Date: Sun, 28 Apr 2019 13:37:33 -0400
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>
Cc: pgsql-translators@postgresql.org
Subject: Re: branching time
Message-ID: <20190428173733.GA12483@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="wRRV7LY7NUeQGEoC"
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <94cfe030-61d0-4b01-2d3a-603e7c326bf1@2ndquadrant.com>
User-Agent: Mutt/1.9.4 (2018-02-28)
Precedence: bulk


--wRRV7LY7NUeQGEoC
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit

On 2019-Apr-26, Peter Eisentraut wrote:
> On 2019-04-26 20:14, Alvaro Herrera wrote:

> > I think we need to write a filter program that compares MSGFILTER_MSGID
> > to MSGFILTER_PREV_MSGID, and if the only difference is the leading "%s: " 
> > and the trailing "\n", then strip those thing from stdin and write that
> > to stdout.
> 
> I didn't do it that way, which explains the above problems.  We can try
> it a few more times. :)

I wrote one today.  I first tried doing it in bash, which is probably
gonna give me nightmares tonight, so it ended up in Perl.  It's below in
case someone wants to give it a go.  You run it like

  msgfilter -i somefile.po ./filter > somefile-2.po
  mv somefile-2.po somefile.po
(or you can use "sponge").

It wasn't as successful as I expected, mostly because msgfilter is
limited; it doesn't let you remove the "fuzzy" flag from the message
when you get a match, so you need to revise the file afterwards each
updated entry to remove the flag from each entry ... but of course there
are fuzzies for other reasons, so it's very error-prone.  Anyway, I
think it's much faster than doing it all by hand.  It took quite some
time to figure out a correct filter, because it's not clear how the '\n'
is represented in each place.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

--wRRV7LY7NUeQGEoC
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename=filter

#!/usr/bin/perl

use warnings;

if (not defined $ENV{MSGFILTER_PREV_MSGID}) {
  while (<>) {
    print $_;
  }
  exit;
}

my $first = 1;
if (("%s: " . $ENV{MSGFILTER_MSGID} . "\n") eq $ENV{MSGFILTER_PREV_MSGID}) {
  while ($l = <>) {
    if ($first) {
      $l =~ s/^%s: (.*)/$1/;
      $first = 0;
    }
    $l =~ s/\n$// if eof(STDIN);
    print $l;
  }
} else {
  while (<>) {
    print $_;
  }
  exit;
}

--wRRV7LY7NUeQGEoC--