Ian Hodson | 2ee91b4 | 2012-05-14 12:29:36 +0100 | [diff] [blame] | 1 | // Copyright 1999-2005 The RE2 Authors. All Rights Reserved. |
| 2 | // Use of this source code is governed by a BSD-style |
| 3 | // license that can be found in the LICENSE file. |
| 4 | |
| 5 | #include "util/util.h" |
| 6 | #include "re2/stringpiece.h" |
| 7 | |
| 8 | namespace re2 { |
| 9 | |
| 10 | // ---------------------------------------------------------------------- |
| 11 | // CEscapeString() |
| 12 | // Copies 'src' to 'dest', escaping dangerous characters using |
| 13 | // C-style escape sequences. 'src' and 'dest' should not overlap. |
| 14 | // Returns the number of bytes written to 'dest' (not including the \0) |
| 15 | // or -1 if there was insufficient space. |
| 16 | // ---------------------------------------------------------------------- |
| 17 | int CEscapeString(const char* src, int src_len, char* dest, |
| 18 | int dest_len) { |
| 19 | const char* src_end = src + src_len; |
| 20 | int used = 0; |
| 21 | |
| 22 | for (; src < src_end; src++) { |
| 23 | if (dest_len - used < 2) // Need space for two letter escape |
| 24 | return -1; |
| 25 | |
| 26 | unsigned char c = *src; |
| 27 | switch (c) { |
| 28 | case '\n': dest[used++] = '\\'; dest[used++] = 'n'; break; |
| 29 | case '\r': dest[used++] = '\\'; dest[used++] = 'r'; break; |
| 30 | case '\t': dest[used++] = '\\'; dest[used++] = 't'; break; |
| 31 | case '\"': dest[used++] = '\\'; dest[used++] = '\"'; break; |
| 32 | case '\'': dest[used++] = '\\'; dest[used++] = '\''; break; |
| 33 | case '\\': dest[used++] = '\\'; dest[used++] = '\\'; break; |
| 34 | default: |
| 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 |
| 37 | // interpreted as part of the character code by C. |
| 38 | if (c < ' ' || c > '~') { |
| 39 | if (dest_len - used < 4) // need space for 4 letter escape |
| 40 | return -1; |
| 41 | sprintf(dest + used, "\\%03o", c); |
| 42 | used += 4; |
| 43 | } else { |
| 44 | dest[used++] = c; break; |
| 45 | } |
| 46 | } |
| 47 | } |
| 48 | |
| 49 | if (dest_len - used < 1) // make sure that there is room for \0 |
| 50 | return -1; |
| 51 | |
| 52 | dest[used] = '\0'; // doesn't count towards return value though |
| 53 | return used; |
| 54 | } |
| 55 | |
| 56 | |
| 57 | // ---------------------------------------------------------------------- |
| 58 | // CEscape() |
| 59 | // Copies 'src' to result, escaping dangerous characters using |
| 60 | // C-style escape sequences. 'src' and 'dest' should not overlap. |
| 61 | // ---------------------------------------------------------------------- |
| 62 | string CEscape(const StringPiece& src) { |
| 63 | const int dest_length = src.size() * 4 + 1; // Maximum possible expansion |
| 64 | char* dest = new char[dest_length]; |
| 65 | const int len = CEscapeString(src.data(), src.size(), |
| 66 | dest, dest_length); |
| 67 | string s = string(dest, len); |
| 68 | delete[] dest; |
| 69 | return s; |
| 70 | } |
| 71 | |
| 72 | string PrefixSuccessor(const StringPiece& prefix) { |
| 73 | // We can increment the last character in the string and be done |
| 74 | // unless that character is 255, in which case we have to erase the |
| 75 | // last character and increment the previous character, unless that |
| 76 | // is 255, etc. If the string is empty or consists entirely of |
| 77 | // 255's, we just return the empty string. |
| 78 | bool done = false; |
| 79 | string limit(prefix.data(), prefix.size()); |
| 80 | int index = limit.length() - 1; |
| 81 | while (!done && index >= 0) { |
| 82 | if ((limit[index]&255) == 255) { |
| 83 | limit.erase(index); |
| 84 | index--; |
| 85 | } else { |
| 86 | limit[index]++; |
| 87 | done = true; |
| 88 | } |
| 89 | } |
| 90 | if (!done) { |
| 91 | return ""; |
| 92 | } else { |
| 93 | return limit; |
| 94 | } |
| 95 | } |
| 96 | |
| 97 | } // namespace re2 |