search

RedSquirrel87 / manga-downloader

A very simple application written in JAVA to download any chapters of any manga from the supported websites
37 stars 1 forks source link

MangaPark - Inability To Choose Version of Chapter to Download #860

Open gokieks opened 2 days ago

gokieks commented 2 days ago

To be honest, I'm not sure if this actually is something that can be done on the Manga Downloader side, but figured I'd post to get the app developer's opinion.

Manga Park's website now seems to only display the first (earliest) uploaded version of a chapter on the manga's main page, which means when Manga Downloader generates the list of chapters to download, it only presents the option of downloading that one. But the problem there is that sometimes the earliest upload is a bad version, or they are inconsistent (from different translation groups), and someone might want to download a different version instead - and there does not appear to be a way to do so.

Example: https://mangapark.net/title/57557-en-kanojo-okarishimasu

As of today (2024-10-15) the latest chapter is 349. On the details page chapter list, it is listed as "Chapter 349", which links to this version of the chapter from uploader "Koala". But the previous chapter is listed as "Ch. 348", which is because they are from different uploaders who used different naming, and links to a different version of that chapter from uploader "Fox". If you go to the actual chapter links, there's a "Files" section at the top, where you can see that for both 348 and 349, there are two files, one from each uploader, just with the order reversed:

image image

If I wanted to have a consistent collection, then it would follow that I would want to download the (for example) "Koala" version of both chapter 348 and 349. But it doesn't seem like it's currently possible to do so using Manga Downloader.

Now, like I mentioned at the start, it seems like this is a result of how the MangaPark details page lists chapters instead of files - I'm not sure if the files are exposed at all when viewing that page, so it may not be possible to get a list of files from that. If that's the case, then maybe Manga Downloader can support choosing the specific file if you give it the URL of a specific chapter rather than the details? As in...

https://mangapark.net/title/57557-en-kanojo-okarishimasu vs https://mangapark.net/title/57557-en-kanojo-okarishimasu/9085613-chapter-349

RedSquirrel87 commented 2 days ago

Thank you for reporting the issue, unfortunately the big problem here is that the information about the dual (or more) versions of each chapter is not visible from the chapter list. Therefore, the only moment where the program could detect the presence of multiple versions is exclusively when it loads the web pages with the images of the individual chapters. I could implement a dialog box asking the user which version to download at that moment, but this would completely break the batch download feature because a dialog would appear for each individual chapter the user tries to download from the list, and that's does not seem to me a good way to implement the support. I need to think carefully about how to implement this in a less obstructive way. Maybe by giving the user the option to enter the name of the translation group, so if the program detects its presence, it gives priority to that one?

gokieks commented 2 days ago

Yeah, there's really no great solution I don't think. The best I could come up with is if you enter the URL of the main details page (where it only lists chapters) then it'll just default to whatever that page links to and allow you to batch download, but if you specifically enter the URL of a single chapter, then it parses through the files list instead and let you choose a specific version. It still would be a mostly manual process, but at the very least having the option to choose a specific version, regardless of how much extra effort it takes, is better than not being able to at all, I think.

I'm not sure the translation group option would really work, because that's not really exposed data either - the only real identifying data of each version of a chapter are the title and uploader, and neither are actually unique (since the uploader name is usually a generic one).