Re: FIELDWIDTHS broken in awk 3.1.5

[Aharon Robbins]

Greetings.  This got sorta buried in my inbox.  Sorry.

> Date: Wed, 25 Jan 2006 20:01:18 -0700
> From: Glenn Zazulia <address@hidden>
> Subject: Re: FIELDWIDTHS broken in awk 3.1.5
> To: address@hidden
> Cc: address@hidden
>
> Aharon,
>
> I found your FIELDWIDTHS patch which fixes a problem that I also noticed 
> with the 3.1.5 gawk release.
>
> I also noticed a related possible problem with FIELDWIDTHS in the gawk 
> 3.1.5 release.  In gawk 3.1.4 and earlier, the statement
>
>     FIELDWIDTHS = ""
>
> used to work -- that is, it was executed without error and a subsequent 
> "print FIELDWIDTHS" statement showed that any previous value was unset.  
> Running that statement through gawk 3.1.5  produces the following fatal 
> error:
>
>     gawk: (FILENAME=- FNR=1) fatal: invalid FIELDWIDTHS value, near `'
>
> Now, one might question what that statement should do, and one might 
> argue that instead of being a new bug, this might be a fix of previously 
> improper behavior.  I don't think so.  Since the FIELDWIDTHS variable is 
> a gawk extension and since the documentation doesn't state what the 
> behavior should be in this case, it's difficult for me to argue one way 
> or the other.  However, this change in behavior in 3.1.5 breaks existing 
> gawk scripts, which is a problem.

I have restored the previous behavior.  I decided not to really think about
what the "correct" thing is, preferring just to not crash and to have things
work as they used to.

Attached is a diff of my current field.c against vanilla 3.1.5.

Thanks!

Arnold

--- ../gawk-3.1.5/field.c       2005-05-11 18:28:15.000000000 +0300
+++ field.c     2007-09-22 22:23:05.000000000 +0200
@@ -3,14 +3,14 @@
  */
 
 /* 
- * Copyright (C) 1986, 1988, 1989, 1991-2005 the Free Software Foundation, Inc.
+ * Copyright (C) 1986, 1988, 1989, 1991-2007 the Free Software Foundation, Inc.
  * 
  * This file is part of GAWK, the GNU implementation of the
  * AWK Programming Language.
  * 
  * GAWK 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
+ * the Free Software Foundation; either version 3 of the License, or
  * (at your option) any later version.
  * 
  * GAWK is distributed in the hope that it will be useful,
@@ -166,6 +166,7 @@
        cops = ops;
        ops[0] = '\0';
        for (i = 1;  i <= NF; i++) {
+               free_wstr(fields_arr[i]);
                tmp = fields_arr[i];
                /* copy field */
                if (tmp->stlen == 1)
@@ -212,6 +213,9 @@
                        n->stptr = cops;
                        unref(fields_arr[i]);
                        fields_arr[i] = n;
+#ifdef MBS_SUPPORT
+                       assert((n->flags & WSTRCUR) == 0);
+#endif
                }
                cops += fields_arr[i]->stlen + ofslen;
        }
@@ -897,7 +901,7 @@
        char *end;
        register int i;
        static int fw_alloc = 4;
-       static int warned = FALSE;
+       static short warned = FALSE;
        extern unsigned long strtoul P((const char *, char **endptr, int base));
 
        if (do_lint && ! warned) {
@@ -916,13 +920,13 @@
 
        parse_field = fw_parse_field;
        scan = force_string(FIELDWIDTHS_node->var_value)->stptr;
-       end = scan + 1;
+
        if (FIELDWIDTHS == NULL)
                emalloc(FIELDWIDTHS, int *, fw_alloc * sizeof(int), 
"set_FIELDWIDTHS");
        FIELDWIDTHS[0] = 0;
        for (i = 1; ; i++) {
                unsigned long int tmp;
-               if (i >= fw_alloc) {
+               if (i + 1 >= fw_alloc) {
                        fw_alloc *= 2;
                        erealloc(FIELDWIDTHS, int *, fw_alloc * sizeof(int), 
"set_FIELDWIDTHS");
                }
@@ -935,13 +939,16 @@
                        fatal(_("invalid FIELDWIDTHS value, near `%s'"),
                              scan);
 
+               if (*scan == '\0')
+                       break;
+
                /* Detect an invalid base-10 integer, a valid value that
                   is followed by something other than a blank or '\0',
                   or a value that is not in the range [1..INT_MAX].  */
                errno = 0;
                tmp = strtoul(scan, &end, 10);
                if (errno != 0
-                   || !(*end == '\0' || is_blank(*end))
+                   || (*end != '\0' && ! is_blank(*end))
                    || !(0 < tmp && tmp <= INT_MAX))
                        fatal(_("invalid FIELDWIDTHS value, near `%s'"),
                              scan);
@@ -954,7 +961,9 @@
                if (*scan == '\0')
                        break;
        }
-       FIELDWIDTHS[i] = -1;
+       if (i == 1)     /* empty string! */
+               i--;
+       FIELDWIDTHS[i+1] = -1;
 
        update_PROCINFO("FS", "FIELDWIDTHS");
 }
@@ -1027,6 +1036,8 @@
                        lintwarn(_("null string for `FS' is a gawk extension"));
                }
        } else if (fs->stlen > 1) {
+               if (do_lint_old)
+                       warning(_("old awk does not support regexps as value of 
`FS'"));
                parse_field = re_parse_field;
        } else if (RS_is_null) {
                /* we know that fs->stlen <= 1 */