improving media page with images and videos

archivers/base_archiver.py

@@ -1,4 +1,4 @@
-import os, datetime, shutil, hashlib, time, requests, re
+import os, datetime, shutil, hashlib, time, requests, re, mimetypes
 from dataclasses import dataclass
 from abc import ABC, abstractmethod
 from urllib.parse import urlparse
@@ -58,7 +58,13 @@ class Archiver(ABC):
             <h3><a href="{url}">{url}</a></h3><ul>'''
 
         for url_info in urls_info:
-            page += f'''<li><a href="{url_info['cdn_url']}">{url_info['key']}</a>: {url_info['hash']}</li>'''
+            mime_global, mime_type = self._guess_file_type(url_info["key"])
+            preview = ""
+            if mime_global == "image":
+                preview = f'<img src="{url_info["cdn_url"]}" style="max-height:200px;max-width:200px;"></img>'
+            elif mime_global == "video":
+                preview = f'<video controls style="max-height:200px;max-width:200px;"><source src="{url_info["cdn_url"]}" type="{mime_type}"></source></video>'
+            page += f'''<li><a href="{url_info['cdn_url']}">{preview}{url_info['key']}</a>: {url_info['hash']}</li>'''
 
         page += f"</ul><h2>{self.name} object data:</h2><code>{object}</code>"
         page += f"</body></html>"
@@ -77,8 +83,20 @@ class Archiver(ABC):
         page_cdn = self.storage.get_cdn_url(page_key)
         return (page_cdn, page_hash, thumbnail)
 
+    def _guess_file_type(self, path:str):
+        """
+        Receives a URL or filename and returns global mimetype like 'image' or 'video' and the specific mimetype as a tuple
+        see https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/MIME_types/Common_types
+        ex: ('audio', 'audio/mp3')
+        """
+        mime = mimetypes.guess_type(path)[0]
+        if mime is not None:
+            return mime.split("/")[0], mime
+        return "", ""
+
+
     # eg images in a tweet save to cloud storage
-    def generate_media_page(self, urls, url, object):
+    def generate_media_page(self, urls, url, object, requester=requests):
         """
         For a list of media urls, fetch them, upload them
         and call self.generate_media_page_html with them
@@ -94,7 +112,7 @@ class Archiver(ABC):
 
             filename = os.path.join(Storage.TMP_FOLDER, key)
 
-            d = requests.get(media_url, headers=headers)
+            d = requester.get(media_url, headers=headers)
             with open(filename, 'wb') as f:
                 f.write(d.content)