list_comments api call improvements

• Improved sql for list_comments
• Removed materialized view from schema: it was not used, it required expensive refreshes
• Some calls are slow comparing to original hived calls
• by_cashout, by_update and by_author_update results are different than expected (waiting for merge from #47 (closed))
• fixed manual tests for list_comments
• Posting a comment will now update active but active is not propagating up at the moment,
• Added post object for database api post
• Added a possibility to convert amount notation to nai notation
• Other minor fixes

hive/db/schema.py

@@ -155,6 +155,11 @@ def build_metadata():
         sa.Index('hive_posts_sc_trend_idx', 'sc_trend'),
         sa.Index('hive_posts_sc_hot_idx', 'sc_hot'),
         sa.Index('hive_posts_created_at_idx', 'created_at'),
+
+        sa.Index('hive_posts_root_author_id', 'root_author_id'),
+        sa.Index('hive_posts_root_permlink_id', 'root_permlink_id'),
+        sa.Index('hive_posts_parent_author_id', 'parent_author_id'),
+        sa.Index('hive_posts_parent_permlink_id', 'parent_permlink_id')
     )
 
     sa.Table(
@@ -567,25 +572,6 @@ def setup(db):
           """
     db.query_no_return(sql)
 
-    sql = """
-        DROP MATERIALIZED VIEW IF EXISTS hive_posts_a_p
-        ;
-        CREATE MATERIALIZED VIEW hive_posts_a_p
-        AS
-        SELECT hp.id AS id,
-            ha_a.name AS author,
-            hpd_p.permlink AS permlink
-        FROM hive_posts hp
-        INNER JOIN hive_accounts ha_a ON ha_a.id = hp.author_id
-        INNER JOIN hive_permlink_data hpd_p ON hpd_p.id = hp.permlink_id
-        WITH DATA
-        ;
-        DROP INDEX IF EXISTS hive_posts_a_p_idx
-        ;
-        CREATE unique index hive_posts_a_p_idx ON hive_posts_a_p (author collate "C", permlink collate "C")
-    """
-    db.query_no_return(sql)
-
     sql = """
         DROP VIEW IF EXISTS public.hive_posts_view;