goodreads-backup/parse.py

139 lines
4.3 KiB
Python
Raw Normal View History

#!/usr/bin/python3
import os
import re
import lxml.html
import jinja2
import json
import sys
import requests
from time import sleep
from pprint import pprint
2024-04-17 09:12:11 +01:00
parser = lxml.html.HTMLParser(encoding="utf-8")
2024-04-17 09:12:11 +01:00
env = jinja2.Environment(loader=jinja2.FileSystemLoader("templates"))
2024-04-17 09:12:11 +01:00
re_book = re.compile(
"function refreshGroupBox(group_id, book_id) \{(.*?)\n *\}", re.DOTALL
)
re_book = re.compile(
"function refreshGroupBox\(group_id, book_id\) \{(.*?)\n *\}", re.DOTALL
)
2024-04-17 09:12:11 +01:00
re_tip = re.compile(rb'var newTip = new Tip\(\$\(\'[^\']+\'\), "(.*?)", {')
2024-04-17 09:12:11 +01:00
dirs = ["shelf", "genre"]
start = "https://www.goodreads.com/book/show/"
existing = {book["title"] for book in json.load(open("calibre_book_list.json"))}
def iter_books():
for d in dirs:
for f in sorted(os.listdir(d)):
filename = os.path.join(d, f)
root = lxml.html.parse(filename, parser=parser).getroot()
2024-04-17 09:12:11 +01:00
for div in root.find_class("bookInformation"):
link = div.find(".//a")
rating = div.find('.//span[@class="minirating"]')
description = div.find('./div[@class="bookDescription"]')
r = rating[0].tail.strip()
2024-04-17 09:12:11 +01:00
cover = div.getnext().find(".//img").get("src")
book = {
2024-04-17 09:12:11 +01:00
"title": link.text,
"url": link.get("href"),
"rating": r,
"r": float(r[:3]),
"cover": cover,
"authors": [a.text for a in div.find_class("authorName")],
}
if description is not None:
index = 1 if len(description) == 3 else 0
2024-04-17 09:12:11 +01:00
book["description"] = description[index].text
yield d, f, book
continue
# print(filename)
2024-04-17 09:12:11 +01:00
for line in open(filename, "rb"):
if b"var newTip" not in line:
continue
print(line)
m = re_tip.search(line)
2024-04-17 09:12:11 +01:00
tip = m.group(1).decode("unicode_escape").replace("\/", "/")
# tip = m.group(1) # .replace('\/', '/')
# print(tip)
if '<ul class="formatting_tips recommendation_tip">' in tip:
continue
2024-04-17 09:12:11 +01:00
if "Recommendations are disabled for that shelf." in tip:
continue
2024-04-17 09:12:11 +01:00
if "Customize by selecting your" in tip:
continue
print(tip)
yield (d, f, lxml.html.fromstring(tip))
2024-04-17 09:12:11 +01:00
template = env.get_template("books.html")
seen = set()
books = []
first_authors = set()
2024-04-17 09:12:11 +01:00
for d, f, book in sorted(iter_books(), key=lambda i: i[2]["r"], reverse=True):
# pprint(book)
# print(repr(book.get('description')))
# continue
# title_link = book.find_class('bookTitle')[0]
2024-04-17 09:12:11 +01:00
url = book["url"]
url = url[: url.find("?")]
if url in seen:
continue
seen.add(url)
2024-04-17 09:12:11 +01:00
title = book["title"]
authors = book["authors"]
first_authors.add(authors[0])
main_title = title
# for sep in ['(']:
2024-04-17 09:12:11 +01:00
for sep in ":", " - ", "(", ",":
if sep in title:
2024-04-17 09:12:11 +01:00
main_title = title[: title.find(sep)]
break
# print((main_title + ' by ' + u', '.join(authors)).encode('utf-8'))
if len(main_title) < 10:
continue
if main_title in existing:
continue
# print(u'{} by {}'.format(main_title, authors[0]).encode('utf-8'))
2024-04-17 09:12:11 +01:00
print("{}".format(main_title))
# print(main_title.encode('utf-8'))
assert url.startswith(start)
2024-04-17 09:12:11 +01:00
filename = "books/" + url[len(start) :] + ".html"
# print(filename)
if False and not os.path.exists(filename):
2024-04-17 09:12:11 +01:00
open(filename, "w").write(requests.get(url).content)
sleep(1)
2024-04-17 09:12:11 +01:00
books.append(
{
"dir": d,
"file": f[:-5],
"title": title,
"main_title": main_title,
"authors": authors,
"url": url,
"rating": book["rating"],
"cover": book["cover"],
"description": book.get("description"),
}
)
page = template.render(books=books)
2024-04-17 09:12:11 +01:00
open("book_list.html", "w").write(page)
sys.exit(0)
for a in sorted(first_authors):
print(a)
# authors = u' OR '.join(u'"{}"'.format(a) for a in sorted(first_authors) if a not in {'Hugh Howey', 'Elizabeth Moon', 'Max Hastings'})
# print(authors.encode('utf-8'))