pixiv downloader 20120114

Changelog:

  • Add BOM handler for list.txt.
  • Add End Page for download new illust from bookmark as requested by Golden Wolf.
  • Add Last Page Detection for download by tags as reported by tribe.
  • Add use wilcard option for download by tags and download by Title/Caption as requested by tribe.
  • Refactor Start/End page number UI.

Download link here, source code here or in the github.

UPDATE: Mediafire mirror here, source code here. please donate :).

22 thoughts on “pixiv downloader 20120114”

  1. I’m having a error using commandline for downloading new illustrations from bookmark
    In command prompt, regardless of whether i put in extra arguments or not i’m getting a error in getStartAndEndNumberFromArgs(). I don’t get any problems when running the exe directly and selecting from the menu.

    http://pastebin.com/jaFCEQVh

  3. A quick fix for the re-checking issue:
    Add the line:
    self.imageList = list()

    to line 404 of PixivModel.py (in the parseTags function of the PixivTags class)

    It’s in the other functions where it builds an image list, but it looks like it was just forgotten in the PixivTags class

  4. Because it’s open source, you can easily add the feature yourself, if you really want it. I did that a while ago when I needed that feature. Just be aware that it’s in Python, where whitespace is important (I hadn’t used Python before, so that part threw me off at first)

    I’m having a problem with it rechecking every file that I’ve downloaded, each time it loads a new page.

  5. I’ve been wondering, is there a way to bypass the 1000 page limit for searches?

    1. No, because the way the program works is depend on the website limitation (it actually get the website page and do parsing based on the page). As far I know, there is no API given for pixiv.

      1. Hmm, I see, what about a feature that uses/exploits the &ecd=yyyy-mm-dd search parameter? It basically searches all the pictures matching the search pattern -up to the date- specified in $ecd, normally it’s only used for pixiv’s “by last day/week/month” filter

        here’s a quick example:
        http://www.pixiv.net/search.php?s_mode=s_tag&word=%E9%B9%BF%E7%9B%AE%E3%81%BE%E3%81%A9%E3%81%8B&order=date_d&ecd=2011-02-21

        Notice how the “newest” pictures date doesn’t go past that february 21st

  6. When I try to download with no limit on end page, the program stops after the first page.

    So start page 1 (or blank, default = 1) and end page 0 (or blank, default = 0) stops after one page. Presumably whichever one I put as page start, although I’ve only done 1 so far.

    My OS is 32-bit Windows 7.

    But this is an amazing program, and I heartily thank you for making it! I encourage everyone who downloads the downloader to donate.

    1. Also, what is the correct way to download images containing multiple specific tags? I tried pasting the encoded tags one after another, separated by spaces, but it seemed to grabbing ones that had any of the tags. I’d like to download only those that have /all/ the tags specified.

    2. Thanks, I have checked the program and it is indeed stop after 1 page. For now you can put 100 as the end page or any big number if you want to download.

      I will fix it in the next release.

      2. Actually, I’m still having problems with the multiple tags.

        For example, inputting %E7%99%BE%E5%90%88%20vocaloid (decoded = η™Ύεˆ vocaloid) returns no images.

        The actual URL for this search is…

        http://www.pixiv.net/search.php?s_mode=s_tag&word=%E7%99%BE%E5%90%88%20vocaloid

        And the URL the program searches is…

        http://www.pixiv.net/tags.php?tag=%E7%99%BE%E5%90%88%20vocaloid&p=1

        Just entering the URLs in the browser, the first one works, but the second one gives pixiv’s no results message.

        When you do the second kind of fromat (tags.php?tag=) with any single tag, it works, but not with tags separated by %20.

        So unless I’m still doing something wrong, the URL format the program uses for single tags doesn’t work for multiple tags.

        Um… tags.php is accessed when you click an image tag, so maybe it doesn’t need to support multiple tags? On the other hand, search.php is the search version of the result page, so it’s meant for any number of tags? Seems right. Basically you can’t click on more than one tag to access tags.php, so it only reads things as a single tag.

        The other one (search.php?word=) should work, though? If the program can be made to look up images through that page it might work.

Comments are closed.