Merge branch 'master' of gitea.moe:lamp/pixiv-popular-downloader

.gitignore

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

README.md

@@ -1,12 +1,11 @@
+This is a python script for downloading original pixiv images from popular search results via a premium account.
 
-### example usage
-```sh
-node pixiv-downloader.js "初音ミク" -p 3
-```
+# Instructions
 
-Chromium window will appear where you'll need to sign in to your pixiv premium account if you haven't already.
+1. Download this repo to your computer of course, and open the terminal in it. Run `pip install -r requirements.txt` if necessary.
 
-### args
-- `-p <num>` for number of pages (default 1)
-- `-s <num>` for page number to start on (default 1)
-- `-r` to include R18
+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.
+
+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"`
+
+4. Check the download folder. If you're getting newest results instead of popular results, then your PHPSESSID failed to work.

junk/.gitignore

@@ -0,0 +1,3 @@
+node_modules
+download
+chromium_data

junk/README.md

@@ -0,0 +1,12 @@
+
+### 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

pixiv-popular-downloader.py

@@ -0,0 +1,51 @@
+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

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

scrap/pixiv-downloader.py

@@ -1,35 +0,0 @@
-# simple downloader in python; i couldn't get cookie to work in order to get sorted by popularity so this is only useful for getting results sorted by newest/oldest and non-r18
-
-import argparse
-import requests
-from urllib.parse import quote as encodeURI
-
-#cookie = open("cookie.txt", 'r').read()
-cookie = None
-
-ap = argparse.ArgumentParser()
-ap.add_argument("tags")
-ap.add_argument("-p", "--pages", type=int, default=1, help="number of pages")
-args = ap.parse_args()
-
-for i in range(1, args.pages+1):
-	url = f"https://www.pixiv.net/ajax/search/artworks/{encodeURI(args.tags, safe='')}?order=popular_d&mode=safe&p={i}"
-	print("get", url)
-	data = requests.get(url, headers={cookie: cookie}).json()
-	if (data['error']):
-		print(data['message'])
-		exit(1)
-	for item in data['body']['illustManga']['data']:
-		url = f"https://www.pixiv.net/ajax/illust/{item['id']}/pages"
-		print("get", url)
-		data = requests.get(url, headers={cookie: cookie}).json()
-		if (data['error']):
-			print(data['message'])
-		else:
-			for image in data['body']:
-				url = image['urls']['original']
-				filename = url.split('/').pop()
-				print("get", url)
-				req = requests.get(url, headers={'referer':'https://www.pixiv.net'})
-				open("download/"+ filename, "wb").write(req.content)
-				print("saved", filename)