IT-Wissen/atom/atom-sql-queries.md

Table of Contents

• AtoM SQL Queries
• recreate DB
• delete all archival descriptions
• get Repository ID by slug
• Get all descriptions recursively (parent --> children)
• Get all descriptions by alternative identifier
• Get all authority records
• get all authority records
• get all places
• delete places
• only places containing "Bahnhof"
• all places
• count all archival descriptions
• get all subjects
• Get all Identifiers
• Update information_object_i18n based on joined data

AtoM SQL Queries

See also here: Common AtoM database queries)

recreate DB

php symfony propel:insert-sql
/* or */
DROP database atom25dbtest;
CREATE DATABASE `my-db` CHARACTER SET utf8 COLLATE utf8_unicode_ci;

dann webinstaller machen. evtl. noch config/config.php löschen.

delete all archival descriptions

DELETE FROM information_object WHERE id <> 1;

get Repository ID by slug

select a.id, a.parent_id, a.repository_id, b.title 
from information_object a 
left join information_object_i18n b on a.id = b.id 
where a.id = (
  select object_id from slug 
  where slug = 'historische-glasdias-2'
);

Get all descriptions recursively (parent --> children)

WITH RECURSIVE cte (id, identifier, parent_id) AS ( 
  SELECT io.id, io.identifier, io.parent_id 
    FROM information_object io
      WHERE io.parent_id = 505795 
  UNION ALL
  SELECT io2.id, io2.identifier, io2.parent_id 
    FROM information_object io2 
      INNER JOIN cte 
        ON io2.parent_id = cte.id 
) 
SELECT cte.id, cte.parent_id, cte.identifier, io18.title, p18.value AS PID, s.slug
  FROM cte
LEFT JOIN information_object_i18n io18 
  ON io18.id = cte.id
LEFT JOIN property as p1
  ON p1.object_id = cte.id
LEFT JOIN property_i18n as p18
  ON p18.id = p1.id
LEFT JOIN slug AS s
  ON s.object_id = cte.id
WHERE p1.name LIKE 'PID';

Get all descriptions by alternative identifier

replace SRG-Test:%

mysql -u root -p -e "use atom; select io.title, s.slug, p2.value from property as p1 left join property_i18n p2 on p1.id = p2.id  left join slug s on s.object_id = p1.object_id   left join information_object_i18n io on io.id = p1.object_id where p1.name like 'PID' and p2.value like 'SRG-Test:%';" > pid_with_SRG-Test.csv

# readable query:
SELECT io.title, s.slug, p2.value 
  FROM property as p1 
LEFT JOIN property_i18n p2 
  ON p1.id = p2.id 
LEFT JOIN slug s 
  ON s.object_id = p1.object_id 
LEFT JOIN information_object_i18n io 
  ON io.id = p1.object_id 
WHERE p1.name LIKE 'PID' 
  AND p2.value LIKE 'SRG-Test:%';

Get all authority records

mysql -e "USE atom25db; SELECT a.*, b.* FROM actor LEFT JOIN actor_i18n B ON A.ID = B.ID;" > /home/docuteam/authorityrecords.csv

Then open CSV, remove 2nd "id"-column.

Entity types:

• 131: Corporate body
• 132: Person

Description status:

• 221: Final
• 222: Revised
• 223: Draft

Level of detail:

• 224: Full
• 225: Partial
• 226: Minimal

get all authority records

select distinct /* with or without distinct? */
actor_i18n.authorized_form_of_name, 
actor.entity_type_id 
from actor_i18n 
join actor on actor.id = actor_i18n.id 
join object on object.id = actor.id 
join event on event.actor_id = actor.id 
where object.class_name = "QubitActor" 
and actor_i18n.culture = "en";

/* query to test on BIS: */
SELECT 
actor_i18n.id AS actor_i18n__id,
actor_i18n.authorized_form_of_name AS actor_i18n__authorized_form_of_name,
actor_i18n.dates_of_existence AS actor_i18n__dates_of_existence,
actor_i18n.history AS actor_i18n__history,
actor_i18n.places AS actor_i18n__places,
actor_i18n.legal_status AS actor_i18n__legal_status,
actor_i18n.functions AS actor_i18n__functions,
actor_i18n.mandates AS actor_i18n__mandates,
actor_i18n.internal_structures AS actor_i18n__internal_structures,
actor_i18n.general_context AS actor_i18n__general_context,
actor_i18n.institution_responsible_identifier AS actor_i18n__institution_responsible_identifier,
actor_i18n.rules AS actor_i18n__rules,
actor_i18n.sources AS actor_i18n__sources,
actor_i18n.culture AS actor_i18n__culture,
actor.id AS actor__id,
actor.entity_type_id AS actor__entity_type_id,
actor.corporate_body_identifiers AS actor__corporate_body_identifiers,
actor.description_status_id AS actor__description_status_id,
actor.description_detail_id AS actor__description_detail_id,
actor.description_identifier AS actor__description_identifier,
actor.source_standard AS actor__source_standard,
actor.parent_id AS actor__parent_id,
actor.lft AS actor__lft,
actor.rgt AS actor__rgt,
actor.source_culture AS actor__source_culture
FROM actor_i18n 
JOIN actor ON actor.id = actor_i18n.id 
JOIN object ON object.id = actor.id 
WHERE object.class_name = "QubitActor";

get all places

Taxonomy_id of subjects is 42

mysql -e "USE atom25db; SELECT a.id, a.taxonomy_id, a.code, a.parent_id, a.source_culture, b.name, b.culture FROM term a LEFT JOIN term_i18n b ON a.id = b.id WHERE a.taxonomy_id = 42 ORDER BY a.id ASC;" > /home/docuteam/places.csv

delete places

only places containing "Bahnhof"

DELETE FROM term 
WHERE term.taxonomy_id = 42 
  AND term.id = (SELECT term_i18n.id FROM term_i18n WHERE term_i18n.name LIKE '%Bahnhof%');

all places

DELETE FROM term WHERE term.taxonomy_id = 42;

Anschliessend build nested set und search index!

count all archival descriptions

SELECT COUNT(*) FROM information_object WHERE id <> 1;

get all subjects

Taxonomy_id of subjects is 35

mysql -e "USE atom25db; SELECT a.id, a.taxonomy_id, a.code, a.parent_id, a.source_culture, b.name, b.culture FROM term a LEFT JOIN term_i18n b ON a.id = b.id WHERE a.taxonomy_id = 35 ORDER BY a.id ASC;" > /home/docuteam/subjects.csv

Get all Identifiers

select identifier, id, parent_id 
from information_object 
where identifier IS NOT NULL 
  and identifier not like '#%' 
  and identifier != '?' 
order by cast(identifier as unsigned) asc;

Update information_object_i18n based on joined data

UPDATE information_object_i18n AS io 
  LEFT JOIN property p1 
    ON p1.object_id = io.id 
  LEFT JOIN property_i18n p2 
    ON p1.id = p2.id 
  SET io.title = CONCAT('recovery_failed_', io.title) 
WHERE p1.name LIKE 'PID' 
AND p2.value = 'CH-000-0:1234';