93 | /* 3-byte escape: tilde followed by two hex digits */ | |
94 | static inline void escape3(char *dest, Py_ssize_t *destlen, size_t destsize, |
|
94 | static inline void escape3(char *dest, Py_ssize_t *destlen, size_t destsize, | |
95 | char c) |
|
95 | char c) | |
96 | { |
|
96 | { | |
97 | charcopy(dest, destlen, destsize, '~'); |
|
97 | charcopy(dest, destlen, destsize, '~'); | |
98 | hexencode(dest, destlen, destsize, c); |
|
98 | hexencode(dest, destlen, destsize, c); | |
99 | } |
|
99 | } | |
100 |
|
100 | |||
101 | static Py_ssize_t _encodedir(char *dest, size_t destsize, |
|
101 | static Py_ssize_t _encodedir(char *dest, size_t destsize, const char *src, | |
102 |
|
|
102 | Py_ssize_t len) | |
103 | { |
|
103 | { | |
104 | enum dir_state state = DDEFAULT; |
|
104 | enum dir_state state = DDEFAULT; | |
105 | Py_ssize_t i = 0, destlen = 0; |
|
105 | Py_ssize_t i = 0, destlen = 0; | |
106 |
|
106 | |||
107 | while (i < len) { |
|
107 | while (i < len) { | |
108 | switch (state) { |
|
108 | switch (state) { | |
109 | case DDOT: |
|
109 | case DDOT: | |
110 | switch (src[i]) { |
|
110 | switch (src[i]) { | |
111 | case 'd': |
|
111 | case 'd': | |
112 | case 'i': |
|
112 | case 'i': | |
113 | state = DHGDI; |
|
113 | state = DHGDI; | |
114 | charcopy(dest, &destlen, destsize, src[i++]); |
|
114 | charcopy(dest, &destlen, destsize, src[i++]); | |
115 | break; |
|
115 | break; | |
116 | case 'h': |
|
116 | case 'h': | |
117 | state = DH; |
|
117 | state = DH; | |
118 | charcopy(dest, &destlen, destsize, src[i++]); |
|
118 | charcopy(dest, &destlen, destsize, src[i++]); | |
119 | break; |
|
119 | break; | |
120 | default: |
|
120 | default: | |
121 | state = DDEFAULT; |
|
121 | state = DDEFAULT; | |
122 | break; |
|
122 | break; | |
123 | } |
|
123 | } | |
124 | break; |
|
124 | break; | |
125 | case DH: |
|
125 | case DH: | |
126 | if (src[i] == 'g') { |
|
126 | if (src[i] == 'g') { | |
127 | state = DHGDI; |
|
127 | state = DHGDI; | |
128 | charcopy(dest, &destlen, destsize, src[i++]); |
|
128 | charcopy(dest, &destlen, destsize, src[i++]); | |
129 | } |
|
129 | } else | |
130 |
|
|
130 | state = DDEFAULT; | |
131 | break; |
|
131 | break; | |
132 | case DHGDI: |
|
132 | case DHGDI: | |
133 | if (src[i] == '/') { |
|
133 | if (src[i] == '/') { | |
134 | memcopy(dest, &destlen, destsize, ".hg", 3); |
|
134 | memcopy(dest, &destlen, destsize, ".hg", 3); | |
135 | charcopy(dest, &destlen, destsize, src[i++]); |
|
135 | charcopy(dest, &destlen, destsize, src[i++]); | |
136 | } |
|
136 | } | |
137 | state = DDEFAULT; |
|
137 | state = DDEFAULT; | |
138 | break; |
|
138 | break; | |
139 | case DDEFAULT: |
|
139 | case DDEFAULT: | |
140 | if (src[i] == '.') |
|
140 | if (src[i] == '.') | |
141 | state = DDOT; |
|
141 | state = DDOT; | |
142 | charcopy(dest, &destlen, destsize, src[i++]); |
|
142 | charcopy(dest, &destlen, destsize, src[i++]); | |
143 | break; |
|
143 | break; | |
144 | } |
|
144 | } | |
145 | } |
|
145 | } | |
146 |
|
146 | |||
147 | return destlen; |
|
147 | return destlen; | |
148 | } |
|
148 | } | |
149 |
|
149 | |||
150 | PyObject *encodedir(PyObject *self, PyObject *args) |
|
150 | PyObject *encodedir(PyObject *self, PyObject *args) | |
151 | { |
|
151 | { | |
152 | Py_ssize_t len, newlen; |
|
152 | Py_ssize_t len, newlen; | |
153 | PyObject *pathobj, *newobj; |
|
153 | PyObject *pathobj, *newobj; | |
154 | char *path; |
|
154 | char *path; | |
155 |
|
155 | |||
156 | if (!PyArg_ParseTuple(args, "O:encodedir", &pathobj)) |
|
156 | if (!PyArg_ParseTuple(args, "O:encodedir", &pathobj)) | |
157 | return NULL; |
|
157 | return NULL; | |
158 |
|
158 | |||
159 | if (PyBytes_AsStringAndSize(pathobj, &path, &len) == -1) { |
|
159 | if (PyBytes_AsStringAndSize(pathobj, &path, &len) == -1) { | |
160 | PyErr_SetString(PyExc_TypeError, "expected a string"); |
|
160 | PyErr_SetString(PyExc_TypeError, "expected a string"); | |
161 | return NULL; |
|
161 | return NULL; | |
162 | } |
|
162 | } | |
163 |
|
163 | |||
164 | newlen = len ? _encodedir(NULL, 0, path, len + 1) : 1; |
|
164 | newlen = len ? _encodedir(NULL, 0, path, len + 1) : 1; | |
165 |
|
165 | |||
166 | if (newlen == len + 1) { |
|
166 | if (newlen == len + 1) { | |
167 | Py_INCREF(pathobj); |
|
167 | Py_INCREF(pathobj); | |
168 | return pathobj; |
|
168 | return pathobj; | |
169 | } |
|
169 | } | |
170 |
|
170 | |||
171 | newobj = PyBytes_FromStringAndSize(NULL, newlen); |
|
171 | newobj = PyBytes_FromStringAndSize(NULL, newlen); | |
172 |
|
172 | |||
173 | if (newobj) { |
|
173 | if (newobj) { | |
174 | assert(PyBytes_Check(newobj)); |
|
174 | assert(PyBytes_Check(newobj)); | |
175 | Py_SIZE(newobj)--; |
|
175 | Py_SIZE(newobj)--; | |
176 | _encodedir(PyBytes_AS_STRING(newobj), newlen, path, |
|
176 | _encodedir(PyBytes_AS_STRING(newobj), newlen, path, len + 1); | |
177 | len + 1); |
|
|||
178 | } |
|
177 | } | |
179 |
|
178 | |||
180 | return newobj; |
|
179 | return newobj; | |
181 | } |
|
180 | } | |
182 |
|
181 | |||
183 | static Py_ssize_t _encode(const uint32_t twobytes[8], const uint32_t onebyte[8], |
|
182 | static Py_ssize_t _encode(const uint32_t twobytes[8], const uint32_t onebyte[8], | |
184 |
|
|
183 | char *dest, Py_ssize_t destlen, size_t destsize, | |
185 |
|
|
184 | const char *src, Py_ssize_t len, int encodedir) | |
186 | int encodedir) |
|
|||
187 | { |
|
185 | { | |
188 | enum path_state state = START; |
|
186 | enum path_state state = START; | |
189 | Py_ssize_t i = 0; |
|
187 | Py_ssize_t i = 0; | |
190 |
|
188 | |||
191 | /* |
|
189 | /* | |
192 | * Python strings end with a zero byte, which we use as a |
|
190 | * Python strings end with a zero byte, which we use as a | |
193 | * terminal token as they are not valid inside path names. |
|
191 | * terminal token as they are not valid inside path names. | |
194 | */ |
|
192 | */ | |
195 |
|
193 | |||
196 | while (i < len) { |
|
194 | while (i < len) { | |
197 | switch (state) { |
|
195 | switch (state) { | |
198 | case START: |
|
196 | case START: | |
199 | switch (src[i]) { |
|
197 | switch (src[i]) { | |
200 | case '/': |
|
198 | case '/': | |
201 | charcopy(dest, &destlen, destsize, src[i++]); |
|
199 | charcopy(dest, &destlen, destsize, src[i++]); | |
202 | break; |
|
200 | break; | |
203 | case '.': |
|
201 | case '.': | |
204 | state = LDOT; |
|
202 | state = LDOT; | |
205 | escape3(dest, &destlen, destsize, src[i++]); |
|
203 | escape3(dest, &destlen, destsize, src[i++]); | |
206 | break; |
|
204 | break; | |
207 | case ' ': |
|
205 | case ' ': | |
208 | state = DEFAULT; |
|
206 | state = DEFAULT; | |
209 | escape3(dest, &destlen, destsize, src[i++]); |
|
207 | escape3(dest, &destlen, destsize, src[i++]); | |
210 | break; |
|
208 | break; | |
211 | case 'a': |
|
209 | case 'a': | |
212 | state = A; |
|
210 | state = A; | |
213 | charcopy(dest, &destlen, destsize, src[i++]); |
|
211 | charcopy(dest, &destlen, destsize, src[i++]); | |
214 | break; |
|
212 | break; | |
215 | case 'c': |
|
213 | case 'c': | |
216 | state = C; |
|
214 | state = C; | |
217 | charcopy(dest, &destlen, destsize, src[i++]); |
|
215 | charcopy(dest, &destlen, destsize, src[i++]); | |
218 | break; |
|
216 | break; | |
219 | case 'l': |
|
217 | case 'l': | |
220 | state = L; |
|
218 | state = L; | |
221 | charcopy(dest, &destlen, destsize, src[i++]); |
|
219 | charcopy(dest, &destlen, destsize, src[i++]); | |
222 | break; |
|
220 | break; | |
223 | case 'n': |
|
221 | case 'n': | |
224 | state = N; |
|
222 | state = N; | |
225 | charcopy(dest, &destlen, destsize, src[i++]); |
|
223 | charcopy(dest, &destlen, destsize, src[i++]); | |
226 | break; |
|
224 | break; | |
227 | case 'p': |
|
225 | case 'p': | |
228 | state = P; |
|
226 | state = P; | |
229 | charcopy(dest, &destlen, destsize, src[i++]); |
|
227 | charcopy(dest, &destlen, destsize, src[i++]); | |
230 | break; |
|
228 | break; | |
231 | default: |
|
229 | default: | |
232 | state = DEFAULT; |
|
230 | state = DEFAULT; | |
233 | break; |
|
231 | break; | |
234 | } |
|
232 | } | |
235 | break; |
|
233 | break; | |
236 | case A: |
|
234 | case A: | |
237 | if (src[i] == 'u') { |
|
235 | if (src[i] == 'u') { | |
238 | state = AU; |
|
236 | state = AU; | |
239 | charcopy(dest, &destlen, destsize, src[i++]); |
|
237 | charcopy(dest, &destlen, destsize, src[i++]); | |
240 | } |
|
238 | } else | |
241 |
|
|
239 | state = DEFAULT; | |
242 | break; |
|
240 | break; | |
243 | case AU: |
|
241 | case AU: | |
244 | if (src[i] == 'x') { |
|
242 | if (src[i] == 'x') { | |
245 | state = THIRD; |
|
243 | state = THIRD; | |
246 | i++; |
|
244 | i++; | |
247 | } |
|
245 | } else | |
248 |
|
|
246 | state = DEFAULT; | |
249 | break; |
|
247 | break; | |
250 | case THIRD: |
|
248 | case THIRD: | |
251 | state = DEFAULT; |
|
249 | state = DEFAULT; | |
252 | switch (src[i]) { |
|
250 | switch (src[i]) { | |
253 | case '.': |
|
251 | case '.': | |
254 | case '/': |
|
252 | case '/': | |
255 | case '\0': |
|
253 | case '\0': | |
256 | escape3(dest, &destlen, destsize, src[i - 1]); |
|
254 | escape3(dest, &destlen, destsize, src[i - 1]); | |
257 | break; |
|
255 | break; | |
258 | default: |
|
256 | default: | |
259 | i--; |
|
257 | i--; | |
260 | break; |
|
258 | break; | |
261 | } |
|
259 | } | |
262 | break; |
|
260 | break; | |
263 | case C: |
|
261 | case C: | |
264 | if (src[i] == 'o') { |
|
262 | if (src[i] == 'o') { | |
265 | state = CO; |
|
263 | state = CO; | |
266 | charcopy(dest, &destlen, destsize, src[i++]); |
|
264 | charcopy(dest, &destlen, destsize, src[i++]); | |
267 | } |
|
265 | } else | |
268 |
|
|
266 | state = DEFAULT; | |
269 | break; |
|
267 | break; | |
270 | case CO: |
|
268 | case CO: | |
271 | if (src[i] == 'm') { |
|
269 | if (src[i] == 'm') { | |
272 | state = COMLPT; |
|
270 | state = COMLPT; | |
273 | i++; |
|
271 | i++; | |
274 | } |
|
272 | } else if (src[i] == 'n') { | |
275 | else if (src[i] == 'n') { |
|
|||
276 | state = THIRD; |
|
273 | state = THIRD; | |
277 | i++; |
|
274 | i++; | |
278 | } |
|
275 | } else | |
279 |
|
|
276 | state = DEFAULT; | |
280 | break; |
|
277 | break; | |
281 | case COMLPT: |
|
278 | case COMLPT: | |
282 | switch (src[i]) { |
|
279 | switch (src[i]) { | |
283 | case '1': case '2': case '3': case '4': case '5': |
|
280 | case '1': | |
284 | case '6': case '7': case '8': case '9': |
|
281 | case '2': | |
|
282 | case '3': | |||
|
283 | case '4': | |||
|
284 | case '5': | |||
|
285 | case '6': | |||
|
286 | case '7': | |||
|
287 | case '8': | |||
|
288 | case '9': | |||
285 | state = COMLPTn; |
|
289 | state = COMLPTn; | |
286 | i++; |
|
290 | i++; | |
287 | break; |
|
291 | break; | |
288 | default: |
|
292 | default: | |
289 | state = DEFAULT; |
|
293 | state = DEFAULT; | |
290 | charcopy(dest, &destlen, destsize, src[i - 1]); |
|
294 | charcopy(dest, &destlen, destsize, src[i - 1]); | |
291 | break; |
|
295 | break; | |
292 | } |
|
296 | } | |
293 | break; |
|
297 | break; | |
294 | case COMLPTn: |
|
298 | case COMLPTn: | |
295 | state = DEFAULT; |
|
299 | state = DEFAULT; | |
296 | switch (src[i]) { |
|
300 | switch (src[i]) { | |
297 | case '.': |
|
301 | case '.': | |
298 | case '/': |
|
302 | case '/': | |
299 | case '\0': |
|
303 | case '\0': | |
300 | escape3(dest, &destlen, destsize, src[i - 2]); |
|
304 | escape3(dest, &destlen, destsize, src[i - 2]); | |
301 | charcopy(dest, &destlen, destsize, src[i - 1]); |
|
305 | charcopy(dest, &destlen, destsize, src[i - 1]); | |
302 | break; |
|
306 | break; | |
303 | default: |
|
307 | default: | |
304 | memcopy(dest, &destlen, destsize, |
|
308 | memcopy(dest, &destlen, destsize, &src[i - 2], | |
305 | &src[i - 2], 2); |
|
309 | 2); | |
306 | break; |
|
310 | break; | |
307 | } |
|
311 | } | |
308 | break; |
|
312 | break; | |
309 | case L: |
|
313 | case L: | |
310 | if (src[i] == 'p') { |
|
314 | if (src[i] == 'p') { | |
311 | state = LP; |
|
315 | state = LP; | |
312 | charcopy(dest, &destlen, destsize, src[i++]); |
|
316 | charcopy(dest, &destlen, destsize, src[i++]); | |
313 | } |
|
317 | } else | |
314 |
|
|
318 | state = DEFAULT; | |
315 | break; |
|
319 | break; | |
316 | case LP: |
|
320 | case LP: | |
317 | if (src[i] == 't') { |
|
321 | if (src[i] == 't') { | |
318 | state = COMLPT; |
|
322 | state = COMLPT; | |
319 | i++; |
|
323 | i++; | |
320 | } |
|
324 | } else | |
321 |
|
|
325 | state = DEFAULT; | |
322 | break; |
|
326 | break; | |
323 | case N: |
|
327 | case N: | |
324 | if (src[i] == 'u') { |
|
328 | if (src[i] == 'u') { | |
325 | state = NU; |
|
329 | state = NU; | |
326 | charcopy(dest, &destlen, destsize, src[i++]); |
|
330 | charcopy(dest, &destlen, destsize, src[i++]); | |
327 | } |
|
331 | } else | |
328 |
|
|
332 | state = DEFAULT; | |
329 | break; |
|
333 | break; | |
330 | case NU: |
|
334 | case NU: | |
331 | if (src[i] == 'l') { |
|
335 | if (src[i] == 'l') { | |
332 | state = THIRD; |
|
336 | state = THIRD; | |
333 | i++; |
|
337 | i++; | |
334 | } |
|
338 | } else | |
335 |
|
|
339 | state = DEFAULT; | |
336 | break; |
|
340 | break; | |
337 | case P: |
|
341 | case P: | |
338 | if (src[i] == 'r') { |
|
342 | if (src[i] == 'r') { | |
339 | state = PR; |
|
343 | state = PR; | |
340 | charcopy(dest, &destlen, destsize, src[i++]); |
|
344 | charcopy(dest, &destlen, destsize, src[i++]); | |
341 | } |
|
345 | } else | |
342 |
|
|
346 | state = DEFAULT; | |
343 | break; |
|
347 | break; | |
344 | case PR: |
|
348 | case PR: | |
345 | if (src[i] == 'n') { |
|
349 | if (src[i] == 'n') { | |
346 | state = THIRD; |
|
350 | state = THIRD; | |
347 | i++; |
|
351 | i++; | |
348 | } |
|
352 | } else | |
349 |
|
|
353 | state = DEFAULT; | |
350 | break; |
|
354 | break; | |
351 | case LDOT: |
|
355 | case LDOT: | |
352 | switch (src[i]) { |
|
356 | switch (src[i]) { | |
353 | case 'd': |
|
357 | case 'd': | |
354 | case 'i': |
|
358 | case 'i': | |
355 | state = HGDI; |
|
359 | state = HGDI; | |
356 | charcopy(dest, &destlen, destsize, src[i++]); |
|
360 | charcopy(dest, &destlen, destsize, src[i++]); | |
357 | break; |
|
361 | break; | |
358 | case 'h': |
|
362 | case 'h': | |
359 | state = H; |
|
363 | state = H; | |
360 | charcopy(dest, &destlen, destsize, src[i++]); |
|
364 | charcopy(dest, &destlen, destsize, src[i++]); | |
361 | break; |
|
365 | break; | |
362 | default: |
|
366 | default: | |
363 | state = DEFAULT; |
|
367 | state = DEFAULT; | |
364 | break; |
|
368 | break; | |
365 | } |
|
369 | } | |
366 | break; |
|
370 | break; | |
367 | case DOT: |
|
371 | case DOT: | |
368 | switch (src[i]) { |
|
372 | switch (src[i]) { | |
369 | case '/': |
|
373 | case '/': | |
370 | case '\0': |
|
374 | case '\0': | |
371 | state = START; |
|
375 | state = START; | |
372 | memcopy(dest, &destlen, destsize, "~2e", 3); |
|
376 | memcopy(dest, &destlen, destsize, "~2e", 3); | |
373 | charcopy(dest, &destlen, destsize, src[i++]); |
|
377 | charcopy(dest, &destlen, destsize, src[i++]); | |
374 | break; |
|
378 | break; | |
375 | case 'd': |
|
379 | case 'd': | |
376 | case 'i': |
|
380 | case 'i': | |
377 | state = HGDI; |
|
381 | state = HGDI; | |
378 | charcopy(dest, &destlen, destsize, '.'); |
|
382 | charcopy(dest, &destlen, destsize, '.'); | |
379 | charcopy(dest, &destlen, destsize, src[i++]); |
|
383 | charcopy(dest, &destlen, destsize, src[i++]); | |
380 | break; |
|
384 | break; | |
381 | case 'h': |
|
385 | case 'h': | |
382 | state = H; |
|
386 | state = H; | |
383 | memcopy(dest, &destlen, destsize, ".h", 2); |
|
387 | memcopy(dest, &destlen, destsize, ".h", 2); | |
384 | i++; |
|
388 | i++; | |
385 | break; |
|
389 | break; | |
386 | default: |
|
390 | default: | |
387 | state = DEFAULT; |
|
391 | state = DEFAULT; | |
388 | charcopy(dest, &destlen, destsize, '.'); |
|
392 | charcopy(dest, &destlen, destsize, '.'); | |
389 | break; |
|
393 | break; | |
390 | } |
|
394 | } | |
391 | break; |
|
395 | break; | |
392 | case H: |
|
396 | case H: | |
393 | if (src[i] == 'g') { |
|
397 | if (src[i] == 'g') { | |
394 | state = HGDI; |
|
398 | state = HGDI; | |
395 | charcopy(dest, &destlen, destsize, src[i++]); |
|
399 | charcopy(dest, &destlen, destsize, src[i++]); | |
396 | } |
|
400 | } else | |
397 |
|
|
401 | state = DEFAULT; | |
398 | break; |
|
402 | break; | |
399 | case HGDI: |
|
403 | case HGDI: | |
400 | if (src[i] == '/') { |
|
404 | if (src[i] == '/') { | |
401 | state = START; |
|
405 | state = START; | |
402 | if (encodedir) |
|
406 | if (encodedir) | |
403 | memcopy(dest, &destlen, destsize, ".hg", |
|
407 | memcopy(dest, &destlen, destsize, ".hg", | |
404 |
|
|
408 | 3); | |
405 | charcopy(dest, &destlen, destsize, src[i++]); |
|
409 | charcopy(dest, &destlen, destsize, src[i++]); | |
406 | } |
|
410 | } else | |
407 |
|
|
411 | state = DEFAULT; | |
408 | break; |
|
412 | break; | |
409 | case SPACE: |
|
413 | case SPACE: | |
410 | switch (src[i]) { |
|
414 | switch (src[i]) { | |
411 | case '/': |
|
415 | case '/': | |
412 | case '\0': |
|
416 | case '\0': | |
413 | state = START; |
|
417 | state = START; | |
414 | memcopy(dest, &destlen, destsize, "~20", 3); |
|
418 | memcopy(dest, &destlen, destsize, "~20", 3); | |
415 | charcopy(dest, &destlen, destsize, src[i++]); |
|
419 | charcopy(dest, &destlen, destsize, src[i++]); | |
416 | break; |
|
420 | break; | |
417 | default: |
|
421 | default: | |
418 | state = DEFAULT; |
|
422 | state = DEFAULT; | |
419 | charcopy(dest, &destlen, destsize, ' '); |
|
423 | charcopy(dest, &destlen, destsize, ' '); | |
420 | break; |
|
424 | break; | |
421 | } |
|
425 | } | |
422 | break; |
|
426 | break; | |
423 | case DEFAULT: |
|
427 | case DEFAULT: | |
424 | while (inset(onebyte, src[i])) { |
|
428 | while (inset(onebyte, src[i])) { | |
425 | charcopy(dest, &destlen, destsize, src[i++]); |
|
429 | charcopy(dest, &destlen, destsize, src[i++]); | |
426 | if (i == len) |
|
430 | if (i == len) | |
427 | goto done; |
|
431 | goto done; | |
428 | } |
|
432 | } | |
429 | switch (src[i]) { |
|
433 | switch (src[i]) { | |
430 | case '.': |
|
434 | case '.': | |
431 | state = DOT; |
|
435 | state = DOT; | |
432 | i++; |
|
436 | i++; | |
433 | break; |
|
437 | break; | |
434 | case ' ': |
|
438 | case ' ': | |
435 | state = SPACE; |
|
439 | state = SPACE; | |
436 | i++; |
|
440 | i++; | |
437 | break; |
|
441 | break; | |
438 | case '/': |
|
442 | case '/': | |
439 | state = START; |
|
443 | state = START; | |
440 | charcopy(dest, &destlen, destsize, '/'); |
|
444 | charcopy(dest, &destlen, destsize, '/'); | |
441 | i++; |
|
445 | i++; | |
442 | break; |
|
446 | break; | |
443 | default: |
|
447 | default: | |
444 | if (inset(onebyte, src[i])) { |
|
448 | if (inset(onebyte, src[i])) { | |
445 | do { |
|
449 | do { | |
446 | charcopy(dest, &destlen, |
|
450 | charcopy(dest, &destlen, | |
447 |
|
|
451 | destsize, src[i++]); | |
448 | } while (i < len && |
|
452 | } while (i < len && | |
449 |
|
|
453 | inset(onebyte, src[i])); | |
450 | } |
|
454 | } else if (inset(twobytes, src[i])) { | |
451 | else if (inset(twobytes, src[i])) { |
|
|||
452 | char c = src[i++]; |
|
455 | char c = src[i++]; | |
453 | charcopy(dest, &destlen, destsize, '_'); |
|
456 | charcopy(dest, &destlen, destsize, '_'); | |
454 | charcopy(dest, &destlen, destsize, |
|
457 | charcopy(dest, &destlen, destsize, | |
455 |
|
|
458 | c == '_' ? '_' : c + 32); | |
456 | } |
|
459 | } else | |
457 | else |
|
|||
458 | escape3(dest, &destlen, destsize, |
|
460 | escape3(dest, &destlen, destsize, | |
459 |
|
|
461 | src[i++]); | |
460 | break; |
|
462 | break; | |
461 | } |
|
463 | } | |
462 | break; |
|
464 | break; | |
463 | } |
|
465 | } | |
464 | } |
|
466 | } | |
465 | done: |
|
467 | done: | |
466 | return destlen; |
|
468 | return destlen; | |
467 | } |
|
469 | } | |
468 |
|
470 | |||
469 | static Py_ssize_t basicencode(char *dest, size_t destsize, |
|
471 | static Py_ssize_t basicencode(char *dest, size_t destsize, const char *src, | |
470 | const char *src, Py_ssize_t len) |
|
472 | Py_ssize_t len) | |
471 | { |
|
473 | { | |
472 |
static const uint32_t twobytes[8] = { |
|
474 | static const uint32_t twobytes[8] = {0, 0, 0x87fffffe}; | |
473 |
|
475 | |||
474 | static const uint32_t onebyte[8] = { |
|
476 | static const uint32_t onebyte[8] = { | |
475 |
|
|
477 | 1, 0x2bff3bfa, 0x68000001, 0x2fffffff, | |
476 | }; |
|
478 | }; | |
477 |
|
479 | |||
478 | Py_ssize_t destlen = 0; |
|
480 | Py_ssize_t destlen = 0; | |
479 |
|
481 | |||
480 | return _encode(twobytes, onebyte, dest, destlen, destsize, |
|
482 | return _encode(twobytes, onebyte, dest, destlen, destsize, src, len, 1); | |
481 | src, len, 1); |
|
|||
482 | } |
|
483 | } | |
483 |
|
484 | |||
484 | static const Py_ssize_t maxstorepathlen = 120; |
|
485 | static const Py_ssize_t maxstorepathlen = 120; | |
485 |
|
486 | |||
486 | static Py_ssize_t _lowerencode(char *dest, size_t destsize, |
|
487 | static Py_ssize_t _lowerencode(char *dest, size_t destsize, const char *src, | |
487 | const char *src, Py_ssize_t len) |
|
488 | Py_ssize_t len) | |
488 | { |
|
489 | { | |
489 | static const uint32_t onebyte[8] = { |
|
490 | static const uint32_t onebyte[8] = {1, 0x2bfffbfb, 0xe8000001, | |
490 | 1, 0x2bfffbfb, 0xe8000001, 0x2fffffff |
|
491 | 0x2fffffff}; | |
491 | }; |
|
|||
492 |
|
492 | |||
493 |
static const uint32_t lower[8] = { |
|
493 | static const uint32_t lower[8] = {0, 0, 0x7fffffe}; | |
494 |
|
494 | |||
495 | Py_ssize_t i, destlen = 0; |
|
495 | Py_ssize_t i, destlen = 0; | |
496 |
|
496 | |||
497 | for (i = 0; i < len; i++) { |
|
497 | for (i = 0; i < len; i++) { | |
498 | if (inset(onebyte, src[i])) |
|
498 | if (inset(onebyte, src[i])) | |
499 | charcopy(dest, &destlen, destsize, src[i]); |
|
499 | charcopy(dest, &destlen, destsize, src[i]); | |
500 | else if (inset(lower, src[i])) |
|
500 | else if (inset(lower, src[i])) | |
501 | charcopy(dest, &destlen, destsize, src[i] + 32); |
|
501 | charcopy(dest, &destlen, destsize, src[i] + 32); | |
502 | else |
|
502 | else | |
503 | escape3(dest, &destlen, destsize, src[i]); |
|
503 | escape3(dest, &destlen, destsize, src[i]); | |
504 | } |
|
504 | } | |
505 |
|
505 | |||
506 | return destlen; |
|
506 | return destlen; | |
507 | } |
|
507 | } | |
508 |
|
508 | |||
509 | PyObject *lowerencode(PyObject *self, PyObject *args) |
|
509 | PyObject *lowerencode(PyObject *self, PyObject *args) | |
510 | { |
|
510 | { | |
511 | char *path; |
|
511 | char *path; | |
512 | Py_ssize_t len, newlen; |
|
512 | Py_ssize_t len, newlen; | |
513 | PyObject *ret; |
|
513 | PyObject *ret; | |
514 |
|
514 | |||
515 | if (!PyArg_ParseTuple(args, "s#:lowerencode", &path, &len)) |
|
515 | if (!PyArg_ParseTuple(args, "s#:lowerencode", &path, &len)) | |
516 | return NULL; |
|
516 | return NULL; | |
517 |
|
517 | |||
518 | newlen = _lowerencode(NULL, 0, path, len); |
|
518 | newlen = _lowerencode(NULL, 0, path, len); | |
519 | ret = PyBytes_FromStringAndSize(NULL, newlen); |
|
519 | ret = PyBytes_FromStringAndSize(NULL, newlen); | |
520 | if (ret) |
|
520 | if (ret) | |
521 | _lowerencode(PyBytes_AS_STRING(ret), newlen, path, len); |
|
521 | _lowerencode(PyBytes_AS_STRING(ret), newlen, path, len); | |
522 |
|
522 | |||
523 | return ret; |
|
523 | return ret; | |
524 | } |
|
524 | } | |
525 |
|
525 | |||
526 | /* See store.py:_auxencode for a description. */ |
|
526 | /* See store.py:_auxencode for a description. */ | |
527 | static Py_ssize_t auxencode(char *dest, size_t destsize, |
|
527 | static Py_ssize_t auxencode(char *dest, size_t destsize, const char *src, | |
528 | const char *src, Py_ssize_t len) |
|
528 | Py_ssize_t len) | |
529 | { |
|
529 | { | |
530 | static const uint32_t twobytes[8]; |
|
530 | static const uint32_t twobytes[8]; | |
531 |
|
531 | |||
532 | static const uint32_t onebyte[8] = { |
|
532 | static const uint32_t onebyte[8] = { | |
533 |
|
|
533 | ~0U, 0xffff3ffe, ~0U, ~0U, ~0U, ~0U, ~0U, ~0U, | |
534 | }; |
|
534 | }; | |
535 |
|
535 | |||
536 | return _encode(twobytes, onebyte, dest, 0, destsize, src, len, 0); |
|
536 | return _encode(twobytes, onebyte, dest, 0, destsize, src, len, 0); | |
537 | } |
|
537 | } | |
538 |
|
538 | |||
539 | static PyObject *hashmangle(const char *src, Py_ssize_t len, const char sha[20]) |
|
539 | static PyObject *hashmangle(const char *src, Py_ssize_t len, const char sha[20]) | |
540 | { |
|
540 | { | |
541 | static const Py_ssize_t dirprefixlen = 8; |
|
541 | static const Py_ssize_t dirprefixlen = 8; | |
542 | static const Py_ssize_t maxshortdirslen = 68; |
|
542 | static const Py_ssize_t maxshortdirslen = 68; | |
543 | char *dest; |
|
543 | char *dest; | |
544 | PyObject *ret; |
|
544 | PyObject *ret; | |
545 |
|
545 | |||
546 | Py_ssize_t i, d, p, lastslash = len - 1, lastdot = -1; |
|
546 | Py_ssize_t i, d, p, lastslash = len - 1, lastdot = -1; | |
547 | Py_ssize_t destsize, destlen = 0, slop, used; |
|
547 | Py_ssize_t destsize, destlen = 0, slop, used; | |
548 |
|
548 | |||
549 | while (lastslash >= 0 && src[lastslash] != '/') { |
|
549 | while (lastslash >= 0 && src[lastslash] != '/') { | |
550 | if (src[lastslash] == '.' && lastdot == -1) |
|
550 | if (src[lastslash] == '.' && lastdot == -1) | |
551 | lastdot = lastslash; |
|
551 | lastdot = lastslash; | |
552 | lastslash--; |
|
552 | lastslash--; | |
553 | } |
|
553 | } | |
554 |
|
554 | |||
555 | #if 0 |
|
555 | #if 0 | |
556 | /* All paths should end in a suffix of ".i" or ".d". |
|
556 | /* All paths should end in a suffix of ".i" or ".d". | |
557 | Unfortunately, the file names in test-hybridencode.py |
|
557 | Unfortunately, the file names in test-hybridencode.py | |
558 | violate this rule. */ |
|
558 | violate this rule. */ | |
559 | if (lastdot != len - 3) { |
|
559 | if (lastdot != len - 3) { | |
560 | PyErr_SetString(PyExc_ValueError, |
|
560 | PyErr_SetString(PyExc_ValueError, | |
561 | "suffix missing or wrong length"); |
|
561 | "suffix missing or wrong length"); | |
562 | return NULL; |
|
562 | return NULL; | |
563 | } |
|
563 | } | |
564 | #endif |
|
564 | #endif | |
565 |
|
565 | |||
566 | /* If src contains a suffix, we will append it to the end of |
|
566 | /* If src contains a suffix, we will append it to the end of | |
567 | the new string, so make room. */ |
|
567 | the new string, so make room. */ | |
568 | destsize = 120; |
|
568 | destsize = 120; | |
569 | if (lastdot >= 0) |
|
569 | if (lastdot >= 0) | |
570 | destsize += len - lastdot - 1; |
|
570 | destsize += len - lastdot - 1; | |
571 |
|
571 | |||
572 | ret = PyBytes_FromStringAndSize(NULL, destsize); |
|
572 | ret = PyBytes_FromStringAndSize(NULL, destsize); | |
573 | if (ret == NULL) |
|
573 | if (ret == NULL) | |
574 | return NULL; |
|
574 | return NULL; | |
575 |
|
575 | |||
576 | dest = PyBytes_AS_STRING(ret); |
|
576 | dest = PyBytes_AS_STRING(ret); | |
577 | memcopy(dest, &destlen, destsize, "dh/", 3); |
|
577 | memcopy(dest, &destlen, destsize, "dh/", 3); | |
578 |
|
578 | |||
579 | /* Copy up to dirprefixlen bytes of each path component, up to |
|
579 | /* Copy up to dirprefixlen bytes of each path component, up to | |
580 | a limit of maxshortdirslen bytes. */ |
|
580 | a limit of maxshortdirslen bytes. */ | |
581 | for (i = d = p = 0; i < lastslash; i++, p++) { |
|
581 | for (i = d = p = 0; i < lastslash; i++, p++) { | |
582 | if (src[i] == '/') { |
|
582 | if (src[i] == '/') { | |
583 | char d = dest[destlen - 1]; |
|
583 | char d = dest[destlen - 1]; | |
584 | /* After truncation, a directory name may end |
|
584 | /* After truncation, a directory name may end | |
585 | in a space or dot, which are unportable. */ |
|
585 | in a space or dot, which are unportable. */ | |
586 | if (d == '.' || d == ' ') |
|
586 | if (d == '.' || d == ' ') | |
587 | dest[destlen - 1] = '_'; |
|
587 | dest[destlen - 1] = '_'; | |
588 | /* The + 3 is to account for "dh/" in the beginning */ |
|
588 | /* The + 3 is to account for "dh/" in the beginning */ | |
589 | if (destlen > maxshortdirslen + 3) |
|
589 | if (destlen > maxshortdirslen + 3) | |
590 | break; |
|
590 | break; | |
591 | charcopy(dest, &destlen, destsize, src[i]); |
|
591 | charcopy(dest, &destlen, destsize, src[i]); | |
592 | p = -1; |
|
592 | p = -1; | |
593 | } |
|
593 | } else if (p < dirprefixlen) | |
594 | else if (p < dirprefixlen) |
|
|||
595 | charcopy(dest, &destlen, destsize, src[i]); |
|
594 | charcopy(dest, &destlen, destsize, src[i]); | |
596 | } |
|
595 | } | |
597 |
|
596 | |||
598 | /* Rewind to just before the last slash copied. */ |
|
597 | /* Rewind to just before the last slash copied. */ | |
599 | if (destlen > maxshortdirslen + 3) |
|
598 | if (destlen > maxshortdirslen + 3) | |
600 | do { |
|
599 | do { | |
601 | destlen--; |
|
600 | destlen--; | |
602 | } while (destlen > 0 && dest[destlen] != '/'); |
|
601 | } while (destlen > 0 && dest[destlen] != '/'); | |
603 |
|
602 | |||
604 | if (destlen > 3) { |
|
603 | if (destlen > 3) { | |
605 | if (lastslash > 0) { |
|
604 | if (lastslash > 0) { | |
606 | char d = dest[destlen - 1]; |
|
605 | char d = dest[destlen - 1]; | |
607 | /* The last directory component may be |
|
606 | /* The last directory component may be | |
608 | truncated, so make it safe. */ |
|
607 | truncated, so make it safe. */ | |
609 | if (d == '.' || d == ' ') |
|
608 | if (d == '.' || d == ' ') | |
610 | dest[destlen - 1] = '_'; |
|
609 | dest[destlen - 1] = '_'; | |
611 | } |
|
610 | } | |
612 |
|
611 | |||
613 | charcopy(dest, &destlen, destsize, '/'); |
|
612 | charcopy(dest, &destlen, destsize, '/'); | |
614 | } |
|
613 | } | |
615 |
|
614 | |||
616 | /* Add a prefix of the original file's name. Its length |
|
615 | /* Add a prefix of the original file's name. Its length | |
617 | depends on the number of bytes left after accounting for |
|
616 | depends on the number of bytes left after accounting for | |
618 | hash and suffix. */ |
|
617 | hash and suffix. */ | |
619 | used = destlen + 40; |
|
618 | used = destlen + 40; | |
620 | if (lastdot >= 0) |
|
619 | if (lastdot >= 0) | |
621 | used += len - lastdot - 1; |
|
620 | used += len - lastdot - 1; | |
622 | slop = maxstorepathlen - used; |
|
621 | slop = maxstorepathlen - used; | |
623 | if (slop > 0) { |
|
622 | if (slop > 0) { | |
624 | Py_ssize_t basenamelen = |
|
623 | Py_ssize_t basenamelen = | |
625 |
|
|
624 | lastslash >= 0 ? len - lastslash - 2 : len - 1; | |
626 |
|
625 | |||
627 | if (basenamelen > slop) |
|
626 | if (basenamelen > slop) | |
628 | basenamelen = slop; |
|
627 | basenamelen = slop; | |
629 | if (basenamelen > 0) |
|
628 | if (basenamelen > 0) | |
630 | memcopy(dest, &destlen, destsize, &src[lastslash + 1], |
|
629 | memcopy(dest, &destlen, destsize, &src[lastslash + 1], | |
631 |
|
|
630 | basenamelen); | |
632 | } |
|
631 | } | |
633 |
|
632 | |||
634 | /* Add hash and suffix. */ |
|
633 | /* Add hash and suffix. */ | |
635 | for (i = 0; i < 20; i++) |
|
634 | for (i = 0; i < 20; i++) | |
636 | hexencode(dest, &destlen, destsize, sha[i]); |
|
635 | hexencode(dest, &destlen, destsize, sha[i]); | |
637 |
|
636 | |||
638 | if (lastdot >= 0) |
|
637 | if (lastdot >= 0) | |
639 | memcopy(dest, &destlen, destsize, &src[lastdot], |
|
638 | memcopy(dest, &destlen, destsize, &src[lastdot], | |
640 |
|
|
639 | len - lastdot - 1); | |
641 |
|
640 | |||
642 | assert(PyBytes_Check(ret)); |
|
641 | assert(PyBytes_Check(ret)); | |
643 | Py_SIZE(ret) = destlen; |
|
642 | Py_SIZE(ret) = destlen; | |
644 |
|
643 | |||
645 | return ret; |
|
644 | return ret; | |
646 | } |
|
645 | } | |
647 |
|
646 | |||
648 | /* |
|
647 | /* | |
649 | * Avoiding a trip through Python would improve performance by 50%, |
|
648 | * Avoiding a trip through Python would improve performance by 50%, | |
650 | * but we don't encounter enough long names to be worth the code. |
|
649 | * but we don't encounter enough long names to be worth the code. | |
651 | */ |
|
650 | */ | |
652 | static int sha1hash(char hash[20], const char *str, Py_ssize_t len) |
|
651 | static int sha1hash(char hash[20], const char *str, Py_ssize_t len) | |
653 | { |
|
652 | { | |
654 | static PyObject *shafunc; |
|
653 | static PyObject *shafunc; | |
655 | PyObject *shaobj, *hashobj; |
|
654 | PyObject *shaobj, *hashobj; | |
656 |
|
655 | |||
657 | if (shafunc == NULL) { |
|
656 | if (shafunc == NULL) { | |
658 | PyObject *hashlib, *name = PyBytes_FromString("hashlib"); |
|
657 | PyObject *hashlib, *name = PyBytes_FromString("hashlib"); | |
659 |
|
658 | |||
660 | if (name == NULL) |
|
659 | if (name == NULL) | |
661 | return -1; |
|
660 | return -1; | |
662 |
|
661 | |||
663 | hashlib = PyImport_Import(name); |
|
662 | hashlib = PyImport_Import(name); | |
664 | Py_DECREF(name); |
|
663 | Py_DECREF(name); | |
665 |
|
664 | |||
666 | if (hashlib == NULL) { |
|
665 | if (hashlib == NULL) { | |
667 | PyErr_SetString(PyExc_ImportError, "hashlib"); |
|
666 | PyErr_SetString(PyExc_ImportError, "hashlib"); | |
668 | return -1; |
|
667 | return -1; | |
669 | } |
|
668 | } | |
670 | shafunc = PyObject_GetAttrString(hashlib, "sha1"); |
|
669 | shafunc = PyObject_GetAttrString(hashlib, "sha1"); | |
671 | Py_DECREF(hashlib); |
|
670 | Py_DECREF(hashlib); | |
672 |
|
671 | |||
673 | if (shafunc == NULL) { |
|
672 | if (shafunc == NULL) { | |
674 | PyErr_SetString(PyExc_AttributeError, |
|
673 | PyErr_SetString(PyExc_AttributeError, | |
675 |
|
|
674 | "module 'hashlib' has no " | |
676 |
|
|
675 | "attribute 'sha1'"); | |
677 | return -1; |
|
676 | return -1; | |
678 | } |
|
677 | } | |
679 | } |
|
678 | } | |
680 |
|
679 | |||
681 | shaobj = PyObject_CallFunction(shafunc, "s#", str, len); |
|
680 | shaobj = PyObject_CallFunction(shafunc, "s#", str, len); | |
682 |
|
681 | |||
683 | if (shaobj == NULL) |
|
682 | if (shaobj == NULL) | |
684 | return -1; |
|
683 | return -1; | |
685 |
|
684 | |||
686 | hashobj = PyObject_CallMethod(shaobj, "digest", ""); |
|
685 | hashobj = PyObject_CallMethod(shaobj, "digest", ""); | |
687 | Py_DECREF(shaobj); |
|
686 | Py_DECREF(shaobj); | |
688 | if (hashobj == NULL) |
|
687 | if (hashobj == NULL) | |
689 | return -1; |
|
688 | return -1; | |
690 |
|
689 | |||
691 | if (!PyBytes_Check(hashobj) || PyBytes_GET_SIZE(hashobj) != 20) { |
|
690 | if (!PyBytes_Check(hashobj) || PyBytes_GET_SIZE(hashobj) != 20) { | |
692 | PyErr_SetString(PyExc_TypeError, |
|
691 | PyErr_SetString(PyExc_TypeError, | |
693 |
|
|
692 | "result of digest is not a 20-byte hash"); | |
694 | Py_DECREF(hashobj); |
|
693 | Py_DECREF(hashobj); | |
695 | return -1; |
|
694 | return -1; | |
696 | } |
|
695 | } | |
697 |
|
696 | |||
698 | memcpy(hash, PyBytes_AS_STRING(hashobj), 20); |
|
697 | memcpy(hash, PyBytes_AS_STRING(hashobj), 20); | |
699 | Py_DECREF(hashobj); |
|
698 | Py_DECREF(hashobj); | |
700 | return 0; |
|
699 | return 0; | |
701 | } |
|
700 | } | |
702 |
|
701 | |||
703 | #define MAXENCODE 4096 * 4 |
|
702 | #define MAXENCODE 4096 * 4 | |
704 |
|
703 | |||
705 | static PyObject *hashencode(const char *src, Py_ssize_t len) |
|
704 | static PyObject *hashencode(const char *src, Py_ssize_t len) | |
706 | { |
|
705 | { | |
707 | char dired[MAXENCODE]; |
|
706 | char dired[MAXENCODE]; | |
708 | char lowered[MAXENCODE]; |
|
707 | char lowered[MAXENCODE]; | |
709 | char auxed[MAXENCODE]; |
|
708 | char auxed[MAXENCODE]; | |
710 | Py_ssize_t dirlen, lowerlen, auxlen, baselen; |
|
709 | Py_ssize_t dirlen, lowerlen, auxlen, baselen; | |
711 | char sha[20]; |
|
710 | char sha[20]; | |
712 |
|
711 | |||
713 | baselen = (len - 5) * 3; |
|
712 | baselen = (len - 5) * 3; | |
714 | if (baselen >= MAXENCODE) { |
|
713 | if (baselen >= MAXENCODE) { | |
715 | PyErr_SetString(PyExc_ValueError, "string too long"); |
|
714 | PyErr_SetString(PyExc_ValueError, "string too long"); | |
716 | return NULL; |
|
715 | return NULL; | |
717 | } |
|
716 | } | |
718 |
|
717 | |||
719 | dirlen = _encodedir(dired, baselen, src, len); |
|
718 | dirlen = _encodedir(dired, baselen, src, len); | |
720 | if (sha1hash(sha, dired, dirlen - 1) == -1) |
|
719 | if (sha1hash(sha, dired, dirlen - 1) == -1) | |
721 | return NULL; |
|
720 | return NULL; | |
722 | lowerlen = _lowerencode(lowered, baselen, dired + 5, dirlen - 5); |
|
721 | lowerlen = _lowerencode(lowered, baselen, dired + 5, dirlen - 5); | |
723 | auxlen = auxencode(auxed, baselen, lowered, lowerlen); |
|
722 | auxlen = auxencode(auxed, baselen, lowered, lowerlen); | |
724 | return hashmangle(auxed, auxlen, sha); |
|
723 | return hashmangle(auxed, auxlen, sha); | |
725 | } |
|
724 | } | |
726 |
|
725 | |||
727 | PyObject *pathencode(PyObject *self, PyObject *args) |
|
726 | PyObject *pathencode(PyObject *self, PyObject *args) | |
728 | { |
|
727 | { | |
729 | Py_ssize_t len, newlen; |
|
728 | Py_ssize_t len, newlen; | |
730 | PyObject *pathobj, *newobj; |
|
729 | PyObject *pathobj, *newobj; | |
731 | char *path; |
|
730 | char *path; | |
732 |
|
731 | |||
733 | if (!PyArg_ParseTuple(args, "O:pathencode", &pathobj)) |
|
732 | if (!PyArg_ParseTuple(args, "O:pathencode", &pathobj)) | |
734 | return NULL; |
|
733 | return NULL; | |
735 |
|
734 | |||
736 | if (PyBytes_AsStringAndSize(pathobj, &path, &len) == -1) { |
|
735 | if (PyBytes_AsStringAndSize(pathobj, &path, &len) == -1) { | |
737 | PyErr_SetString(PyExc_TypeError, "expected a string"); |
|
736 | PyErr_SetString(PyExc_TypeError, "expected a string"); | |
738 | return NULL; |
|
737 | return NULL; | |
739 | } |
|
738 | } | |
740 |
|
739 | |||
741 | if (len > maxstorepathlen) |
|
740 | if (len > maxstorepathlen) | |
742 | newlen = maxstorepathlen + 2; |
|
741 | newlen = maxstorepathlen + 2; | |
743 | else |
|
742 | else | |
744 | newlen = len ? basicencode(NULL, 0, path, len + 1) : 1; |
|
743 | newlen = len ? basicencode(NULL, 0, path, len + 1) : 1; | |
745 |
|
744 | |||
746 | if (newlen <= maxstorepathlen + 1) { |
|
745 | if (newlen <= maxstorepathlen + 1) { | |
747 | if (newlen == len + 1) { |
|
746 | if (newlen == len + 1) { | |
748 | Py_INCREF(pathobj); |
|
747 | Py_INCREF(pathobj); | |
749 | return pathobj; |
|
748 | return pathobj; | |
750 | } |
|
749 | } | |
751 |
|
750 | |||
752 | newobj = PyBytes_FromStringAndSize(NULL, newlen); |
|
751 | newobj = PyBytes_FromStringAndSize(NULL, newlen); | |
753 |
|
752 | |||
754 | if (newobj) { |
|
753 | if (newobj) { | |
755 | assert(PyBytes_Check(newobj)); |
|
754 | assert(PyBytes_Check(newobj)); | |
756 | Py_SIZE(newobj)--; |
|
755 | Py_SIZE(newobj)--; | |
757 | basicencode(PyBytes_AS_STRING(newobj), newlen, path, |
|
756 | basicencode(PyBytes_AS_STRING(newobj), newlen, path, | |
758 |
|
|
757 | len + 1); | |
759 | } |
|
758 | } | |
760 | } |
|
759 | } else | |
761 | else |
|
|||
762 | newobj = hashencode(path, len + 1); |
|
760 | newobj = hashencode(path, len + 1); | |
763 |
|
761 | |||
764 | return newobj; |
|
762 | return newobj; | |
765 | } |
|
763 | } |
General Comments 0
You need to be logged in to leave comments.
Login now