From 88f337f2cc40824ed3eaf32b1fec17c3b053bfdf Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@archive.org>
Date: Thu, 19 Mar 2020 16:10:40 -0700
Subject: persist grobid: add option to skip S3 upload

Motivation for this is that current S3 target (minio) is overloaded,
with too many files on a single partition (80 million+). Going to look
in to seaweedfs and other options, but for now stopping minio persist.
Data is all stored in kafka anyways.
---
 python/persist_tool.py        |  4 ++++
 python/sandcrawler/persist.py | 17 ++++++++++-------
 2 files changed, 14 insertions(+), 7 deletions(-)

diff --git a/python/persist_tool.py b/python/persist_tool.py
index 19e6dd7..869af06 100755
--- a/python/persist_tool.py
+++ b/python/persist_tool.py
@@ -41,6 +41,7 @@ def run_grobid(args):
         s3_access_key=args.s3_access_key,
         s3_secret_key=args.s3_secret_key,
         s3_only=args.s3_only,
+        db_only=args.db_only,
     )
     pusher = JsonLinePusher(
         worker,
@@ -135,6 +136,9 @@ def main():
     sub_grobid.add_argument('--s3-only',
         action='store_true',
         help="only upload TEI-XML to S3 (don't write to database)")
+    sub_grobid.add_argument('--db-only',
+        action='store_true',
+        help="only write status to sandcrawler-db (don't save TEI-XML to S3)")
 
     sub_grobid_disk = subparsers.add_parser('grobid-disk',
         help="dump GRBOID output to (local) files on disk")
diff --git a/python/sandcrawler/persist.py b/python/sandcrawler/persist.py
index f5de44a..379fd8b 100644
--- a/python/sandcrawler/persist.py
+++ b/python/sandcrawler/persist.py
@@ -240,6 +240,8 @@ class PersistGrobidWorker(SandcrawlerWorker):
             default_bucket=kwargs['s3_bucket'],
         )
         self.s3_only = kwargs.get('s3_only', False)
+        self.db_only = kwargs.get('db_only', False)
+        assert not (self.s3_only and self.db_only), "Only one of s3_only and db_only allowed"
 
     def process(self, record):
         """
@@ -264,13 +266,14 @@ class PersistGrobidWorker(SandcrawlerWorker):
                 continue
 
             assert len(r['key']) == 40
-            resp = self.s3.put_blob(
-                folder="grobid",
-                blob=r['tei_xml'],
-                sha1hex=r['key'],
-                extension=".tei.xml",
-            )
-            self.counts['s3-put'] += 1
+            if not self.db_only:
+                resp = self.s3.put_blob(
+                    folder="grobid",
+                    blob=r['tei_xml'],
+                    sha1hex=r['key'],
+                    extension=".tei.xml",
+                )
+                self.counts['s3-put'] += 1
 
             # enhance with teixml2json metadata, if available
             try:
-- 
cgit v1.2.3