From 2a1bb6978c6e58995e89e055bd3a16042f5c5636 Mon Sep 17 00:00:00 2001
From: Mort Yao <soi@mort.ninja>
Date: Mon, 27 Aug 2018 14:49:37 +0200
Subject: [PATCH] [twitter] match correct screen_name and item_id in a
 conversation

---
 src/you_get/extractors/twitter.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/you_get/extractors/twitter.py b/src/you_get/extractors/twitter.py
index 8ed400db..1c027973 100644
--- a/src/you_get/extractors/twitter.py
+++ b/src/you_get/extractors/twitter.py
@@ -30,9 +30,9 @@ def twitter_download(url, output_dir='.', merge=True, info_only=False, **kwargs)
         return
 
     html = get_html(url, faker=True)
-    screen_name = r1(r'data-screen-name="([^"]*)"', html) or \
+    screen_name = r1(r'twitter\.com/([^/]+)', url) or r1(r'data-screen-name="([^"]*)"', html) or \
         r1(r'<meta name="twitter:title" content="([^"]*)"', html)
-    item_id = r1(r'data-item-id="([^"]*)"', html) or \
+    item_id = r1(r'twitter\.com/[^/]+/status/(\d+)', url) or r1(r'data-item-id="([^"]*)"', html) or \
         r1(r'<meta name="twitter:site:id" content="([^"]*)"', html)
     page_title = "{} [{}]".format(screen_name, item_id)