From 69b931b9bb79d161633f9028963c3b6e2bf8293d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Sat, 4 Nov 2023 17:06:46 +0100
Subject: [PATCH] [exhentai] provide fallback URLs (#1021, #4745)

---
 gallery_dl/extractor/exhentai.py | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/gallery_dl/extractor/exhentai.py b/gallery_dl/extractor/exhentai.py
index 549b8b2b..5185fe31 100644
--- a/gallery_dl/extractor/exhentai.py
+++ b/gallery_dl/extractor/exhentai.py
@@ -275,15 +275,19 @@ class ExhentaiGalleryExtractor(ExhentaiExtractor):
 
         self.key_next = extr("'", "'")
         iurl = extr('<img id="img" src="', '"')
+        nl = extr(" nl(", ")").strip("\"'")
         orig = extr('hentai.org/fullimg', '"')
 
         try:
             if self.original and orig:
                 url = self.root + "/fullimg" + text.unescape(orig)
                 data = self._parse_original_info(extr('ownload original', '<'))
+                data["_fallback"] = ("{}?nl={}".format(url, nl),)
             else:
                 url = iurl
                 data = self._parse_image_info(url)
+                data["_fallback"] = self._fallback(
+                    None, self.image_num, nl)
         except IndexError:
             self.log.debug("Page content:\n%s", page)
             raise exception.StopExtraction(
@@ -317,6 +321,8 @@ class ExhentaiGalleryExtractor(ExhentaiExtractor):
             imgkey = nextkey
             nextkey, pos = text.extract(i3, "'", "'")
             imgurl , pos = text.extract(i3, 'id="img" src="', '"', pos)
+            nl     , pos = text.extract(i3, " nl(", ")", pos)
+            nl = (nl or "").strip("\"'")
 
             try:
                 pos = i6.find("hentai.org/fullimg")
@@ -325,9 +331,12 @@ class ExhentaiGalleryExtractor(ExhentaiExtractor):
                     url = text.unescape(origurl)
                     data = self._parse_original_info(text.extract(
                         i6, "ownload original", "<", pos)[0])
+                    data["_fallback"] = ("{}?nl={}".format(url, nl),)
                 else:
                     url = imgurl
                     data = self._parse_image_info(url)
+                    data["_fallback"] = self._fallback(
+                        imgkey, request["page"], nl)
             except IndexError:
                 self.log.debug("Page content:\n%s", page)
                 raise exception.StopExtraction(
@@ -401,6 +410,14 @@ class ExhentaiGalleryExtractor(ExhentaiExtractor):
             raise exception.NotFoundError("image page")
         return page
 
+    def _fallback(self, imgkey, num, nl):
+        url = "{}/s/{}/{}-{}?nl={}".format(
+            self.root, imgkey or self.key_start, self.gallery_id, num, nl)
+        page = self.request(url, fatal=False).text
+        if page.startswith(("Invalid page", "Keep trying")):
+            return
+        yield self.image_from_page(page)[0]
+
     @staticmethod
     def _parse_image_info(url):
         for part in url.split("/")[4:]: