Add notes for 2019-04-30

content/posts/2019-04.md

@@ -1016,4 +1016,46 @@ dspace=# SELECT * FROM item WHERE item_id=74648;
 
 - I even tried to "expunge" the item using an [action in CSV](https://wiki.duraspace.org/display/DSDOC5x/Batch+Metadata+Editing#BatchMetadataEditing-Performing'actions'onitems), and it said "EXPUNGED!" but the item is still there...
 
+## 2019-04-30
+
+- Send mail to the dspace-tech mailing list to ask about the item expunge issue
+- Delete and re-create Podman container for dspacedb after pulling a new PostgreSQL container:
+
+```
+$ podman run --name dspacedb -v dspacedb_data:/var/lib/postgresql/data -e POSTGRES_PASSWORD=postgres -p 5432:5432 -d postgres:9.6-alpine
+```
+
+- Carlos from LandPortal asked if I could export CGSpace in a machine-readable format so I think I'll try to do a CSV
+  - In order to make it easier for him to understand the CSV I will normalize the text languages (minus the provenance field) on my local development instance before exporting:
+
+```
+dspace=# SELECT DISTINCT text_lang, count(*) FROM metadatavalue WHERE resource_type_id = 2 AND metadata_field_id != 28 GROUP BY text_lang;
+ text_lang |  count
+-----------+---------
+           |  358647
+ *         |      11
+ E.        |       1
+ en        |    1635
+ en_US     |  602312
+ es        |      12
+ es_ES     |       2
+ ethnob    |       1
+ fr        |       2
+ spa       |       2
+           | 1074345
+(11 rows)
+dspace=# UPDATE metadatavalue SET text_lang='en_US' WHERE resource_type_id=2 AND metadata_field_id != 28 AND text_lang IN ('ethnob', 'en', '*', 'E.', '');
+UPDATE 360295
+dspace=# UPDATE metadatavalue SET text_lang='en_US' WHERE resource_type_id=2 AND metadata_field_id != 28 AND text_lang IS NULL;
+UPDATE 1074345
+dspace=# UPDATE metadatavalue SET text_lang='es_ES' WHERE resource_type_id=2 AND metadata_field_id != 28 AND text_lang IN ('es', 'spa');
+UPDATE 14
+```
+
+- Then I exported the whole repository as CSV, imported it into OpenRefine, removed a few unneeded columns, exported it, zipped it down to 36MB, and emailed a link to Carlos
+- In other news, while I was looking through the CSV in OpenRefine I saw lots of weird values in some fields... we should check, for example:
+  - issue dates
+  - items missing handles
+  - authorship types
+
 <!-- vim: set sw=2 ts=2: -->