From d661c95480abd61f7ef8877d8dbcb827534aa54d Mon Sep 17 00:00:00 2001
From: Mort Yao <soi@mort.ninja>
Date: Fri, 1 Jul 2022 22:21:47 +0200
Subject: [PATCH] [instagram] fix extraction

---
 src/you_get/extractors/instagram.py | 74 +++++++++++++----------------
 1 file changed, 32 insertions(+), 42 deletions(-)

diff --git a/src/you_get/extractors/instagram.py b/src/you_get/extractors/instagram.py
index 4167b226..604c534c 100755
--- a/src/you_get/extractors/instagram.py
+++ b/src/you_get/extractors/instagram.py
@@ -10,60 +10,50 @@ def instagram_download(url, output_dir='.', merge=True, info_only=False, **kwarg
 
     vid = r1(r'instagram.com/\w+/([^/]+)', url)
     description = r1(r'<meta property="og:title" content="([^"]*)"', cont) or \
-        r1(r'<title>\s([^<]*)</title>', cont) # with logged-in cookies
+        r1(r'<title>([^<]*)</title>', cont) # with logged-in cookies
     title = "{} [{}]".format(description.replace("\n", " "), vid)
 
-    stream = r1(r'<meta property="og:video" content="([^"]*)"', cont)
-    if stream:
-        _, ext, size = url_info(stream)
+    appId = r1(r'"appId":"(\d+)"', cont)
+    media_id = r1(r'"media_id":"(\d+)"', cont)
 
-        print_info(site_info, title, ext, size)
-        if not info_only:
-            download_urls([stream], title, ext, size, output_dir, merge=merge)
-    else:
-        data = re.search(r'window\._sharedData\s*=\s*(.*);</script>', cont)
-        try:
-            info = json.loads(data.group(1))
-            post = info['entry_data']['PostPage'][0]
-            assert post['items']
-        except:
-            # with logged-in cookies
-            data = re.search(r'window\.__additionalDataLoaded\(\'[^\']+\',(.*)\);</script>', cont)
-            if data is not None:
-                log.e('[Warning] Cookies needed.')
-            post = json.loads(data.group(1))
+    api_url = 'https://i.instagram.com/api/v1/media/%s/info/' % media_id
+    try:
+        api_cont = get_content(api_url, headers={**fake_headers, **{'x-ig-app-id': appId}})
+    except:
+        log.wtf('[Error] Please specify a cookie file.')
+    post = json.loads(api_cont)
 
-        for item in post['items']:
-            code = item['code']
-            carousel_media = item.get('carousel_media') or [item]
-            for i, media in enumerate(carousel_media):
-                title = '%s [%s]' % (code, i)
-                image_url = media['image_versions2']['candidates'][0]['url']
-                ext = image_url.split('?')[0].split('.')[-1]
-                size = int(get_head(image_url)['Content-Length'])
+    for item in post['items']:
+        code = item['code']
+        carousel_media = item.get('carousel_media') or [item]
+        for i, media in enumerate(carousel_media):
+            title = '%s [%s]' % (code, i)
+            image_url = media['image_versions2']['candidates'][0]['url']
+            ext = image_url.split('?')[0].split('.')[-1]
+            size = int(get_head(image_url)['Content-Length'])
+
+            print_info(site_info, title, ext, size)
+            if not info_only:
+                download_urls(urls=[image_url],
+                              title=title,
+                              ext=ext,
+                              total_size=size,
+                              output_dir=output_dir)
+
+            # download videos (if any)
+            if 'video_versions' in media:
+                video_url = media['video_versions'][0]['url']
+                ext = video_url.split('?')[0].split('.')[-1]
+                size = int(get_head(video_url)['Content-Length'])
 
                 print_info(site_info, title, ext, size)
                 if not info_only:
-                    download_urls(urls=[image_url],
+                    download_urls(urls=[video_url],
                                   title=title,
                                   ext=ext,
                                   total_size=size,
                                   output_dir=output_dir)
 
-                # download videos (if any)
-                if 'video_versions' in media:
-                    video_url = media['video_versions'][0]['url']
-                    ext = video_url.split('?')[0].split('.')[-1]
-                    size = int(get_head(video_url)['Content-Length'])
-
-                    print_info(site_info, title, ext, size)
-                    if not info_only:
-                        download_urls(urls=[video_url],
-                                      title=title,
-                                      ext=ext,
-                                      total_size=size,
-                                      output_dir=output_dir)
-
 site_info = "Instagram.com"
 download = instagram_download
 download_playlist = playlist_not_supported('instagram')