Add support for file names longer than MAX_FILE.

NEWS

@@ -5,6 +5,11 @@ Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
 See the end for copying conditions.
 
 Please send GNU Wget bug reports to <bug-wget@gnu.org>.
+
+* Changes in Wget X.Y.Z
+
+** Add support for file names longer than MAX_FILE.
+
 
 * Changes in Wget 1.14
 

src/ChangeLog

@@ -1,3 +1,11 @@
+2012-09-29  Tim Ruehsen  <tim.ruehsen@gmx.de>
+
+	* url.h (CHOMP_BUFFER): Add definition.
+	* url.c (url_file_name): New local variables `fname_len_check' and
+	`max_length'. Check that the length of the file name is acceptable.
+	* utils.h (get_max_length): Declare function.
+	* utils.c (get_max_length): New function.
+
 2012-09-28  Steven Schubiger  <stsc@member.fsf.org>
 
 	* src/recur.c (retrieve_tree): Combine duplicated code.

src/url.c

@@ -1361,6 +1361,7 @@ UWC,  C,  C,  C,   C,  C,  C,  C,   /* NUL SOH STX ETX  EOT ENQ ACK BEL */
    query, normally '?'.  Since Windows cannot handle '?' as part of
    file name, we use '@' instead there.  */
 #define FN_QUERY_SEP (opt.restrict_files_os != restrict_windows ? '?' : '@')
+#define FN_QUERY_SEP_STR (opt.restrict_files_os != restrict_windows ? "?" : "@")
 
 /* Quote path element, characters in [b, e), as file name, and append
    the quoted string to DEST.  Each character is quoted as per
@@ -1494,22 +1495,28 @@ append_dir_structure (const struct url *u, struct growable *dest)
     }
 }
 
-/* Return a unique file name that matches the given URL as good as
+/* Return a unique file name that matches the given URL as well as
    possible.  Does not create directories on the file system.  */
 
 char *
 url_file_name (const struct url *u, char *replaced_filename)
 {
   struct growable fnres;        /* stands for "file name result" */
+  struct growable temp_fnres;
 
   const char *u_file;
-  char *fname, *unique;
+  char *fname, *unique, *fname_len_check;
   const char *index_filename = "index.html"; /* The default index file is index.html */
+  size_t max_length;
 
   fnres.base = NULL;
   fnres.size = 0;
   fnres.tail = 0;
 
+  temp_fnres.base = NULL;
+  temp_fnres.size = 0;
+  temp_fnres.tail = 0;
+
   /* If an alternative index file was defined, change index_filename */
   if (opt.default_page)
     index_filename = opt.default_page;
@@ -1555,33 +1562,60 @@ url_file_name (const struct url *u, char *replaced_filename)
 
   if (!replaced_filename)
     {
-      /* Add the file name. */
-      if (fnres.tail)
-	append_char ('/', &fnres);
+      /* Create the filename. */
       u_file = *u->file ? u->file : index_filename;
-      append_uri_pathel (u_file, u_file + strlen (u_file), false, &fnres);
 
-      /* Append "?query" to the file name, even if empty */
+      /* Append "?query" to the file name, even if empty,
+       * and create fname_len_check. */
       if (u->query)
-	{
-	  append_char (FN_QUERY_SEP, &fnres);
-	  append_uri_pathel (u->query, u->query + strlen (u->query),
-			     true, &fnres);
-	}
+        fname_len_check = concat_strings (u_file, FN_QUERY_SEP_STR, u->query, NULL);
+      else
+        fname_len_check = strdupdelim (u_file, u_file + strlen (u_file));
     }
   else
     {
-      if (fnres.tail)
-	append_char ('/', &fnres);
       u_file = replaced_filename;
-      append_uri_pathel (u_file, u_file + strlen (u_file), false, &fnres);
+      fname_len_check = strdupdelim (u_file, u_file + strlen (u_file));
     }
 
+  append_uri_pathel (fname_len_check,
+    fname_len_check + strlen (fname_len_check), false, &temp_fnres);
+
+  /* Zero-terminate the temporary file name. */
+  append_char ('\0', &temp_fnres);
+
+  /* Check that the length of the file name is acceptable. */
+  max_length = get_max_length (fnres.base, fnres.tail, _PC_NAME_MAX) - CHOMP_BUFFER;
+  if (max_length > 0 && strlen (temp_fnres.base) > max_length)
+    {
+      logprintf (LOG_NOTQUIET, "The name is too long, %lu chars total.\n",
+          (unsigned long) strlen (temp_fnres.base));
+      logprintf (LOG_NOTQUIET, "Trying to shorten...\n");
+
+      /* Shorten the file name. */
+      temp_fnres.base[max_length] = '\0';
+
+      logprintf (LOG_NOTQUIET, "New name is %s.\n", temp_fnres.base);
+    }
+
+  free (fname_len_check);
+
+  /* The filename has already been 'cleaned' by append_uri_pathel() above.  So,
+   * just append it. */
+  if (fnres.tail)
+    append_char ('/', &fnres);
+  append_string (temp_fnres.base, &fnres);
+
   /* Zero-terminate the file name. */
   append_char ('\0', &fnres);
 
   fname = fnres.base;
 
+  /* Make a final check that the path length is acceptable? */
+  /* TODO: check fnres.base for path length problem */
+
+  free (temp_fnres.base);
+
   /* Check the cases in which the unique extensions are not used:
      1) Clobbering is turned off (-nc).
      2) Retrieval with regetting.

src/url.h

@@ -37,6 +37,16 @@ as that of the covered work.  */
 #define DEFAULT_FTP_PORT 21
 #define DEFAULT_HTTPS_PORT 443
 
+/* This represents how many characters less than the OS max name length a file
+ * should be.  More precisely, a file name should be at most
+ * (NAME_MAX - CHOMP_BUFFER) characters in length.  This number was arrived at
+ * by adding the lengths of all possible strings that could be appended to a
+ * file name later in the code (e.g. ".orig", ".html", etc.).  This is
+ * hopefully plenty of extra characters, but I am not guaranteeing that a file
+ * name will be of the proper length by the time the code wants to open a
+ * file descriptor. */
+#define CHOMP_BUFFER 19
+
 /* Specifies how, or whether, user auth information should be included
  * in URLs regenerated from URL parse structures. */
 enum url_auth_mode {

src/utils.c

@@ -2494,6 +2494,56 @@ print_decimal (double number)
   return buf;
 }
 
+/* Get the maximum name length for the given path. */
+/* Return 0 if length is unknown. */
+size_t
+get_max_length (const char *path, int length, int name)
+{
+  long ret;
+  char *p, *d;
+
+  /* Make a copy of the path that we can modify. */
+  p = path ? strdupdelim (path, path + length) : strdup ("");
+
+  for (;;)
+    {
+      errno = 0;
+      /* For an empty path query the current directory. */
+      ret = pathconf (*p ? p : ".", name);
+      if (!(ret < 0 && errno == ENOENT))
+        break;
+
+      /* The path does not exist yet, but may be created. */
+      /* Already at current or root directory, give up. */
+      if (!*p || strcmp (p, "/") == 0)
+        break;
+
+      /* Remove one directory level and try again. */
+      d = strrchr (p, '/');
+      if (d == p)
+        p[1] = '\0';  /* check root directory */
+      else if (d)
+        *d = '\0';  /* remove last directory part */
+      else
+        *p = '\0';  /* check current directory */
+    }
+
+  xfree (p);
+
+  if (ret < 0)
+    {
+      /* pathconf() has a message for us. */
+      if (errno != 0)
+          perror ("pathconf");
+
+      /* If (errno == 0) then there is no max length.
+         Even on error return 0 so the caller can continue. */
+      return 0;
+    }
+
+  return ret;
+}
+
 #ifdef TESTING
 
 const char *

src/utils.h

@@ -155,6 +155,8 @@ void stable_sort (void *, size_t, size_t, int (*) (const void *, const void *));
 
 const char *print_decimal (double);
 
+size_t get_max_length (const char *path, int length, int name);
+
 extern unsigned char char_prop[];
 
 #endif /* UTILS_H */