3 # Converts a Unicorn users.data file to a tab-separated file.
4 # 2009-08-10 Ben Ostrowsky <ben@esilibrary.com>
7 #$Data::Dumper::Sortkeys = true;
15 my @output_fields = qw(
16 l_user_id l_user_altid l_user_pin l_user_profile l_user_status l_user_library l_user_priv_granted
17 l_user_priv_expires l_user_mailingaddr l_birthdate l_prefix_name l_last_name l_first_name l_middle_name
18 l_suffix_name l_note l_note1 l_patron l_comment l_staff l_webcatpref l_user_category1 l_user_category2
19 l_user_category3 l_user_category4 l_dept l_guardian l_license l_ssn l_misc l_aup l_photo l_notify_via
20 l_user_claims_ret l_user_environment l_user_department l_ums_id l_user_last_activity l_placcard l_user_email
21 l_addr1_std_line1 l_addr1_std_line2 l_addr1_std_city l_addr1_std_state l_addr1_std_zip l_addr1_country
22 l_addr1_township l_addr1_room l_addr1_company l_addr1_office l_addr1_phone l_addr1_dayphone l_addr1_homephone
23 l_addr1_workphone l_addr1_cellphone l_addr1_fax l_addr1_email l_addr1_location l_addr1_usefor l_addr1_care_of
24 l_addr1_known_bad l_addr1_ums_addrid l_addr2_std_line1 l_addr2_std_line2 l_addr2_std_city l_addr2_std_state
25 l_addr2_std_zip l_addr2_country l_addr2_township l_addr2_room l_addr2_company l_addr2_office l_addr2_phone
26 l_addr2_dayphone l_addr2_homephone l_addr2_workphone l_addr2_cellphone l_addr2_fax l_addr2_email
27 l_addr2_locationl_addr2_usefor l_addr2_care_of l_addr2_known_bad l_addr2_ums_addrid l_addr3_std_line1
28 l_addr3_std_line2 l_addr3_std_city l_addr3_std_state l_addr3_std_zip l_addr3_country l_addr3_township
29 l_addr3_room l_addr3_company l_addr3_office l_addr3_phone l_addr3_dayphone l_addr3_homephone l_addr3_workphone
30 l_addr3_cellphone l_addr3_fax l_addr3_email l_addr3_location l_addr3_usefor l_addr3_care_of l_addr3_known_bad
31 l_addr3_ums_addrid l_identific l_noempl l_profession l_program l_represent l_userid_active l_inactive_barcode1
38 # print STDERR "Loaded this line: " . $_;
40 # Is this the start of a new record?
41 if ( /^... DOCUMENT BOUNDARY ...$/ ) {
44 $section = ''; # just in case this didn't get reset in the previous record
45 # print STDERR "Processing record $serial.\n";
49 # Is this a FORM= line (which can be ignored)?
54 # If this isn't the start of the new record, it's a new line in the present record.
57 # Is this line the beginning of a block of data (typically an address or a note)?
58 if ( /^\.(.*?)_BEGIN.$/ ) {
59 # print STDERR "I think this might be the beginning of a beautiful " . $1 . ".\n";
64 # Is this line the end of a block of data (typically an address or a note)?
65 if ( /^\.(.*?)_END.$/ ) {
66 if ("$1." ne $section) {
67 print STDERR "Error in record $serial, line $line (input line $.): got an end-of-$1 but I thought I was in a $section block!\n";
69 # print STDERR "It's been fun, guys, but... this is the end of the " . $1 . ".\n";
74 # Looks like we've got some actual data! Let's store it.
75 # FIXME: For large batches of data, we may run out of memory and should store this on disk.
76 if ( /^\.(.*?).\s+(\|a)?(.*)$/ ) {
78 # Build the name of this field (taking note of whether we're in a named section of data)
85 # Store the field as a key of an array. If it already exists, oh well, now it still exists.
86 $unique_fields{$field} = 1;
88 # Now we can actually store this line of data!
89 $records[$serial]{$field} = $3;
91 # print STDERR "Data extracted: \$records[$serial]{'$field'} = '$3'\n";
96 # This is the continuation of the previous line.
99 $records[$serial]{$field} .= ' ' . $_;
100 # print STDERR "Appended data to previous field. \$records[$serial]{'$field'} is now '" . $records[$serial]{$field} . "'.\n";
105 print STDERR "Loaded " . scalar(@records) . " records.\n";
108 # Process the records:
109 for (my $u = 0; $u < @records; $u++) {
111 #print STDERR "Before processing:\n\n";
112 #print STDERR Dumper($records[$u]);
114 # Some fields can be mapped straightforwardly:
115 foreach $f (qw( user_id user_alt_id user_pin user_profile user_status user_library user_priv_granted user_priv_expires user_mailingaddr user_claims_ret user_environment user_department user_last_activity user_category1 user_category2 user_category3 user_category4 )) {
116 $records[$u]{uc($f)} = '' unless defined $records[$u]{uc($f)};
117 $records[$u]{'l_' . $f} = $records[$u]{uc($f)};
120 # Addresses are a bit different:
121 foreach $a (qw( addr1 addr2 addr3 )) {
122 foreach $f (qw( std_line1 std_line2 std_city std_state std_zip country township room company office phone dayphone homephone workphone cellphone fax email location usefor care_of known_bad ums_addrid )) {
123 $records[$u]{uc('USER_' . $a . '.' . $f)} = '' unless defined $records[$u]{uc('USER_' . $a . '.' . $f)};
124 $records[$u]{'l_' . $a . '_' . $f} = $records[$u]{uc('USER_' . $a . '.' . $f)};
126 $records[$u]{'l_' . $a . '_std_line1'} = $records[$u]{'USER_' . uc($a) . '.STREET'};
127 if ((defined $records[$u]{'USER_' . uc($a) . '.CITY/STATE'}) && ($records[$u]{'USER_' . uc($a) . '.CITY/STATE'} =~ m/^(.*), (.*)$/)) {
128 $records[$u]{'l_' . $a . '_std_city'} = $1;
129 $records[$u]{'l_' . $a . '_std_state'} = $2;
131 $records[$u]{'l_' . $a . '_std_zip'} = $records[$u]{'USER_' . uc($a) . '.ZIP'};
135 # Handle fields that don't exactly match (e.g. parse USER_NAME into l_last_name etc.)
137 $records[$u]{'l_birthdate'} = $records[$u]{'USER_BIRTH_DATE'};
138 $records[$u]{'l_note'} = $records[$u]{'USER_XINFO.NOTE'};
139 $records[$u]{'l_note1'} = '';
140 $records[$u]{'l_patron'} = '';
141 $records[$u]{'l_comment'} = $records[$u]{'USER_XINFO.COMMENT'};
142 $records[$u]{'l_staff'} = $records[$u]{'USER_XINFO.STAFF'};
143 $records[$u]{'l_webcatpref'} = $records[$u]{'USER_XINFO.WEBCATPREF'};
144 $records[$u]{'l_dept'} = $records[$u]{'USER_DEPARTMENT'};
145 $records[$u]{'l_guardian'} = '';
146 $records[$u]{'l_license'} = $records[$u]{'USER_XINFO.LICENSE'};
147 $records[$u]{'l_ssn'} = $records[$u]{'USER_XINFO.SSN'};
148 $records[$u]{'l_misc'} = '';
149 $records[$u]{'l_aup'} = '';
150 $records[$u]{'l_photo'} = '';
151 $records[$u]{'l_notify_via'} = $records[$u]{'USER_XINFO.NOTIFY_VIA'};
152 $records[$u]{'l_ums_id'} = '';
153 $records[$u]{'l_placcard'} = '';
154 $records[$u]{'l_user_email'} = '';
155 $records[$u]{'l_identific'} = '';
156 $records[$u]{'l_noempl'} = '';
157 $records[$u]{'l_profession'} = '';
158 $records[$u]{'l_program'} = '';
159 $records[$u]{'l_represent'} = '';
160 $records[$u]{'l_userid_active'} = '';
161 $records[$u]{'l_inactive_barcode1'} = $records[$u]{'USER_XINFO.PREV_ID'};
162 $records[$u]{'l_inactive_barcode2'} = $records[$u]{'USER_XINFO.PREV_ID2'};
163 $records[$u]{'l_user_category1'} = $records[$u]{'USER_ROUTING_FLAG'};
164 $records[$u]{'l_user_category2'} = $records[$u]{'USER_WEB_AUTH'};
166 # We can parse the name like so:
168 # Copy the name to a temp value
169 $temp_name = $records[$u]{'USER_NAME'};
171 # If there's no comma, don't try to parse the name
172 unless ($temp_name =~ m/,/) {
173 $records[$u]{'l_last_name'} = $temp_name;
177 # Strip off a prefix, if there is one
178 foreach $prefix (qw( Ms\. Mrs\. Mr\. Dr\. )) {
179 if ($temp_name =~ /$prefix /i) {
180 $records[$u]{'l_prefix_name'} = $prefix;
181 $temp_name =~ s/$prefix //i;
185 # Strip off a suffix, if there is one
186 foreach $suffix (qw( Jr\. Jr Sr\. Sr III II IV )) {
187 if ($temp_name =~ / $suffix/i) {
188 $records[$u]{'l_suffix_name'} = $suffix;
189 $temp_name =~ s/ $suffix//i;
193 # Strip off the family name (before the comma)
194 # Of what remains, whatever is before the first space is the first name and the rest is the middle name
195 $temp_name =~ m/^([^,]*)\s*,.*$/;
196 $records[$u]{'l_last_name'} = $1;
197 $temp_name =~ m/^[^,]*\s*,\s*([^,\s]*)\s*(.*)$/;
198 $records[$u]{'l_first_name'} = $1;
199 $records[$u]{'l_middle_name'} = $2;
201 #print STDERR "\n\nAfter processing:\n\n";
202 #print STDERR Dumper($records[$u]);
209 for (my $u = 0; $u < @records; $u++) {
210 foreach $f (@output_fields) {
211 if (defined $records[$u]{$f}) {
212 print $records[$u]{$f};
219 print STDERR "Wrote " . scalar(@records) . " records.\n";
220 # uh-bdee-uh-bdee-uh-bdee-uh- THAT'S ALL, FOLKS