[incubator-ponymail-foal] branch master updated: Try to locate a better date for the From line.

[humbedooh]

This is an automated email from the ASF dual-hosted git repository.

humbedooh pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-ponymail-foal.git


The following commit(s) were added to refs/heads/master by this push:
     new df57669  Try to locate a better date for the From line.
df57669 is described below

commit df57669628d5287b854e20ed7245a46fd101af1a
Author: Daniel Gruno <humbed...@apache.org>
AuthorDate: Fri Sep 17 21:13:03 2021 -0500

    Try to locate a better date for the From line.
    
    This could probably get molded into an external function, but for now it is 
only needed here.
---
 server/endpoints/mbox.py | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/server/endpoints/mbox.py b/server/endpoints/mbox.py
index 59cfc01..ec67eae 100644
--- a/server/endpoints/mbox.py
+++ b/server/endpoints/mbox.py
@@ -26,6 +26,9 @@ import typing
 import aiohttp.web
 import asyncio.exceptions
 import time
+import email.utils
+import datetime
+import dateutil.tz
 
 
 async def convert_source(session: plugins.session.SessionObject, email: dict):
@@ -34,7 +37,16 @@ async def convert_source(session: 
plugins.session.SessionObject, email: dict):
         source_as_text = source["_source"]["source"]
         # Ensure it starts with "From "...or fake it
         if not source_as_text.startswith("From "):
-            from_line = "From MAILER-DAEMON %s\n" % time.strftime("%a %b %d 
%H:%M:%S %Y")
+            from_line = "From MAILER-DAEMON %s\n" % time.strftime("%a %b %d 
%H:%M:%S %Y", time.gmtime(0))  # Fallback in case no date found
+            # If we have any Received: headers, we can extrapolate an 
approximate time from the last (top) one.
+            from_match = re.search(r"(?:[\r\n]|^)Received:\s+from[^;]+?; 
(.+?)[\r\n]", source_as_text)
+            if from_match:
+                recv_time = email.utils.parsedate_tz(from_match.group(1))
+                if recv_time:
+                    dt_tuple = datetime.datetime(*recv_time[:7])
+                    if recv_time[9]:  # If we have an offset, set timezone
+                        dt_tuple = 
dt_tuple.replace(tzinfo=dateutil.tz.tzoffset("Offset", recv_time[9]))
+                    from_line = "From MAILER-DAEMON %s\n" % 
dt_tuple.strftime("%a %b %d %H:%M:%S %Y %z")
             source_as_text = from_line + source_as_text
         # Convert to mboxrd format
         mboxrd_source = ""