A year ago I made a pool downloader with regex/cache/php/etc, but it was a real mess and it failed like 75% of the times I try to download. (it also needed tweaks to work on e926/wc/etc.... so that script is kinda useless and I quit it. :P)

A week ago I noticed this site got an API, so I wanted to give a try to remake the pool downloader. The only problem is: I have no idea how to use this site's API. (Yes, I did read the api help and wiki.)

I was able to get the XML from pool, but I am unable to parse XML attributes. (XML elements are nicer and easier)

Can anyone give me some directions?

It depends on what platform you're doing it

NSFW said:

I was able to get the XML from pool, but I am unable to parse XML attributes. (XML elements are nicer and easier)

Can anyone give me some directions?

Yes. Use your language's libxml2 bindings, or whatever other library it has available . Don't roll your own, that's just the path to blowing up repeatedly.

That said, XML is annoying to parse even when you have great libraries. I suggest going with JSON instead, as e621 supports it for many types of request -- for example, https://e621.net/pool/show.json?id=12&page=1
. Pretty much every language has decent JSON support by now, and IME it's much easier to handle than XML.

In my case, I use Python's standard 'json' library to parse the JSON, and for getting the JSON from the e621 servers I use the amazingly easy-to-use 'requests' library.

For me, that looks like (quick example that just gets the pool name)

import json
import requests
data = json.loads(requests.get('https://e621.net/pool/show.json?id=12&page=1').text)
pool_name = data['name']
print (pool_name)

I was using php's simpleXML for this. XML itself isn't hard, but XML attributes are annoying to parse.

Indeed... JSON seems to be a lot easier to parse (or even regex it). Now all I need to do is to figure how to use JSON, but nothing on php is hard to learn.

ht tp s://e621.net/pool/show.json?id=$1&page=$2 returns the pool and pages. Not that hard. :D

Thank you

NSFW said:
I was using php's simpleXML for this. XML itself isn't hard, but XML attributes are annoying to parse.

Indeed... JSON seems to be a lot easier to parse (or even regex it). Now all I need to do is to figure how to use JSON, but nothing on php is hard to learn.

https://e621.net/pool/show.json?id=$1&page=$2 returns the pool and pages. Not that hard. :D

Thank you

JSON is really easy to understand, it's just

{"key1":value1, "key2":value2, "keyN":valueN}

This should come in handy for you (taken from JSON.org site)

Almost done.
The only problem I got is: It doesn't download all imgs.
I'm testing the script using the pool #12. This pool has 18 imgs, but the script downloads 17 only. Even the JSON it returns 17 imgs only and there is no page 2, but if you check https://e621.net/pool/show/12 , this pool has 18 imgs. :P

Well, that's funny. This pool doesn't work right, but 1188 seems to work correctly. =P

NSFW said:
Almost done.
The only problem I got is: It doesn't download all imgs.
I'm testing the script using the pool #12. This pool has 18 imgs, but the script downloads 17 only. Even the JSON it returns 17 imgs only and there is no page 2, but if you check https://e621.net/pool/show/12 , this pool has 18 imgs. :P

Well, that's funny. This pool doesn't work right, but 1188 seems to work correctly. =P

That's because there's a deleted image in pool #12.

parasprite said:
That's because there's a deleted image in pool #12.

Oh, I see. I spent last 30 minutes trying to figure what happened. lolz

Well the script: http://pastebin.com/Hxv3krKj
For now, it'll no user/web interface, but this isn't hard to use: script.php?id=pool_id&pg=page_id

All you'll need is: http://www.wampserver.com/en/ (or if you're using Linux, you could try LAMP. I'm not 100% sure, but it should work with linux. You also may need to chmod 777 the folder/script.)

Have fun :P