OLD | NEW |
1 // Copyright 1999-2005 The RE2 Authors. All Rights Reserved. | 1 // Copyright 1999-2005 The RE2 Authors. All Rights Reserved. |
2 // Use of this source code is governed by a BSD-style | 2 // Use of this source code is governed by a BSD-style |
3 // license that can be found in the LICENSE file. | 3 // license that can be found in the LICENSE file. |
4 | 4 |
5 #include "util/util.h" | 5 #include "util/util.h" |
6 #include "re2/stringpiece.h" | 6 #include "re2/stringpiece.h" |
7 | 7 |
8 namespace re2 { | 8 namespace re2 { |
9 | 9 |
10 // ---------------------------------------------------------------------- | 10 // ---------------------------------------------------------------------- |
11 // CEscapeString() | 11 // CEscapeString() |
12 // Copies 'src' to 'dest', escaping dangerous characters using | 12 // Copies 'src' to 'dest', escaping dangerous characters using |
13 // C-style escape sequences. 'src' and 'dest' should not overlap. | 13 // C-style escape sequences. 'src' and 'dest' should not overlap. |
14 // Returns the number of bytes written to 'dest' (not including the \0) | 14 // Returns the number of bytes written to 'dest' (not including the \0) |
15 // or -1 if there was insufficient space. | 15 // or -1 if there was insufficient space. |
16 // ---------------------------------------------------------------------- | 16 // ---------------------------------------------------------------------- |
17 int CEscapeString(const char* src, int src_len, char* dest, | 17 int CEscapeString(const char* src, int src_len, char* dest, |
18 int dest_len) { | 18 int dest_len) { |
19 const char* src_end = src + src_len; | 19 const char* src_end = src + src_len; |
20 int used = 0; | 20 int used = 0; |
21 | 21 |
22 for (; src < src_end; src++) { | 22 for (; src < src_end; src++) { |
23 if (dest_len - used < 2) // space for two-character escape | 23 if (dest_len - used < 2) // Need space for two letter escape |
24 return -1; | 24 return -1; |
25 | 25 |
26 unsigned char c = *src; | 26 unsigned char c = *src; |
27 switch (c) { | 27 switch (c) { |
28 case '\n': dest[used++] = '\\'; dest[used++] = 'n'; break; | 28 case '\n': dest[used++] = '\\'; dest[used++] = 'n'; break; |
29 case '\r': dest[used++] = '\\'; dest[used++] = 'r'; break; | 29 case '\r': dest[used++] = '\\'; dest[used++] = 'r'; break; |
30 case '\t': dest[used++] = '\\'; dest[used++] = 't'; break; | 30 case '\t': dest[used++] = '\\'; dest[used++] = 't'; break; |
31 case '\"': dest[used++] = '\\'; dest[used++] = '\"'; break; | 31 case '\"': dest[used++] = '\\'; dest[used++] = '\"'; break; |
32 case '\'': dest[used++] = '\\'; dest[used++] = '\''; break; | 32 case '\'': dest[used++] = '\\'; dest[used++] = '\''; break; |
33 case '\\': dest[used++] = '\\'; dest[used++] = '\\'; break; | 33 case '\\': dest[used++] = '\\'; dest[used++] = '\\'; break; |
34 default: | 34 default: |
35 // Note that if we emit \xNN and the src character after that is a hex | 35 // Note that if we emit \xNN and the src character after that is a hex |
36 // digit then that digit must be escaped too to prevent it being | 36 // digit then that digit must be escaped too to prevent it being |
37 // interpreted as part of the character code by C. | 37 // interpreted as part of the character code by C. |
38 if (c < ' ' || c > '~') { | 38 if (c < ' ' || c > '~') { |
39 if (dest_len - used < 5) // space for four-character escape + \0 | 39 if (dest_len - used < 4) // need space for 4 letter escape |
40 return -1; | 40 return -1; |
41 #if !defined(_WIN32) | 41 sprintf(dest + used, "\\%03o", c); |
42 snprintf(dest + used, 5, "\\%03o", c); | |
43 #else | |
44 // On Windows, the function takes 4+VA arguments, not 3+VA. With an | |
45 // array, the buffer size will be inferred, but not with a pointer. | |
46 snprintf(dest + used, 5, _TRUNCATE, "\\%03o", c); | |
47 #endif | |
48 used += 4; | 42 used += 4; |
49 } else { | 43 } else { |
50 dest[used++] = c; break; | 44 dest[used++] = c; break; |
51 } | 45 } |
52 } | 46 } |
53 } | 47 } |
54 | 48 |
55 if (dest_len - used < 1) // make sure that there is room for \0 | 49 if (dest_len - used < 1) // make sure that there is room for \0 |
56 return -1; | 50 return -1; |
57 | 51 |
58 dest[used] = '\0'; // doesn't count towards return value though | 52 dest[used] = '\0'; // doesn't count towards return value though |
59 return used; | 53 return used; |
60 } | 54 } |
61 | 55 |
62 | 56 |
63 // ---------------------------------------------------------------------- | 57 // ---------------------------------------------------------------------- |
64 // CEscape() | 58 // CEscape() |
65 // Copies 'src' to result, escaping dangerous characters using | 59 // Copies 'src' to result, escaping dangerous characters using |
66 // C-style escape sequences. 'src' and 'dest' should not overlap. | 60 // C-style escape sequences. 'src' and 'dest' should not overlap. |
67 // ---------------------------------------------------------------------- | 61 // ---------------------------------------------------------------------- |
68 string CEscape(const StringPiece& src) { | 62 string CEscape(const StringPiece& src) { |
69 const int dest_length = src.size() * 4 + 1; // Maximum possible expansion | 63 const int dest_length = src.size() * 4 + 1; // Maximum possible expansion |
70 char* dest = new char[dest_length]; | 64 char* dest = new char[dest_length]; |
71 const int len = CEscapeString(src.data(), src.size(), | 65 const int len = CEscapeString(src.data(), src.size(), |
72 dest, dest_length); | 66 dest, dest_length); |
73 string s = string(dest, len); | 67 string s = string(dest, len); |
74 delete[] dest; | 68 delete[] dest; |
75 return s; | 69 return s; |
76 } | 70 } |
77 | 71 |
78 string PrefixSuccessor(const StringPiece& prefix) { | 72 string PrefixSuccessor(const StringPiece& prefix) { |
79 // We can increment the last character in the string and be done | 73 // We can increment the last character in the string and be done |
80 // unless that character is 255, in which case we have to erase the | 74 // unless that character is 255, in which case we have to erase the |
81 // last character and increment the previous character, unless that | 75 // last character and increment the previous character, unless that |
82 // is 255, etc. If the string is empty or consists entirely of | 76 // is 255, etc. If the string is empty or consists entirely of |
83 // 255's, we just return the empty string. | 77 // 255's, we just return the empty string. |
84 bool done = false; | 78 bool done = false; |
85 string limit(prefix.data(), prefix.size()); | 79 string limit(prefix.data(), prefix.size()); |
86 int index = static_cast<int>(limit.size()) - 1; | 80 int index = limit.length() - 1; |
87 while (!done && index >= 0) { | 81 while (!done && index >= 0) { |
88 if ((limit[index]&255) == 255) { | 82 if ((limit[index]&255) == 255) { |
89 limit.erase(index); | 83 limit.erase(index); |
90 index--; | 84 index--; |
91 } else { | 85 } else { |
92 limit[index]++; | 86 limit[index]++; |
93 done = true; | 87 done = true; |
94 } | 88 } |
95 } | 89 } |
96 if (!done) { | 90 if (!done) { |
97 return ""; | 91 return ""; |
98 } else { | 92 } else { |
99 return limit; | 93 return limit; |
100 } | 94 } |
101 } | 95 } |
102 | 96 |
103 } // namespace re2 | 97 } // namespace re2 |
OLD | NEW |