mirrors/netsurf
1
0
Fork 0
mirror of https://github.com/netsurf-browser/netsurf synced 2024-12-21 03:32:35 +03:00
Code Issues Packages Projects Releases Wiki Activity
ad273a41e8
netsurf/content/fetchers/curl.c
Michael Drake 4055ecf796 Tiny simplification of curl debug logging.
2016-01-21 09:27:27 +00:00

1406 lines
37 KiB
C
Raw Blame History

/*
* Copyright 2006 Daniel Silverstone <dsilvers@digital-scurf.org>
* Copyright 2007 James Bursa <bursa@users.sourceforge.net>
* Copyright 2003 Phil Mellor <monkeyson@users.sourceforge.net>
*
* This file is part of NetSurf.
*
* NetSurf is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; version 2 of the License.
*
* NetSurf 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/>.
*/
/**
* \file
* implementation of fetching of data from http and https schemes.
*
* This implementation uses libcurl's 'multi' interface.
*
* The CURL handles are cached in the curl_handle_ring.
*/
#include "utils/config.h"
#include <assert.h>
#include <errno.h>
#include <inttypes.h>
#include <stdbool.h>
#include <string.h>
#include <strings.h>
#include <time.h>
#include <sys/stat.h>
#include <openssl/ssl.h>
#include <libwapcaplet/libwapcaplet.h>
#include "utils/corestrings.h"
#include "utils/nsoption.h"
#include "utils/log.h"
#include "utils/messages.h"
#include "utils/utils.h"
#include "utils/ring.h"
#include "utils/useragent.h"
#include "utils/file.h"
#include "desktop/gui_fetch.h"
#include "desktop/gui_internal.h"
#include "content/fetch.h"
#include "content/fetchers.h"
#include "content/fetchers/curl.h"
#include "content/urldb.h"
/** SSL certificate info */
struct cert_info {
X509 *cert; /**< Pointer to certificate */
long err; /**< OpenSSL error code */
};
/** Information for a single fetch. */
struct curl_fetch_info {
struct fetch *fetch_handle; /**< The fetch handle we're parented by. */
CURL * curl_handle; /**< cURL handle if being fetched, or 0. */
bool had_headers; /**< Headers have been processed. */
bool abort; /**< Abort requested. */
bool stopped; /**< Download stopped on purpose. */
bool only_2xx; /**< Only HTTP 2xx responses acceptable. */
bool downgrade_tls; /**< Downgrade to TLS <= 1.0 */
nsurl *url; /**< URL of this fetch. */
lwc_string *host; /**< The hostname of this fetch. */
struct curl_slist *headers; /**< List of request headers. */
char *location; /**< Response Location header, or 0. */
unsigned long content_length; /**< Response Content-Length, or 0. */
char *cookie_string; /**< Cookie string for this fetch */
char *realm; /**< HTTP Auth Realm */
char *post_urlenc; /**< Url encoded POST string, or 0. */
long http_code; /**< HTTP result code from cURL. */
struct curl_httppost *post_multipart; /**< Multipart post data, or 0. */
#define MAX_CERTS 10
struct cert_info cert_data[MAX_CERTS]; /**< HTTPS certificate data */
unsigned int last_progress_update; /**< Time of last progress update */
};
struct cache_handle {
CURL *handle; /**< The cached cURL handle */
lwc_string *host; /**< The host for which this handle is cached */
struct cache_handle *r_prev; /**< Previous cached handle in ring. */
struct cache_handle *r_next; /**< Next cached handle in ring. */
};
CURLM *fetch_curl_multi; /**< Global cURL multi handle. */
/** Curl handle with default options set; not used for transfers. */
static CURL *fetch_blank_curl;
static struct cache_handle *curl_handle_ring = 0; /**< Ring of cached handles */
static int curl_fetchers_registered = 0;
static bool curl_with_openssl;
static char fetch_error_buffer[CURL_ERROR_SIZE]; /**< Error buffer for cURL. */
static char fetch_proxy_userpwd[100]; /**< Proxy authentication details. */
/**
* Initialise a cURL fetcher.
*/
static bool fetch_curl_initialise(lwc_string *scheme)
{
LOG("Initialise cURL fetcher for %s", lwc_string_data(scheme));
curl_fetchers_registered++;
return true; /* Always succeeds */
}
/**
* Finalise a cURL fetcher.
*
* \param scheme The scheme to finalise.
*/
static void fetch_curl_finalise(lwc_string *scheme)
{
struct cache_handle *h;
curl_fetchers_registered--;
LOG("Finalise cURL fetcher %s", lwc_string_data(scheme));
if (curl_fetchers_registered == 0) {
CURLMcode codem;
/* All the fetchers have been finalised. */
LOG("All cURL fetchers finalised, closing down cURL");
curl_easy_cleanup(fetch_blank_curl);
codem = curl_multi_cleanup(fetch_curl_multi);
if (codem != CURLM_OK)
LOG("curl_multi_cleanup failed: ignoring");
curl_global_cleanup();
}
/* Free anything remaining in the cached curl handle ring */
while (curl_handle_ring != NULL) {
h = curl_handle_ring;
RING_REMOVE(curl_handle_ring, h);
lwc_string_unref(h->host);
curl_easy_cleanup(h->handle);
free(h);
}
}
/**
* Check if this fetcher can fetch a url.
*
* \param url The url to check.
* \return true if the fetcher supports the url else false.
*/
static bool fetch_curl_can_fetch(const nsurl *url)
{
return nsurl_has_component(url, NSURL_HOST);
}
/**
* Convert a list of struct ::fetch_multipart_data to a list of
* struct curl_httppost for libcurl.
*/
static struct curl_httppost *
fetch_curl_post_convert(const struct fetch_multipart_data *control)
{
struct curl_httppost *post = 0, *last = 0;
CURLFORMcode code;
nserror ret;
for (; control; control = control->next) {
if (control->file) {
char *leafname = NULL;
ret = guit->file->basename(control->value, &leafname, NULL);
if (ret != NSERROR_OK) {
continue;
}
/* We have to special case filenames of "", so curl
* a) actually attempts the fetch and
* b) doesn't attempt to open the file ""
*/
if (control->value[0] == '\0') {
/* dummy buffer - needs to be static so
* pointer's still valid when we go out
* of scope (not that libcurl should be
* attempting to access it, of course).
*/
static char buf;
code = curl_formadd(&post, &last,
CURLFORM_COPYNAME, control->name,
CURLFORM_BUFFER, control->value,
/* needed, as basename("") == "." */
CURLFORM_FILENAME, "",
CURLFORM_BUFFERPTR, &buf,
CURLFORM_BUFFERLENGTH, 0,
CURLFORM_CONTENTTYPE,
"application/octet-stream",
CURLFORM_END);
if (code != CURL_FORMADD_OK)
LOG("curl_formadd: %d (%s)", code, control->name);
} else {
char *mimetype = guit->fetch->mimetype(control->value);
code = curl_formadd(&post, &last,
CURLFORM_COPYNAME, control->name,
CURLFORM_FILE, control->rawfile,
CURLFORM_FILENAME, leafname,
CURLFORM_CONTENTTYPE,
(mimetype != 0 ? mimetype : "text/plain"),
CURLFORM_END);
if (code != CURL_FORMADD_OK)
LOG("curl_formadd: %d (%s=%s)", code, control->name, control->value);
free(mimetype);
}
free(leafname);
}
else {
code = curl_formadd(&post, &last,
CURLFORM_COPYNAME, control->name,
CURLFORM_COPYCONTENTS, control->value,
CURLFORM_END);
if (code != CURL_FORMADD_OK)
LOG("curl_formadd: %d (%s=%s)", code, control->name, control->value);
}
}
return post;
}
/**
* Start fetching data for the given URL.
*
* The function returns immediately. The fetch may be queued for later
* processing.
*
* A pointer to an opaque struct curl_fetch_info is returned, which can be
* passed to fetch_abort() to abort the fetch at any time. Returns 0 if memory
* is exhausted (or some other fatal error occurred).
*
* The caller must supply a callback function which is called when anything
* interesting happens. The callback function is first called with msg
* FETCH_HEADER, with the header in data, then one or more times
* with FETCH_DATA with some data for the url, and finally with
* FETCH_FINISHED. Alternatively, FETCH_ERROR indicates an error occurred:
* data contains an error message. FETCH_REDIRECT may replace the FETCH_HEADER,
* FETCH_DATA, FETCH_FINISHED sequence if the server sends a replacement URL.
*
* Some private data can be passed as the last parameter to fetch_start, and
* callbacks will contain this.
*/
static void *
fetch_curl_setup(struct fetch *parent_fetch,
nsurl *url,
bool only_2xx,
bool downgrade_tls,
const char *post_urlenc,
const struct fetch_multipart_data *post_multipart,
const char **headers)
{
struct curl_fetch_info *fetch;
struct curl_slist *slist;
int i;
fetch = malloc(sizeof (*fetch));
if (fetch == NULL)
return 0;
fetch->fetch_handle = parent_fetch;
LOG("fetch %p, url '%s'", fetch, nsurl_access(url));
/* construct a new fetch structure */
fetch->curl_handle = NULL;
fetch->had_headers = false;
fetch->abort = false;
fetch->stopped = false;
fetch->only_2xx = only_2xx;
fetch->downgrade_tls = downgrade_tls;
fetch->headers = NULL;
fetch->url = nsurl_ref(url);
fetch->host = nsurl_get_component(url, NSURL_HOST);
fetch->location = NULL;
fetch->content_length = 0;
fetch->http_code = 0;
fetch->cookie_string = NULL;
fetch->realm = NULL;
fetch->post_urlenc = NULL;
fetch->post_multipart = NULL;
if (post_urlenc)
fetch->post_urlenc = strdup(post_urlenc);
else if (post_multipart)
fetch->post_multipart = fetch_curl_post_convert(post_multipart);
memset(fetch->cert_data, 0, sizeof(fetch->cert_data));
fetch->last_progress_update = 0;
if (fetch->host == NULL ||
(post_multipart != NULL && fetch->post_multipart == NULL) ||
(post_urlenc != NULL && fetch->post_urlenc == NULL))
goto failed;
#define APPEND(list, value) \
slist = curl_slist_append(list, value); \
if (slist == NULL) \
goto failed; \
list = slist;
/* remove curl default headers */
APPEND(fetch->headers, "Pragma:");
/* when doing a POST libcurl sends Expect: 100-continue" by default
* which fails with lighttpd, so disable it (see bug 1429054) */
APPEND(fetch->headers, "Expect:");
if ((nsoption_charp(accept_language) != NULL) &&
(nsoption_charp(accept_language)[0] != '\0')) {
char s[80];
snprintf(s, sizeof s, "Accept-Language: %s, *;q=0.1",
nsoption_charp(accept_language));
s[sizeof s - 1] = 0;
APPEND(fetch->headers, s);
}
if (nsoption_charp(accept_charset) != NULL &&
nsoption_charp(accept_charset)[0] != '\0') {
char s[80];
snprintf(s, sizeof s, "Accept-Charset: %s, *;q=0.1",
nsoption_charp(accept_charset));
s[sizeof s - 1] = 0;
APPEND(fetch->headers, s);
}
if (nsoption_bool(do_not_track) == true) {
APPEND(fetch->headers, "DNT: 1");
}
/* And add any headers specified by the caller */
for (i = 0; headers[i] != NULL; i++) {
APPEND(fetch->headers, headers[i]);
}
return fetch;
#undef APPEND
failed:
if (fetch->host != NULL)
lwc_string_unref(fetch->host);
nsurl_unref(fetch->url);
free(fetch->post_urlenc);
if (fetch->post_multipart)
curl_formfree(fetch->post_multipart);
curl_slist_free_all(fetch->headers);
free(fetch);
return NULL;
}
/**
* OpenSSL Certificate verification callback
* Stores certificate details in fetch struct.
*/
static int
fetch_curl_verify_callback(int preverify_ok, X509_STORE_CTX *x509_ctx)
{
X509 *cert = X509_STORE_CTX_get_current_cert(x509_ctx);
int depth = X509_STORE_CTX_get_error_depth(x509_ctx);
int err = X509_STORE_CTX_get_error(x509_ctx);
struct curl_fetch_info *f = X509_STORE_CTX_get_app_data(x509_ctx);
/* save the certificate by incrementing the reference count and
* keeping a pointer.
*/
if (depth < MAX_CERTS && !f->cert_data[depth].cert) {
f->cert_data[depth].cert = cert;
f->cert_data[depth].err = err;
cert->references++;
}
return preverify_ok;
}
/**
* OpenSSL certificate chain verification callback
* Verifies certificate chain, setting up context for fetch_curl_verify_callback
*/
static int fetch_curl_cert_verify_callback(X509_STORE_CTX *x509_ctx, void *parm)
{
int ok;
/* Store fetch struct in context for verify callback */
ok = X509_STORE_CTX_set_app_data(x509_ctx, parm);
/* and verify the certificate chain */
if (ok)
ok = X509_verify_cert(x509_ctx);
return ok;
}
/**
* cURL SSL setup callback
*
* \param curl_handle The curl handle to perform the ssl operation on.
* \param _sslctx The ssl context.
* \param parm The callback context.
* \return A curl result code.
*/
static CURLcode
fetch_curl_sslctxfun(CURL *curl_handle, void *_sslctx, void *parm)
{
struct curl_fetch_info *f = (struct curl_fetch_info *) parm;
SSL_CTX *sslctx = _sslctx;
long options = SSL_OP_ALL | SSL_OP_NO_SSLv2 | SSL_OP_NO_SSLv3;
SSL_CTX_set_verify(sslctx, SSL_VERIFY_PEER, fetch_curl_verify_callback);
SSL_CTX_set_cert_verify_callback(sslctx,
fetch_curl_cert_verify_callback,
parm);
if (f->downgrade_tls) {
/* Disable TLS 1.1/1.2 if the server can't cope with them */
#ifdef SSL_OP_NO_TLSv1_1
options |= SSL_OP_NO_TLSv1_1;
#endif
#ifdef SSL_OP_NO_TLSv1_2
options |= SSL_OP_NO_TLSv1_2;
#endif
#ifdef SSL_MODE_SEND_FALLBACK_SCSV
/* Ensure server rejects the connection if downgraded too far */
SSL_CTX_set_mode(sslctx, SSL_MODE_SEND_FALLBACK_SCSV);
#endif
}
SSL_CTX_set_options(sslctx, options);
return CURLE_OK;
}
/**
* Set options specific for a fetch.
*
* \param f The fetch to set options on.
* \return A curl result code.
*/
static CURLcode fetch_curl_set_options(struct curl_fetch_info *f)
{
CURLcode code;
const char *auth;
#undef SETOPT
#define SETOPT(option, value) { \
code = curl_easy_setopt(f->curl_handle, option, value); \
if (code != CURLE_OK) \
return code; \
}
SETOPT(CURLOPT_URL, nsurl_access(f->url));
SETOPT(CURLOPT_PRIVATE, f);
SETOPT(CURLOPT_WRITEDATA, f);
SETOPT(CURLOPT_WRITEHEADER, f);
SETOPT(CURLOPT_PROGRESSDATA, f);
SETOPT(CURLOPT_REFERER, fetch_get_referer_to_send(f->fetch_handle));
SETOPT(CURLOPT_HTTPHEADER, f->headers);
if (f->post_urlenc) {
SETOPT(CURLOPT_HTTPPOST, NULL);
SETOPT(CURLOPT_HTTPGET, 0L);
SETOPT(CURLOPT_POSTFIELDS, f->post_urlenc);
} else if (f->post_multipart) {
SETOPT(CURLOPT_POSTFIELDS, NULL);
SETOPT(CURLOPT_HTTPGET, 0L);
SETOPT(CURLOPT_HTTPPOST, f->post_multipart);
} else {
SETOPT(CURLOPT_POSTFIELDS, NULL);
SETOPT(CURLOPT_HTTPPOST, NULL);
SETOPT(CURLOPT_HTTPGET, 1L);
}
f->cookie_string = urldb_get_cookie(f->url, true);
if (f->cookie_string) {
SETOPT(CURLOPT_COOKIE, f->cookie_string);
} else {
SETOPT(CURLOPT_COOKIE, NULL);
}
if ((auth = urldb_get_auth_details(f->url, NULL)) != NULL) {
SETOPT(CURLOPT_HTTPAUTH, CURLAUTH_ANY);
SETOPT(CURLOPT_USERPWD, auth);
} else {
SETOPT(CURLOPT_USERPWD, NULL);
}
/* set up proxy options */
if (nsoption_bool(http_proxy) &&
(nsoption_charp(http_proxy_host) != NULL) &&
(strncmp(nsurl_access(f->url), "file:", 5) != 0)) {
SETOPT(CURLOPT_PROXY, nsoption_charp(http_proxy_host));
SETOPT(CURLOPT_PROXYPORT, (long) nsoption_int(http_proxy_port));
#if LIBCURL_VERSION_NUM >= 0x071304
/* Added in 7.19.4 */
/* setup the omission list */
SETOPT(CURLOPT_NOPROXY, nsoption_charp(http_proxy_noproxy));
#endif
if (nsoption_int(http_proxy_auth) != OPTION_HTTP_PROXY_AUTH_NONE) {
SETOPT(CURLOPT_PROXYAUTH,
nsoption_int(http_proxy_auth) ==
OPTION_HTTP_PROXY_AUTH_BASIC ?
(long) CURLAUTH_BASIC :
(long) CURLAUTH_NTLM);
snprintf(fetch_proxy_userpwd,
sizeof fetch_proxy_userpwd,
"%s:%s",
nsoption_charp(http_proxy_auth_user),
nsoption_charp(http_proxy_auth_pass));
SETOPT(CURLOPT_PROXYUSERPWD, fetch_proxy_userpwd);
}
} else {
SETOPT(CURLOPT_PROXY, NULL);
}
/* Disable SSL session ID caching, as some servers can't cope. */
SETOPT(CURLOPT_SSL_SESSIONID_CACHE, 0);
if (urldb_get_cert_permissions(f->url)) {
/* Disable certificate verification */
SETOPT(CURLOPT_SSL_VERIFYPEER, 0L);
SETOPT(CURLOPT_SSL_VERIFYHOST, 0L);
if (curl_with_openssl) {
SETOPT(CURLOPT_SSL_CTX_FUNCTION, NULL);
SETOPT(CURLOPT_SSL_CTX_DATA, NULL);
}
} else {
/* do verification */
SETOPT(CURLOPT_SSL_VERIFYPEER, 1L);
SETOPT(CURLOPT_SSL_VERIFYHOST, 2L);
if (curl_with_openssl) {
SETOPT(CURLOPT_SSL_CTX_FUNCTION, fetch_curl_sslctxfun);
SETOPT(CURLOPT_SSL_CTX_DATA, f);
}
}
return CURLE_OK;
}
/**
* Initiate a fetch from the queue.
*
* \param fetch fetch to use to fetch content.
* \param handle CURL handle to be used to fetch the content.
* \return true if the fetch was successfully initiated else false.
*/
static bool
fetch_curl_initiate_fetch(struct curl_fetch_info *fetch, CURL *handle)
{
CURLcode code;
CURLMcode codem;
fetch->curl_handle = handle;
/* Initialise the handle */
code = fetch_curl_set_options(fetch);
if (code != CURLE_OK) {
fetch->curl_handle = 0;
return false;
}
/* add to the global curl multi handle */
codem = curl_multi_add_handle(fetch_curl_multi, fetch->curl_handle);
assert(codem == CURLM_OK || codem == CURLM_CALL_MULTI_PERFORM);
return true;
}
/**
* Find a CURL handle to use to dispatch a job
*/
static CURL *fetch_curl_get_handle(lwc_string *host)
{
struct cache_handle *h;
CURL *ret;
RING_FINDBYLWCHOST(curl_handle_ring, h, host);
if (h) {
ret = h->handle;
lwc_string_unref(h->host);
RING_REMOVE(curl_handle_ring, h);
free(h);
} else {
ret = curl_easy_duphandle(fetch_blank_curl);
}
return ret;
}
/**
* Dispatch a single job
*/
static bool fetch_curl_start(void *vfetch)
{
struct curl_fetch_info *fetch = (struct curl_fetch_info*)vfetch;
return fetch_curl_initiate_fetch(fetch,
fetch_curl_get_handle(fetch->host));
}
/**
* Cache a CURL handle for the provided host (if wanted)
*/
static void fetch_curl_cache_handle(CURL *handle, lwc_string *host)
{
#if LIBCURL_VERSION_NUM >= 0x071e00
/* 7.30.0 or later has its own connection caching; suppress ours */
curl_easy_cleanup(handle);
return;
#else
struct cache_handle *h = 0;
int c;
RING_FINDBYLWCHOST(curl_handle_ring, h, host);
if (h) {
/* Already have a handle cached for this hostname */
curl_easy_cleanup(handle);
return;
}
/* We do not have a handle cached, first up determine if the cache is full */
RING_GETSIZE(struct cache_handle, curl_handle_ring, c);
if (c >= nsoption_int(max_cached_fetch_handles)) {
/* Cache is full, so, we rotate the ring by one and
* replace the oldest handle with this one. We do this
* without freeing/allocating memory (except the
* hostname) and without removing the entry from the
* ring and then re-inserting it, in order to be as
* efficient as we can.
*/
if (curl_handle_ring != NULL) {
h = curl_handle_ring;
curl_handle_ring = h->r_next;
curl_easy_cleanup(h->handle);
h->handle = handle;
lwc_string_unref(h->host);
h->host = lwc_string_ref(host);
} else {
/* Actually, we don't want to cache any handles */
curl_easy_cleanup(handle);
}
return;
}
/* The table isn't full yet, so make a shiny new handle to add to the ring */
h = (struct cache_handle*)malloc(sizeof(struct cache_handle));
h->handle = handle;
h->host = lwc_string_ref(host);
RING_INSERT(curl_handle_ring, h);
#endif
}
/**
* Abort a fetch.
*/
static void fetch_curl_abort(void *vf)
{
struct curl_fetch_info *f = (struct curl_fetch_info *)vf;
assert(f);
LOG("fetch %p, url '%s'", f, nsurl_access(f->url));
if (f->curl_handle) {
f->abort = true;
} else {
fetch_remove_from_queues(f->fetch_handle);
fetch_free(f->fetch_handle);
}
}
/**
* Clean up the provided fetch object and free it.
*
* Will prod the queue afterwards to allow pending requests to be initiated.
*/
static void fetch_curl_stop(struct curl_fetch_info *f)
{
CURLMcode codem;
assert(f);
LOG("fetch %p, url '%s'", f, nsurl_access(f->url));
if (f->curl_handle) {
/* remove from curl multi handle */
codem = curl_multi_remove_handle(fetch_curl_multi,
f->curl_handle);
assert(codem == CURLM_OK);
/* Put this curl handle into the cache if wanted. */
fetch_curl_cache_handle(f->curl_handle, f->host);
f->curl_handle = 0;
}
fetch_remove_from_queues(f->fetch_handle);
}
/**
* Free a fetch structure and associated resources.
*/
static void fetch_curl_free(void *vf)
{
struct curl_fetch_info *f = (struct curl_fetch_info *)vf;
int i;
if (f->curl_handle)
curl_easy_cleanup(f->curl_handle);
nsurl_unref(f->url);
lwc_string_unref(f->host);
free(f->location);
free(f->cookie_string);
free(f->realm);
if (f->headers)
curl_slist_free_all(f->headers);
free(f->post_urlenc);
if (f->post_multipart)
curl_formfree(f->post_multipart);
for (i = 0; i < MAX_CERTS && f->cert_data[i].cert; i++) {
f->cert_data[i].cert->references--;
if (f->cert_data[i].cert->references == 0)
X509_free(f->cert_data[i].cert);
}
free(f);
}
/**
* Find the status code and content type and inform the caller.
*
* Return true if the fetch is being aborted.
*/
static bool fetch_curl_process_headers(struct curl_fetch_info *f)
{
long http_code;
CURLcode code;
fetch_msg msg;
f->had_headers = true;
if (!f->http_code)
{
code = curl_easy_getinfo(f->curl_handle, CURLINFO_HTTP_CODE,
&f->http_code);
fetch_set_http_code(f->fetch_handle, f->http_code);
assert(code == CURLE_OK);
}
http_code = f->http_code;
LOG("HTTP status code %li", http_code);
if (http_code == 304 && !f->post_urlenc && !f->post_multipart) {
/* Not Modified && GET request */
msg.type = FETCH_NOTMODIFIED;
fetch_send_callback(&msg, f->fetch_handle);
return true;
}
/* handle HTTP redirects (3xx response codes) */
if (300 <= http_code && http_code < 400 && f->location != 0) {
LOG("FETCH_REDIRECT, '%s'", f->location);
msg.type = FETCH_REDIRECT;
msg.data.redirect = f->location;
fetch_send_callback(&msg, f->fetch_handle);
return true;
}
/* handle HTTP 401 (Authentication errors) */
if (http_code == 401) {
msg.type = FETCH_AUTH;
msg.data.auth.realm = f->realm;
fetch_send_callback(&msg, f->fetch_handle);
return true;
}
/* handle HTTP errors (non 2xx response codes) */
if (f->only_2xx && strncmp(nsurl_access(f->url), "http", 4) == 0 &&
(http_code < 200 || 299 < http_code)) {
msg.type = FETCH_ERROR;
msg.data.error = messages_get("Not2xx");
fetch_send_callback(&msg, f->fetch_handle);
return true;
}
if (f->abort)
return true;
return false;
}
/**
* Handle a completed fetch (CURLMSG_DONE from curl_multi_info_read()).
*
* \param curl_handle curl easy handle of fetch
* \param result The result code of the completed fetch.
*/
static void fetch_curl_done(CURL *curl_handle, CURLcode result)
{
fetch_msg msg;
bool finished = false;
bool error = false;
bool cert = false;
bool abort_fetch;
struct curl_fetch_info *f;
char **_hideous_hack = (char **) (void *) &f;
CURLcode code;
struct cert_info certs[MAX_CERTS];
memset(certs, 0, sizeof(certs));
/* find the structure associated with this fetch */
/* For some reason, cURL thinks CURLINFO_PRIVATE should be a string?! */
code = curl_easy_getinfo(curl_handle, CURLINFO_PRIVATE, _hideous_hack);
assert(code == CURLE_OK);
abort_fetch = f->abort;
LOG("done %s", nsurl_access(f->url));
if (abort_fetch == false && (result == CURLE_OK ||
(result == CURLE_WRITE_ERROR && f->stopped == false))) {
/* fetch completed normally or the server fed us a junk gzip
* stream (usually in the form of garbage at the end of the
* stream). Curl will have fed us all but the last chunk of
* decoded data, which is sad as, if we'd received the last
* chunk, too, we'd be able to render the whole object.
* As is, we'll just have to accept that the end of the
* object will be truncated in this case and leave it to
* the content handlers to cope.
*/
if (f->stopped ||
(!f->had_headers &&
fetch_curl_process_headers(f)))
; /* redirect with no body or similar */
else
finished = true;
} else if (result == CURLE_PARTIAL_FILE) {
/* CURLE_PARTIAL_FILE occurs if the received body of a
* response is smaller than that specified in the
* Content-Length header. */
if (!f->had_headers && fetch_curl_process_headers(f))
; /* redirect with partial body, or similar */
else {
finished = true;
}
} else if (result == CURLE_WRITE_ERROR && f->stopped) {
/* CURLE_WRITE_ERROR occurs when fetch_curl_data
* returns 0, which we use to abort intentionally */
;
} else if (result == CURLE_SSL_PEER_CERTIFICATE ||
result == CURLE_SSL_CACERT) {
memcpy(certs, f->cert_data, sizeof(certs));
memset(f->cert_data, 0, sizeof(f->cert_data));
cert = true;
} else {
LOG("Unknown cURL response code %d", result);
error = true;
}
fetch_curl_stop(f);
if (abort_fetch)
; /* fetch was aborted: no callback */
else if (finished) {
msg.type = FETCH_FINISHED;
fetch_send_callback(&msg, f->fetch_handle);
} else if (cert) {
int i;
BIO *mem;
BUF_MEM *buf;
struct ssl_cert_info ssl_certs[MAX_CERTS];
for (i = 0; i < MAX_CERTS && certs[i].cert; i++) {
ssl_certs[i].version =
X509_get_version(certs[i].cert);
mem = BIO_new(BIO_s_mem());
ASN1_TIME_print(mem,
X509_get_notBefore(certs[i].cert));
BIO_get_mem_ptr(mem, &buf);
(void) BIO_set_close(mem, BIO_NOCLOSE);
BIO_free(mem);
memcpy(ssl_certs[i].not_before,
buf->data,
min(sizeof(ssl_certs[i].not_before) - 1,
(unsigned)buf->length));
ssl_certs[i].not_before[min(sizeof(ssl_certs[i].not_before) - 1,
(unsigned)buf->length)] = 0;
BUF_MEM_free(buf);
mem = BIO_new(BIO_s_mem());
ASN1_TIME_print(mem,
X509_get_notAfter(certs[i].cert));
BIO_get_mem_ptr(mem, &buf);
(void) BIO_set_close(mem, BIO_NOCLOSE);
BIO_free(mem);
memcpy(ssl_certs[i].not_after,
buf->data,
min(sizeof(ssl_certs[i].not_after) - 1,
(unsigned)buf->length));
ssl_certs[i].not_after[min(sizeof(ssl_certs[i].not_after) - 1,
(unsigned)buf->length)] = 0;
BUF_MEM_free(buf);
ssl_certs[i].sig_type =
X509_get_signature_type(certs[i].cert);
ssl_certs[i].serial =
ASN1_INTEGER_get(
X509_get_serialNumber(certs[i].cert));
mem = BIO_new(BIO_s_mem());
X509_NAME_print_ex(mem,
X509_get_issuer_name(certs[i].cert),
0, XN_FLAG_SEP_CPLUS_SPC |
XN_FLAG_DN_REV | XN_FLAG_FN_NONE);
BIO_get_mem_ptr(mem, &buf);
(void) BIO_set_close(mem, BIO_NOCLOSE);
BIO_free(mem);
memcpy(ssl_certs[i].issuer,
buf->data,
min(sizeof(ssl_certs[i].issuer) - 1,
(unsigned) buf->length));
ssl_certs[i].issuer[min(sizeof(ssl_certs[i].issuer) - 1,
(unsigned) buf->length)] = 0;
BUF_MEM_free(buf);
mem = BIO_new(BIO_s_mem());
X509_NAME_print_ex(mem,
X509_get_subject_name(certs[i].cert),
0,
XN_FLAG_SEP_CPLUS_SPC |
XN_FLAG_DN_REV |
XN_FLAG_FN_NONE);
BIO_get_mem_ptr(mem, &buf);
(void) BIO_set_close(mem, BIO_NOCLOSE);
BIO_free(mem);
memcpy(ssl_certs[i].subject,
buf->data,
min(sizeof(ssl_certs[i].subject) - 1,
(unsigned)buf->length));
ssl_certs[i].subject[min(sizeof(ssl_certs[i].subject) - 1,
(unsigned) buf->length)] = 0;
BUF_MEM_free(buf);
ssl_certs[i].cert_type =
X509_certificate_type(certs[i].cert,
X509_get_pubkey(certs[i].cert));
/* and clean up */
certs[i].cert->references--;
if (certs[i].cert->references == 0)
X509_free(certs[i].cert);
}
msg.type = FETCH_CERT_ERR;
msg.data.cert_err.certs = ssl_certs;
msg.data.cert_err.num_certs = i;
fetch_send_callback(&msg, f->fetch_handle);
} else if (error) {
switch (result) {
case CURLE_SSL_CONNECT_ERROR:
msg.type = FETCH_SSL_ERR;
break;
case CURLE_OPERATION_TIMEDOUT:
msg.type = FETCH_TIMEDOUT;
msg.data.error = curl_easy_strerror(result);
break;
default:
msg.type = FETCH_ERROR;
msg.data.error = curl_easy_strerror(result);
}
fetch_send_callback(&msg, f->fetch_handle);
}
fetch_free(f->fetch_handle);
}
/**
* Do some work on current fetches.
*
* Must be called regularly to make progress on fetches.
*/
static void fetch_curl_poll(lwc_string *scheme_ignored)
{
int running, queue;
CURLMcode codem;
CURLMsg *curl_msg;
if (nsoption_bool(suppress_curl_debug) == false) {
fd_set read_fd_set, write_fd_set, exc_fd_set;
int max_fd = -1;
int i;
FD_ZERO(&read_fd_set);
FD_ZERO(&write_fd_set);
FD_ZERO(&exc_fd_set);
codem = curl_multi_fdset(fetch_curl_multi,
&read_fd_set, &write_fd_set,
&exc_fd_set, &max_fd);
assert(codem == CURLM_OK);
LOG("Curl file descriptor states (maxfd=%i):", max_fd);
for (i = 0; i <= max_fd; i++) {
bool read = false;
bool write = false;
bool error = false;
if (FD_ISSET(i, &read_fd_set)) {
read = true;
}
if (FD_ISSET(i, &write_fd_set)) {
write = true;
}
if (FD_ISSET(i, &exc_fd_set)) {
error = true;
}
if (read || write || error) {
LOG(" fd %i: %s %s %s", i,
read ? "read" : " ",
write ? "write" : " ",
error ? "error" : " ");
}
}
}
/* do any possible work on the current fetches */
do {
codem = curl_multi_perform(fetch_curl_multi, &running);
if (codem != CURLM_OK && codem != CURLM_CALL_MULTI_PERFORM) {
LOG("curl_multi_perform: %i %s", codem, curl_multi_strerror(codem));
warn_user("MiscError", curl_multi_strerror(codem));
return;
}
} while (codem == CURLM_CALL_MULTI_PERFORM);
/* process curl results */
curl_msg = curl_multi_info_read(fetch_curl_multi, &queue);
while (curl_msg) {
switch (curl_msg->msg) {
case CURLMSG_DONE:
fetch_curl_done(curl_msg->easy_handle,
curl_msg->data.result);
break;
default:
break;
}
curl_msg = curl_multi_info_read(fetch_curl_multi, &queue);
}
}
/**
* Callback function for fetch progress.
*/
static int fetch_curl_progress(void *clientp, double dltotal, double dlnow,
double ultotal, double ulnow)
{
static char fetch_progress_buffer[256]; /**< Progress buffer for cURL */
struct curl_fetch_info *f = (struct curl_fetch_info *) clientp;
unsigned int time_now_cs;
fetch_msg msg;
if (f->abort)
return 0;
msg.type = FETCH_PROGRESS;
msg.data.progress = fetch_progress_buffer;
/* Rate limit each fetch's progress notifications to 2 a second */
#define UPDATES_PER_SECOND 2
#define UPDATE_DELAY_CS (100 / UPDATES_PER_SECOND)
time_now_cs = wallclock();
if (time_now_cs - f->last_progress_update < UPDATE_DELAY_CS)
return 0;
f->last_progress_update = time_now_cs;
#undef UPDATE_DELAY_CS
#undef UPDATES_PERS_SECOND
if (dltotal > 0) {
snprintf(fetch_progress_buffer, 255,
messages_get("Progress"),
human_friendly_bytesize(dlnow),
human_friendly_bytesize(dltotal));
fetch_send_callback(&msg, f->fetch_handle);
} else {
snprintf(fetch_progress_buffer, 255,
messages_get("ProgressU"),
human_friendly_bytesize(dlnow));
fetch_send_callback(&msg, f->fetch_handle);
}
return 0;
}
/**
* Ignore everything given to it.
*
* Used to ignore cURL debug.
*/
static int fetch_curl_ignore_debug(CURL *handle,
curl_infotype type,
char *data,
size_t size,
void *userptr)
{
return 0;
}
/**
* Callback function for cURL.
*/
static size_t fetch_curl_data(char *data, size_t size, size_t nmemb, void *_f)
{
struct curl_fetch_info *f = _f;
CURLcode code;
fetch_msg msg;
/* ensure we only have to get this information once */
if (!f->http_code)
{
code = curl_easy_getinfo(f->curl_handle, CURLINFO_HTTP_CODE,
&f->http_code);
fetch_set_http_code(f->fetch_handle, f->http_code);
assert(code == CURLE_OK);
}
/* ignore body if this is a 401 reply by skipping it and reset
* the HTTP response code to enable follow up fetches.
*/
if (f->http_code == 401)
{
f->http_code = 0;
return size * nmemb;
}
if (f->abort || (!f->had_headers && fetch_curl_process_headers(f))) {
f->stopped = true;
return 0;
}
/* send data to the caller */
msg.type = FETCH_DATA;
msg.data.header_or_data.buf = (const uint8_t *) data;
msg.data.header_or_data.len = size * nmemb;
fetch_send_callback(&msg, f->fetch_handle);
if (f->abort) {
f->stopped = true;
return 0;
}
return size * nmemb;
}
/**
* Callback function for headers.
*
* See RFC 2616 4.2.
*/
static size_t fetch_curl_header(char *data, size_t size, size_t nmemb,
void *_f)
{
struct curl_fetch_info *f = _f;
int i;
fetch_msg msg;
size *= nmemb;
if (f->abort) {
f->stopped = true;
return 0;
}
msg.type = FETCH_HEADER;
msg.data.header_or_data.buf = (const uint8_t *) data;
msg.data.header_or_data.len = size;
fetch_send_callback(&msg, f->fetch_handle);
#define SKIP_ST(o) for (i = (o); i < (int) size && (data[i] == ' ' || data[i] == '\t'); i++)
if (12 < size && strncasecmp(data, "Location:", 9) == 0) {
/* extract Location header */
free(f->location);
f->location = malloc(size);
if (!f->location) {
LOG("malloc failed");
return size;
}
SKIP_ST(9);
strncpy(f->location, data + i, size - i);
f->location[size - i] = '\0';
for (i = size - i - 1; i >= 0 &&
(f->location[i] == ' ' ||
f->location[i] == '\t' ||
f->location[i] == '\r' ||
f->location[i] == '\n'); i--)
f->location[i] = '\0';
} else if (15 < size && strncasecmp(data, "Content-Length:", 15) == 0) {
/* extract Content-Length header */
SKIP_ST(15);
if (i < (int)size && '0' <= data[i] && data[i] <= '9')
f->content_length = atol(data + i);
} else if (17 < size && strncasecmp(data, "WWW-Authenticate:", 17) == 0) {
/* extract the first Realm from WWW-Authenticate header */
SKIP_ST(17);
while (i < (int) size - 5 &&
strncasecmp(data + i, "realm", 5))
i++;
while (i < (int) size - 1 && data[++i] != '"')
/* */;
i++;
if (i < (int) size) {
size_t end = i;
while (end < size && data[end] != '"')
++end;
if (end < size) {
free(f->realm);
f->realm = malloc(end - i + 1);
if (f->realm != NULL) {
strncpy(f->realm, data + i, end - i);
f->realm[end - i] = '\0';
}
}
}
} else if (11 < size && strncasecmp(data, "Set-Cookie:", 11) == 0) {
/* extract Set-Cookie header */
SKIP_ST(11);
fetch_set_cookie(f->fetch_handle, &data[i]);
}
return size;
#undef SKIP_ST
}
/* exported function documented in content/fetchers/curl.h */
nserror fetch_curl_register(void)
{
CURLcode code;
curl_version_info_data *data;
int i;
lwc_string *scheme;
const struct fetcher_operation_table fetcher_ops = {
.initialise = fetch_curl_initialise,
.acceptable = fetch_curl_can_fetch,
.setup = fetch_curl_setup,
.start = fetch_curl_start,
.abort = fetch_curl_abort,
.free = fetch_curl_free,
.poll = fetch_curl_poll,
.finalise = fetch_curl_finalise
};
LOG("curl_version %s", curl_version());
code = curl_global_init(CURL_GLOBAL_ALL);
if (code != CURLE_OK) {
LOG("curl_global_init failed.");
return NSERROR_INIT_FAILED;
}
fetch_curl_multi = curl_multi_init();
if (!fetch_curl_multi) {
LOG("curl_multi_init failed.");
return NSERROR_INIT_FAILED;
}
#if LIBCURL_VERSION_NUM >= 0x071e00
/* built against 7.30.0 or later: configure caching */
{
CURLMcode mcode;
int maxconnects = nsoption_int(max_fetchers) +
nsoption_int(max_cached_fetch_handles);
#undef SETOPT
#define SETOPT(option, value) \
mcode = curl_multi_setopt(fetch_curl_multi, option, value); \
if (mcode != CURLM_OK) \
goto curl_multi_setopt_failed;
SETOPT(CURLMOPT_MAXCONNECTS, maxconnects);
SETOPT(CURLMOPT_MAX_TOTAL_CONNECTIONS, maxconnects);
SETOPT(CURLMOPT_MAX_HOST_CONNECTIONS, nsoption_int(max_fetchers_per_host));
}
#endif
/* Create a curl easy handle with the options that are common to all
fetches.
*/
fetch_blank_curl = curl_easy_init();
if (!fetch_blank_curl) {
LOG("curl_easy_init failed");
return NSERROR_INIT_FAILED;
}
#undef SETOPT
#define SETOPT(option, value) \
code = curl_easy_setopt(fetch_blank_curl, option, value); \
if (code != CURLE_OK) \
goto curl_easy_setopt_failed;
if (verbose_log) {
SETOPT(CURLOPT_VERBOSE, 1);
} else {
SETOPT(CURLOPT_VERBOSE, 0);
}
SETOPT(CURLOPT_ERRORBUFFER, fetch_error_buffer);
if (nsoption_bool(suppress_curl_debug)) {
SETOPT(CURLOPT_DEBUGFUNCTION, fetch_curl_ignore_debug);
}
SETOPT(CURLOPT_WRITEFUNCTION, fetch_curl_data);
SETOPT(CURLOPT_HEADERFUNCTION, fetch_curl_header);
SETOPT(CURLOPT_PROGRESSFUNCTION, fetch_curl_progress);
SETOPT(CURLOPT_NOPROGRESS, 0);
SETOPT(CURLOPT_USERAGENT, user_agent_string());
SETOPT(CURLOPT_ENCODING, "gzip");
SETOPT(CURLOPT_LOW_SPEED_LIMIT, 1L);
SETOPT(CURLOPT_LOW_SPEED_TIME, 180L);
SETOPT(CURLOPT_NOSIGNAL, 1L);
SETOPT(CURLOPT_CONNECTTIMEOUT, nsoption_uint(curl_fetch_timeout));
if (nsoption_charp(ca_bundle) &&
strcmp(nsoption_charp(ca_bundle), "")) {
LOG("ca_bundle: '%s'", nsoption_charp(ca_bundle));
SETOPT(CURLOPT_CAINFO, nsoption_charp(ca_bundle));
}
if (nsoption_charp(ca_path) && strcmp(nsoption_charp(ca_path), "")) {
LOG("ca_path: '%s'", nsoption_charp(ca_path));
SETOPT(CURLOPT_CAPATH, nsoption_charp(ca_path));
}
/* Detect whether the SSL CTX function API works */
curl_with_openssl = true;
code = curl_easy_setopt(fetch_blank_curl,
CURLOPT_SSL_CTX_FUNCTION, NULL);
if (code != CURLE_OK) {
curl_with_openssl = false;
}
LOG("cURL %slinked against openssl", curl_with_openssl ? "" : "not ");
/* cURL initialised okay, register the fetchers */
data = curl_version_info(CURLVERSION_NOW);
for (i = 0; data->protocols[i]; i++) {
if (strcmp(data->protocols[i], "http") == 0) {
scheme = lwc_string_ref(corestring_lwc_http);
} else if (strcmp(data->protocols[i], "https") == 0) {
scheme = lwc_string_ref(corestring_lwc_https);
} else {
/* Ignore non-http(s) protocols */
continue;
}
if (fetcher_add(scheme, &fetcher_ops) != NSERROR_OK) {
LOG("Unable to register cURL fetcher for %s", data->protocols[i]);
}
}
return NSERROR_OK;
curl_easy_setopt_failed:
LOG("curl_easy_setopt failed.");
return NSERROR_INIT_FAILED;
#if LIBCURL_VERSION_NUM >= 0x071e00
curl_multi_setopt_failed:
LOG("curl_multi_setopt failed.");
return NSERROR_INIT_FAILED;
#endif
}
Reference in New Issue View Git Blame Copy Permalink