Attachment #10076 for bug #6679

View | Details | Raw Unified | Return to bug 6679
Collapse All | Expand All





###############################################################################

sub FATAL_P		 {'fatal-p'}
sub SYNTAXERROR_P	 {'syntaxerror-p'}

sub FILENAME		 {'input'}
#sub HANDLE		{'handle'}

#sub READAHEAD		 {'readahead'}
sub LINENUM_START	 {'lc_0'}
sub LINENUM		 {'lc'}
sub CDATA_MODE_P	 {'cdata-mode-p'}
sub CDATA_CLOSE		 {'cdata-close'}
#sub PCDATA_MODE_P	 {'pcdata-mode-p'}	# additional submode for CDATA
sub JS_MODE_P		 {'js-mode-p'}	# cdata-mode-p must also be true

sub ALLOW_CFORMAT_P	 {'allow-cformat-p'}

sub new {
    shift;

    # Perl quoting is really screwed up, but this common subexp is way too long
    $js_EscapeSequence = q{\\\\(?:['"\\\\bfnrt]|[^0-7xu]|[0-3]?[0-7]{1,2}|x[\da-fA-F]{2}|u[\da-fA-F]{4})};
}
sub parenleft   { '(' }
sub parenright  { ')' }

sub _split_js  {
    my ($s0) = @_;
    my @it = ();
    while (length $s0) {

    return @it;
}

sub STATE_UNDERSCORE      { 1 }
sub STATE_PARENLEFT       { 2 }
sub STATE_STRING_LITERAL  { 3 }

# XXX This is a crazy hack. I don't want to write an ECMAScript parser.
# XXX A scanner is one thing; a parser another thing.
sub _identify_js_translatables  {
    my @input = @_;
    my @output = ();
    # We mark a JavaScript translatable string as in C, i.e., _("literal")


###############################################################################

sub string_canon ) {
  my $s = shift;
  # Fold all whitespace into single blanks
  $s =~ s/\s+/ /g;

}

# safer version used internally, preserves new lines
sub string_canon_safe  {
  my $s = shift;
  # fold tabs and spaces into single spaces
  $s =~ s/[\ \t]+/ /gs;


# function taken from old version
# used by tmpl_process3
sub parametrize  {
    my($fmt_0, $cformat_p, $t, $f) = @_;
    my $it = '';
    if ($cformat_p) {


# Other simple functions (These are not methods)

sub blank_p  {
    my($s) = @_;
    return $s =~ /^(?:\s|\&nbsp$re_end_entity|$re_tmpl_var|$re_xsl)*$/osi;
}

sub trim  {
    my($s0) = @_;
    my $l0 = length $s0;
    my $s = $s0;

    return wantarray? (substr($s0, 0, $l1), $s, substr($s0, $l0 - $l2)): $s;
}

sub quote_po  {
    my($s) = @_;
    # Locale::PO->quote is buggy, it doesn't quote newlines :-/
    $s =~ s/([\\"])/\\\1/gs;

    return "\"$s\"";
}

sub charset_canon  {
    my($charset) = @_;
    $charset = uc($charset);
    $charset = "$1-$2" if $charset =~ /^(ISO|UTF)(\d.*)/i;

    "\303\270", "\303\271", "\303\272", "\303\273", "\303\274", "\303\275",
    "\303\276", "\303\277" );

sub charset_convert {
    my($s, $charset_in, $charset_out) = @_;
    if ($s !~ /[\200-\377]/s) { # FIXME: don't worry about iso2022 for now
	;
- 

Return to bug 6679

Lines 48-68 use vars qw( $serial ); Link Here

(-)a/misc/translator/TmplTokenizer.pm (-28 / +27 lines)
48		48
49	###############################################################################	49	###############################################################################
50		50
51	sub FATAL_P () {'fatal-p'}	51	sub FATAL_P {'fatal-p'}
52	sub SYNTAXERROR_P () {'syntaxerror-p'}	52	sub SYNTAXERROR_P {'syntaxerror-p'}
53		53
54	sub FILENAME () {'input'}	54	sub FILENAME {'input'}
55	#sub HANDLE () {'handle'}	55	#sub HANDLE {'handle'}
56		56
57	#sub READAHEAD () {'readahead'}	57	#sub READAHEAD {'readahead'}
58	sub LINENUM_START () {'lc_0'}	58	sub LINENUM_START {'lc_0'}
59	sub LINENUM () {'lc'}	59	sub LINENUM {'lc'}
60	sub CDATA_MODE_P () {'cdata-mode-p'}	60	sub CDATA_MODE_P {'cdata-mode-p'}
61	sub CDATA_CLOSE () {'cdata-close'}	61	sub CDATA_CLOSE {'cdata-close'}
62	#sub PCDATA_MODE_P () {'pcdata-mode-p'} # additional submode for CDATA	62	#sub PCDATA_MODE_P {'pcdata-mode-p'} # additional submode for CDATA
63	sub JS_MODE_P () {'js-mode-p'} # cdata-mode-p must also be true	63	sub JS_MODE_P {'js-mode-p'} # cdata-mode-p must also be true
64		64
65	sub ALLOW_CFORMAT_P () {'allow-cformat-p'}	65	sub ALLOW_CFORMAT_P {'allow-cformat-p'}
66		66
67	sub new {	67	sub new {
68	shift;	68	shift;
Lines 137-146 BEGIN { Link Here
137	# Perl quoting is really screwed up, but this common subexp is way too long	137	# Perl quoting is really screwed up, but this common subexp is way too long
138	$js_EscapeSequence = q{\\\\(?:['"\\\\bfnrt]\|[^0-7xu]\|[0-3]?[0-7]{1,2}\|x[\da-fA-F]{2}\|u[\da-fA-F]{4})};	138	$js_EscapeSequence = q{\\\\(?:['"\\\\bfnrt]\|[^0-7xu]\|[0-3]?[0-7]{1,2}\|x[\da-fA-F]{2}\|u[\da-fA-F]{4})};
139	}	139	}
140	sub parenleft () { '(' }	140	sub parenleft { '(' }
141	sub parenright () { ')' }	141	sub parenright { ')' }
142		142
143	sub _split_js ($) {	143	sub _split_js {
144	my ($s0) = @_;	144	my ($s0) = @_;
145	my @it = ();	145	my @it = ();
146	while (length $s0) {	146	while (length $s0) {
Lines 186-198 sub _split_js ($) { Link Here
186	return @it;	186	return @it;
187	}	187	}
188		188
189	sub STATE_UNDERSCORE () { 1 }	189	sub STATE_UNDERSCORE { 1 }
190	sub STATE_PARENLEFT () { 2 }	190	sub STATE_PARENLEFT { 2 }
191	sub STATE_STRING_LITERAL () { 3 }	191	sub STATE_STRING_LITERAL { 3 }
192		192
193	# XXX This is a crazy hack. I don't want to write an ECMAScript parser.	193	# XXX This is a crazy hack. I don't want to write an ECMAScript parser.
194	# XXX A scanner is one thing; a parser another thing.	194	# XXX A scanner is one thing; a parser another thing.
195	sub _identify_js_translatables (@) {	195	sub _identify_js_translatables {
196	my @input = @_;	196	my @input = @_;
197	my @output = ();	197	my @output = ();
198	# We mark a JavaScript translatable string as in C, i.e., _("literal")	198	# We mark a JavaScript translatable string as in C, i.e., _("literal")
Lines 229-235 sub _identify_js_translatables (@) { Link Here
229		229
230	###############################################################################	230	###############################################################################
231		231
232	sub string_canon ($) {	232	sub string_canon ) {
233	my $s = shift;	233	my $s = shift;
234	# Fold all whitespace into single blanks	234	# Fold all whitespace into single blanks
235	$s =~ s/\s+/ /g;	235	$s =~ s/\s+/ /g;
Lines 238-244 sub string_canon ($) { Link Here
238	}	238	}
239		239
240	# safer version used internally, preserves new lines	240	# safer version used internally, preserves new lines
241	sub string_canon_safe ($) {	241	sub string_canon_safe {
242	my $s = shift;	242	my $s = shift;
243	# fold tabs and spaces into single spaces	243	# fold tabs and spaces into single spaces
244	$s =~ s/[\ \t]+/ /gs;	244	$s =~ s/[\ \t]+/ /gs;
Lines 361-367 sub next_token { Link Here
361		361
362	# function taken from old version	362	# function taken from old version
363	# used by tmpl_process3	363	# used by tmpl_process3
364	sub parametrize ($$$$) {	364	sub parametrize {
365	my($fmt_0, $cformat_p, $t, $f) = @_;	365	my($fmt_0, $cformat_p, $t, $f) = @_;
366	my $it = '';	366	my $it = '';
367	if ($cformat_p) {	367	if ($cformat_p) {
Lines 454-465 sub parametrize ($$$$) { Link Here
454		454
455	# Other simple functions (These are not methods)	455	# Other simple functions (These are not methods)
456		456
457	sub blank_p ($) {	457	sub blank_p {
458	my($s) = @_;	458	my($s) = @_;
459	return $s =~ /^(?:\s\|\&nbsp$re_end_entity\|$re_tmpl_var\|$re_xsl)*$/osi;	459	return $s =~ /^(?:\s\|\&nbsp$re_end_entity\|$re_tmpl_var\|$re_xsl)*$/osi;
460	}	460	}
461		461
462	sub trim ($) {	462	sub trim {
463	my($s0) = @_;	463	my($s0) = @_;
464	my $l0 = length $s0;	464	my $l0 = length $s0;
465	my $s = $s0;	465	my $s = $s0;
Lines 468-474 sub trim ($) { Link Here
468	return wantarray? (substr($s0, 0, $l1), $s, substr($s0, $l0 - $l2)): $s;	468	return wantarray? (substr($s0, 0, $l1), $s, substr($s0, $l0 - $l2)): $s;
469	}	469	}
470		470
471	sub quote_po ($) {	471	sub quote_po {
472	my($s) = @_;	472	my($s) = @_;
473	# Locale::PO->quote is buggy, it doesn't quote newlines :-/	473	# Locale::PO->quote is buggy, it doesn't quote newlines :-/
474	$s =~ s/([\\"])/\\\1/gs;	474	$s =~ s/([\\"])/\\\1/gs;
Lines 477-483 sub quote_po ($) { Link Here
477	return "\"$s\"";	477	return "\"$s\"";
478	}	478	}
479		479
480	sub charset_canon ($) {	480	sub charset_canon {
481	my($charset) = @_;	481	my($charset) = @_;
482	$charset = uc($charset);	482	$charset = uc($charset);
483	$charset = "$1-$2" if $charset =~ /^(ISO\|UTF)(\d.*)/i;	483	$charset = "$1-$2" if $charset =~ /^(ISO\|UTF)(\d.*)/i;
Lines 510-516 use vars qw( @latin1_utf8 ); Link Here
510	"\303\270", "\303\271", "\303\272", "\303\273", "\303\274", "\303\275",	510	"\303\270", "\303\271", "\303\272", "\303\273", "\303\274", "\303\275",
511	"\303\276", "\303\277" );	511	"\303\276", "\303\277" );
512		512
513	sub charset_convert ($$$) {	513	sub charset_convert {
514	my($s, $charset_in, $charset_out) = @_;	514	my($s, $charset_in, $charset_out) = @_;
515	if ($s !~ /[\200-\377]/s) { # FIXME: don't worry about iso2022 for now	515	if ($s !~ /[\200-\377]/s) { # FIXME: don't worry about iso2022 for now
516	;	516	;
517	-