Ted Mielczarek's CSS wonder-patch, applied against the source from around the time...

[wget] / src / retr.c

diff --git a/src/retr.c b/src/retr.c
index 37d63273d199b26f1d3f8bf88569a9eca304a89e..245eb129f40d5f49644d0bb69de1e7489f54cf1d 100644 (file)
--- a/src/retr.c
+++ b/src/retr.c
@@ -1,5 +1,5 @@
 /* File retrieval.
-   Copyright (C) 1996-2005 Free Software Foundation, Inc.
+   Copyright (C) 1996-2006 Free Software Foundation, Inc.
 
 This file is part of GNU Wget.
 
@@ -51,6 +51,7 @@ so, delete this exception statement from your version.  */
 #include "hash.h"
 #include "convert.h"
 #include "ptimer.h"
+#include "html-url.h"
 
 /* Total size of downloaded files.  Used to enforce quota.  */
 SUM_SIZE_INT total_downloaded_bytes;
@@ -515,13 +516,13 @@ fd_read_line (int fd)
    the units appropriate for the download speed.  */
 
 const char *
-retr_rate (wgint bytes, double msecs)
+retr_rate (wgint bytes, double secs)
 {
   static char res[20];
   static const char *rate_names[] = {"B/s", "KB/s", "MB/s", "GB/s" };
   int units;
 
-  double dlrate = calc_rate (bytes, msecs, &units);
+  double dlrate = calc_rate (bytes, secs, &units);
   /* Use more digits for smaller numbers (regardless of unit used),
      e.g. "1022", "247", "12.5", "2.38".  */
   sprintf (res, "%.*f %s",
@@ -602,7 +603,7 @@ static char *getproxy (struct url *);
 
 uerr_t
 retrieve_url (const char *origurl, char **file, char **newloc,
-             const char *refurl, int *dt)
+             const char *refurl, int *dt, bool recursive)
 {
   uerr_t result;
   char *url;
@@ -684,13 +685,12 @@ retrieve_url (const char *origurl, char **file, char **newloc,
       /* If this is a redirection, temporarily turn off opt.ftp_glob
         and opt.recursive, both being undesirable when following
         redirects.  */
-      bool oldrec = opt.recursive, oldglob = opt.ftp_glob;
+      bool oldrec = recursive, glob = opt.ftp_glob;
       if (redirection_count)
-       opt.recursive = opt.ftp_glob = false;
+       oldrec = glob = false;
 
-      result = ftp_loop (u, dt, proxy_url);
-      opt.recursive = oldrec;
-      opt.ftp_glob = oldglob;
+      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
@@ -785,6 +785,8 @@ retrieve_url (const char *origurl, char **file, char **newloc,
            register_redirection (origurl, u->url);
          if (*dt & TEXTHTML)
            register_html (u->url, local_file);
+         if (*dt & TEXTCSS)
+           register_css (u->url, local_file);
        }
     }
 
@@ -845,10 +847,20 @@ retrieve_from_file (const char *file, bool html, int *count)
          break;
        }
       if ((opt.recursive || opt.page_requisites)
-         && cur_url->url->scheme != SCHEME_FTP)
-       status = retrieve_tree (cur_url->url->url);
+         && (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;
+       }
       else
-       status = retrieve_url (cur_url->url->url, &filename, &new_file, NULL, &dt);
+       status = retrieve_url (cur_url->url->url, &filename, &new_file, NULL, &dt, opt.recursive);
 
       if (filename && opt.delete_after && file_exists_p (filename))
        {
@@ -915,9 +927,9 @@ sleep_between_retrievals (int count)
       else
        {
          /* Sleep a random amount of time averaging in opt.wait
-            seconds.  The sleeping amount ranges from 0 to
-            opt.wait*2, inclusive.  */
-         double waitsecs = 2 * opt.wait * random_float ();
+            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);
@@ -978,7 +990,7 @@ getproxy (struct url *u)
 
   if (!opt.use_proxy)
     return NULL;
-  if (!no_proxy_match (u->host, (const char **)opt.no_proxy))
+  if (no_proxy_match (u->host, (const char **)opt.no_proxy))
     return NULL;
 
   switch (u->scheme)
@@ -1013,12 +1025,26 @@ getproxy (struct url *u)
   return proxy;
 }
 
+/* Returns true if URL would be downloaded through a proxy. */
+
+bool
+url_uses_proxy (const char *url)
+{
+  bool ret;
+  struct url *u = url_parse (url, NULL);
+  if (!u)
+    return false;
+  ret = getproxy (u) != NULL;
+  url_free (u);
+  return ret;
+}
+
 /* Should a host be accessed through proxy, concerning no_proxy?  */
 static bool
 no_proxy_match (const char *host, const char **no_proxy)
 {
   if (!no_proxy)
-    return true;
+    return false;
   else
-    return !sufmatch (no_proxy, host);
+    return sufmatch (no_proxy, host);
 }