Re: Problem (multiline records) with gawk 3.0.6

[Aharon Robbins]

Greetings. Re the bug report below, you have indeed
found a bug.  Thanks for the cogent bug report.

Here is a patch.

Thanks,

Arnold
----------------------------------------------------------
*** io.c.save   Sun Jul 16 06:13:59 2000
--- io.c        Wed Mar  7 11:10:55 2001
***************
*** 1722,1732 ****
                        /*
                         * Leading newlines at the beginning of the file
                         * should be ignored. Whew!
-                        *
-                        * Is this code ever executed?
                         */
!                       if (RS_is_null && RESTART(rsre, start) == 0) {
!                               start += REEND(rsre, start);
                                goto again;
                        }
                        bp = start + RESTART(rsre, start);
--- 1722,1737 ----
                        /*
                         * Leading newlines at the beginning of the file
                         * should be ignored. Whew!
                         */
!                       if (RS_is_null && *start == '\n') {
!                               /*
!                                * have to catch the case of a
!                                * single newline at the front of
!                                * the record, which the regex
!                                * doesn't. gurr.
!                                */
!                               while (*start == '\n' && start < iop->end)
!                                       start++;
                                goto again;
                        }
                        bp = start + RESTART(rsre, start);

------------------------------------------------------------------------
> From: "Volker Kiefel" <address@hidden>
> To: <address@hidden>
> Subject: Problem (multiline records) with gawk 3.0.6
> Date: Mon, 5 Mar 2001 12:39:29 +0100
>
> Possible bug in gawk v3.0.6
>
> Hello dear maintainer(s) of GAWK,
> hello Arnold Robbins,
>
> Thank you for the excellent GAWK tool which I often use
> (under Windows 95 SE) instead of commercial spreadsheet
> software.  Recently I encountered some problems in an
> attempt to process multiple line records. The problem is
> best exemplified with the example in "Effective
> AWK-programming" (chapter 5.7: multiple-line
> records). The data file:
>
> Jane Doe
> 123 Main Street
> Anywhere, SE 12345-6789
>
> John Smith
> 456 Tree-lined Avenue
> Smallville, MW 98765-4321
>
> is processed with the AWK script
>
> BEGIN {
>     RS = ""; FS = "\n"
> }
>
> {
>     print "Name is: ", $1
>     print "Address is: ", $2
>     print "City and State are: ", $3
>     print ""
> }
>
> A problem arises with one leading newline (empty line)
> in the "addresses" data file before "Jane Doe", the
> (unexpected) output is:
>
> Name is:  
> Address is:  Jane Doe
> City and State are:  123 Main Street
>
> Name is:  John Smith
> Address is:  456 Tree-lined Avenue
> City and State are:  Smallville, MW 98765-4321
>
> i. e. there is a shift in output in the first record.
> With RS="" leading empty lines should not influence the
> result. 0 or 2 leading lines produce the expected
> result: 
>
> Name is:  Jane Doe
> Address is:  123 Main Street
> City and State are:  Anywhere, SE 12345-6789
>
> Name is:  John Smith
> Address is:  456 Tree-lined Avenue
> City and State are:  Smallville, MW 98765-4321
>
> Kerninghan's AWK in its current version does not show
> this unexpected behaviour.
>
> I use gawk v5.0.6 compiled with the mingw32
> implementation of the GCC (v2.95.2) and with the
> djgpp-implementation of GCC. Can this problem be
> addressed in a future version of GAWK?
>
> Yours sincerely 
>   Volker Kiefel
> (Rostock, Germany)