From 23713e9e62a00e69b28dd7da69b0e8b68828224d Mon Sep 17 00:00:00 2001 From: Ben Ostrowsky Date: Tue, 27 Oct 2009 19:57:22 +0000 Subject: [PATCH] Finished the bit that processes the records. Next up: actual output. git-svn-id: svn://nox.esilibrary.com/migration-tools@626 eee7cc8d-164e-4af6-8e1b-092a69004917 --- unicorn/unicorn_patrons_to_tsv.pl | 74 +++++++++++++++++++++++++++++------- 1 files changed, 59 insertions(+), 15 deletions(-) diff --git a/unicorn/unicorn_patrons_to_tsv.pl b/unicorn/unicorn_patrons_to_tsv.pl index fb36114..1eff026 100755 --- a/unicorn/unicorn_patrons_to_tsv.pl +++ b/unicorn/unicorn_patrons_to_tsv.pl @@ -83,20 +83,23 @@ while (<>) { print STDERR "Loaded " . scalar(@records) . " records.\n"; -# We're aiming to produce output that can be slurped in by this SQL code: \COPY m_anderson.actor_usr_unicorn ( l_user_id, l_user_altid, l_user_pin, l_user_profile, l_user_status, l_user_library, l_user_priv_granted, l_user_priv_expires, l_user_mailingaddr, -# l_birthdate, l_last_name, l_first_name, l_middle_name, l_suffix_name, l_note, l_note1, l_patron, l_comment, l_staff, l_webcatpref, l_user_category1, l_user_category2, l_user_category3, l_user_category4, l_dept, l_guardian, l_license, l_ssn, l_misc, l_aup, l_photo, -# l_notify_via, l_user_claims_ret, l_user_environment, l_user_department, l_ums_id, l_user_last_activity, l_placcard, l_user_email, l_addr1_std_line1, l_addr1_std_line2, l_addr1_std_city, l_addr1_std_state, l_addr1_std_zip, l_addr1_country, l_addr1_township, -# l_addr1_room, l_addr1_company, l_addr1_office, l_addr1_phone, l_addr1_dayphone, l_addr1_homephone, l_addr1_workphone, l_addr1_cellphone, l_addr1_fax, l_addr1_email, l_addr1_location, l_addr1_usefor, l_addr1_care_of, l_addr1_known_bad, l_addr1_ums_addrid, -# l_addr2_std_line1, l_addr2_std_line2, l_addr2_std_city, l_addr2_std_state, l_addr2_std_zip, l_addr2_country, l_addr2_township, l_addr2_room, l_addr2_company, l_addr2_office, l_addr2_phone, l_addr2_dayphone, l_addr2_homephone, l_addr2_workphone, l_addr2_cellphone, -# l_addr2_fax, l_addr2_email, l_addr2_location,l_addr2_usefor, l_addr2_care_of, l_addr2_known_bad, l_addr2_ums_addrid, l_addr3_std_line1, l_addr3_std_line2, l_addr3_std_city, l_addr3_std_state, l_addr3_std_zip, l_addr3_country, l_addr3_township, l_addr3_room, -# l_addr3_company, l_addr3_office, l_addr3_phone, l_addr3_dayphone, l_addr3_homephone, l_addr3_workphone, l_addr3_cellphone, l_addr3_fax, l_addr3_email, l_addr3_location, l_addr3_usefor, l_addr3_care_of, l_addr3_known_bad, l_addr3_ums_addrid, l_identific, l_noempl, -# l_profession, l_program, l_represent, l_userid_active, l_inactive_barcode1, l_inactive_barcode2 ) FROM './users.data.tsv' + +# We're aiming to produce output that can be slurped in by this SQL code: \COPY m_anderson.actor_usr_unicorn ( l_user_id, l_user_altid, l_user_pin, l_user_profile, l_user_status, l_user_library, +# l_user_priv_granted, l_user_priv_expires, l_user_mailingaddr, l_birthdate, l_prefix_name, l_last_name, l_first_name, l_middle_name, l_suffix_name, l_note, l_note1, l_patron, l_comment, l_staff, +# l_webcatpref, l_user_category1, l_user_category2, l_user_category3, l_user_category4, l_dept, l_guardian, l_license, l_ssn, l_misc, l_aup, l_photo, l_notify_via, l_user_claims_ret, l_user_environment, +# l_user_department, l_ums_id, l_user_last_activity, l_placcard, l_user_email, l_addr1_std_line1, l_addr1_std_line2, l_addr1_std_city, l_addr1_std_state, l_addr1_std_zip, l_addr1_country, l_addr1_township, +# l_addr1_room, l_addr1_company, l_addr1_office, l_addr1_phone, l_addr1_dayphone, l_addr1_homephone, l_addr1_workphone, l_addr1_cellphone, l_addr1_fax, l_addr1_email, l_addr1_location, l_addr1_usefor, +# l_addr1_care_of, l_addr1_known_bad, l_addr1_ums_addrid, l_addr2_std_line1, l_addr2_std_line2, l_addr2_std_city, l_addr2_std_state, l_addr2_std_zip, l_addr2_country, l_addr2_township, l_addr2_room, +# l_addr2_company, l_addr2_office, l_addr2_phone, l_addr2_dayphone, l_addr2_homephone, l_addr2_workphone, l_addr2_cellphone, l_addr2_fax, l_addr2_email, l_addr2_location,l_addr2_usefor, l_addr2_care_of, +# l_addr2_known_bad, l_addr2_ums_addrid, l_addr3_std_line1, l_addr3_std_line2, l_addr3_std_city, l_addr3_std_state, l_addr3_std_zip, l_addr3_country, l_addr3_township, l_addr3_room, l_addr3_company, +# l_addr3_office, l_addr3_phone, l_addr3_dayphone, l_addr3_homephone, l_addr3_workphone, l_addr3_cellphone, l_addr3_fax, l_addr3_email, l_addr3_location, l_addr3_usefor, l_addr3_care_of, l_addr3_known_bad, +# l_addr3_ums_addrid, l_identific, l_noempl, l_profession, l_program, l_represent, l_userid_active, l_inactive_barcode1, l_inactive_barcode2 ) FROM './users.data.tsv' # Process the records: for (my $u = 0; $u < @records; $u++) { # Some fields can be mapped straightforwardly: - 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 )) { + 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 )) { $records[$u]{uc($f)} '' unless defined $records[$u]{uc($f)}; $records[$u]{'l_' + $f} = $records[$u]{uc($f)}; } @@ -109,23 +112,64 @@ for (my $u = 0; $u < @records; $u++) { } } - # FIXME: handle fields that don't exactly match (e.g. parse USER_NAME into l_last_name etc.) - # l_birthdate, l_last_name, l_first_name, l_middle_name, l_suffix_name, l_note, l_note1, l_patron, l_comment, l_staff, l_webcatpref, l_user_category1, l_user_category2, l_user_category3, l_user_category4, l_dept, l_guardian, l_license, l_ssn, l_misc, l_aup, - # l_photo, l_notify_via, l_ums_id, l_placcard, l_user_email, l_identific, l_noempl, l_profession, l_program, l_represent, l_userid_active, l_inactive_barcode1, l_inactive_barcode2 + # Handle fields that don't exactly match (e.g. parse USER_NAME into l_last_name etc.) + + $records[$u]{'l_birthdate'} = $records[$u]{'USER.BIRTH_DATE'}'; + $records[$u]{'l_note'} = $records[$u]{'USER_XINFO.NOTE'}'; + $records[$u]{'l_note1'} = ''; + $records[$u]{'l_patron'} = ''; + $records[$u]{'l_comment'} = $records[$u]{'USER_XINFO.COMMENT'}'; + $records[$u]{'l_staff'} = $records[$u]{'USER_XINFO.STAFF'}'; + $records[$u]{'l_webcatpref'} = $records[$u]{'USER_XINFO.WEBCATPREF'}'; + $records[$u]{'l_dept'} = $records[$u]{'USER_DEPARTMENT'}'; + $records[$u]{'l_guardian'} = ''; + $records[$u]{'l_license'} = $records[$u]{'USER_XINFO.LICENSE'}'; + $records[$u]{'l_ssn'} = $records[$u]{'USER_XINFO.SSN'}'; + $records[$u]{'l_misc'} = ''; + $records[$u]{'l_aup'} = ''; + $records[$u]{'l_photo'} = ''; + $records[$u]{'l_notify_via'} = $records[$u]{'USER_XINFO.NOTIFY_VIA'}'; + $records[$u]{'l_ums_id'} = ''; + $records[$u]{'l_placcard'} = ''; + $records[$u]{'l_user_email'} = ''; + $records[$u]{'l_identific'} = ''; + $records[$u]{'l_noempl'} = ''; + $records[$u]{'l_profession'} = ''; + $records[$u]{'l_program'} = ''; + $records[$u]{'l_represent'} = ''; + $records[$u]{'l_userid_active'} = ''; + $records[$u]{'l_inactive_barcode1'} = $records[$u]{'USER_XINFO.PREV_ID'}'; + $records[$u]{'l_inactive_barcode2'} = $records[$u]{'USER_XINFO.PREV_ID2'}'; # We can parse the name like so: + # Copy the name to a temp value + $temp_name = $records[$u]{'USER_NAME'}; + # Strip off a prefix, if there is one + foreach $prefix (qw( Ms. Mrs. Mr. Dr. )) { + if ($temp_name =~ /$prefix /i) { + $records[$u]{'l_prefix_name'} = $prefix; + $temp_name =~ s/$prefix //i; + } + } + # Strip off a suffix, if there is one + foreach $suffix (qw( Jr. Sr. II III IV )) { + if ($temp_name =~ /$suffix /i) { + $records[$u]{'l_suffix_name'} = $suffix; + $temp_name =~ s/$suffix //i; + } + } + # Strip off the family name (before the comma) # Of what remains, whatever is before the first space is the first name and the rest is the middle name - + $records[$u]{'l_last_name'} = $temp_name =~ /^([^,])*\s*,.*$/; + $records[$u]{'l_first_given_name', 'l_second_given_name'} = $temp_name =~ /^[^,]*\s*,\s*([^,\s]*)\s*(.*)$/; } - - # Print the records # ----------------- # Print a header line -- 1.7.2.5