Add a sqlite make target.

This commits adds a new `sqlite` target to Makefile.
When invoked, it will load IMA data into a sqlite database.

Makefile

@@ -11,5 +11,8 @@ data:	download
 	tar xvjf ${dataset_archive} -C ${dataset}
 	cat ${dataset}/prod* | shuf > ${dataset}/matches.tsv
 	rm ${dataset}/prod*
+sqlite: 
+	test -d ${dataset} || make data
+	sqlite3 < ddl/imagerec.sqlite ${dataset}/imagerec.db
 clean:
 	rm -r ${dataset} ${dataset_archive}

README.md

@@ -23,3 +23,9 @@ $ docker-compose <up|down> [--build] cassandra-load-imagerec
 
 Rows not imported will be stored under `ingestion_status/import_imagerec_matches.err`.
 
+
+# Other targets
+Run
+`make sqlite` 
+
+to load IMA data into a sqlite database under `imagerec_prod/matches.db`.

ddl/imagerec.sqlite

+DROP TABLE IF EXISTS matches;
+CREATE TABLE matches(page_id INTEGER,
+               page_title TEXT,
+               image_id TEXT,
+               confidence_rating TEXT,
+               source TEXT,
+               dataset_id TEXT,
+               insertion_ts REAL,
+               wiki TEXT,
+               found_on TEXT);
+UPDATE matches SET image_id = NULL WHERE image_id = '';
+UPDATE matches SET confidence_rating = NULL WHERE confidence_rating = '';
+UPDATE matches SET source = NULL WHERE source = '';
+UPDATE matches SET found_on = NULL WHERE found_on = '';
+
+
+DROP INDEX IF EXISTS matches_wiki_page_id;
+CREATE INDEX matches_wiki_page_id ON matches(wiki, page_id);
+
+.mode ascii
+.separator "\t" "\n"
+.timer on
+.import imagerec_prod/matches.tsv matches
+