pdfextract_tool fixes from prod usage

2 files changed, 6 insertions, 3 deletions

diff --git a/python/pdfextract_tool.py b/python/pdfextract_tool.py
index e05d48d..0d33ec9 100755
--- a/python/pdfextract_tool.py
+++ b/python/pdfextract_tool.py
@@ -61,8 +61,11 @@ def run_single(args):
     worker = PdfExtractBlobWorker(sink=args.sink, thumbnail_sink=args.thumbnail_sink)
     with open(args.pdf_file, 'rb') as pdf_file:
         pdf_bytes = pdf_file.read()
-    result = worker.process(pdf_bytes)
-    print(json.dumps(result.to_pdftext_dict(), sort_keys=True))
+    worker.push_record(pdf_bytes)
+    worker.finish()
+    if args.thumbnail_sink:
+        args.thumbnail_sink.finish()
+
 
 
 def main():
diff --git a/python/sandcrawler/pdfextract.py b/python/sandcrawler/pdfextract.py
index efb6cca..4606632 100644
--- a/python/sandcrawler/pdfextract.py
+++ b/python/sandcrawler/pdfextract.py
@@ -259,5 +259,5 @@ class PdfExtractBlobWorker(SandcrawlerWorker):
         if self.thumbnail_sink and result.page0_thumbnail is not None:
             self.thumbnail_sink.push_record(result.page0_thumbnail, key=result.sha1hex)
 
-        return result
+        return result.to_pdftext_dict()