From 139edd6104f6824a0b94a7577fd23d6ae7bf2d28 Mon Sep 17 00:00:00 2001
From: smitcona <conal.smith@hpe.com>
Date: Wed, 1 Feb 2017 17:16:30 +0000
Subject: [PATCH] Add new method which marks as splitlines, lines which are
 splitlines but start with email quotation indents ("> ")

---
 talon/quotations.py | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/talon/quotations.py b/talon/quotations.py
index 8e2b2b7..514617f 100644
--- a/talon/quotations.py
+++ b/talon/quotations.py
@@ -469,12 +469,33 @@ def split_emails(msg):
     lines = msg_body.splitlines()[:MAX_LINES_COUNT]
     markers = mark_message_lines(lines)
 
+    markers = _mark_quoted_email_splitlines(markers, lines)
+
     # we don't want splitlines in header blocks
     markers = _correct_splitlines_in_headers(markers, lines)
 
     return markers
 
 
+def _mark_quoted_email_splitlines(markers, lines):
+    """
+    When there are headers indented with '>' characters, we will attempt to identify if the header is a splitline header
+    using a slightly altered SPLITTER_PATTERNS list and mark it as 's'.
+    """
+    # Create a list of markers to easily alter specific characters
+    markerlist = list(markers)
+    for i, line in enumerate(lines):
+        if markerlist[i] != 'm':
+            continue
+        for pattern in SPLITTER_PATTERNS:
+            matcher = re.search(pattern, line)
+            if matcher:
+                markerlist[i] = 's'
+                break
+
+    return "".join(markerlist)
+
+
 def _correct_splitlines_in_headers(markers, lines):
     """Corrects markers by removing splitlines deemed to be inside header blocks"""
     updated_markers = ""