From 5fa1e13866b9dcbf8bb919d36a6ead96e7de9a20 Mon Sep 17 00:00:00 2001
From: prowlguru <183935626+prowlguru@users.noreply.github.com>
Date: Mon, 28 Apr 2025 15:14:41 +0000
Subject: [PATCH] [pixiv:novel] fix 'embeds' extraction with AJAX request
 (#7422 #7435)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

* [pixiv] fix novel embeds with AJAX request

All novel embeds seem to be broken since
`meta-preload-data` is no longer present in the
html of the php response after a site change.

* [pixiv] remove headers used for old novel embed request

* [pixiv] cleanup novel embed fix

* [pixiv] handle novel embed exceptions

* [pixiv] fix PEP formatting for novel embeds fix

* Fix error message

* [pixiv] include exception info in novel error message

---------

Co-authored-by: prowlguru <prowlguru@github.com>
Co-authored-by: Mike Fährmann <mike_faehrmann@web.de>
---
 gallery_dl/extractor/pixiv.py | 25 ++++++-------------------
 1 file changed, 6 insertions(+), 19 deletions(-)

diff --git a/gallery_dl/extractor/pixiv.py b/gallery_dl/extractor/pixiv.py
index dfed1aa5..c063216a 100644
--- a/gallery_dl/extractor/pixiv.py
+++ b/gallery_dl/extractor/pixiv.py
@@ -866,16 +866,6 @@ class PixivNovelExtractor(PixivExtractor):
         embeds = self.config("embeds")
         covers = self.config("covers")
 
-        if embeds:
-            headers = {
-                "User-Agent"    : "Mozilla/5.0",
-                "App-OS"        : None,
-                "App-OS-Version": None,
-                "App-Version"   : None,
-                "Referer"       : self.root + "/",
-                "Authorization" : None,
-            }
-
         novels = self.novels()
         if self.max_posts:
             novels = itertools.islice(novels, self.max_posts)
@@ -935,15 +925,12 @@ class PixivNovelExtractor(PixivExtractor):
 
                 if desktop:
                     try:
-                        novel_id = str(novel["id"])
-                        url = "{}/novel/show.php?id={}".format(
-                            self.root, novel_id)
-                        data = util.json_loads(text.extr(
-                            self.request(url, headers=headers).text,
-                            "id=\"meta-preload-data\" content='", "'"))
-                        images = (data["novel"][novel_id]
-                                  ["textEmbeddedImages"]).values()
-                    except Exception:
+                        body = self._request_ajax("/novel/" + str(novel["id"]))
+                        images = body["textEmbeddedImages"].values()
+                    except Exception as exc:
+                        self.log.warning(
+                            "%s: Failed to get embedded novel images (%s: %s)",
+                            novel["id"], exc.__class__.__name__, exc)
                         images = ()
 
                     for image in images: