1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
|
/*****************************************************************************
* _ _ ____ _
* Project ___| | | | _ \| |
* / __| | | | |_) | |
* | (__| |_| | _ <| |___
* \___|\___/|_| \_\_____|
*
* $Id$
*
* This example source code introduces an fopen()/fread()/fclose() emulation
* for URL reads. Using an approach similar to this, you could replace your
* program's fopen() with this url_fopen() and fread() with url_fread() and
* it should be possible to read remote streams instead of (only) local files.
*
* See the main() function at the bottom that shows a tiny app in action.
*
* This source code is a proof of concept. It will need further attention to
* become production-use useful and solid.
*
* This example requires libcurl 7.9.7 or later.
*/
#include <stdio.h>
#include <string.h>
#include <sys/time.h>
#include <curl/curl.h>
#include <curl/types.h>
#include <curl/easy.h>
struct data {
int type;
union {
CURL *curl;
FILE *file;
} handle;
/* TODO: We should perhaps document the biggest possible buffer chunk we can
get from libcurl in one single callback... */
char buffer[CURL_MAX_WRITE_SIZE];
char *readptr; /* read from here */
int bytes; /* bytes available from read pointer */
CURLMcode m; /* stored from a previous url_fread() */
};
typedef struct data URL_FILE;
/* we use a global one for convenience */
CURLM *multi_handle;
static
size_t write_callback(char *buffer,
size_t size,
size_t nitems,
void *userp)
{
URL_FILE *url = (URL_FILE *)userp;
size *= nitems;
memcpy(url->readptr, buffer, size);
url->readptr += size;
url->bytes += size;
return size;
}
URL_FILE *url_fopen(char *url, char *operation)
{
/* this code could check for URLs or types in the 'url' and
basicly use the real fopen() for standard files */
URL_FILE *file;
int still_running;
file = (URL_FILE *)malloc(sizeof(URL_FILE));
if(!file)
return NULL;
memset(file, 0, sizeof(URL_FILE));
file->type = 1; /* marked as URL, use 0 for plain file */
file->handle.curl = curl_easy_init();
curl_easy_setopt(file->handle.curl, CURLOPT_URL, url);
curl_easy_setopt(file->handle.curl, CURLOPT_FILE, file);
curl_easy_setopt(file->handle.curl, CURLOPT_VERBOSE, FALSE);
curl_easy_setopt(file->handle.curl, CURLOPT_WRITEFUNCTION, write_callback);
if(!multi_handle)
multi_handle = curl_multi_init();
curl_multi_add_handle(multi_handle, file->handle.curl);
while(CURLM_CALL_MULTI_PERFORM ==
curl_multi_perform(multi_handle, &still_running));
/* if still_running would be 0 now, we should return NULL */
return file;
}
void url_fclose(URL_FILE *file)
{
/* make sure the easy handle is not in the multi handle anymore */
curl_multi_remove_handle(multi_handle, file->handle.curl);
/* cleanup */
curl_easy_cleanup(file->handle.curl);
}
size_t url_fread(void *ptr, size_t size, size_t nmemb, URL_FILE *file)
{
fd_set fdread;
fd_set fdwrite;
fd_set fdexcep;
int maxfd;
struct timeval timeout;
int rc;
int still_running = 0;
if(!file->bytes) { /* no data available at this point */
file->readptr = file->buffer; /* reset read pointer */
if(CURLM_CALL_MULTI_PERFORM == file->m) {
while(CURLM_CALL_MULTI_PERFORM ==
curl_multi_perform(multi_handle, &still_running)) {
if(file->bytes) {
printf("(fread) WOAH! THis happened!\n");
break;
}
}
if(!still_running) {
printf("NO MORE RUNNING AROUND!\n");
return 0;
}
}
FD_ZERO(&fdread);
FD_ZERO(&fdwrite);
FD_ZERO(&fdexcep);
/* set a suitable timeout to fail on */
timeout.tv_sec = 500; /* 5 minutes */
timeout.tv_usec = 0;
/* get file descriptors from the transfers */
curl_multi_fdset(multi_handle, &fdread, &fdwrite, &fdexcep, &maxfd);
rc = select(maxfd+1, &fdread, &fdwrite, &fdexcep, &timeout);
switch(rc) {
case -1:
/* select error */
break;
case 0:
break;
default:
/* timeout or readable/writable sockets */
do {
file->m = curl_multi_perform(multi_handle, &still_running);
if(file->bytes)
/* we have received data, return that now */
break;
} while(CURLM_CALL_MULTI_PERFORM == file->m);
if(!still_running)
printf("NO MORE RUNNING AROUND!\n");
break;
}
}
else
printf("(fread) Skip network read\n");
if(file->bytes) {
/* data already available, return that */
int want = size * nmemb;
if(file->bytes < want)
want = file->bytes;
memcpy(ptr, file->readptr, want);
file->readptr += want;
file->bytes -= want;
printf("(fread) return %d bytes\n", want);
return want;
}
return 0; /* no data available to return */
}
int main(int argc, char *argv[])
{
URL_FILE *handle;
int nread;
char buffer[256];
handle = url_fopen("http://www.haxx.se", "r");
if(!handle) {
printf("couldn't url_fopen()\n");
}
do {
nread = url_fread(buffer, sizeof(buffer), 1, handle);
printf("We got: %d bytes\n", nread);
} while(nread);
url_fclose(handle);
return 0;
}
|
