From 7b845b34ce18863e519ad3cce8e53431ba41664d Mon Sep 17 00:00:00 2001
From: Mort Yao <soi@mort.ninja>
Date: Sun, 11 Dec 2022 17:43:07 +0100
Subject: [PATCH] [tiktok] fix extraction for alternative URLs

---
 src/you_get/common.py            | 15 +++++++++------
 src/you_get/extractors/tiktok.py | 12 ++++++------
 2 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/src/you_get/common.py b/src/you_get/common.py
index 1558baf6..c337a2a2 100755
--- a/src/you_get/common.py
+++ b/src/you_get/common.py
@@ -344,21 +344,24 @@ def undeflate(data):
 
 # an http.client implementation of get_content()
 # because urllib does not support "Connection: keep-alive"
-def getHttps(host, url, headers, gzip=True, deflate=False, debuglevel=0):
+def getHttps(host, url, headers, debuglevel=0):
     import http.client
 
     conn = http.client.HTTPSConnection(host)
     conn.set_debuglevel(debuglevel)
     conn.request("GET", url, headers=headers)
     resp = conn.getresponse()
+    set_cookie = resp.getheader('set-cookie')
 
     data = resp.read()
-    if gzip:
-        data = ungzip(data)
-    if deflate:
-        data = undeflate(data)
+    try:
+        data = ungzip(data)  # gzip
+        data = undeflate(data)  # deflate
+    except:
+        pass
 
-    return str(data, encoding='utf-8'), resp.getheader('set-cookie')
+    conn.close()
+    return str(data, encoding='utf-8'), set_cookie
 
 
 # DEPRECATED in favor of get_content()
diff --git a/src/you_get/extractors/tiktok.py b/src/you_get/extractors/tiktok.py
index 641e5e97..2c4892f6 100644
--- a/src/you_get/extractors/tiktok.py
+++ b/src/you_get/extractors/tiktok.py
@@ -16,12 +16,12 @@ def tiktok_download(url, output_dir='.', merge=True, info_only=False, **kwargs):
     m = re.match('(https?://)?([^/]+)(/.*)', url)
     host = m.group(2)
     if host != 'www.tiktok.com':  # non-canonical URL
-        url = get_location(url, headers=headers)
-        m = re.match('(https?://)?([^/]+)(/.*)', url)
-        host = m.group(2)
-
-    url = m.group(3).split('?')[0]
-    vid = url.split('/')[3]  # should be a string of numbers
+        vid = r1(r'/video/(\d+)', url)
+        url = 'https://www.tiktok.com/@/video/%s/' % vid
+        host = 'www.tiktok.com'
+    else:
+        url = m.group(3).split('?')[0]
+        vid = url.split('/')[3]  # should be a string of numbers
 
     html, set_cookie = getHttps(host, url, headers=headers)
     tt_chain_token = r1('tt_chain_token=([^;]+);', set_cookie)