From c2d0910e593faf34bae37f5c9b5cfaae509bdf8c Mon Sep 17 00:00:00 2001
From: Tai Dickerson <tdickers@akamai.com>
Date: Tue, 14 Jun 2016 12:47:10 -0400
Subject: Make har_extractor.py output HAR 1.2 spec-compliant HAR files were
 failing to load in harviewer (http://www.softwareishard.com/har/viewer/) due
 to: - ISO 8601 dates for startedDateTime missing timezone
 (http://www.softwareishard.com/blog/har-12-spec/#pages) -- Used UTC but could
 add detection of default system timezone if desired - pages object missing
 pageTimings (http://www.softwareishard.com/blog/har-12-spec/#pageTimings) --
 Used {} because all child fields are optional

---
 examples/har_extractor.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'examples')

diff --git a/examples/har_extractor.py b/examples/har_extractor.py
index c21f1a8f..d6b50c21 100644
--- a/examples/har_extractor.py
+++ b/examples/har_extractor.py
@@ -4,6 +4,7 @@
 """
 import six
 import sys
+import pytz
 from harparser import HAR
 
 from datetime import datetime
@@ -120,7 +121,7 @@ def response(context, flow):
     full_time = sum(v for v in timings.values() if v > -1)
 
     started_date_time = datetime.utcfromtimestamp(
-        flow.request.timestamp_start).isoformat()
+        flow.request.timestamp_start).replace(tzinfo=pytz.timezone("UTC")).isoformat()
 
     request_query_string = [{"name": k, "value": v}
                             for k, v in flow.request.query or {}]
@@ -174,6 +175,7 @@ def response(context, flow):
                 "startedDateTime": entry['startedDateTime'],
                 "id": page_id,
                 "title": flow.request.url,
+                "pageTimings": {}
             })
         )
         context.HARLog.set_page_ref(flow.request.url, page_id)
-- 
cgit v1.2.3