1 Commits

Author SHA1 Message Date
lamp 6a442bb5bf delete scrap 2021-09-24 10:55:06 -07:00
8 changed files with 12 additions and 74 deletions
+3 -2
View File
@@ -1,2 +1,3 @@
PHPSESSID.txt node_modules
download download
chromium_data
+9 -10
View File
@@ -1,13 +1,12 @@
This is a python script for downloading original pixiv images from popular search results via a premium account.
# Instructions ### example usage
```sh
node pixiv-downloader.js "初音ミク" -p 3
```
1. Add `210.140.131.219 www.pixiv.net` to your hosts file to bypass Cloudflare, or the script will be blocked. (`nslookup pixiv.net` in case the address needs to be changed) Chromium window will appear where you'll need to sign in to your pixiv premium account if you haven't already.
2. Download this repo to your computer of course, and open the terminal in it. ### args
- `-p <num>` for number of pages (default 1)
3. In your browser, on Pixiv logged in to a premium account, in dev tools Application tab, copy the **value** of the `PHPSESSID` cookie, and paste it into a new file named `PHPSESSID.txt` in this folder. - `-s <num>` for page number to start on (default 1)
- `-r` to include R18
4. Run `python pixiv-popular-downloader.py -h` for usage information. Example usage to download 10 pages of 初音ミク tag, including r18: `python pixiv-popular-downloader.py -r -p 10 "初音ミク"`
5. Check the download folder. If you're getting newest results instead of popular results, then your PHPSESSID failed to work.
-3
View File
@@ -1,3 +0,0 @@
node_modules
download
chromium_data
-12
View File
@@ -1,12 +0,0 @@
### example usage
```sh
node pixiv-downloader.js "初音ミク" -p 3
```
Chromium window will appear where you'll need to sign in to your pixiv premium account if you haven't already.
### args
- `-p <num>` for number of pages (default 1)
- `-s <num>` for page number to start on (default 1)
- `-r` to include R18
View File
View File
-47
View File
@@ -1,47 +0,0 @@
import argparse
import requests
from urllib.parse import quote as encodeURI
import os
ap = argparse.ArgumentParser()
ap.add_argument("tag", help="Pixiv tag(s) to search")
ap.add_argument("-p", dest="numpages", type=int, default=1, help="number of pages to download (default 1)")
ap.add_argument("-s", dest="startpagenum", type=int, default=1, help="page number to start at")
ap.add_argument("-r", action='store_true', help="include r18 posts")
args = ap.parse_args()
PHPSESSID = None
with open("PHPSESSID.txt", 'r') as f:
PHPSESSID = f.read()
download_count = 1
for i in range(args.startpagenum, args.numpages+1):
page_url = f"https://www.pixiv.net/ajax/search/artworks/{encodeURI(args.tag, safe='')}?order=popular_d&mode={'all' if args.r else 'safe'}&p={i}"
print("get", page_url)
page_data = requests.get(page_url, cookies={"PHPSESSID": PHPSESSID}).json()
if (page_data['error']):
print(page_data['message'])
exit(1)
for illust in page_data['body']['illustManga']['data']:
illust_r18 = bool(illust['xRestrict'])
illust_url = f"https://www.pixiv.net/ajax/illust/{illust['id']}/pages"
print("get", illust_url)
illust_data = requests.get(illust_url).json()
if (illust_data['error']):
print(illust_data['message'])
else:
for image in illust_data['body']:
image_url = image['urls']['original']
download_dir = f"download/{args.tag}/"
os.makedirs(download_dir, exist_ok=True)
download_filename = str(download_count) + '_' + ('x_' if illust_r18 else '') + image_url.split('/').pop()
download_path = download_dir + download_filename
if os.path.exists(download_path):
print(download_path, "already exists")
continue
print("get", image_url)
req = requests.get(image_url, headers={'referer':'https://www.pixiv.net'})
with open(download_path, "wb") as f:
f.write(req.content)
print("saved", download_filename)
download_count = download_count + 1