1/*
2 *  Copyright 2004 The WebRTC Project Authors. All rights reserved.
3 *
4 *  Use of this source code is governed by a BSD-style license
5 *  that can be found in the LICENSE file in the root of the source
6 *  tree. An additional intellectual property rights grant can be found
7 *  in the file PATENTS.  All contributing project authors may
8 *  be found in the AUTHORS file in the root of the source tree.
9 */
10
11#ifndef WEBRTC_BASE_STRINGENCODE_H_
12#define WEBRTC_BASE_STRINGENCODE_H_
13
14#include <sstream>
15#include <string>
16#include <vector>
17
18#include "webrtc/base/checks.h"
19
20namespace rtc {
21
22//////////////////////////////////////////////////////////////////////
23// String Encoding Utilities
24//////////////////////////////////////////////////////////////////////
25
26// Convert an unsigned value to it's utf8 representation.  Returns the length
27// of the encoded string, or 0 if the encoding is longer than buflen - 1.
28size_t utf8_encode(char* buffer, size_t buflen, unsigned long value);
29// Decode the utf8 encoded value pointed to by source.  Returns the number of
30// bytes used by the encoding, or 0 if the encoding is invalid.
31size_t utf8_decode(const char* source, size_t srclen, unsigned long* value);
32
33// Escaping prefixes illegal characters with the escape character.  Compact, but
34// illegal characters still appear in the string.
35size_t escape(char * buffer, size_t buflen,
36              const char * source, size_t srclen,
37              const char * illegal, char escape);
38// Note: in-place unescaping (buffer == source) is allowed.
39size_t unescape(char * buffer, size_t buflen,
40                const char * source, size_t srclen,
41                char escape);
42
43// Encoding replaces illegal characters with the escape character and 2 hex
44// chars, so it's a little less compact than escape, but completely removes
45// illegal characters.  note that hex digits should not be used as illegal
46// characters.
47size_t encode(char * buffer, size_t buflen,
48              const char * source, size_t srclen,
49              const char * illegal, char escape);
50// Note: in-place decoding (buffer == source) is allowed.
51size_t decode(char * buffer, size_t buflen,
52              const char * source, size_t srclen,
53              char escape);
54
55// Returns a list of characters that may be unsafe for use in the name of a
56// file, suitable for passing to the 'illegal' member of escape or encode.
57const char* unsafe_filename_characters();
58
59// url_encode is an encode operation with a predefined set of illegal characters
60// and escape character (for use in URLs, obviously).
61size_t url_encode(char * buffer, size_t buflen,
62                  const char * source, size_t srclen);
63// Note: in-place decoding (buffer == source) is allowed.
64size_t url_decode(char * buffer, size_t buflen,
65                  const char * source, size_t srclen);
66
67// html_encode prevents data embedded in html from containing markup.
68size_t html_encode(char * buffer, size_t buflen,
69                   const char * source, size_t srclen);
70// Note: in-place decoding (buffer == source) is allowed.
71size_t html_decode(char * buffer, size_t buflen,
72                   const char * source, size_t srclen);
73
74// xml_encode makes data suitable for inside xml attributes and values.
75size_t xml_encode(char * buffer, size_t buflen,
76                  const char * source, size_t srclen);
77// Note: in-place decoding (buffer == source) is allowed.
78size_t xml_decode(char * buffer, size_t buflen,
79                  const char * source, size_t srclen);
80
81// Convert an unsigned value from 0 to 15 to the hex character equivalent...
82char hex_encode(unsigned char val);
83// ...and vice-versa.
84bool hex_decode(char ch, unsigned char* val);
85
86// hex_encode shows the hex representation of binary data in ascii.
87size_t hex_encode(char* buffer, size_t buflen,
88                  const char* source, size_t srclen);
89
90// hex_encode, but separate each byte representation with a delimiter.
91// |delimiter| == 0 means no delimiter
92// If the buffer is too short, we return 0
93size_t hex_encode_with_delimiter(char* buffer, size_t buflen,
94                                 const char* source, size_t srclen,
95                                 char delimiter);
96
97// Helper functions for hex_encode.
98std::string hex_encode(const std::string& str);
99std::string hex_encode(const char* source, size_t srclen);
100std::string hex_encode_with_delimiter(const char* source, size_t srclen,
101                                      char delimiter);
102
103// hex_decode converts ascii hex to binary.
104size_t hex_decode(char* buffer, size_t buflen,
105                  const char* source, size_t srclen);
106
107// hex_decode, assuming that there is a delimiter between every byte
108// pair.
109// |delimiter| == 0 means no delimiter
110// If the buffer is too short or the data is invalid, we return 0.
111size_t hex_decode_with_delimiter(char* buffer, size_t buflen,
112                                 const char* source, size_t srclen,
113                                 char delimiter);
114
115// Helper functions for hex_decode.
116size_t hex_decode(char* buffer, size_t buflen, const std::string& source);
117size_t hex_decode_with_delimiter(char* buffer, size_t buflen,
118                                 const std::string& source, char delimiter);
119
120// Apply any suitable string transform (including the ones above) to an STL
121// string.  Stack-allocated temporary space is used for the transformation,
122// so value and source may refer to the same string.
123typedef size_t (*Transform)(char * buffer, size_t buflen,
124                            const char * source, size_t srclen);
125size_t transform(std::string& value, size_t maxlen, const std::string& source,
126                 Transform t);
127
128// Return the result of applying transform t to source.
129std::string s_transform(const std::string& source, Transform t);
130
131// Convenience wrappers.
132inline std::string s_url_encode(const std::string& source) {
133  return s_transform(source, url_encode);
134}
135inline std::string s_url_decode(const std::string& source) {
136  return s_transform(source, url_decode);
137}
138
139// Splits the source string into multiple fields separated by delimiter,
140// with duplicates of delimiter creating empty fields.
141size_t split(const std::string& source, char delimiter,
142             std::vector<std::string>* fields);
143
144// Splits the source string into multiple fields separated by delimiter,
145// with duplicates of delimiter ignored.  Trailing delimiter ignored.
146size_t tokenize(const std::string& source, char delimiter,
147                std::vector<std::string>* fields);
148
149// Tokenize, including the empty tokens.
150size_t tokenize_with_empty_tokens(const std::string& source,
151                                  char delimiter,
152                                  std::vector<std::string>* fields);
153
154// Tokenize and append the tokens to fields. Return the new size of fields.
155size_t tokenize_append(const std::string& source, char delimiter,
156                       std::vector<std::string>* fields);
157
158// Splits the source string into multiple fields separated by delimiter, with
159// duplicates of delimiter ignored. Trailing delimiter ignored. A substring in
160// between the start_mark and the end_mark is treated as a single field. Return
161// the size of fields. For example, if source is "filename
162// \"/Library/Application Support/media content.txt\"", delimiter is ' ', and
163// the start_mark and end_mark are '"', this method returns two fields:
164// "filename" and "/Library/Application Support/media content.txt".
165size_t tokenize(const std::string& source, char delimiter, char start_mark,
166                char end_mark, std::vector<std::string>* fields);
167
168// Extract the first token from source as separated by delimiter, with
169// duplicates of delimiter ignored. Return false if the delimiter could not be
170// found, otherwise return true.
171bool tokenize_first(const std::string& source,
172                    const char delimiter,
173                    std::string* token,
174                    std::string* rest);
175
176// Safe sprintf to std::string
177//void sprintf(std::string& value, size_t maxlen, const char * format, ...)
178//     PRINTF_FORMAT(3);
179
180// Convert arbitrary values to/from a string.
181
182template <class T>
183static bool ToString(const T &t, std::string* s) {
184  RTC_DCHECK(s);
185  std::ostringstream oss;
186  oss << std::boolalpha << t;
187  *s = oss.str();
188  return !oss.fail();
189}
190
191template <class T>
192static bool FromString(const std::string& s, T* t) {
193  RTC_DCHECK(t);
194  std::istringstream iss(s);
195  iss >> std::boolalpha >> *t;
196  return !iss.fail();
197}
198
199// Inline versions of the string conversion routines.
200
201template<typename T>
202static inline std::string ToString(const T& val) {
203  std::string str; ToString(val, &str); return str;
204}
205
206template<typename T>
207static inline T FromString(const std::string& str) {
208  T val; FromString(str, &val); return val;
209}
210
211template<typename T>
212static inline T FromString(const T& defaultValue, const std::string& str) {
213  T val(defaultValue); FromString(str, &val); return val;
214}
215
216// simple function to strip out characters which shouldn't be
217// used in filenames
218char make_char_safe_for_filename(char c);
219
220//////////////////////////////////////////////////////////////////////
221
222}  // namespace rtc
223
224#endif  // WEBRTC_BASE_STRINGENCODE_H__
225