From d73c79b2ba1cdfa206d65a2b605e4d028e42af26 Mon Sep 17 00:00:00 2001
From: Chaiwat Suttipongsakul <cwt@bashell.com>
Date: Sun, 24 Feb 2019 21:36:06 +0700
Subject: [PATCH 1/2] fix unicode hashtag

---
 little_boxes/content_helper.py | 8 ++++++--
 requirements.txt               | 1 +
 setup.py                       | 4 +++-
 3 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/little_boxes/content_helper.py b/little_boxes/content_helper.py
index f670849..cd7ee30 100644
--- a/little_boxes/content_helper.py
+++ b/little_boxes/content_helper.py
@@ -1,10 +1,11 @@
-import re
 from typing import Dict
 from typing import List
 from typing import Tuple
 
 from markdown import markdown
 
+import regex as re
+
 from .activitypub import get_backend
 from .webfinger import get_actor_url
 
@@ -24,7 +25,10 @@ MENTION_REGEX = re.compile(r"@[\d\w_.+-]+@[\d\w-]+\.[\d\w\-.]+")
 def hashtagify(content: str) -> Tuple[str, List[Dict[str, str]]]:
     base_url = get_backend().base_url()
     tags = []
-    for hashtag in re.findall(HASHTAG_REGEX, content):
+    hashtags = re.findall(HASHTAG_REGEX, content)
+    hashtags.sort()
+    hashtags.reverse()  # replace longest tags first
+    for hashtag in hashtags:
         tag = hashtag[1:]
         link = f'<a href="{base_url}/tags/{tag}" class="mention hashtag" rel="tag">#<span>{tag}</span></a>'
         tags.append(dict(href=f"{base_url}/tags/{tag}", name=hashtag, type="Hashtag"))
diff --git a/requirements.txt b/requirements.txt
index c5e4c7c..785d00b 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -5,3 +5,4 @@ markdown
 pyld
 pycryptodome
 html2text
+regex
diff --git a/setup.py b/setup.py
index 7872ad7..f19afe3 100644
--- a/setup.py
+++ b/setup.py
@@ -1,11 +1,12 @@
 #!/usr/bin/env python
 
+from distutils.core import setup
 import io
 import os
-from distutils.core import setup
 
 from setuptools import find_packages
 
+
 here = os.path.abspath(os.path.dirname(__file__))
 
 
@@ -29,6 +30,7 @@ REQUIRED = [
     "pycryptodome",
     "html2text",
     "mdx_linkify",
+    "regex",
 ]
 
 DEPENDENCY_LINKS = []

From 9ec30df3f67bceec5bc859dd0e0bfdf5642779f3 Mon Sep 17 00:00:00 2001
From: Chaiwat Suttipongsakul <cwt@bashell.com>
Date: Sun, 24 Feb 2019 22:02:28 +0700
Subject: [PATCH 2/2] same hashtag should be replaced only one time

---
 little_boxes/content_helper.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/little_boxes/content_helper.py b/little_boxes/content_helper.py
index cd7ee30..c3b342c 100644
--- a/little_boxes/content_helper.py
+++ b/little_boxes/content_helper.py
@@ -26,8 +26,9 @@ def hashtagify(content: str) -> Tuple[str, List[Dict[str, str]]]:
     base_url = get_backend().base_url()
     tags = []
     hashtags = re.findall(HASHTAG_REGEX, content)
+    hashtags = list(set(hashtags))  # unique tags
     hashtags.sort()
-    hashtags.reverse()  # replace longest tags first
+    hashtags.reverse()  # replace longest tag first
     for hashtag in hashtags:
         tag = hashtag[1:]
         link = f'<a href="{base_url}/tags/{tag}" class="mention hashtag" rel="tag">#<span>{tag}</span></a>'