From 250d35107ccaaaa5bb76f6544263e4af03726cb7 Mon Sep 17 00:00:00 2001
From: ClosedPort22 <44864697+ClosedPort22@users.noreply.github.com>
Date: Sat, 24 Dec 2022 12:35:44 +0800
Subject: [PATCH] [twitter] prioritize tweet type checks (#3439)

Do not consider a tweet seen before applying 'retweet', 'quote' and
'reply' checks. Otherwise the original tweets will also be skipped if
the "derivative" tweets and the original tweets are from the same user.
---
 gallery_dl/extractor/twitter.py | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/gallery_dl/extractor/twitter.py b/gallery_dl/extractor/twitter.py
index 22aa78e6..2bd8c484 100644
--- a/gallery_dl/extractor/twitter.py
+++ b/gallery_dl/extractor/twitter.py
@@ -75,11 +75,6 @@ class TwitterExtractor(Extractor):
             else:
                 data = tweet
 
-            if seen_tweets is not None:
-                if data["id_str"] in seen_tweets:
-                    continue
-                seen_tweets.add(data["id_str"])
-
             if not self.retweets and "retweeted_status_id_str" in data:
                 self.log.debug("Skipping %s (retweet)", data["id_str"])
                 continue
@@ -97,6 +92,13 @@ class TwitterExtractor(Extractor):
                 self.log.debug("Skipping %s (reply)", data["id_str"])
                 continue
 
+            if seen_tweets is not None:
+                if data["id_str"] in seen_tweets:
+                    self.log.debug(
+                        "Skipping %s (previously seen)", data["id_str"])
+                    continue
+                seen_tweets.add(data["id_str"])
+
             files = []
             if "extended_entities" in data:
                 self._extract_media(