sanitizing markdown and getting social media names (#63)

* sanitizing markdown and getting social media names

Haven't gotten around to test this but I think it should work. Would solve #61 and #62

* Making flake happy

* Minor optimization by adding a break

* Rewrote as a single function

* A (poor) attempt to make flake happy

* Trying to make flake happy again also formatting

* Fixed unclosed parenthesis

* Properly labelling a function

* Add test for formatted network provider

* Update README

I think with the github outage it didn't save my latest commit so adding
this commit here to make sure the previous commit is added

Co-authored-by: Martijn Boers <martijn@plebian.nl>

README.md

@@ -33,7 +33,12 @@ And that's it! If it doesn't work you can create an issue
 
 ## How to use?  
 Please set the `SUBREDDIT` in `conf/config.ini` to select the desired subreddit to moderate. This only works when the Reddit account is moderator of the selected subreddit
-
+
+## Features
+- [x] Using Youtube, Last.fm and Soundcloud for detecting submissions with exceeding listen count
+- [x] Report self-promotion posts using username and submission title
+- [x] Post artist tags, album information and social media links using Musicbrainz
+
 ## Contribute  
 
 If you have features or bugfixes don't be shy to create a pull requests or bug report!

blottertrax/description_provider.py

@@ -43,8 +43,8 @@ def get_reply(self, parsed_submission: ParsedSubmission) -> str:
         album_release_date = '' if album_release_date is False else '({})'.format(album_release_date)
         life_span = '' if has_life_span is False else '({} to {})'.format(life_span_begin, life_span_end)
         tags = ', '.join(map(lambda t: t['name'], artist['tag-list'][:5])) if has_tags else 'none'
-        socials = ', '.join(map(lambda u: '[{}]({})'.format(u['type'], u['target']),
-                                artist['url-relation-list'])) if has_socials else 'none'
+        socials = ', '.join(
+            map(self.format_network_to_friendly_name, artist['url-relation-list'])) if has_socials else 'none'
 
         return templates.musicbrainz_artist_info.strip().format(
             artist['name'],
@@ -60,3 +60,17 @@ def _get_artist_by_id(self, artist_id: str):
         return self.musicbrainz.get_artist_by_id(
             id=artist_id, includes=['tags', 'ratings', 'annotation', 'url-rels', 'user-tags']
         )['artist']
+
+    @staticmethod
+    def format_network_to_friendly_name(info) -> str:
+        social_network = ['twitter.com', 'facebook.com', 'instagram.com']
+        link_type = info['type']
+        target = info['target']
+        target = target.replace('(', '\(').replace(')', '\)')
+
+        if link_type == 'social network':
+            for domain in social_network:
+                if domain in target:
+                    link_type = domain.split('.')[0]
+                    break
+        return '[{}]({})'.format(link_type, target)

blottertrax/services/youtube.py

@@ -1,6 +1,5 @@
 from typing import List
 from urllib import parse
-
 from urllib.parse import urlparse
 
 import googleapiclient.discovery
@@ -11,8 +10,6 @@
 from blottertrax.value_objects.parsed_submission import ParsedSubmission
 from blottertrax.value_objects.service_result import ThresholdServiceResult
 
-scopes = ["https://www.googleapis.com/auth/youtube.readonly"]
-
 
 class Youtube(ThresholdService):
     config: Config = Config()

tests/test_description_provider.py

@@ -0,0 +1,22 @@
+import unittest
+
+from blottertrax.description_provider import DescriptionProvider
+
+
+class MyTestCase(unittest.TestCase):
+    def test_format_network_to_friendly_name(self):
+        networks = [
+            ('social network', 'https://facebook.com/linkedpage', '[facebook](https://facebook.com/linkedpage)'),
+            ('social network', 'https://twitter.com/linkedpage', '[twitter](https://twitter.com/linkedpage)'),
+            ('social network', 'https://instagram.com/linkedpage', '[instagram](https://instagram.com/linkedpage)'),
+            ('discogs', 'https://discogs.com/linkedpage', '[discogs](https://discogs.com/linkedpage)'),
+            ('some other database', 'https://some-other-database.com/linkedpage', '[some other database](https://some-other-database.com/linkedpage)'),
+            ('wikipedia', 'https://en.wikipedia.org/wiki/Caravan_(band)', '[wikipedia](https://en.wikipedia.org/wiki/Caravan_\(band\))'),
+        ]
+        for network_type, target, expected in networks:
+            formatted = DescriptionProvider.format_network_to_friendly_name({
+                'target': target,
+                'type': network_type
+            })
+
+            self.assertEqual(expected, formatted)