[svn] Better version of read_whole_line().

Published in <sxsr94jd7z4.fsf@florida.arsdigita.de>.

src/ChangeLog

@@ -1,3 +1,19 @@
+2000-11-10  Hrvoje Niksic  <hniksic@arsdigita.com>
+
+	* init.c (run_wgetrc): Don't bother killing off '\r' since
+	pars_line() skips whitespace at end of line anyway.
+	(parse_line): Oops, it didn't.  Now it does.
+
+	* recur.c (parse_robots): Ditto here.
+
+	* ftp-ls.c (ftp_parse_unix_ls): Kill off the newline character
+	manually because read_whole_line no longer does.
+
+	* utils.c (read_whole_line): Rewrite to: a) use less memory
+	(reallocates to needed size after work), b) work faster -->
+	fgets() instead of getc, c) be more correct --> doesn't kill the
+	newline character at the end of line.
+
 2000-11-10  Hrvoje Niksic  <hniksic@arsdigita.com>
 
 	* init.c (comind): Initialize MAX to array size - 1.

src/ftp-ls.c

@@ -95,7 +95,9 @@ ftp_parse_unix_ls (const char *file)
     {
       DEBUGP (("%s\n", line));
       len = strlen (line);
-      /* Destroy <CR> if there is one.  */
+      /* Destroy <CR><LF> if present.  */
+      if (len && line[len - 1] == '\n')
+	line[--len] = '\0';
       if (len && line[len - 1] == '\r')
 	line[--len] = '\0';
 

src/init.c

@@ -336,10 +336,7 @@ run_wgetrc (const char *file)
     {
       char *com, *val;
       int status;
-      int length = strlen (line);
 
-      if (length && line[length - 1] == '\r')
-	line[length - 1] = '\0';
       /* Parse the line.  */
       status = parse_line (line, &com, &val);
       free (line);
@@ -411,12 +408,12 @@ parse_line (const char *line, char **com, char **val)
   const char *orig_comptr, *end;
   char *new_comptr;
 
-  /* Skip spaces.  */
-  while (*p == ' ' || *p == '\t')
+  /* Skip whitespace.  */
+  while (*p && ISSPACE (*p))
     ++p;
 
   /* Don't process empty lines.  */
-  if (!*p || *p == '\n' || *p == '#')
+  if (!*p || *p == '#')
     return -1;
 
   for (orig_comptr = p; ISALPHA (*p) || *p == '_' || *p == '-'; p++)
@@ -424,6 +421,8 @@ parse_line (const char *line, char **com, char **val)
   /* The next char should be space or '='.  */
   if (!ISSPACE (*p) && (*p != '='))
     return 0;
+  /* Here we cannot use strdupdelim() as we normally would because we
+     want to skip the `-' and `_' characters in the input string.  */
   *com = (char *)xmalloc (p - orig_comptr + 1);
   for (new_comptr = *com; orig_comptr < p; orig_comptr++)
     {
@@ -449,10 +448,12 @@ parse_line (const char *line, char **com, char **val)
     }
   /* Skip spaces after '='.  */
   for (++p; ISSPACE (*p); p++);
-  /* Get the ending position.  */
-  for (end = p; *end && *end != '\n'; end++);
-  /* Allocate *val, and copy from line.  */
-  *val = strdupdelim (p, end);
+  /* Get the ending position for VAL by starting with the end of the
+     line and skipping whitespace.  */
+  end = line + strlen (line) - 1;
+  while (end > p && ISSPACE (*end))
+    --end;
+  *val = strdupdelim (p, end + 1);
   return 1;
 }
 

src/recur.c

@@ -702,9 +702,11 @@ parse_robots (const char *robots_filename)
   while ((line = read_whole_line (fp)))
     {
       len = strlen (line);
-      /* Destroy <CR> if there is one.  */
+      /* Destroy <CR><LF> if present.  */
+      if (len && line[len - 1] == '\n')
+	line[--len] = '\0';
       if (len && line[len - 1] == '\r')
-	line[len - 1] = '\0';
+	line[--len] = '\0';
       /* According to specifications, optional space may be at the
 	 end...  */
       DEBUGP (("Line: %s\n", line));

src/utils.c

@@ -45,6 +45,7 @@ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.  */
 #ifdef NeXT
 # include <libc.h>		/* for access() */
 #endif
+#include <assert.h>
 
 #include "wget.h"
 #include "utils.h"
@@ -696,37 +697,43 @@ suffix (const char *str)
 /* Read a line from FP.  The function reallocs the storage as needed
    to accomodate for any length of the line.  Reallocs are done
    storage exponentially, doubling the storage after each overflow to
-   minimize the number of calls to realloc().
+   minimize the number of calls to realloc() and fgets().  The newline
+   character at the end of line is retained.
+
+   After end-of-file is encountered without anything being read, NULL
+   is returned.  NULL is also returned on error.  To distinguish
+   between these two cases, use the stdio function ferror().  */
 
-   It is not an exemplary of correctness, since it kills off the
-   newline (and no, there is no way to know if there was a newline at
-   EOF).  */
 char *
 read_whole_line (FILE *fp)
 {
-  char *line;
-  int i, bufsize, c;
+  int length = 0;
+  int bufsize = 81;
+  char *line = (char *)xmalloc (bufsize);
 
-  i = 0;
-  bufsize = 40;
-  line = (char *)xmalloc (bufsize);
-  /* Construct the line.  */
-  while ((c = getc (fp)) != EOF && c != '\n')
+  while (fgets (line + length, bufsize - length, fp))
     {
-      if (i > bufsize - 1)
-	line = (char *)xrealloc (line, (bufsize <<= 1));
-      line[i++] = c;
+      length += strlen (line + length);
+      assert (length > 0);
+      if (line[length - 1] == '\n')
+	break;
+      /* fgets() guarantees to read the whole line, or to use up the
+         space we've given it.  We can double the buffer
+         unconditionally.  */
+      bufsize <<= 1;
+      line = xrealloc (line, bufsize);
     }
-  if (c == EOF && !i)
+  if (length == 0 || ferror (fp))
     {
       free (line);
       return NULL;
     }
-  /* Check for overflow at zero-termination (no need to double the
-     buffer in this case.  */
-  if (i == bufsize)
-    line = (char *)xrealloc (line, i + 1);
-  line[i] = '\0';
+  if (length + 1 < bufsize)
+    /* Relieve the memory from our exponential greediness.  We say
+       `length + 1' because the terminating \0 is not included in
+       LENGTH.  We don't need to zero-terminate the string ourselves,
+       though, because fgets() does that.  */
+    line = xrealloc (line, length + 1);
   return line;
 }