NeoMutt
Teaching an old dog new tricks
DOXYGEN
Loading...
Searching...
No Matches
url.h File Reference

Parse and identify different URL schemes. More...

#include <stddef.h>
#include <stdint.h>
#include "mutt/lib.h"
+ Include dependency graph for url.h:
+ This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Data Structures

struct  UrlQuery
 Parsed Query String. More...
 
struct  Url
 A parsed URL proto://user:password@host:port/path?a=1&b=2 More...
 

Macros

#define U_NO_FLAGS   0
 
#define U_PATH   (1 << 1)
 

Enumerations

enum  UrlScheme {
  U_UNKNOWN , U_FILE , U_POP , U_POPS ,
  U_IMAP , U_IMAPS , U_NNTP , U_NNTPS ,
  U_SMTP , U_SMTPS , U_MAILTO , U_NOTMUCH
}
 All recognised Url types. More...
 

Functions

 STAILQ_HEAD (UrlQueryList, UrlQuery)
 
enum UrlScheme url_check_scheme (const char *s)
 Check the protocol of a URL.
 
void url_free (struct Url **ptr)
 Free the contents of a URL.
 
struct Urlurl_parse (const char *src)
 Fill in Url.
 
int url_pct_decode (char *s)
 Decode a percent-encoded string.
 
void url_pct_encode (char *buf, size_t buflen, const char *src)
 Percent-encode a string.
 
int url_tobuffer (struct Url *url, struct Buffer *dest, uint8_t flags)
 Output the URL string for a given Url object.
 
int url_tostring (struct Url *url, char *buf, size_t buflen, uint8_t flags)
 Output the URL string for a given Url object.
 

Detailed Description

Parse and identify different URL schemes.

Authors
  • Thomas Roessler

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

Definition in file url.h.

Macro Definition Documentation

◆ U_NO_FLAGS

#define U_NO_FLAGS   0

Definition at line 49 of file url.h.

◆ U_PATH

#define U_PATH   (1 << 1)

Definition at line 50 of file url.h.

Enumeration Type Documentation

◆ UrlScheme

enum UrlScheme

All recognised Url types.

Enumerator
U_UNKNOWN 

Url wasn't recognised.

U_FILE 

Url is file://.

U_POP 

Url is pop://.

U_POPS 

Url is pops://.

U_IMAP 

Url is imap://.

U_IMAPS 

Url is imaps://.

U_NNTP 

Url is nntp://.

U_NNTPS 

Url is nntps://.

U_SMTP 

Url is smtp://.

U_SMTPS 

Url is smtps://.

U_MAILTO 

Url is mailto://.

U_NOTMUCH 

Url is notmuch://.

Definition at line 33 of file url.h.

34{
35 U_UNKNOWN,
36 U_FILE,
37 U_POP,
38 U_POPS,
39 U_IMAP,
40 U_IMAPS,
41 U_NNTP,
42 U_NNTPS,
43 U_SMTP,
44 U_SMTPS,
45 U_MAILTO,
46 U_NOTMUCH,
47};
@ U_NOTMUCH
Url is notmuch://.
Definition: url.h:46
@ U_UNKNOWN
Url wasn't recognised.
Definition: url.h:35
@ U_FILE
Url is file://.
Definition: url.h:36
@ U_NNTPS
Url is nntps://.
Definition: url.h:42
@ U_MAILTO
Url is mailto://.
Definition: url.h:45
@ U_SMTPS
Url is smtps://.
Definition: url.h:44
@ U_SMTP
Url is smtp://.
Definition: url.h:43
@ U_NNTP
Url is nntp://.
Definition: url.h:41
@ U_IMAP
Url is imap://.
Definition: url.h:39
@ U_POPS
Url is pops://.
Definition: url.h:38
@ U_IMAPS
Url is imaps://.
Definition: url.h:40
@ U_POP
Url is pop://.
Definition: url.h:37

Function Documentation

◆ STAILQ_HEAD()

STAILQ_HEAD ( UrlQueryList  ,
UrlQuery   
)

◆ url_check_scheme()

enum UrlScheme url_check_scheme ( const char *  str)

Check the protocol of a URL.

Parameters
strString to check
Return values
enumUrlScheme, e.g. U_IMAPS

Definition at line 225 of file url.c.

226{
227 return get_scheme(str, mutt_prex_capture(PREX_URL, str));
228}
regmatch_t * mutt_prex_capture(enum Prex which, const char *str)
Match a precompiled regex against a string.
Definition: prex.c:289
@ PREX_URL
[imaps://user:pass@example.com/INBOX?foo=bar]
Definition: prex.h:33
static enum UrlScheme get_scheme(const char *src, const regmatch_t *match)
Extract the scheme part from a matched URL.
Definition: url.c:96
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ url_free()

void url_free ( struct Url **  ptr)

Free the contents of a URL.

Parameters
ptrUrl to free

Definition at line 123 of file url.c.

124{
125 if (!ptr || !*ptr)
126 return;
127
128 struct Url *url = *ptr;
129
130 struct UrlQueryList *l = &url->query_strings;
131 while (!STAILQ_EMPTY(l))
132 {
133 struct UrlQuery *np = STAILQ_FIRST(l);
134 STAILQ_REMOVE_HEAD(l, entries);
135 // Don't free 'name', 'value': they are pointers into the 'src' string
136 FREE(&np);
137 }
138
139 FREE(&url->src);
140 FREE(ptr);
141}
#define FREE(x)
Definition: memory.h:45
#define STAILQ_REMOVE_HEAD(head, field)
Definition: queue.h:422
#define STAILQ_FIRST(head)
Definition: queue.h:350
#define STAILQ_EMPTY(head)
Definition: queue.h:348
Parsed Query String.
Definition: url.h:58
A parsed URL proto://user:password@host:port/path?a=1&b=2
Definition: url.h:69
struct UrlQueryList query_strings
List of query strings.
Definition: url.h:76
char * src
Raw URL string.
Definition: url.h:77
+ Here is the caller graph for this function:

◆ url_parse()

struct Url * url_parse ( const char *  src)

Fill in Url.

Parameters
srcString to parse
Return values
ptrParsed URL
NULLError
Note
Caller must free returned Url with url_free()

Definition at line 238 of file url.c.

239{
240 const regmatch_t *match = mutt_prex_capture(PREX_URL, src);
241 if (!match)
242 return NULL;
243
244 enum UrlScheme scheme = get_scheme(src, match);
245 if (scheme == U_UNKNOWN)
246 return NULL;
247
248 const regmatch_t *userinfo = &match[PREX_URL_MATCH_USERINFO];
249 const regmatch_t *user = &match[PREX_URL_MATCH_USER];
250 const regmatch_t *pass = &match[PREX_URL_MATCH_PASS];
251 const regmatch_t *host = &match[PREX_URL_MATCH_HOSTNAME];
252 const regmatch_t *ipvx = &match[PREX_URL_MATCH_HOSTIPVX];
253 const regmatch_t *port = &match[PREX_URL_MATCH_PORT];
254 const regmatch_t *path = &match[PREX_URL_MATCH_PATH];
255 const regmatch_t *query = &match[PREX_URL_MATCH_QUERY];
256 const regmatch_t *pathonly = &match[PREX_URL_MATCH_PATH_ONLY];
257
258 struct Url *url = url_new();
259 url->scheme = scheme;
260 url->src = mutt_str_dup(src);
261
262 /* If the scheme is not followed by two forward slashes, then it's a simple
263 * path (see https://tools.ietf.org/html/rfc3986#section-3). */
264 if (mutt_regmatch_start(pathonly) != -1)
265 {
266 url->src[mutt_regmatch_end(pathonly)] = '\0';
267 url->path = url->src + mutt_regmatch_start(pathonly);
268 if (url_pct_decode(url->path) < 0)
269 goto err;
270 }
271
272 /* separate userinfo part */
273 if (mutt_regmatch_end(userinfo) != -1)
274 {
275 url->src[mutt_regmatch_end(userinfo) - 1] = '\0';
276 }
277
278 /* user */
279 if (mutt_regmatch_end(user) != -1)
280 {
281 url->src[mutt_regmatch_end(user)] = '\0';
282 url->user = url->src + mutt_regmatch_start(user);
283 if (url_pct_decode(url->user) < 0)
284 goto err;
285 }
286
287 /* pass */
288 if (mutt_regmatch_end(pass) != -1)
289 {
290 url->pass = url->src + mutt_regmatch_start(pass);
291 if (url_pct_decode(url->pass) < 0)
292 goto err;
293 }
294
295 /* host */
296 if (mutt_regmatch_len(host) != 0)
297 {
298 url->host = url->src + mutt_regmatch_start(host);
299 url->src[mutt_regmatch_end(host)] = '\0';
300 }
301 else if (mutt_regmatch_end(ipvx) != -1)
302 {
303 url->host = url->src + mutt_regmatch_start(ipvx) + 1; /* skip opening '[' */
304 url->src[mutt_regmatch_end(ipvx) - 1] = '\0'; /* skip closing ']' */
305 }
306
307 /* port */
308 if (mutt_regmatch_end(port) != -1)
309 {
310 url->src[mutt_regmatch_end(port)] = '\0';
311 const char *ports = url->src + mutt_regmatch_start(port);
312 unsigned short num;
313 if (!mutt_str_atous_full(ports, &num))
314 {
315 goto err;
316 }
317 url->port = num;
318 }
319
320 /* path */
321 if (mutt_regmatch_end(path) != -1)
322 {
323 url->src[mutt_regmatch_end(path)] = '\0';
324 url->path = url->src + mutt_regmatch_start(path);
325 if (!url->host)
326 {
327 /* If host is not provided, restore the '/': this is an absolute path */
328 *(--url->path) = '/';
329 }
330 if (url_pct_decode(url->path) < 0)
331 goto err;
332 }
333
334 /* query */
335 if (mutt_regmatch_end(query) != -1)
336 {
337 char *squery = url->src + mutt_regmatch_start(query);
338 if (!parse_query_string(&url->query_strings, squery))
339 goto err;
340 }
341
342 return url;
343
344err:
345 url_free(&url);
346 return NULL;
347}
char * mutt_str_dup(const char *str)
Copy a string, safely.
Definition: string.c:251
@ PREX_URL_MATCH_USER
...//[user]:pass@...
Definition: prex.h:58
@ PREX_URL_MATCH_QUERY
...Inbox?[foo=bar&baz=value]
Definition: prex.h:70
@ PREX_URL_MATCH_HOSTNAME
imaps://...[host.com]...
Definition: prex.h:62
@ PREX_URL_MATCH_PORT
imaps://host.com:[993]/...
Definition: prex.h:65
@ PREX_URL_MATCH_PATH_ONLY
mailto:[me@example.com]?foo=bar
Definition: prex.h:68
@ PREX_URL_MATCH_USERINFO
...//[user:pass@]...
Definition: prex.h:57
@ PREX_URL_MATCH_PATH
...:993/[Inbox]
Definition: prex.h:67
@ PREX_URL_MATCH_HOSTIPVX
imaps://...[127.0.0.1]...
Definition: prex.h:63
@ PREX_URL_MATCH_PASS
...//user:[pass]@...
Definition: prex.h:60
static size_t mutt_regmatch_len(const regmatch_t *match)
Return the length of a match.
Definition: regex3.h:80
static regoff_t mutt_regmatch_end(const regmatch_t *match)
Return the end of a match.
Definition: regex3.h:70
static regoff_t mutt_regmatch_start(const regmatch_t *match)
Return the start of a match.
Definition: regex3.h:60
char * user
Username.
Definition: url.h:71
unsigned short port
Port.
Definition: url.h:74
char * host
Host.
Definition: url.h:73
char * pass
Password.
Definition: url.h:72
char * path
Path.
Definition: url.h:75
enum UrlScheme scheme
Scheme, e.g. U_SMTPS.
Definition: url.h:70
int url_pct_decode(char *s)
Decode a percent-encoded string.
Definition: url.c:189
static bool parse_query_string(struct UrlQueryList *list, char *src)
Parse a URL query string.
Definition: url.c:55
void url_free(struct Url **ptr)
Free the contents of a URL.
Definition: url.c:123
static struct Url * url_new(void)
Create a Url.
Definition: url.c:112
UrlScheme
All recognised Url types.
Definition: url.h:34
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ url_pct_decode()

int url_pct_decode ( char *  s)

Decode a percent-encoded string.

Parameters
sString to decode
Return values
0Success
-1Error

e.g. turn "hello%20world" into "hello world" The string is decoded in-place.

Definition at line 189 of file url.c.

190{
191 if (!s)
192 return -1;
193
194 char *d = NULL;
195
196 for (d = s; *s; s++)
197 {
198 if (*s == '%')
199 {
200 if ((s[1] != '\0') && (s[2] != '\0') && isxdigit((unsigned char) s[1]) &&
201 isxdigit((unsigned char) s[2]) && (hexval(s[1]) >= 0) && (hexval(s[2]) >= 0))
202 {
203 *d++ = (hexval(s[1]) << 4) | (hexval(s[2]));
204 s += 2;
205 }
206 else
207 {
208 return -1;
209 }
210 }
211 else
212 {
213 *d++ = *s;
214 }
215 }
216 *d = '\0';
217 return 0;
218}
#define hexval(ch)
Definition: mime.h:80
+ Here is the caller graph for this function:

◆ url_pct_encode()

void url_pct_encode ( char *  buf,
size_t  buflen,
const char *  src 
)

Percent-encode a string.

Parameters
bufBuffer for the result
buflenLength of buffer
srcString to encode

e.g. turn "hello world" into "hello%20world"

Definition at line 151 of file url.c.

152{
153 static const char *hex = "0123456789ABCDEF";
154
155 if (!buf)
156 return;
157
158 *buf = '\0';
159 buflen--;
160 while (src && *src && (buflen != 0))
161 {
162 if (strchr(" /:&%=", *src))
163 {
164 if (buflen < 3)
165 break;
166
167 *buf++ = '%';
168 *buf++ = hex[(*src >> 4) & 0xf];
169 *buf++ = hex[*src & 0xf];
170 src++;
171 buflen -= 3;
172 continue;
173 }
174 *buf++ = *src++;
175 buflen--;
176 }
177 *buf = '\0';
178}
+ Here is the caller graph for this function:

◆ url_tobuffer()

int url_tobuffer ( struct Url url,
struct Buffer buf,
uint8_t  flags 
)

Output the URL string for a given Url object.

Parameters
urlUrl to turn into a string
bufBuffer for the result
flagsFlags, e.g. U_PATH
Return values
0Success
-1Error

Definition at line 357 of file url.c.

358{
359 if (!url || !buf)
360 return -1;
361 if (url->scheme == U_UNKNOWN)
362 return -1;
363
364 buf_printf(buf, "%s:", mutt_map_get_name(url->scheme, UrlMap));
365
366 if (url->host)
367 {
368 if (!(flags & U_PATH))
369 buf_addstr(buf, "//");
370
371 if (url->user && (url->user[0] || !(flags & U_PATH)))
372 {
373 char str[256] = { 0 };
374 url_pct_encode(str, sizeof(str), url->user);
375 buf_add_printf(buf, "%s@", str);
376 }
377
378 if (strchr(url->host, ':'))
379 buf_add_printf(buf, "[%s]", url->host);
380 else
381 buf_addstr(buf, url->host);
382
383 if (url->port)
384 buf_add_printf(buf, ":%hu/", url->port);
385 else
386 buf_addstr(buf, "/");
387 }
388
389 if (url->path)
390 buf_addstr(buf, url->path);
391
392 if (STAILQ_FIRST(&url->query_strings))
393 {
394 buf_addstr(buf, "?");
395
396 char str[256] = { 0 };
397 struct UrlQuery *np = NULL;
398 STAILQ_FOREACH(np, &url->query_strings, entries)
399 {
400 url_pct_encode(str, sizeof(str), np->name);
401 buf_addstr(buf, str);
402 buf_addstr(buf, "=");
403 url_pct_encode(str, sizeof(str), np->value);
404 buf_addstr(buf, str);
405 if (STAILQ_NEXT(np, entries))
406 buf_addstr(buf, "&");
407 }
408 }
409
410 return 0;
411}
int buf_printf(struct Buffer *buf, const char *fmt,...)
Format a string overwriting a Buffer.
Definition: buffer.c:173
int buf_add_printf(struct Buffer *buf, const char *fmt,...)
Format a string appending a Buffer.
Definition: buffer.c:216
size_t buf_addstr(struct Buffer *buf, const char *s)
Add a string to a Buffer.
Definition: buffer.c:238
const char * mutt_map_get_name(int val, const struct Mapping *map)
Lookup a string for a constant.
Definition: mapping.c:42
#define STAILQ_FOREACH(var, head, field)
Definition: queue.h:352
#define STAILQ_NEXT(elm, field)
Definition: queue.h:400
char * name
Query name.
Definition: url.h:59
char * value
Query value.
Definition: url.h:60
static const struct Mapping UrlMap[]
Constants for URL protocols.
Definition: url.c:40
void url_pct_encode(char *buf, size_t buflen, const char *src)
Percent-encode a string.
Definition: url.c:151
#define U_PATH
Definition: url.h:50
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ url_tostring()

int url_tostring ( struct Url url,
char *  dest,
size_t  len,
uint8_t  flags 
)

Output the URL string for a given Url object.

Parameters
urlUrl to turn into a string
destBuffer for the result
lenLength of buffer
flagsFlags, e.g. U_PATH
Return values
0Success
-1Error

Definition at line 422 of file url.c.

423{
424 if (!url || !dest)
425 return -1;
426
427 struct Buffer *dest_buf = buf_pool_get();
428
429 int rc = url_tobuffer(url, dest_buf, flags);
430 if (rc == 0)
431 mutt_str_copy(dest, buf_string(dest_buf), len);
432
433 buf_pool_release(&dest_buf);
434
435 return rc;
436}
static const char * buf_string(const struct Buffer *buf)
Convert a buffer to a const char * "string".
Definition: buffer.h:93
size_t mutt_str_copy(char *dest, const char *src, size_t dsize)
Copy a string into a buffer (guaranteeing NUL-termination)
Definition: string.c:653
struct Buffer * buf_pool_get(void)
Get a Buffer from the pool.
Definition: pool.c:81
void buf_pool_release(struct Buffer **ptr)
Return a Buffer to the pool.
Definition: pool.c:94
String manipulation buffer.
Definition: buffer.h:34
int url_tobuffer(struct Url *url, struct Buffer *buf, uint8_t flags)
Output the URL string for a given Url object.
Definition: url.c:357
+ Here is the call graph for this function:
+ Here is the caller graph for this function: