Update reports.md

datadryad · Aug 31, 2020 · 0fa500d · 0fa500d
1 parent 4ccc6f2
commit 0fa500d
Showing 1 changed file with 55 additions and 0 deletions.
diff --git a/documentation/reports.md b/documentation/reports.md
@@ -95,3 +95,58 @@ Fields in the admin screen CSV report
 - views
 - downloads
 - citations
+
+Authors at an Institution Report (from SQL)
+-------------------------------------------
+
+This SQL may be more complicated than it needs to be, but it seems to work.  It gives information about
+published and embargoed items with authors at an institution based on the ROR ids you put into the query
+(look and replace in two separate places).
+It can then be exported from SQL as TSV/CSV or whatever.  There are duplicate rows per dataset if more
+than one author is from the institution that contributed to the same dataset.  The views and downloads may
+be off if we go to a model where we retrieve them in real-time from DataCite because they come from
+multiple sources (like us and Zenodo) and we don't pre-populate them all the time.
+
+It includes published and embargoed (shows the landing page but no downloads) which is why some
+publication dates are in the future.
+
+Sorry, IDK why the markdown seems to do it's own thing when indenting, even if I do it preformatted.
+
+<pre>
+SELECT se_id3.identifier, se_res3.title, se_auth3.author_first_name, se_auth3.author_last_name, dcs_affil3.long_name, se_res3.publication_date,
+(stash_engine_counter_stats.unique_investigation_count - stash_engine_counter_stats.unique_request_count) as unique_views, stash_engine_counter_stats.unique_request_count as unique_downloads
+FROM
+  /* get only the earliest published/embagoed one */
+  (SELECT unique_ids.identifier_id, min(res2.id) first_pub_resource FROM
+    /* only get distinct identifiers from all the ror_ids working back through zillions of joined tables */
+    (SELECT DISTINCT se_id.id as identifier_id
+	    FROM dcs_affiliations affil
+	    JOIN dcs_affiliations_authors affil_auth
+	    ON affil.id = affil_auth.`affiliation_id`
+	    JOIN stash_engine_authors auth
+	    ON affil_auth.`author_id` = auth.`id`
+	    JOIN stash_engine_resources res
+	    ON auth.`resource_id` = res.id
+	    JOIN stash_engine_identifiers se_id
+	    ON se_id.id = res.identifier_id
+	    WHERE affil.ror_id IN ('https://ror.org/02y3ad647', 'https://ror.org/0419bgt07', 'https://ror.org/04tk2gy88')
+	    AND se_id.pub_state IN ('published', 'embargoed')) as unique_ids
+  JOIN stash_engine_resources res2
+  ON unique_ids.identifier_id = res2.identifier_id
+  WHERE res2.publication_date IS NOT NULL
+  GROUP BY unique_ids.identifier_id) as ident_and_res	
+JOIN stash_engine_identifiers se_id3
+ON se_id3.id = ident_and_res.identifier_id
+JOIN stash_engine_resources se_res3
+ON se_res3.id = ident_and_res.first_pub_resource
+JOIN stash_engine_authors se_auth3
+ON se_res3.id = se_auth3.`resource_id`
+JOIN dcs_affiliations_authors dcs_affils_authors3
+ON se_auth3.`id` = dcs_affils_authors3.`author_id`
+JOIN dcs_affiliations dcs_affil3
+ON dcs_affils_authors3.`affiliation_id` = dcs_affil3.`id`
+LEFT JOIN stash_engine_counter_stats
+ON se_id3.id = stash_engine_counter_stats.`identifier_id`
+WHERE dcs_affil3.ror_id IN ('https://ror.org/02y3ad647', 'https://ror.org/0419bgt07', 'https://ror.org/04tk2gy88')
+ORDER BY se_res3.publication_date, se_id3.identifier, se_res3.title;
+</pre>