diff options
author | redpony <redpony@ec762483-ff6d-05da-a07a-a48fb63a330f> | 2010-06-22 05:12:27 +0000 |
---|---|---|
committer | redpony <redpony@ec762483-ff6d-05da-a07a-a48fb63a330f> | 2010-06-22 05:12:27 +0000 |
commit | 0172721855098ca02b207231a654dffa5e4eb1c9 (patch) | |
tree | 8069c3a62e2d72bd64a2cdeee9724b2679c8a56b /word-aligner/ortho-norm/ur.pl | |
parent | 37728b8be4d0b3df9da81fdda2198ff55b4b2d91 (diff) |
initial checkin
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@2 ec762483-ff6d-05da-a07a-a48fb63a330f
Diffstat (limited to 'word-aligner/ortho-norm/ur.pl')
-rwxr-xr-x | word-aligner/ortho-norm/ur.pl | 34 |
1 files changed, 34 insertions, 0 deletions
diff --git a/word-aligner/ortho-norm/ur.pl b/word-aligner/ortho-norm/ur.pl new file mode 100755 index 00000000..34953dd3 --- /dev/null +++ b/word-aligner/ortho-norm/ur.pl @@ -0,0 +1,34 @@ +#!/usr/bin/perl -w +use strict; +use utf8; + +binmode(STDIN, ":utf8"); +binmode(STDOUT, ":utf8"); + +while(<STDIN>) { + chomp; + my @out = (); + my @words = split /\s+/; + for my $of (@words) { + if (length($of) > 1 && !($of =~ /\d/)) { + $of =~ s/\$/sh/g; + } + $of =~ s/([a-z])\~/$1$1/g; + $of =~ s/E/'/g; + $of =~ s/^Aw/o/; + $of =~ s/\|/a/g; + $of =~ s/@/h/g; + $of =~ s/c/ch/g; + $of =~ s/x/kh/g; + $of =~ s/\*/dh/g; + $of =~ s/p$/a/; + $of =~ s/w/o/g; + $of =~ s/Z/dh/g; + $of =~ s/y/i/g; + $of =~ s/Y/a/g; + $of = lc $of; + push @out, $of; + } + print "@out\n"; +} + |