From 5ba0e38090001bfe543e73729c54247bb4a65047 Mon Sep 17 00:00:00 2001
From: Bob Mottram <bob@freedombone.net>
Date: Sun, 14 Jun 2020 14:25:38 +0100
Subject: [PATCH] Remove formatting such as bold and italics from posts

This is to be conformant with what Mastodon is doing. Probably markdown should be an optional extra
---
 content.py      | 15 +++++++++++++++
 posts.py        |  5 +++++
 tests.py        | 12 ++++++++++++
 webinterface.py |  2 ++
 4 files changed, 34 insertions(+)

diff --git a/content.py b/content.py
index 87fd69a6..c08b1301 100644
--- a/content.py
+++ b/content.py
@@ -385,6 +385,21 @@ def replaceContentDuplicates(content: str) -> str:
     return content
 
 
+def removeTextFormatting(content: str) -> str:
+    """Removes markup for bold, italics, etc
+    """
+    if '<' not in content:
+        return content
+    content = content.replace('<b>', '').replace('</b>', '')
+    content = content.replace('<i>', '').replace('</i>', '')
+    content = content.replace('<ul>', '').replace('</ul>', '')
+    content = content.replace('<ol>', '').replace('</ol>', '')
+    for level in range(1, 5):
+        content = content.replace('<h' + str(level) + '>', '')
+        content = content.replace('</h' + str(level) + '>', '')
+    return content
+
+
 def removeLongWords(content: str, maxWordLength: int,
                     longWordsList: []) -> str:
     """Breaks up long words so that on mobile screens this doesn't
diff --git a/posts.py b/posts.py
index 32f0b7b2..36dea320 100644
--- a/posts.py
+++ b/posts.py
@@ -46,6 +46,7 @@ from media import replaceYouTube
 from content import removeLongWords
 from content import addHtmlTags
 from content import replaceEmojiFromTags
+from content import removeTextFormatting
 from auth import createBasicAuthHeader
 from config import getConfigParam
 from blocking import isBlocked
@@ -3086,6 +3087,10 @@ def downloadAnnounce(session, baseDir: str, httpPrefix: str,
         announcedJson['content'] = \
             removeLongWords(announcedJson['content'], 40, [])
 
+        # remove text formatting, such as bold/italics
+        announcedJson['content'] = \
+            removeTextFormatting(announcedJson['content'])
+
         # wrap in create to be consistent with other posts
         announcedJson = \
             outboxMessageCreateWrap(httpPrefix,
diff --git a/tests.py b/tests.py
index fe90d9e4..86d16aac 100644
--- a/tests.py
+++ b/tests.py
@@ -67,6 +67,7 @@ from content import replaceEmojiFromTags
 from content import addHtmlTags
 from content import removeLongWords
 from content import replaceContentDuplicates
+from content import removeTextFormatting
 from theme import setCSSparam
 from semantic import isAccusatory
 
@@ -1798,8 +1799,19 @@ def testAccusatory():
     assert(not isAccusatory(testStr, None, 3))
 
 
+def testRemoveTextFormatting():
+    print('testRemoveTextFormatting')
+    testStr = '<p>Text without formatting</p>'
+    resultStr = removeTextFormatting(testStr)
+    assert(resultStr == testStr)
+    testStr = '<p>Text <i>with</i> <h3>formatting</h3></p>'
+    resultStr = removeTextFormatting(testStr)
+    assert(resultStr == '<p>Text with formatting</p>')
+
+
 def runAllTests():
     print('Running tests...')
+    testRemoveTextFormatting()
     testAccusatory()
     testWebLinks()
     testRecentPostsCache()
diff --git a/webinterface.py b/webinterface.py
index 9e90ffa8..ad1c609b 100644
--- a/webinterface.py
+++ b/webinterface.py
@@ -56,6 +56,7 @@ from bookmarks import bookmarkedByPerson
 from announce import announcedByPerson
 from blocking import isBlocked
 from blocking import isBlockedHashtag
+from content import removeTextFormatting
 from content import switchWords
 from content import getMentionsFromHtml
 from content import addHtmlTags
@@ -4099,6 +4100,7 @@ def individualPostAsHtml(recentPostsCache: {}, maxRecentPosts: int,
     if not isPatch:
         objectContent = \
             removeLongWords(postJsonObject['object']['content'], 40, [])
+        objectContent = removeTextFormatting(objectContent)
         objectContent = \
             switchWords(baseDir, nickname, domain, objectContent)
     else: