delete scrap

.gitignore

@@ -1,2 +1,3 @@
-PHPSESSID.txt
+node_modules
-download
+download
+chromium_data

README.md

@@ -1,11 +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. Download this repo to your computer of course, and open the terminal in it. Run `pip install -r requirements.txt` if necessary.
+Chromium window will appear where you'll need to sign in to your pixiv premium account if you haven't already.
 
-2. 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.
+### args
-
+- `-p <num>` for number of pages (default 1)
-3. 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"`
+- `-s <num>` for page number to start on (default 1)
-
+- `-r` to include R18
-4. Check the download folder. If you're getting newest results instead of popular results, then your PHPSESSID failed to work.

package.json

@@ -0,0 +1 @@
+{"dependencies":{"download":"^8.0.0","minimist":"^1.2.5","puppeteer":"^10.2.0"}}

pixiv-downloader.js

@@ -0,0 +1,65 @@
+// node pixiv-downloader.js <tags> -p [numpages] -s [startpage] [-r]
+var argv = require("minimist")(process.argv.slice(2), {
+	default: {
+		"p": 1,
+		"s": 1
+	}
+});
+var query = argv._.join(" ");
+var puppeteer = require("puppeteer");
+var download = require("download");
+var fs = require("fs");
+
+(async function(){
+	var browser = await puppeteer.launch({
+		headless: false,
+		userDataDir: process.cwd() + "/chromium_data"
+	});
+
+	var page = await browser.newPage();
+	await page.setContent(`login to <a target="_blank" href="https://accounts.pixiv.net/login">pixiv</a> if necessary, then close this page to continue.`);
+	await new Promise(r => page.on("close", r));
+
+	var page = await browser.newPage();
+	for (let i = argv.s; i < argv.s + argv.p; i++) {
+		let url = `https://www.pixiv.net/ajax/search/artworks/${encodeURIComponent(query)}?order=popular_d&mode=${argv.r ? "all" : "safe"}&p=${i}`;
+		console.log("get page", i, url);
+		await page.goto(url);
+		let data = JSON.parse(await page.evaluate(() => document.querySelector("body").innerText));
+		if (data.error) throw data.message;
+		for (let item of data.body.illustManga.data) {
+			let url = `https://www.pixiv.net/ajax/illust/${item.id}/pages`;
+			console.log("get", url);
+			await page.goto(url);
+			let data = JSON.parse(await page.evaluate(() => document.querySelector("body").innerText));
+			if (data.error) {
+				console.error(error.message);
+			} else {
+				for (let image of data.body) {
+					let url = image.urls.original;
+					console.log("download", url);
+					if (!fs.existsSync("download")) fs.mkdirSync("download");
+					let filename = url.split('/').pop();
+					let filedir = "download/" + query + "/";
+					if (!fs.existsSync(filedir)) fs.mkdirSync(filedir);
+					let filepath = filedir + filename;
+					if (fs.existsSync(filepath)) {
+						console.log(filename, "already exists");
+						continue;
+					}
+					let write = fs.createWriteStream(filepath);
+					download(url, {
+						headers: {
+							"Referer": "https://www.pixiv.net"
+						}
+					}).pipe(write);
+					await new Promise(r => write.on("close", r));
+					console.log("saved", filename)
+				}
+			}
+		}
+	}
+	console.log("complete");
+	await browser.close();
+})();
+

pixiv-popular-downloader.py

@@ -1,51 +0,0 @@
-import argparse
-import requests
-from requests_toolbelt.adapters import host_header_ssl
-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()
-
-rqs = requests.Session()
-rqs.mount('https://', host_header_ssl.HostHeaderSSLAdapter())
-
-download_count = 1
-for i in range(args.startpagenum, args.numpages+1):
-	page_url = f"https://210.140.131.219/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 = rqs.get(page_url, cookies={"PHPSESSID": PHPSESSID}, headers={"host":"www.pixiv.net"}).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://210.140.131.219/ajax/illust/{illust['id']}/pages"
-		print("get", illust_url)
-		illust_data = rqs.get(illust_url, headers={"host":"www.pixiv.net"}).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)
-				res = rqs.get(image_url, headers={'referer':'https://www.pixiv.net'})
-				with open(download_path, "wb") as f:
-					f.write(res.content)
-				print("saved", download_filename)
-				download_count = download_count + 1

requirements.txt

@@ -1,3 +0,0 @@
-requests==2.26.0
-requests-toolbelt==0.9.1
-urllib3==1.26.6