X-Git-Url: http://sjero.net/git/?p=wget;a=blobdiff_plain;f=src%2Fretr.c;h=b667ca2ff3cf6ecb4edca9c1148e856f68b0ecbe;hp=245eb129f40d5f49644d0bb69de1e7489f54cf1d;hb=d763f8bf6d6e13ce006ffab616cc8a77e747a633;hpb=a0d0f332d5f230e40fe7fff8fc76839c4f4704ce
diff --git a/src/retr.c b/src/retr.c
index 245eb129..b667ca2f 100644
--- a/src/retr.c
+++ b/src/retr.c
@@ -1,11 +1,12 @@
/* File retrieval.
- Copyright (C) 1996-2006 Free Software Foundation, Inc.
+ Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003,
+ 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
This file is part of GNU Wget.
GNU Wget 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; either version 2 of the License, or (at
+the Free Software Foundation; either version 3 of the License, or (at
your option) any later version.
GNU Wget is distributed in the hope that it will be useful,
@@ -14,20 +15,20 @@ 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 Wget; if not, write to the Free Software Foundation, Inc.,
-51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+along with Wget. If not, see .
-In addition, as a special exception, the Free Software Foundation
-gives permission to link the code of its release of Wget with the
-OpenSSL project's "OpenSSL" library (or with modified versions of it
-that use the same license as the "OpenSSL" library), and distribute
-the linked executables. You must obey the GNU General Public License
-in all respects for all of the code used other than "OpenSSL". If you
-modify this file, you may extend this exception to your version of the
-file, but you are not obligated to do so. If you do not wish to do
-so, delete this exception statement from your version. */
+Additional permission under GNU GPL version 3 section 7
-#include
+If you modify this program, or any covered work, by linking or
+combining it with the OpenSSL project's OpenSSL library (or a
+modified version of that library), containing parts covered by the
+terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
+grants you additional permission to convey the resulting work.
+Corresponding Source for a non-source form of such a combination
+shall include the source code for the parts of OpenSSL used as well
+as that of the covered work. */
+
+#include "wget.h"
#include
#include
@@ -38,7 +39,6 @@ so, delete this exception statement from your version. */
#include
#include
-#include "wget.h"
#include "utils.h"
#include "retr.h"
#include "progress.h"
@@ -52,6 +52,7 @@ so, delete this exception statement from your version. */
#include "convert.h"
#include "ptimer.h"
#include "html-url.h"
+#include "iri.h"
/* Total size of downloaded files. Used to enforce quota. */
SUM_SIZE_INT total_downloaded_bytes;
@@ -101,31 +102,31 @@ limit_bandwidth (wgint bytes, struct ptimer *timer)
double slp = expected - delta_t + limit_data.sleep_adjust;
double t0, t1;
if (slp < 0.2)
- {
- DEBUGP (("deferring a %.2f ms sleep (%s/%.2f).\n",
- slp * 1000, number_to_static_string (limit_data.chunk_bytes),
- delta_t));
- return;
- }
+ {
+ DEBUGP (("deferring a %.2f ms sleep (%s/%.2f).\n",
+ slp * 1000, number_to_static_string (limit_data.chunk_bytes),
+ delta_t));
+ return;
+ }
DEBUGP (("\nsleeping %.2f ms for %s bytes, adjust %.2f ms\n",
- slp * 1000, number_to_static_string (limit_data.chunk_bytes),
- limit_data.sleep_adjust));
+ slp * 1000, number_to_static_string (limit_data.chunk_bytes),
+ limit_data.sleep_adjust));
t0 = ptimer_read (timer);
xsleep (slp);
t1 = ptimer_measure (timer);
/* Due to scheduling, we probably slept slightly longer (or
- shorter) than desired. Calculate the difference between the
- desired and the actual sleep, and adjust the next sleep by
- that amount. */
+ shorter) than desired. Calculate the difference between the
+ desired and the actual sleep, and adjust the next sleep by
+ that amount. */
limit_data.sleep_adjust = slp - (t1 - t0);
/* If sleep_adjust is very large, it's likely due to suspension
- and not clock inaccuracy. Don't enforce those. */
+ and not clock inaccuracy. Don't enforce those. */
if (limit_data.sleep_adjust > 0.5)
- limit_data.sleep_adjust = 0.5;
+ limit_data.sleep_adjust = 0.5;
else if (limit_data.sleep_adjust < -0.5)
- limit_data.sleep_adjust = -0.5;
+ limit_data.sleep_adjust = -0.5;
}
limit_data.chunk_bytes = 0;
@@ -142,7 +143,7 @@ limit_bandwidth (wgint bytes, struct ptimer *timer)
static int
write_data (FILE *out, const char *buf, int bufsize, wgint *skip,
- wgint *written)
+ wgint *written)
{
if (!out)
return 1;
@@ -157,7 +158,7 @@ write_data (FILE *out, const char *buf, int bufsize, wgint *skip,
bufsize -= *skip;
*skip = 0;
if (bufsize == 0)
- return 1;
+ return 1;
}
fwrite (buf, 1, bufsize, out);
@@ -167,7 +168,18 @@ write_data (FILE *out, const char *buf, int bufsize, wgint *skip,
performance: fast downloads will arrive in large 16K chunks
(which stdio would write out immediately anyway), and slow
downloads wouldn't be limited by disk speed. */
+
+ /* 2005-04-20 SMS.
+ Perhaps it shouldn't hinder performance, but it sure does, at least
+ on VMS (more than 2X). Rather than speculate on what it should or
+ shouldn't do, it might make more sense to test it. Even better, it
+ might be nice to explain what possible benefit it could offer, as
+ it appears to be a clear invitation to poor performance with no
+ actual justification. (Also, why 16K? Anyone test other values?)
+ */
+#ifndef __VMS
fflush (out);
+#endif /* ndef __VMS */
return !ferror (out);
}
@@ -192,7 +204,7 @@ write_data (FILE *out, const char *buf, int bufsize, wgint *skip,
int
fd_read_body (int fd, FILE *out, wgint toread, wgint startpos,
- wgint *qtyread, wgint *qtywritten, double *elapsed, int flags)
+ wgint *qtyread, wgint *qtywritten, double *elapsed, int flags)
{
int ret = 0;
@@ -224,9 +236,10 @@ fd_read_body (int fd, FILE *out, wgint toread, wgint startpos,
if (opt.verbose)
{
/* If we're skipping STARTPOS bytes, pass 0 as the INITIAL
- argument to progress_create because the indicator doesn't
- (yet) know about "skipping" data. */
- progress = progress_create (skip ? 0 : startpos, startpos + toread);
+ argument to progress_create because the indicator doesn't
+ (yet) know about "skipping" data. */
+ wgint start = skip ? 0 : startpos;
+ progress = progress_create (start, start + toread);
progress_interactive = progress_interactive_p (progress);
}
@@ -258,61 +271,61 @@ fd_read_body (int fd, FILE *out, wgint toread, wgint startpos,
int rdsize = exact ? MIN (toread - sum_read, dlbufsize) : dlbufsize;
double tmout = opt.read_timeout;
if (progress_interactive)
- {
- /* For interactive progress gauges, always specify a ~1s
- timeout, so that the gauge can be updated regularly even
- when the data arrives very slowly or stalls. */
- tmout = 0.95;
- if (opt.read_timeout)
- {
- double waittm;
- waittm = ptimer_read (timer) - last_successful_read_tm;
- if (waittm + tmout > opt.read_timeout)
- {
- /* Don't let total idle time exceed read timeout. */
- tmout = opt.read_timeout - waittm;
- if (tmout < 0)
- {
- /* We've already exceeded the timeout. */
- ret = -1, errno = ETIMEDOUT;
- break;
- }
- }
- }
- }
+ {
+ /* For interactive progress gauges, always specify a ~1s
+ timeout, so that the gauge can be updated regularly even
+ when the data arrives very slowly or stalls. */
+ tmout = 0.95;
+ if (opt.read_timeout)
+ {
+ double waittm;
+ waittm = ptimer_read (timer) - last_successful_read_tm;
+ if (waittm + tmout > opt.read_timeout)
+ {
+ /* Don't let total idle time exceed read timeout. */
+ tmout = opt.read_timeout - waittm;
+ if (tmout < 0)
+ {
+ /* We've already exceeded the timeout. */
+ ret = -1, errno = ETIMEDOUT;
+ break;
+ }
+ }
+ }
+ }
ret = fd_read (fd, dlbuf, rdsize, tmout);
if (progress_interactive && ret < 0 && errno == ETIMEDOUT)
- ret = 0; /* interactive timeout, handled above */
+ ret = 0; /* interactive timeout, handled above */
else if (ret <= 0)
- break; /* EOF or read error */
+ break; /* EOF or read error */
if (progress || opt.limit_rate)
- {
- ptimer_measure (timer);
- if (ret > 0)
- last_successful_read_tm = ptimer_read (timer);
- }
+ {
+ ptimer_measure (timer);
+ if (ret > 0)
+ last_successful_read_tm = ptimer_read (timer);
+ }
if (ret > 0)
- {
- sum_read += ret;
- if (!write_data (out, dlbuf, ret, &skip, &sum_written))
- {
- ret = -2;
- goto out;
- }
- }
+ {
+ sum_read += ret;
+ if (!write_data (out, dlbuf, ret, &skip, &sum_written))
+ {
+ ret = -2;
+ goto out;
+ }
+ }
if (opt.limit_rate)
- limit_bandwidth (ret, timer);
+ limit_bandwidth (ret, timer);
if (progress)
- progress_update (progress, ret, ptimer_read (timer));
+ progress_update (progress, ret, ptimer_read (timer));
#ifdef WINDOWS
if (toread > 0 && !opt.quiet)
- ws_percenttitle (100.0 *
- (startpos + sum_read) / (startpos + toread));
+ ws_percenttitle (100.0 *
+ (startpos + sum_read) / (startpos + toread));
#endif
}
if (ret < -1)
@@ -391,9 +404,9 @@ fd_read_hunk (int fd, hunk_terminator_t terminator, long sizehint, long maxsize)
{
long bufsize = sizehint;
char *hunk = xmalloc (bufsize);
- int tail = 0; /* tail position in HUNK */
+ int tail = 0; /* tail position in HUNK */
- assert (maxsize >= bufsize);
+ assert (!maxsize || maxsize >= bufsize);
while (1)
{
@@ -404,82 +417,82 @@ fd_read_hunk (int fd, hunk_terminator_t terminator, long sizehint, long maxsize)
pklen = fd_peek (fd, hunk + tail, bufsize - 1 - tail, -1);
if (pklen < 0)
- {
- xfree (hunk);
- return NULL;
- }
+ {
+ xfree (hunk);
+ return NULL;
+ }
end = terminator (hunk, hunk + tail, pklen);
if (end)
- {
- /* The data contains the terminator: we'll drain the data up
- to the end of the terminator. */
- remain = end - (hunk + tail);
- assert (remain >= 0);
- if (remain == 0)
- {
- /* No more data needs to be read. */
- hunk[tail] = '\0';
- return hunk;
- }
- if (bufsize - 1 < tail + remain)
- {
- bufsize = tail + remain + 1;
- hunk = xrealloc (hunk, bufsize);
- }
- }
+ {
+ /* The data contains the terminator: we'll drain the data up
+ to the end of the terminator. */
+ remain = end - (hunk + tail);
+ assert (remain >= 0);
+ if (remain == 0)
+ {
+ /* No more data needs to be read. */
+ hunk[tail] = '\0';
+ return hunk;
+ }
+ if (bufsize - 1 < tail + remain)
+ {
+ bufsize = tail + remain + 1;
+ hunk = xrealloc (hunk, bufsize);
+ }
+ }
else
- /* No terminator: simply read the data we know is (or should
- be) available. */
- remain = pklen;
+ /* No terminator: simply read the data we know is (or should
+ be) available. */
+ remain = pklen;
/* Now, read the data. Note that we make no assumptions about
- how much data we'll get. (Some TCP stacks are notorious for
- read returning less data than the previous MSG_PEEK.) */
+ how much data we'll get. (Some TCP stacks are notorious for
+ read returning less data than the previous MSG_PEEK.) */
rdlen = fd_read (fd, hunk + tail, remain, 0);
if (rdlen < 0)
- {
- xfree_null (hunk);
- return NULL;
- }
+ {
+ xfree_null (hunk);
+ return NULL;
+ }
tail += rdlen;
hunk[tail] = '\0';
if (rdlen == 0)
- {
- if (tail == 0)
- {
- /* EOF without anything having been read */
- xfree (hunk);
- errno = 0;
- return NULL;
- }
- else
- /* EOF seen: return the data we've read. */
- return hunk;
- }
+ {
+ if (tail == 0)
+ {
+ /* EOF without anything having been read */
+ xfree (hunk);
+ errno = 0;
+ return NULL;
+ }
+ else
+ /* EOF seen: return the data we've read. */
+ return hunk;
+ }
if (end && rdlen == remain)
- /* The terminator was seen and the remaining data drained --
- we got what we came for. */
- return hunk;
+ /* The terminator was seen and the remaining data drained --
+ we got what we came for. */
+ return hunk;
/* Keep looping until all the data arrives. */
if (tail == bufsize - 1)
- {
- /* Double the buffer size, but refuse to allocate more than
- MAXSIZE bytes. */
- if (maxsize && bufsize >= maxsize)
- {
- xfree (hunk);
- errno = ENOMEM;
- return NULL;
- }
- bufsize <<= 1;
- if (maxsize && bufsize > maxsize)
- bufsize = maxsize;
- hunk = xrealloc (hunk, bufsize);
- }
+ {
+ /* Double the buffer size, but refuse to allocate more than
+ MAXSIZE bytes. */
+ if (maxsize && bufsize >= maxsize)
+ {
+ xfree (hunk);
+ errno = ENOMEM;
+ return NULL;
+ }
+ bufsize <<= 1;
+ if (maxsize && bufsize > maxsize)
+ bufsize = maxsize;
+ hunk = xrealloc (hunk, bufsize);
+ }
}
}
@@ -526,8 +539,8 @@ retr_rate (wgint bytes, double secs)
/* Use more digits for smaller numbers (regardless of unit used),
e.g. "1022", "247", "12.5", "2.38". */
sprintf (res, "%.*f %s",
- dlrate >= 99.95 ? 0 : dlrate >= 9.995 ? 1 : 2,
- dlrate, rate_names[units]);
+ dlrate >= 99.95 ? 0 : dlrate >= 9.995 ? 1 : 2,
+ dlrate, rate_names[units]);
return res;
}
@@ -569,28 +582,22 @@ calc_rate (wgint bytes, double secs, int *units)
return dlrate;
}
-/* Maximum number of allowed redirections. 20 was chosen as a
- "reasonable" value, which is low enough to not cause havoc, yet
- high enough to guarantee that normal retrievals will not be hurt by
- the check. */
-
-#define MAX_REDIRECTIONS 20
-
-#define SUSPEND_POST_DATA do { \
- post_data_suspended = true; \
- saved_post_data = opt.post_data; \
- saved_post_file_name = opt.post_file_name; \
- opt.post_data = NULL; \
- opt.post_file_name = NULL; \
+
+#define SUSPEND_POST_DATA do { \
+ post_data_suspended = true; \
+ saved_post_data = opt.post_data; \
+ saved_post_file_name = opt.post_file_name; \
+ opt.post_data = NULL; \
+ opt.post_file_name = NULL; \
} while (0)
-#define RESTORE_POST_DATA do { \
- if (post_data_suspended) \
- { \
- opt.post_data = saved_post_data; \
- opt.post_file_name = saved_post_file_name; \
- post_data_suspended = false; \
- } \
+#define RESTORE_POST_DATA do { \
+ if (post_data_suspended) \
+ { \
+ opt.post_data = saved_post_data; \
+ opt.post_file_name = saved_post_file_name; \
+ post_data_suspended = false; \
+ } \
} while (0)
static char *getproxy (struct url *);
@@ -602,16 +609,18 @@ static char *getproxy (struct url *);
multiple points. */
uerr_t
-retrieve_url (const char *origurl, char **file, char **newloc,
- const char *refurl, int *dt, bool recursive)
+retrieve_url (struct url * orig_parsed, const char *origurl, char **file,
+ char **newloc, const char *refurl, int *dt, bool recursive,
+ struct iri *iri)
{
uerr_t result;
char *url;
bool location_changed;
+ bool iri_fallbacked = 0;
int dummy;
char *mynewloc, *proxy;
- struct url *u, *proxy_url;
- int up_error_code; /* url parse error code */
+ struct url *u = orig_parsed, *proxy_url;
+ int up_error_code; /* url parse error code */
char *local_file;
int redirection_count = 0;
@@ -631,18 +640,11 @@ retrieve_url (const char *origurl, char **file, char **newloc,
if (file)
*file = NULL;
- u = url_parse (url, &up_error_code);
- if (!u)
- {
- logprintf (LOG_NOTQUIET, "%s: %s.\n", url, url_error (up_error_code));
- xfree (url);
- return URLERROR;
- }
-
if (!refurl)
refurl = opt.referer;
redirected:
+ /* (also for IRI fallbacking) */
result = NOCONERROR;
mynewloc = NULL;
@@ -652,24 +654,30 @@ retrieve_url (const char *origurl, char **file, char **newloc,
proxy = getproxy (u);
if (proxy)
{
+ struct iri *pi = iri_new ();
+ set_uri_encoding (pi, opt.locale, true);
+ pi->utf8_encode = false;
+
/* Parse the proxy URL. */
- proxy_url = url_parse (proxy, &up_error_code);
+ proxy_url = url_parse (proxy, &up_error_code, NULL, true);
if (!proxy_url)
- {
- logprintf (LOG_NOTQUIET, _("Error parsing proxy URL %s: %s.\n"),
- proxy, url_error (up_error_code));
- xfree (url);
- RESTORE_POST_DATA;
- return PROXERR;
- }
+ {
+ char *error = url_error (proxy, up_error_code);
+ logprintf (LOG_NOTQUIET, _("Error parsing proxy URL %s: %s.\n"),
+ proxy, error);
+ xfree (url);
+ xfree (error);
+ RESTORE_POST_DATA;
+ return PROXERR;
+ }
if (proxy_url->scheme != SCHEME_HTTP && proxy_url->scheme != u->scheme)
- {
- logprintf (LOG_NOTQUIET, _("Error in proxy URL %s: Must be HTTP.\n"), proxy);
- url_free (proxy_url);
- xfree (url);
- RESTORE_POST_DATA;
- return PROXERR;
- }
+ {
+ logprintf (LOG_NOTQUIET, _("Error in proxy URL %s: Must be HTTP.\n"), proxy);
+ url_free (proxy_url);
+ xfree (url);
+ RESTORE_POST_DATA;
+ return PROXERR;
+ }
}
if (u->scheme == SCHEME_HTTP
@@ -678,29 +686,29 @@ retrieve_url (const char *origurl, char **file, char **newloc,
#endif
|| (proxy_url && proxy_url->scheme == SCHEME_HTTP))
{
- result = http_loop (u, &mynewloc, &local_file, refurl, dt, proxy_url);
+ result = http_loop (u, &mynewloc, &local_file, refurl, dt, proxy_url, iri);
}
else if (u->scheme == SCHEME_FTP)
{
/* If this is a redirection, temporarily turn off opt.ftp_glob
- and opt.recursive, both being undesirable when following
- redirects. */
+ and opt.recursive, both being undesirable when following
+ redirects. */
bool oldrec = recursive, glob = opt.ftp_glob;
if (redirection_count)
- oldrec = glob = false;
+ oldrec = glob = false;
result = ftp_loop (u, dt, proxy_url, recursive, glob);
recursive = oldrec;
/* There is a possibility of having HTTP being redirected to
- FTP. In these cases we must decide whether the text is HTML
- according to the suffix. The HTML suffixes are `.html',
- `.htm' and a few others, case-insensitive. */
+ FTP. In these cases we must decide whether the text is HTML
+ according to the suffix. The HTML suffixes are `.html',
+ `.htm' and a few others, case-insensitive. */
if (redirection_count && local_file && u->scheme == SCHEME_FTP)
- {
- if (has_html_suffix_p (local_file))
- *dt |= TEXTHTML;
- }
+ {
+ if (has_html_suffix_p (local_file))
+ *dt |= TEXTHTML;
+ }
}
if (proxy_url)
@@ -718,28 +726,39 @@ retrieve_url (const char *origurl, char **file, char **newloc,
assert (mynewloc != NULL);
if (local_file)
- xfree (local_file);
+ xfree (local_file);
/* The HTTP specs only allow absolute URLs to appear in
- redirects, but a ton of boneheaded webservers and CGIs out
- there break the rules and use relative URLs, and popular
- browsers are lenient about this, so wget should be too. */
+ redirects, but a ton of boneheaded webservers and CGIs out
+ there break the rules and use relative URLs, and popular
+ browsers are lenient about this, so wget should be too. */
construced_newloc = uri_merge (url, mynewloc);
xfree (mynewloc);
mynewloc = construced_newloc;
+ /* Reset UTF-8 encoding state, keep the URI encoding and reset
+ the content encoding. */
+ iri->utf8_encode = opt.enable_iri;
+ set_content_encoding (iri, NULL);
+ xfree_null (iri->orig_url);
+
/* Now, see if this new location makes sense. */
- newloc_parsed = url_parse (mynewloc, &up_error_code);
+ newloc_parsed = url_parse (mynewloc, &up_error_code, iri, true);
if (!newloc_parsed)
- {
- logprintf (LOG_NOTQUIET, "%s: %s.\n", escnonprint_uri (mynewloc),
- url_error (up_error_code));
- url_free (u);
- xfree (url);
- xfree (mynewloc);
- RESTORE_POST_DATA;
- return result;
- }
+ {
+ char *error = url_error (mynewloc, up_error_code);
+ logprintf (LOG_NOTQUIET, "%s: %s.\n", escnonprint_uri (mynewloc),
+ error);
+ if (orig_parsed != u)
+ {
+ url_free (u);
+ }
+ xfree (url);
+ xfree (mynewloc);
+ xfree (error);
+ RESTORE_POST_DATA;
+ return result;
+ }
/* Now mynewloc will become newloc_parsed->url, because if the
Location contained relative paths like .././something, we
@@ -748,46 +767,77 @@ retrieve_url (const char *origurl, char **file, char **newloc,
mynewloc = xstrdup (newloc_parsed->url);
/* Check for max. number of redirections. */
- if (++redirection_count > MAX_REDIRECTIONS)
- {
- logprintf (LOG_NOTQUIET, _("%d redirections exceeded.\n"),
- MAX_REDIRECTIONS);
- url_free (newloc_parsed);
- url_free (u);
- xfree (url);
- xfree (mynewloc);
- RESTORE_POST_DATA;
- return WRONGCODE;
- }
+ if (++redirection_count > opt.max_redirect)
+ {
+ logprintf (LOG_NOTQUIET, _("%d redirections exceeded.\n"),
+ opt.max_redirect);
+ url_free (newloc_parsed);
+ if (orig_parsed != u)
+ {
+ url_free (u);
+ }
+ xfree (url);
+ xfree (mynewloc);
+ RESTORE_POST_DATA;
+ return WRONGCODE;
+ }
xfree (url);
url = mynewloc;
- url_free (u);
+ if (orig_parsed != u)
+ {
+ url_free (u);
+ }
u = newloc_parsed;
/* If we're being redirected from POST, we don't want to POST
- again. Many requests answer POST with a redirection to an
- index page; that redirection is clearly a GET. We "suspend"
- POST data for the duration of the redirections, and restore
- it when we're done. */
+ again. Many requests answer POST with a redirection to an
+ index page; that redirection is clearly a GET. We "suspend"
+ POST data for the duration of the redirections, and restore
+ it when we're done. */
if (!post_data_suspended)
- SUSPEND_POST_DATA;
+ SUSPEND_POST_DATA;
goto redirected;
}
- if (local_file)
+ /* Try to not encode in UTF-8 if fetching failed */
+ if (!(*dt & RETROKF) && iri->utf8_encode)
+ {
+ iri->utf8_encode = false;
+ if (orig_parsed != u)
+ {
+ url_free (u);
+ }
+ u = url_parse (origurl, NULL, iri, true);
+ if (u)
+ {
+ DEBUGP (("[IRI fallbacking to non-utf8 for %s\n", quote (url)));
+ url = xstrdup (u->url);
+ iri_fallbacked = 1;
+ goto redirected;
+ }
+ else
+ DEBUGP (("[Couldn't fallback to non-utf8 for %s\n", quote (url)));
+ }
+
+ if (local_file && *dt & RETROKF)
{
+ register_download (u->url, local_file);
+ if (redirection_count && 0 != strcmp (origurl, u->url))
+ register_redirection (origurl, u->url);
+ if (*dt & TEXTHTML)
+ register_html (u->url, local_file);
if (*dt & RETROKF)
- {
- register_download (u->url, local_file);
- if (redirection_count && 0 != strcmp (origurl, u->url))
- register_redirection (origurl, u->url);
- if (*dt & TEXTHTML)
- register_html (u->url, local_file);
- if (*dt & TEXTCSS)
- register_css (u->url, local_file);
- }
+ {
+ register_download (u->url, local_file);
+ if (redirection_count && 0 != strcmp (origurl, u->url))
+ register_redirection (origurl, u->url);
+ if (*dt & TEXTHTML)
+ register_html (u->url, local_file);
+ if (*dt & TEXTCSS)
+ register_css (u->url, local_file);
+ }
}
if (file)
@@ -795,19 +845,22 @@ retrieve_url (const char *origurl, char **file, char **newloc,
else
xfree_null (local_file);
- url_free (u);
+ if (orig_parsed != u)
+ {
+ url_free (u);
+ }
- if (redirection_count)
+ if (redirection_count || iri_fallbacked)
{
if (newloc)
- *newloc = url;
+ *newloc = url;
else
- xfree (url);
+ xfree (url);
}
else
{
if (newloc)
- *newloc = NULL;
+ *newloc = NULL;
xfree (url);
}
@@ -827,58 +880,121 @@ retrieve_from_file (const char *file, bool html, int *count)
{
uerr_t status;
struct urlpos *url_list, *cur_url;
+ struct iri *iri = iri_new();
+
+ char *input_file = NULL;
+ const char *url = file;
- url_list = (html ? get_urls_html (file, NULL, NULL)
- : get_urls_file (file));
status = RETROK; /* Suppose everything is OK. */
*count = 0; /* Reset the URL count. */
+ /* sXXXav : Assume filename and links in the file are in the locale */
+ set_uri_encoding (iri, opt.locale, true);
+ set_content_encoding (iri, opt.locale);
+
+ if (url_has_scheme (url))
+ {
+ int dt,url_err;
+ uerr_t status;
+ struct url * url_parsed = url_parse(url, &url_err, iri, true);
+
+ if (!url_parsed)
+ {
+ char *error = url_error (url, url_err);
+ logprintf (LOG_NOTQUIET, "%s: %s.\n", url, error);
+ xfree (error);
+ return URLERROR;
+ }
+
+ if (!opt.base_href)
+ opt.base_href = xstrdup (url);
+
+ status = retrieve_url (url_parsed, url, &input_file, NULL, NULL, &dt,
+ false, iri);
+ if (status != RETROK)
+ return status;
+
+ if (dt & TEXTHTML)
+ html = true;
+
+ /* If we have a found a content encoding, use it.
+ * ( == is okay, because we're checking for identical object) */
+ if (iri->content_encoding != opt.locale)
+ set_uri_encoding (iri, iri->content_encoding, false);
+
+ /* Reset UTF-8 encode status */
+ iri->utf8_encode = opt.enable_iri;
+ xfree_null (iri->orig_url);
+ iri->orig_url = NULL;
+ }
+ else
+ input_file = (char *) file;
+
+ url_list = (html ? get_urls_html (input_file, NULL, NULL, iri)
+ : get_urls_file (input_file));
+
for (cur_url = url_list; cur_url; cur_url = cur_url->next, ++*count)
{
char *filename = NULL, *new_file = NULL;
int dt;
+ struct iri *tmpiri = iri_dup (iri);
+ struct url *parsed_url = NULL;
if (cur_url->ignore_when_downloading)
- continue;
+ continue;
if (opt.quota && total_downloaded_bytes > opt.quota)
- {
- status = QUOTEXC;
- break;
- }
+ {
+ status = QUOTEXC;
+ break;
+ }
+
+ /* Need to reparse the url, since it didn't have iri information. */
+ if (opt.enable_iri)
+ parsed_url = url_parse (cur_url->url->url, NULL, tmpiri, true);
+
if ((opt.recursive || opt.page_requisites)
- && (cur_url->url->scheme != SCHEME_FTP || getproxy (cur_url->url)))
- {
- int old_follow_ftp = opt.follow_ftp;
-
- /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
- if (cur_url->url->scheme == SCHEME_FTP)
- opt.follow_ftp = 1;
-
- status = retrieve_tree (cur_url->url->url);
-
- opt.follow_ftp = old_follow_ftp;
- }
+ && (cur_url->url->scheme != SCHEME_FTP || getproxy (cur_url->url)))
+ {
+ int old_follow_ftp = opt.follow_ftp;
+
+ /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
+ if (cur_url->url->scheme == SCHEME_FTP)
+ opt.follow_ftp = 1;
+
+ status = retrieve_tree (parsed_url ? parsed_url : cur_url->url,
+ tmpiri);
+
+ opt.follow_ftp = old_follow_ftp;
+ }
else
- status = retrieve_url (cur_url->url->url, &filename, &new_file, NULL, &dt, opt.recursive);
+ status = retrieve_url (parsed_url ? parsed_url : cur_url->url,
+ cur_url->url->url, &filename,
+ &new_file, NULL, &dt, opt.recursive, tmpiri);
+
+ if (parsed_url)
+ url_free (parsed_url);
if (filename && opt.delete_after && file_exists_p (filename))
- {
- DEBUGP (("\
+ {
+ DEBUGP (("\
Removing file due to --delete-after in retrieve_from_file():\n"));
- logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
- if (unlink (filename))
- logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
- dt &= ~RETROKF;
- }
+ logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
+ if (unlink (filename))
+ logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
+ dt &= ~RETROKF;
+ }
xfree_null (new_file);
xfree_null (filename);
+ iri_free (tmpiri);
}
/* Free the linked list of URL-s. */
free_urlpos (url_list);
+ iri_free (iri);
+
return status;
}
@@ -911,29 +1027,29 @@ sleep_between_retrievals (int count)
if (opt.waitretry && count > 1)
{
/* If opt.waitretry is specified and this is a retry, wait for
- COUNT-1 number of seconds, or for opt.waitretry seconds. */
+ COUNT-1 number of seconds, or for opt.waitretry seconds. */
if (count <= opt.waitretry)
- xsleep (count - 1);
+ xsleep (count - 1);
else
- xsleep (opt.waitretry);
+ xsleep (opt.waitretry);
}
else if (opt.wait)
{
if (!opt.random_wait || count > 1)
- /* If random-wait is not specified, or if we are sleeping
- between retries of the same download, sleep the fixed
- interval. */
- xsleep (opt.wait);
+ /* If random-wait is not specified, or if we are sleeping
+ between retries of the same download, sleep the fixed
+ interval. */
+ xsleep (opt.wait);
else
- {
- /* Sleep a random amount of time averaging in opt.wait
- seconds. The sleeping amount ranges from 0.5*opt.wait to
- 1.5*opt.wait. */
- double waitsecs = (0.5 + random_float ()) * opt.wait;
- DEBUGP (("sleep_between_retrievals: avg=%f,sleep=%f\n",
- opt.wait, waitsecs));
- xsleep (waitsecs);
- }
+ {
+ /* Sleep a random amount of time averaging in opt.wait
+ seconds. The sleeping amount ranges from 0.5*opt.wait to
+ 1.5*opt.wait. */
+ double waitsecs = (0.5 + random_float ()) * opt.wait;
+ DEBUGP (("sleep_between_retrievals: avg=%f,sleep=%f\n",
+ opt.wait, waitsecs));
+ xsleep (waitsecs);
+ }
}
}
@@ -945,7 +1061,7 @@ free_urlpos (struct urlpos *l)
{
struct urlpos *next = l->next;
if (l->url)
- url_free (l->url);
+ url_free (l->url);
xfree_null (l->local_name);
xfree (l);
l = next;
@@ -1028,14 +1144,12 @@ getproxy (struct url *u)
/* Returns true if URL would be downloaded through a proxy. */
bool
-url_uses_proxy (const char *url)
+url_uses_proxy (struct url * u)
{
bool ret;
- struct url *u = url_parse (url, NULL);
if (!u)
return false;
ret = getproxy (u) != NULL;
- url_free (u);
return ret;
}
@@ -1048,3 +1162,16 @@ no_proxy_match (const char *host, const char **no_proxy)
else
return sufmatch (no_proxy, host);
}
+
+/* Set the file parameter to point to the local file string. */
+void
+set_local_file (const char **file, const char *default_file)
+{
+ if (opt.output_document)
+ {
+ if (output_stream_regular)
+ *file = opt.output_document;
+ }
+ else
+ *file = default_file;
+}