Danbooru Downloader 20100722

UPDATE: see here.

Updated, small bugfixing and,

Update the DanbooruProviderList.xml as suggested by max and trick.

  • moe.imouto.org
  • gelbooru.com, different api
  • nekobooru.net
  • safebooru.org, different api
  • hijiribe.donmai.us
  • sonohara.donmai.us
  • ichijou.org

Please note that gelbooru and safebooru use different api from standard danbooru, and I don’t know if they provide json api (in the documentation only have xml).

Download link here.

13 thoughts on “Danbooru Downloader 20100722”

  1. Hey nandaka, I love your work! But I have some bugs to report to you and some tweaks I’d like to suggest.

    When I edit the settings under the “Downloads” section and press “Save” it doesn’t actually save my settings, only my “Internet Settings”

    The tag for %md5% doesn’t work correctly, you must input %md5%% in order for it to work.

    When I have %tags% in my filenames, some times the file won’t save, instead, it will say “File Completed!” when it actually just skipped it.

    “Auto Load Next Page” doesn’t seem to work correctly… The next page will only load if I scroll the list all the way down. So basically, what I currently have to do is hold PgDn and keep waiting for the next page until all pages have be loaded.

    Could you make the “Save” section a text-type input box so that we have a choice of either typing our save path or press “Browse” and going through folders. It’s just that it would be easier for some of us to just copy and paste a path instead of looking through a billion folders.

    Make it possible to stop a search. Currently, the only way is to press clear list, and that clears the whole list.

    “Save Query” is a great idea, but I think if you could optimize it so that it can save a whole list of a tag you input in your search and save the whole list, that would be much better. This way, if you want to update a tag you downloaded, then you can just load your saved query instead of doing a new search. If this is a little confusing, here is an example:

    I tag search “long hair” with “Save Query” and “Auto Load Next Page,” on. When it list is done loading, it will automatically save the query to a location that you’ve specified. Then when you want to update your list for new files, you can simply load the saved query, then update it by doing a search of the tag again.

    Currently, if you select save query, you have to first create a .xml file, and then it will only save one page. And if you have “Auto Load Next Page” on, then it will prompt you to save each page, which isnt that useful because pages on Danbooru is always changing because if new uploads.

    1. >> Bugs #1, #2, and #3:
      will check, for #3 can you give me the link?

      >> #4
      That was the designed behavior, user must scroll down to the last image (sankaku complex style), if you want to load more than default number of images, you can set the limit (usually 1000 is the maximum).

      >> Suggestion #1, #3:
      can be done. For #3 maybe a little bit difficult, because I’m only saving the raw reply from the server. I must change the data structure, but it can be done.

      >> #2:
      you mean terminating the list download before displayed on the table?

      1. >>Hmm, I’m unable to recreate bug #3… I’m not sure why.

        >>Thanks for explaining #4, works perfectly!

        >>Suggestion #2
        Hmm, I guess this suggestion wouldn’t be that necessary, since if you search for a new tag, it will be added to the list, in which we are still unable to delete certain files from the list (though I doubt many people would be deleting certain files from the list, seems like a useless feature).

        >>I guess you can forget about my suggestion #2, but here are some new ones I just thought of after using your Danbooru downloader some more:

        If I were to search for a tag, download some files, then attempt to clear the list to search for a different tag and download more files, it wouldn’t be possible. To circumvent this, people would just open a new client. If you could somehow make it so that you can just queue up a list of files to download, and keep adding to it, we would only need one client.

        Also wanted to ask if you could automatically change the “Page” to 1 after searching a new tag. This feature would be helping those that keep forgetting (including me) to delete the page of the previous search. If I’m unclear, here is an example:

        I input my Tag search as “green_hair”. The pages shown after doing so will be 30. If I input a new tag and press “Get” the search will begin at page 30 instead of starting back at 1.

        1. >> 4
          Can be done by making separate datagridview for downloading.

          >> 5
          Ok, will clear the page number if the tags is changed.

      2. Wanted to report another bug:

        When your list comes to a page in which it requires you to download, the program will say “Input string was not in a correct format.” I’m not sure what that means exactly, but I know it is because of these weird files.

        Here is an example you can try:

        Provider: Danbooru
        tag: carnelian
        limit: 20 (doesn’t really matter, but when it comes to the page which has the file, it will give you the error)
        page: 37 (in this case, it is on page 37 when each page is 20)

        This is the page in which those files are on: http://danbooru.donmai.us/post/index?tags=carnelian&commit=Search&page=37

      3. Forgot to mention, that last bug i reported, #5, when you get this error, the program will skip all files on the page in which the weird files are on.

      1. Thank you. I already tried that, but I tried it again just now, and I got the same error. It says that the framework is already installed and it refuses to let me install again.

        Screenshot: http://i28.tinypic.com/15z4ghv.png

        I’m searching around to see if I can find a solution.

        1. Seeing from your screen shot, you are trying to install .Net Framework version 2, but it need the version 4.

          It is different framework.

Comments are closed.