Updated config.guess, config.sub, install.sh.

[wget] / src / retr.c

diff --git a/src/retr.c b/src/retr.c
index 2a9dec7bbb658aa6bdb12365813c0121142e7cdb..b667ca2ff3cf6ecb4edca9c1148e856f68b0ecbe 100644 (file)
--- a/src/retr.c
+++ b/src/retr.c
@@ -52,6 +52,7 @@ as that of the covered work.  */
 #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;
@@ -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);
 }
 
@@ -604,6 +616,7 @@ retrieve_url (struct url * orig_parsed, const char *origurl, char **file,
   uerr_t result;
   char *url;
   bool location_changed;
+  bool iri_fallbacked = 0;
   int dummy;
   char *mynewloc, *proxy;
   struct url *u = orig_parsed, *proxy_url;
@@ -627,15 +640,11 @@ retrieve_url (struct url * orig_parsed, const char *origurl, char **file,
   if (file)
     *file = NULL;
 
- second_try:
-  DEBUGP (("[IRI Retrieving %s with %s (UTF-8=%d)\n", quote_n (0, url),
-           iri->uri_encoding ? quote_n (1, iri->uri_encoding) : "None",
-           iri->utf8_encode));
-
   if (!refurl)
     refurl = opt.referer;
 
  redirected:
+  /* (also for IRI fallbacking) */
 
   result = NOCONERROR;
   mynewloc = NULL;
@@ -804,7 +813,9 @@ retrieve_url (struct url * orig_parsed, const char *origurl, char **file,
       if (u)
         {
           DEBUGP (("[IRI fallbacking to non-utf8 for %s\n", quote (url)));
-          goto second_try;
+          url = xstrdup (u->url);
+          iri_fallbacked = 1;
+          goto redirected;
         }
       else
           DEBUGP (("[Couldn't fallback to non-utf8 for %s\n", quote (url)));
@@ -839,7 +850,7 @@ retrieve_url (struct url * orig_parsed, const char *origurl, char **file,
       url_free (u);
     }
 
-  if (redirection_count)
+  if (redirection_count || iri_fallbacked)
     {
       if (newloc)
         *newloc = url;
@@ -885,7 +896,7 @@ retrieve_from_file (const char *file, bool html, int *count)
     {
       int dt,url_err;
       uerr_t status;
-      struct url * url_parsed = url_parse(url, &url_err, NULL, true);
+      struct url * url_parsed = url_parse(url, &url_err, iri, true);
 
       if (!url_parsed)
         {
@@ -906,9 +917,15 @@ retrieve_from_file (const char *file, bool html, int *count)
       if (dt & TEXTHTML)
         html = true;
 
-      /* If we have a found a content encoding, use it */
-      if (iri->content_encoding)
+      /* 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;
@@ -920,6 +937,8 @@ retrieve_from_file (const char *file, bool html, int *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;
@@ -930,10 +949,9 @@ retrieve_from_file (const char *file, bool html, int *count)
           break;
         }
 
-      /* Reset UTF-8 encode status */
-      iri->utf8_encode = opt.enable_iri;
-      xfree_null (iri->orig_url);
-      iri->orig_url = NULL;
+      /* 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)))
@@ -944,13 +962,18 @@ retrieve_from_file (const char *file, bool html, int *count)
           if (cur_url->url->scheme == SCHEME_FTP)
             opt.follow_ftp = 1;
 
-          status = retrieve_tree (cur_url->url, iri);
+          status = retrieve_tree (parsed_url ? parsed_url : cur_url->url,
+                                  tmpiri);
 
           opt.follow_ftp = old_follow_ftp;
         }
       else
-        status = retrieve_url (cur_url->url, cur_url->url->url, &filename,
-                               &new_file, NULL, &dt, opt.recursive, iri);
+        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))
         {
@@ -964,6 +987,7 @@ Removing file due to --delete-after in retrieve_from_file():\n"));
 
       xfree_null (new_file);
       xfree_null (filename);
+      iri_free (tmpiri);
     }
 
   /* Free the linked list of URL-s.  */