From 413435b7fb688f8e0cada81fbec63e80800953da Mon Sep 17 00:00:00 2001
From: JustAnotherArchivist <JustAnotherArchivist@users.noreply.github.com>
Date: Sun, 8 Sep 2019 20:45:46 +0000
Subject: [PATCH] Work around warcio not writing the correct WARC-Profile
 header for revisit records on WARC/1.1

https://github.com/webrecorder/warcio/issues/94
---
 qwarc/warc.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/qwarc/warc.py b/qwarc/warc.py
index 13687de..83cc7ce 100644
--- a/qwarc/warc.py
+++ b/qwarc/warc.py
@@ -154,6 +154,8 @@ class WARC:
 					      'WARC-Warcinfo-ID': self._metaWarcinfoRecordID,
 					    }
 					  )
+					# Workaround for https://github.com/webrecorder/warcio/issues/94
+					responseRecord.rec_headers.replace_header('WARC-Profile', 'http://netpreserve.org/warc/1.1/revisit/identical-payload-digest')
 				else:
 					self._dedupeMap[payloadDigest] = (responseRecord.rec_headers.get_header('WARC-Record-ID'), str(r.url), requestDate)
 			self._warcWriter.write_record(requestRecord)