From 21573b6ca7fa0ea3df84b0f0883c1fea90711b9b Mon Sep 17 00:00:00 2001
From: Bob Mottram <bob@freedombone.net>
Date: Mon, 14 Sep 2020 10:33:42 +0100
Subject: [PATCH] Turn email style quotes into blockquotes

---
 content.py | 21 +++++++++++++++++++++
 tests.py   | 26 ++++++++++++++++++++++++++
 2 files changed, 47 insertions(+)

diff --git a/content.py b/content.py
index 3f85eff5b..3db21aa92 100644
--- a/content.py
+++ b/content.py
@@ -14,6 +14,25 @@ from utils import fileLastModified
 from utils import getLinkPrefixes
 
 
+def htmlReplaceEmailQuote(content: str) -> str:
+    """Replaces an email style quote "> Some quote" with html blockquote
+    """
+    if '>&gt; ' not in content:
+        return content
+    contentStr = content.replace('<p>', '')
+    contentLines = contentStr.split('</p>')
+    newContent = ''
+    for lineStr in contentLines:
+        if not lineStr:
+            continue
+        if '>&gt; ' not in lineStr:
+            newContent += '<p>' + lineStr + '</p>'
+        else:
+            lineStr = lineStr.replace('>&gt; ', '><blockquote>')
+            newContent += '<p>' + lineStr + '</blockquote></p>'
+    return newContent
+
+
 def htmlReplaceQuoteMarks(content: str) -> str:
     """Replaces quotes with html formatting
     "hello" becomes <q>hello</q>
@@ -612,6 +631,7 @@ def addHtmlTags(baseDir: str, httpPrefix: str,
     by matching against known following accounts
     """
     if content.startswith('<p>'):
+        content = htmlReplaceEmailQuote(content)
         return htmlReplaceQuoteMarks(content)
     maxWordLength = 40
     content = content.replace('\r', '')
@@ -718,6 +738,7 @@ def addHtmlTags(baseDir: str, httpPrefix: str,
     if longWordsList:
         content = removeLongWords(content, maxWordLength, longWordsList)
     content = content.replace(' --linebreak-- ', '</p><p>')
+    content = htmlReplaceEmailQuote(content)
     return '<p>' + htmlReplaceQuoteMarks(content) + '</p>'
 
 
diff --git a/tests.py b/tests.py
index 51b00cbd4..ecb58134a 100644
--- a/tests.py
+++ b/tests.py
@@ -68,6 +68,7 @@ from delete import sendDeleteViaServer
 from inbox import jsonPostAllowsComments
 from inbox import validInbox
 from inbox import validInboxFilenames
+from content import htmlReplaceEmailQuote
 from content import htmlReplaceQuoteMarks
 from content import dangerousMarkup
 from content import removeHtml
@@ -2124,8 +2125,33 @@ def testConstantTimeStringCheck():
     assert timeDiffMicroseconds < 10
 
 
+def testReplaceEmailQuote():
+    print('testReplaceEmailQuote')
+    testStr='<p>This content has no quote.</p>'
+    assert htmlReplaceEmailQuote(testStr) == testStr
+
+    testStr='<p>This content has no quote.\nWith multiple\nlines</p>'
+    assert htmlReplaceEmailQuote(testStr) == testStr
+
+    testStr = "<p><span class=\"h-card\">" + \
+        "<a href=\"https://somewebsite/@nickname\" " + \
+        "class=\"u-url mention\">@<span>nickname</span></a></span> " + \
+        "<br />&gt; This is a quote</p><p>Some other text.</p>"
+    expectedStr = "<p><span class=\"h-card\">" + \
+        "<a href=\"https://somewebsite/@nickname\" " + \
+        "class=\"u-url mention\">@<span>nickname</span></a></span> " + \
+        "<br /><blockquote>This is a quote</blockquote></p>" + \
+        "<p>Some other text.</p>"
+    resultStr = htmlReplaceEmailQuote(testStr)
+    if resultStr != expectedStr:
+        print('Result: ' + resultStr)
+        print('Expect: ' + expectedStr)
+    assert resultStr == expectedStr
+
+
 def runAllTests():
     print('Running tests...')
+    testReplaceEmailQuote()
     testConstantTimeStringCheck()
     testTranslations()
     testValidContentWarning()