summaryrefslogtreecommitdiffabout
path: root/lib
authorMichael Krelin <hacker@klever.net>2008-03-02 23:52:14 (UTC)
committer Michael Krelin <hacker@klever.net>2008-03-02 23:52:14 (UTC)
commit5fd5ecad8c2bd1e8846c11fa9b281f0f4ab8a4a7 (patch) (unidiff)
tree76ba24c589d0ce7de6cab113787cf7a70572f994 /lib
parentf6482fb9003e4953f838ba4ef2c110190355622f (diff)
downloadlibopkele-5fd5ecad8c2bd1e8846c11fa9b281f0f4ab8a4a7.zip
libopkele-5fd5ecad8c2bd1e8846c11fa9b281f0f4ab8a4a7.tar.gz
libopkele-5fd5ecad8c2bd1e8846c11fa9b281f0f4ab8a4a7.tar.bz2
aded util::url_decode()
Signed-off-by: Michael Krelin <hacker@klever.net>
Diffstat (limited to 'lib') (more/less context) (ignore whitespace changes)
-rw-r--r--lib/util.cc28
1 files changed, 28 insertions, 0 deletions
diff --git a/lib/util.cc b/lib/util.cc
index a6e08e2..3e7f3aa 100644
--- a/lib/util.cc
+++ b/lib/util.cc
@@ -147,96 +147,124 @@ namespace opkele {
147 tm_t.tm_year-=1900; 147 tm_t.tm_year-=1900;
148 time_t rv = mktime(&tm_t); 148 time_t rv = mktime(&tm_t);
149 if(rv==(time_t)-1) 149 if(rv==(time_t)-1)
150 throw failed_conversion(OPKELE_CP_ "failed to mktime()"); 150 throw failed_conversion(OPKELE_CP_ "failed to mktime()");
151 return rv-timezone; 151 return rv-timezone;
152 } 152 }
153 153
154 /* 154 /*
155 * 155 *
156 */ 156 */
157 157
158 static inline bool isrfc3986unreserved(int c) { 158 static inline bool isrfc3986unreserved(int c) {
159 if(c<'-') return false; 159 if(c<'-') return false;
160 if(c<='.') return true; 160 if(c<='.') return true;
161 if(c<'0') return false; if(c<='9') return true; 161 if(c<'0') return false; if(c<='9') return true;
162 if(c<'A') return false; if(c<='Z') return true; 162 if(c<'A') return false; if(c<='Z') return true;
163 if(c<'_') return false; 163 if(c<'_') return false;
164 if(c=='_') return true; 164 if(c=='_') return true;
165 if(c<'a') return false; if(c<='z') return true; 165 if(c<'a') return false; if(c<='z') return true;
166 if(c=='~') return true; 166 if(c=='~') return true;
167 return false; 167 return false;
168 } 168 }
169 169
170 struct __url_encoder : public unary_function<char,void> { 170 struct __url_encoder : public unary_function<char,void> {
171 public: 171 public:
172 string& rv; 172 string& rv;
173 173
174 __url_encoder(string& r) : rv(r) { } 174 __url_encoder(string& r) : rv(r) { }
175 175
176 result_type operator()(argument_type c) { 176 result_type operator()(argument_type c) {
177 if(isrfc3986unreserved(c)) 177 if(isrfc3986unreserved(c))
178 rv += c; 178 rv += c;
179 else{ 179 else{
180 char tmp[4]; 180 char tmp[4];
181 snprintf(tmp,sizeof(tmp),"%%%02X", 181 snprintf(tmp,sizeof(tmp),"%%%02X",
182 (c&0xff)); 182 (c&0xff));
183 rv += tmp; 183 rv += tmp;
184 } 184 }
185 } 185 }
186 }; 186 };
187 187
188 string url_encode(const string& str) { 188 string url_encode(const string& str) {
189 string rv; 189 string rv;
190 for_each(str.begin(),str.end(), 190 for_each(str.begin(),str.end(),
191 __url_encoder(rv)); 191 __url_encoder(rv));
192 return rv; 192 return rv;
193 } 193 }
194 194
195 string url_decode(const string& str) {
196 string rv;
197 back_insert_iterator<string> ii(rv);
198 for(string::const_iterator i=str.begin(),ie=str.end();
199 i!=ie;++i) {
200 switch(*i) {
201 case '+':
202 *(ii++) = ' '; break;
203 case '%':
204 ++i;
205 static char tmp[3] = {0,0,0};
206 if(i==ie)
207 throw failed_conversion(OPKELE_CP_ "trailing percent in the url-encoded string");
208 tmp[0] == *(i++);
209 if(i==ie)
210 throw failed_conversion(OPKELE_CP_ "not enough hexadecimals after the percent sign in url-encoded string");
211 tmp[1] == *i;
212 if(!(isxdigit(tmp[0]) && isxdigit(tmp[1])))
213 throw failed_conversion(OPKELE_CP_ "non-hex follows percent in url-encoded string");
214 *(ii++) = strtol(tmp,0,16);
215 break;
216 default:
217 *(ii++) = *i; break;
218 }
219 }
220 return rv;
221 }
222
195 string attr_escape(const string& str) { 223 string attr_escape(const string& str) {
196 static const char *unsafechars = "<>&\n\"'"; 224 static const char *unsafechars = "<>&\n\"'";
197 string rv; 225 string rv;
198 string::size_type p=0; 226 string::size_type p=0;
199 while(true) { 227 while(true) {
200 string::size_type us = str.find_first_of(unsafechars,p); 228 string::size_type us = str.find_first_of(unsafechars,p);
201 if(us==string::npos) { 229 if(us==string::npos) {
202 if(p!=str.length()) 230 if(p!=str.length())
203 rv.append(str,p,str.length()-p); 231 rv.append(str,p,str.length()-p);
204 return rv; 232 return rv;
205 } 233 }
206 rv.append(str,p,us-p); 234 rv.append(str,p,us-p);
207 rv += "&#"; 235 rv += "&#";
208 rv += long_to_string((long)str[us]); 236 rv += long_to_string((long)str[us]);
209 rv += ';'; 237 rv += ';';
210 p = us+1; 238 p = us+1;
211 } 239 }
212 } 240 }
213 241
214 string long_to_string(long l) { 242 string long_to_string(long l) {
215 char rv[32]; 243 char rv[32];
216 int r=snprintf(rv,sizeof(rv),"%ld",l); 244 int r=snprintf(rv,sizeof(rv),"%ld",l);
217 if(r<0 || r>=(int)sizeof(rv)) 245 if(r<0 || r>=(int)sizeof(rv))
218 throw failed_conversion(OPKELE_CP_ "failed to snprintf()"); 246 throw failed_conversion(OPKELE_CP_ "failed to snprintf()");
219 return rv; 247 return rv;
220 } 248 }
221 249
222 long string_to_long(const string& s) { 250 long string_to_long(const string& s) {
223 char *endptr = 0; 251 char *endptr = 0;
224 long rv = strtol(s.c_str(),&endptr,10); 252 long rv = strtol(s.c_str(),&endptr,10);
225 if((!endptr) || endptr==s.c_str()) 253 if((!endptr) || endptr==s.c_str())
226 throw failed_conversion(OPKELE_CP_ "failed to strtol()"); 254 throw failed_conversion(OPKELE_CP_ "failed to strtol()");
227 return rv; 255 return rv;
228 } 256 }
229 257
230 /* 258 /*
231 * Normalize URL according to the rules, described in rfc 3986, section 6 259 * Normalize URL according to the rules, described in rfc 3986, section 6
232 * 260 *
233 * - uppercase hex triplets (e.g. %ab -> %AB) 261 * - uppercase hex triplets (e.g. %ab -> %AB)
234 * - lowercase scheme and host 262 * - lowercase scheme and host
235 * - decode %-encoded characters, specified as unreserved in rfc 3986, section 2.3, 263 * - decode %-encoded characters, specified as unreserved in rfc 3986, section 2.3,
236 * that is - [:alpha:][:digit:]._~- 264 * that is - [:alpha:][:digit:]._~-
237 * - remove dot segments 265 * - remove dot segments
238 * - remove empty and default ports 266 * - remove empty and default ports
239 * - if there's no path component, add '/' 267 * - if there's no path component, add '/'
240 */ 268 */
241 string rfc_3986_normalize_uri(const string& uri) { 269 string rfc_3986_normalize_uri(const string& uri) {
242 string rv; 270 string rv;