From 6e499dd5672e5b22ecbfe202146041d3715fc584 Mon Sep 17 00:00:00 2001 From: Josef Moravec Date: Wed, 3 Oct 2018 10:39:58 +0000 Subject: [PATCH] Bug 21481: Remove obsolete files from translation toolkit Signed-off-by: Owen Leonard Signed-off-by: Julian Maurice Conflicts: misc/translator/text-extract.pl misc/translator/text-extract2.pl Signed-off-by: Nick Clemens --- misc/translator/Makefile | 7 -- misc/translator/text-extract.pl | 39 ------- misc/translator/text-extract2.pl | 170 ------------------------------- misc/translator/xgettext.pl | 3 +- 4 files changed, 1 insertion(+), 218 deletions(-) delete mode 100644 misc/translator/Makefile delete mode 100755 misc/translator/text-extract.pl delete mode 100755 misc/translator/text-extract2.pl diff --git a/misc/translator/Makefile b/misc/translator/Makefile deleted file mode 100644 index d97a666bee..0000000000 --- a/misc/translator/Makefile +++ /dev/null @@ -1,7 +0,0 @@ -all: translator_doc.txt - -%.txt: %.html - set -x; sed -e 's/
/&
/g' -e 's/<\/blockquote>/<\/pre><\/table>&/g' $<|LANG=C w3m -T text/html -cols 80 | sed -e 's/  *$$//g' | cat -s > $@
-
-clean:
-	rm translator_doc.txt
diff --git a/misc/translator/text-extract.pl b/misc/translator/text-extract.pl
deleted file mode 100755
index 10d35200e3..0000000000
--- a/misc/translator/text-extract.pl
+++ /dev/null
@@ -1,39 +0,0 @@
-#!/usr/bin/perl
-#use strict;
-#use warnings; FIXME - Bug 2505
-
-use FindBin;
-use lib $FindBin::Bin;
-
-use HTML::Tree;
-use Getopt::Std;
-getopt("f:");
-	my $tree = HTML::TreeBuilder->new; # empty tree
-
-	$tree->parse_file($opt_f);
-	sub give_id {
-		my $x = $_[0];
-		foreach my $c ($x->content_list) {
-			next if (ref($c) && $c->tag() eq "~comment");
-			next if (ref($c) && $c->tag() eq "script");
-			next if (ref($c) && $c->tag() eq "style");
-			if (!ref($c)) {
-				print "$c\n";
-			}
-			if (ref($c) && $c->attr('alt')) {
-				print $c->attr('alt')."\n";
-			}
-			if (ref($c) && $c->attr('title')) {
-				print $c->attr('title')."\n";
-			}
-			if (ref($c) && $c->tag() eq "input" && $c->attr('value')) {
-				print $c->attr('value')."\n";
-			}
-			if (ref($c) && $c->tag() eq 'meta') {
-				print $c->attr('content')."\n ";
-			}
-			give_id($c) if ref $c; # ignore text nodes
-		}
-	};
-	give_id($tree);
-	$tree = $tree->delete;
diff --git a/misc/translator/text-extract2.pl b/misc/translator/text-extract2.pl
deleted file mode 100755
index f7a5c83e0e..0000000000
--- a/misc/translator/text-extract2.pl
+++ /dev/null
@@ -1,170 +0,0 @@
-#!/usr/bin/perl
-
-# Test filter partially based on Ambrose's hideous subst.pl code
-# The idea is that the .tt files are not valid HTML, and as a result
-# HTML::Parse would be completely confused by these templates.
-# This is just a simple scanner (not a parser) & should give better results.
-
-# This script is meant to be a drop-in replacement of text-extract.pl
-
-# A grander plan: Code could be written to detect template variables and
-# construct gettext-c-format-string-like meta-strings (e.g., "Results %s
-# through %s of %s records" that will be more likely to be translatable
-# to languages where word order is very unlike English word order.
-# --> This will be relatively major rework, and requires corresponding
-# rework in tmpl_process.pl
-
-use FindBin;
-use lib $FindBin::Bin;
-
-use strict;
-#use warnings; FIXME - Bug 2505
-use Getopt::Long;
-use TmplTokenizer;
-use VerboseWarnings;
-
-use vars qw( $input );
-use vars qw( $debug_dump_only_p );
-use vars qw( $pedantic_p );
-use vars qw( $allow_cformat_p ); # FOR TESTING PURPOSES ONLY!!
-
-###############################################################################
-
-sub underline ($) { # for testing only
-    my($s) = @_;
-    join('', map {/[\0-\37]/? $_: "$_\b$_"} split(//, $s));
-}
-
-sub debug_dump ($) { # for testing only
-    my($h) = @_;
-    print "re_tag_compat is /", TmplTokenizer::re_tag(1), "/\n";
-    for (;;) {
-	my $s = TmplTokenizer::next_token $h;
-    last unless defined $s;
-	printf "%s\n", ('-' x 79);
-	my($kind, $t, $attr) = ($s->type, $s->string, $s->attributes);
-	printf "%s [line %d]:\n", $kind->to_string, $s->line_number;
-	printf "%4dH%s\n", length($t), underline($t);
-    if ($kind == TmplTokenType::TAG() && %$attr) {
-	    printf "Attributes:\n";
-	    for my $a (keys %$attr) {
-		my($key, $val, $val_orig, $order) = @{$attr->{$a}};
-		printf "%s = %dH%s -- %s\n", $a, length $val, underline $val,
-		$val_orig;
-	    }
-	}
-    if ($kind == TmplTokenType::TEXT_PARAMETRIZED()) {
-	    printf "Form (c-format string):\n";
-	    printf "%dH%s\n", length $s->form, underline $s->form;
-	    printf "Parameters:\n";
-	    my $i = 1;
-	    for my $a ($s->parameters) {
-		my $t = $a->string;
-		printf "%%%d\$s = %dH%s\n", $i, length $t, underline $t;
-		$i += 1;
-	    }
-	}
-	if ($s->has_js_data) {
-	    printf "JavaScript translatable strings:\n";
-	    for my $t (@{$s->js_data}) {
-		printf "%dH%s\n", length $t->[3], underline $t->[3] if $t->[0]; # FIXME
-	    }
-	}
-    }
-}
-
-###############################################################################
-
-sub text_extract ($) {
-    my($h) = @_;
-    my %text = ();
-    for (;;) {
-	my $s = TmplTokenizer::next_token $h;
-    last unless defined $s;
-	my($kind, $t, $attr) = ($s->type, $s->string, $s->attributes);
-    if ($kind == TmplTokenType::TEXT()) {
-	    $t = TmplTokenizer::trim $t;
-	    $text{$t} = 1 if $t =~ /\S/s;
-    } elsif ($kind == TmplTokenType::TAG() && %$attr) {
-	    # value [tag=input], meta
-	    my $tag = lc($1) if $t =~ /^<(\S+)/s;
-	    for my $a ('alt', 'content', 'title', 'value') {
-		if ($attr->{$a}) {
-		    next if $a eq 'content' && $tag ne 'meta';
-		    next if $a eq 'value' && ($tag ne 'input'
-			|| (ref $attr->{'type'} && $attr->{'type'}->[1] =~ /^(?:hidden|radio)$/)); # FIXME
-		    my($key, $val, $val_orig, $order) = @{$attr->{$a}}; #FIXME
-		    $val = TmplTokenizer::trim $val;
-		    $text{$val} = 1 if $val =~ /\S/s;
-		}
-	    }
-	} elsif ($s->has_js_data) {
-	    for my $t (@{$s->js_data}) {
-		remember( $s, $t->[3] ) if $t->[0]; # FIXME
-	    }
-	}
-    }
-    # Emit all extracted strings.
-    # Don't emit pure whitespace, pure numbers, or TMPL_VAR's.
-    for my $t (keys %text) {
-	printf "%s\n", $t
-	    unless TmplTokenizer::blank_p($t) || $t =~ /^\d+$/;
-    }
-}
-
-###############################################################################
-
-sub usage ($) {
-    my($exitcode) = @_;
-    my $h = $exitcode? *STDERR: *STDOUT;
-    print $h < \$allow_cformat_p,
-    'f|file=s'		=> \$input,
-    'debug-dump-only'	=> \$debug_dump_only_p,
-    'pedantic-warnings'	=> sub { $pedantic_p = 1 },
-    'help'		=> sub { usage(0) },
-) || usage_error;
-
-VerboseWarnings::set_application_name $0;
-VerboseWarnings::set_input_file_name $input;
-VerboseWarnings::set_pedantic_mode $pedantic_p;
-
-usage_error('Missing mandatory option -f') unless defined $input;
-
-my $h = TmplTokenizer->new( $input );
-$h->set_allow_cformat( 1 ) if $allow_cformat_p;
-if ($debug_dump_only_p) {
-    debug_dump( $h );
-} else {
-    text_extract( $h );
-}
-
-warn "This input will not work with Mozilla standards-compliant mode\n", undef
-	if TmplTokenizer::syntaxerror_p;
-
-close INPUT;
-
-exit(-1) if TmplTokenizer::fatal_p;
diff --git a/misc/translator/xgettext.pl b/misc/translator/xgettext.pl
index 508fd87855..0343ed29c2 100755
--- a/misc/translator/xgettext.pl
+++ b/misc/translator/xgettext.pl
@@ -396,8 +396,7 @@ exit(-1) if TmplTokenizer::fatal_p;
 
 =head1 DESCRIPTION
 
-This is an experimental script based on the modularized
-text-extract2.pl script.  It has behaviour similar to
+This script has behaviour similar to
 xgettext(1), and generates gettext-compatible output files.
 
 A gettext-like format provides the following advantages:
-- 
2.20.1