From 4ac646952b8c40fc8cf445e77e76287b28da36b8 Mon Sep 17 00:00:00 2001 From: Mike Rylander Date: Tue, 11 Oct 2011 14:25:36 -0400 Subject: [PATCH] Move the NACO normalization script to earlier in the baseline schema installation process, as it is now used in the authority schema Signed-off-by: Mike Rylander Signed-off-by: Bill Erickson --- Open-ILS/src/sql/Pg/002.functions.config.sql | 72 ++++++++++++++++++++++++++ Open-ILS/src/sql/Pg/020.schema.functions.sql | 72 -------------------------- 2 files changed, 72 insertions(+), 72 deletions(-) diff --git a/Open-ILS/src/sql/Pg/002.functions.config.sql b/Open-ILS/src/sql/Pg/002.functions.config.sql index 353c7f2..4836384 100644 --- a/Open-ILS/src/sql/Pg/002.functions.config.sql +++ b/Open-ILS/src/sql/Pg/002.functions.config.sql @@ -635,5 +635,77 @@ CREATE OR REPLACE FUNCTION evergreen.lpad_number_substrings( TEXT, TEXT, INT ) R return $string; $$ LANGUAGE PLPERLU; +CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT, TEXT ) RETURNS TEXT AS $func$ + + use strict; + use Unicode::Normalize; + use Encode; + + my $str = decode_utf8(shift); + my $sf = shift; + + # Apply NACO normalization to input string; based on + # http://www.loc.gov/catdir/pcc/naco/SCA_PccNormalization_Final_revised.pdf + # + # Note that unlike a strict reading of the NACO normalization rules, + # output is returned as lowercase instead of uppercase for compatibility + # with previous versions of the Evergreen naco_normalize routine. + + # Convert to upper-case first; even though final output will be lowercase, doing this will + # ensure that the German eszett (ß) and certain ligatures (ff, fi, ffl, etc.) will be handled correctly. + # If there are any bugs in Perl's implementation of upcasing, they will be passed through here. + $str = uc $str; + + # remove non-filing strings + $str =~ s/\x{0098}.*?\x{009C}//g; + + $str = NFKD($str); + + # additional substitutions - 3.6. + $str =~ s/\x{00C6}/AE/g; + $str =~ s/\x{00DE}/TH/g; + $str =~ s/\x{0152}/OE/g; + $str =~ tr/\x{0110}\x{00D0}\x{00D8}\x{0141}\x{2113}\x{02BB}\x{02BC}]['/DDOLl/d; + + # transformations based on Unicode category codes + $str =~ s/[\p{Cc}\p{Cf}\p{Co}\p{Cs}\p{Lm}\p{Mc}\p{Me}\p{Mn}]//g; + + if ($sf && $sf =~ /^a/o) { + my $commapos = index($str, ','); + if ($commapos > -1) { + if ($commapos != length($str) - 1) { + $str =~ s/,/\x07/; # preserve first comma + } + } + } + + # since we've stripped out the control characters, we can now + # use a few as placeholders temporarily + $str =~ tr/+&@\x{266D}\x{266F}#/\x01\x02\x03\x04\x05\x06/; + $str =~ s/[\p{Pc}\p{Pd}\p{Pe}\p{Pf}\p{Pi}\p{Po}\p{Ps}\p{Sk}\p{Sm}\p{So}\p{Zl}\p{Zp}\p{Zs}]/ /g; + $str =~ tr/\x01\x02\x03\x04\x05\x06\x07/+&@\x{266D}\x{266F}#,/; + + # decimal digits + $str =~ tr/\x{0660}-\x{0669}\x{06F0}-\x{06F9}\x{07C0}-\x{07C9}\x{0966}-\x{096F}\x{09E6}-\x{09EF}\x{0A66}-\x{0A6F}\x{0AE6}-\x{0AEF}\x{0B66}-\x{0B6F}\x{0BE6}-\x{0BEF}\x{0C66}-\x{0C6F}\x{0CE6}-\x{0CEF}\x{0D66}-\x{0D6F}\x{0E50}-\x{0E59}\x{0ED0}-\x{0ED9}\x{0F20}-\x{0F29}\x{1040}-\x{1049}\x{1090}-\x{1099}\x{17E0}-\x{17E9}\x{1810}-\x{1819}\x{1946}-\x{194F}\x{19D0}-\x{19D9}\x{1A80}-\x{1A89}\x{1A90}-\x{1A99}\x{1B50}-\x{1B59}\x{1BB0}-\x{1BB9}\x{1C40}-\x{1C49}\x{1C50}-\x{1C59}\x{A620}-\x{A629}\x{A8D0}-\x{A8D9}\x{A900}-\x{A909}\x{A9D0}-\x{A9D9}\x{AA50}-\x{AA59}\x{ABF0}-\x{ABF9}\x{FF10}-\x{FF19}/0-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-9/; + + # intentionally skipping step 8 of the NACO algorithm; if the string + # gets normalized away, that's fine. + + # leading and trailing spaces + $str =~ s/\s+/ /g; + $str =~ s/^\s+//; + $str =~ s/\s+$//g; + + return lc $str; +$func$ LANGUAGE 'plperlu' STRICT IMMUTABLE; + +CREATE OR REPLACE FUNCTION public.naco_normalize_keep_comma( TEXT ) RETURNS TEXT AS $func$ + SELECT public.naco_normalize($1,'a'); +$func$ LANGUAGE SQL STRICT IMMUTABLE; + +CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT ) RETURNS TEXT AS $func$ + SELECT public.naco_normalize($1,''); +$func$ LANGUAGE 'sql' STRICT IMMUTABLE; + COMMIT; diff --git a/Open-ILS/src/sql/Pg/020.schema.functions.sql b/Open-ILS/src/sql/Pg/020.schema.functions.sql index 8c6a0c8..ca4306e 100644 --- a/Open-ILS/src/sql/Pg/020.schema.functions.sql +++ b/Open-ILS/src/sql/Pg/020.schema.functions.sql @@ -33,82 +33,10 @@ CREATE OR REPLACE FUNCTION public.non_filing_normalize ( TEXT, "char" ) RETURNS ); $$ LANGUAGE SQL STRICT IMMUTABLE; -CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT, TEXT ) RETURNS TEXT AS $func$ - - use strict; - use Unicode::Normalize; - use Encode; - - my $str = decode_utf8(shift); - my $sf = shift; - - # Apply NACO normalization to input string; based on - # http://www.loc.gov/catdir/pcc/naco/SCA_PccNormalization_Final_revised.pdf - # - # Note that unlike a strict reading of the NACO normalization rules, - # output is returned as lowercase instead of uppercase for compatibility - # with previous versions of the Evergreen naco_normalize routine. - - # Convert to upper-case first; even though final output will be lowercase, doing this will - # ensure that the German eszett (ß) and certain ligatures (ff, fi, ffl, etc.) will be handled correctly. - # If there are any bugs in Perl's implementation of upcasing, they will be passed through here. - $str = uc $str; - - # remove non-filing strings - $str =~ s/\x{0098}.*?\x{009C}//g; - - $str = NFKD($str); - - # additional substitutions - 3.6. - $str =~ s/\x{00C6}/AE/g; - $str =~ s/\x{00DE}/TH/g; - $str =~ s/\x{0152}/OE/g; - $str =~ tr/\x{0110}\x{00D0}\x{00D8}\x{0141}\x{2113}\x{02BB}\x{02BC}]['/DDOLl/d; - - # transformations based on Unicode category codes - $str =~ s/[\p{Cc}\p{Cf}\p{Co}\p{Cs}\p{Lm}\p{Mc}\p{Me}\p{Mn}]//g; - - if ($sf && $sf =~ /^a/o) { - my $commapos = index($str, ','); - if ($commapos > -1) { - if ($commapos != length($str) - 1) { - $str =~ s/,/\x07/; # preserve first comma - } - } - } - - # since we've stripped out the control characters, we can now - # use a few as placeholders temporarily - $str =~ tr/+&@\x{266D}\x{266F}#/\x01\x02\x03\x04\x05\x06/; - $str =~ s/[\p{Pc}\p{Pd}\p{Pe}\p{Pf}\p{Pi}\p{Po}\p{Ps}\p{Sk}\p{Sm}\p{So}\p{Zl}\p{Zp}\p{Zs}]/ /g; - $str =~ tr/\x01\x02\x03\x04\x05\x06\x07/+&@\x{266D}\x{266F}#,/; - - # decimal digits - $str =~ tr/\x{0660}-\x{0669}\x{06F0}-\x{06F9}\x{07C0}-\x{07C9}\x{0966}-\x{096F}\x{09E6}-\x{09EF}\x{0A66}-\x{0A6F}\x{0AE6}-\x{0AEF}\x{0B66}-\x{0B6F}\x{0BE6}-\x{0BEF}\x{0C66}-\x{0C6F}\x{0CE6}-\x{0CEF}\x{0D66}-\x{0D6F}\x{0E50}-\x{0E59}\x{0ED0}-\x{0ED9}\x{0F20}-\x{0F29}\x{1040}-\x{1049}\x{1090}-\x{1099}\x{17E0}-\x{17E9}\x{1810}-\x{1819}\x{1946}-\x{194F}\x{19D0}-\x{19D9}\x{1A80}-\x{1A89}\x{1A90}-\x{1A99}\x{1B50}-\x{1B59}\x{1BB0}-\x{1BB9}\x{1C40}-\x{1C49}\x{1C50}-\x{1C59}\x{A620}-\x{A629}\x{A8D0}-\x{A8D9}\x{A900}-\x{A909}\x{A9D0}-\x{A9D9}\x{AA50}-\x{AA59}\x{ABF0}-\x{ABF9}\x{FF10}-\x{FF19}/0-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-90-9/; - - # intentionally skipping step 8 of the NACO algorithm; if the string - # gets normalized away, that's fine. - - # leading and trailing spaces - $str =~ s/\s+/ /g; - $str =~ s/^\s+//; - $str =~ s/\s+$//g; - - return lc $str; -$func$ LANGUAGE 'plperlu' STRICT IMMUTABLE; - -CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT ) RETURNS TEXT AS $func$ - SELECT public.naco_normalize($1,''); -$func$ LANGUAGE 'sql' STRICT IMMUTABLE; - CREATE OR REPLACE FUNCTION public.first_word ( TEXT ) RETURNS TEXT AS $$ SELECT COALESCE(SUBSTRING( $1 FROM $_$^\S+$_$), ''); $$ LANGUAGE SQL STRICT IMMUTABLE; -CREATE OR REPLACE FUNCTION public.naco_normalize_keep_comma( TEXT ) RETURNS TEXT AS $func$ - SELECT public.naco_normalize($1,'a'); -$func$ LANGUAGE SQL STRICT IMMUTABLE; - CREATE OR REPLACE FUNCTION public.normalize_space( TEXT ) RETURNS TEXT AS $$ SELECT regexp_replace(regexp_replace(regexp_replace($1, E'\\n', ' ', 'g'), E'(?:^\\s+)|(\\s+$)', '', 'g'), E'\\s+', ' ', 'g'); $$ LANGUAGE SQL STRICT IMMUTABLE; -- 1.7.2.5