From 6c598d48cbe06ca01f6907b16df5e4a9c4d27999 Mon Sep 17 00:00:00 2001 From: Michele Morgan Date: Fri, 12 Oct 2018 16:01:40 -0400 Subject: [PATCH] =?utf8?q?LP#1657171=C2=A0-=20Normalize=20right=20and=20le?= =?utf8?q?ft=20single=20and=20double=20quotes?= MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Changes search_normalize and naco_normalize to replace right and left leaning single quotes with U+0027, double quotes with U=0022 Signed-off-by: Michele Morgan --- .../perlmods/lib/OpenILS/Utils/Normalize.pm | 4 + Open-ILS/src/sql/Pg/002.functions.config.sql | 8 + .../upgrade/xxxx.schema.normalize_quotes.sql | 141 ++++++++++++++++++ 3 files changed, 153 insertions(+) create mode 100644 Open-ILS/src/sql/Pg/upgrade/xxxx.schema.normalize_quotes.sql diff --git a/Open-ILS/src/perlmods/lib/OpenILS/Utils/Normalize.pm b/Open-ILS/src/perlmods/lib/OpenILS/Utils/Normalize.pm index 4770b43ef3..4a217d7636 100644 --- a/Open-ILS/src/perlmods/lib/OpenILS/Utils/Normalize.pm +++ b/Open-ILS/src/perlmods/lib/OpenILS/Utils/Normalize.pm @@ -54,6 +54,10 @@ sub _normalize_substitutions { # remove non-filing strings $str =~ s/\x{0098}.*?\x{009C}//g; + # Replace curly single and double quote-like characters with straight single and double quotes + $str =~ s/[\x{2018}\x{2019}\x{201B}\x{FF07}\x{201A}]/\x{0027}/g; + $str =~ s/[\x{201C}\x{201D}\x{201F}\x{FF0C}\x{201E}\x{2E42}]/\x{0022}/g; + $str = NFKD($str); # additional substitutions - 3.6. diff --git a/Open-ILS/src/sql/Pg/002.functions.config.sql b/Open-ILS/src/sql/Pg/002.functions.config.sql index f0982db504..9a0d549cbb 100644 --- a/Open-ILS/src/sql/Pg/002.functions.config.sql +++ b/Open-ILS/src/sql/Pg/002.functions.config.sql @@ -625,6 +625,10 @@ CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT, TEXT ) RETURNS TEXT AS $ # remove non-filing strings $str =~ s/\x{0098}.*?\x{009C}//g; + # Replace curly single and double quote-like characters with straight single and double quotes + $str =~ s/[\x{2018}\x{2019}\x{201B}\x{FF07}\x{201A}]/\x{0027}/g; + $str =~ s/[\x{201C}\x{201D}\x{201F}\x{FF0C}\x{201E}\x{2E42}]/\x{0022}/g; + $str = NFKD($str); # additional substitutions - 3.6. @@ -691,6 +695,10 @@ CREATE OR REPLACE FUNCTION public.search_normalize( TEXT, TEXT ) RETURNS TEXT AS # remove non-filing strings $str =~ s/\x{0098}.*?\x{009C}//g; + # Replace curly single and double quote-like characters with straight single and double quotes + $str =~ s/[\x{2018}\x{2019}\x{201B}\x{FF07}\x{201A}]/\x{0027}/g; + $str =~ s/[\x{201C}\x{201D}\x{201F}\x{FF0C}\x{201E}\x{2E42}]/\x{0022}/g; + $str = NFKD($str); # additional substitutions - 3.6. diff --git a/Open-ILS/src/sql/Pg/upgrade/xxxx.schema.normalize_quotes.sql b/Open-ILS/src/sql/Pg/upgrade/xxxx.schema.normalize_quotes.sql new file mode 100644 index 0000000000..6b52a86255 --- /dev/null +++ b/Open-ILS/src/sql/Pg/upgrade/xxxx.schema.normalize_quotes.sql @@ -0,0 +1,141 @@ +BEGIN; + +CREATE OR REPLACE FUNCTION public.naco_normalize( TEXT, TEXT ) RETURNS TEXT AS $func$ + + use strict; + use Unicode::Normalize; + use Encode; + + my $str = shift; + my $sf = shift; + + # Apply NACO normalization to input string; based on + # https://www.loc.gov/aba/pcc/naco/documents/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; + + + # Replace unicode curly single and double quote-like characters with straight + $str =~ s/[\x{2018}\x{2019}\x{201B}\x{FF07}\x{201A}]/\x{0027}/g; + $str =~ s/[\x{201C}\x{201D}\x{201F}\x{FF0C}\x{201E}\x{2E42}]/\x{0022}/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.search_normalize( TEXT, TEXT ) RETURNS TEXT AS $func$ + + use strict; + use Unicode::Normalize; + use Encode; + + my $str = shift; + my $sf = shift; + + # Apply NACO normalization to input string; based on + # https://www.loc.gov/aba/pcc/naco/documents/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; + + # Replace unicode curly single and double quote-like characters with straight + $str =~ s/[\x{2018}\x{2019}\x{201B}\x{FF07}\x{201A}]/\x{0027}/g; + $str =~ s/[\x{201C}\x{201D}\x{201F}\x{FF0C}\x{201E}\x{2E42}]/\x{0022}/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; + +COMMIT; -- 2.43.2