mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-07 18:24:30 -04:00
Implement live loading of common amazon code
This commit is contained in:
Kovid Goyal
parent
61b24aef56
commit
c98ad9a233
@ -2,13 +2,32 @@
|
||||
# vim:fileencoding=utf-8
|
||||
# License: GPL v3 Copyright: 2022, Kovid Goyal <kovid at kovidgoyal.net>
|
||||
|
||||
from lxml import etree, html
|
||||
from qt.core import QUrl
|
||||
from urllib.parse import urlencode
|
||||
from threading import Lock
|
||||
from time import monotonic
|
||||
|
||||
from calibre.gui2 import open_url
|
||||
from calibre.gui2.store.search_result import SearchResult
|
||||
from calibre.scraper.simple import read_url
|
||||
|
||||
|
||||
lock = Lock()
|
||||
cached_mod = None
|
||||
cached_time = -10000000
|
||||
|
||||
|
||||
def live_module():
|
||||
global cached_time, cached_mod
|
||||
with lock:
|
||||
now = monotonic()
|
||||
if now - cached_time > 3600:
|
||||
cached_mod = None
|
||||
if cached_mod is None:
|
||||
from calibre.live import load_module, Strategy
|
||||
cached_mod = load_module('calibre.gui2.store.amazon_live', strategy=Strategy.fast)
|
||||
return cached_mod
|
||||
|
||||
|
||||
def get_method(name):
|
||||
return getattr(live_module(), name)
|
||||
|
||||
|
||||
class AmazonStore:
|
||||
@ -24,92 +43,18 @@ class AmazonStore:
|
||||
DRM_FREE_TEXT = 'Unlimited'
|
||||
FIELD_KEYWORDS = 'k'
|
||||
|
||||
def search_amazon(self, query, max_results=10, timeout=60, write_html_to=None):
|
||||
field_keywords = self.FIELD_KEYWORDS
|
||||
uquery = self.SEARCH_BASE_QUERY.copy()
|
||||
uquery[field_keywords] = query
|
||||
|
||||
def asbytes(x):
|
||||
if isinstance(x, type('')):
|
||||
x = x.encode('utf-8')
|
||||
return x
|
||||
uquery = {asbytes(k):asbytes(v) for k, v in uquery.items()}
|
||||
url = self.SEARCH_BASE_URL + '?' + urlencode(uquery)
|
||||
|
||||
counter = max_results
|
||||
raw = read_url(self.scraper_storage, url, timeout=timeout)
|
||||
if write_html_to is not None:
|
||||
with open(write_html_to, 'w') as f:
|
||||
f.write(raw)
|
||||
doc = html.fromstring(raw)
|
||||
for result in doc.xpath('//div[contains(@class, "s-result-list")]//div[@data-index and @data-asin]'):
|
||||
kformat = ''.join(result.xpath('.//a[contains(text(), "{}")]//text()'.format(self.KINDLE_EDITION)))
|
||||
# Even though we are searching digital-text only Amazon will still
|
||||
# put in results for non Kindle books (author pages). So we need
|
||||
# to explicitly check if the item is a Kindle book and ignore it
|
||||
# if it isn't.
|
||||
if 'kindle' not in kformat.lower():
|
||||
continue
|
||||
asin = result.get('data-asin')
|
||||
if not asin:
|
||||
continue
|
||||
|
||||
cover_url = ''.join(result.xpath('.//img/@src'))
|
||||
title = etree.tostring(result.xpath('.//h2')[0], method='text', encoding='unicode')
|
||||
adiv = result.xpath('.//div[contains(@class, "a-color-secondary")]')[0]
|
||||
aparts = etree.tostring(adiv, method='text', encoding='unicode').split()
|
||||
idx = aparts.index(self.BY)
|
||||
author = ' '.join(aparts[idx+1:]).split('|')[0].strip()
|
||||
price = ''
|
||||
for span in result.xpath('.//span[contains(@class, "a-price")]/span[contains(@class, "a-offscreen")]'):
|
||||
q = ''.join(span.xpath('./text()'))
|
||||
if q:
|
||||
price = q
|
||||
break
|
||||
|
||||
counter -= 1
|
||||
|
||||
s = SearchResult()
|
||||
s.cover_url = cover_url.strip()
|
||||
s.title = title.strip()
|
||||
s.author = author.strip()
|
||||
s.detail_item = asin.strip()
|
||||
s.price = price.strip()
|
||||
s.formats = 'Kindle'
|
||||
|
||||
yield s
|
||||
|
||||
def get_details_amazon(self, search_result, timeout):
|
||||
url = self.DETAILS_URL + search_result.detail_item
|
||||
raw = read_url(self.scraper_storage, url, timeout=timeout)
|
||||
idata = html.fromstring(raw)
|
||||
return self.parse_details_amazon(idata, search_result)
|
||||
|
||||
def parse_details_amazon(self, idata, search_result):
|
||||
if idata.xpath('boolean(//div[@class="content"]//li/b[contains(text(), "' +
|
||||
self.DRM_SEARCH_TEXT + '")])'):
|
||||
if idata.xpath('boolean(//div[@class="content"]//li[contains(., "' +
|
||||
self.DRM_FREE_TEXT + '") and contains(b, "' +
|
||||
self.DRM_SEARCH_TEXT + '")])'):
|
||||
search_result.drm = SearchResult.DRM_UNLOCKED
|
||||
else:
|
||||
search_result.drm = SearchResult.DRM_UNKNOWN
|
||||
else:
|
||||
search_result.drm = SearchResult.DRM_LOCKED
|
||||
return True
|
||||
|
||||
def open(self, parent=None, detail_item=None, external=False):
|
||||
store_link = (self.DETAILS_URL + detail_item) if detail_item else self.STORE_LINK
|
||||
store_link = get_method('get_store_link_amazon')(self, detail_item)
|
||||
open_url(QUrl(store_link))
|
||||
|
||||
def search(self, query, max_results=10, timeout=60):
|
||||
for result in self.search_amazon(query, max_results=max_results, timeout=timeout):
|
||||
for result in get_method('search_amazon')(self, query, max_results=max_results, timeout=timeout):
|
||||
yield result
|
||||
|
||||
def get_details(self, search_result, timeout):
|
||||
return self.get_details_amazon(search_result, timeout)
|
||||
return get_method('get_details_amazon')(self, search_result, timeout)
|
||||
|
||||
def develop_plugin(self):
|
||||
import sys
|
||||
for result in self.search_amazon(' '.join(sys.argv[1:]), write_html_to='/t/amazon.html'):
|
||||
for result in get_method('search_amazon')(self, ' '.join(sys.argv[1:]), write_html_to='/t/amazon.html'):
|
||||
print(result)
|
||||
|
||||
94
src/calibre/gui2/store/amazon_live.py
Normal file
94
src/calibre/gui2/store/amazon_live.py
Normal file
@ -0,0 +1,94 @@
|
||||
#!/usr/bin/env python
|
||||
# vim:fileencoding=utf-8
|
||||
# License: GPL v3 Copyright: 2022, Kovid Goyal <kovid at kovidgoyal.net>
|
||||
|
||||
|
||||
from lxml import etree, html
|
||||
from urllib.parse import urlencode
|
||||
|
||||
from calibre.scraper.simple import read_url
|
||||
from calibre.gui2.store.search_result import SearchResult
|
||||
|
||||
|
||||
module_version = 1 # needed for live updates
|
||||
|
||||
|
||||
def search_amazon(self, query, max_results=10, timeout=60, write_html_to=None):
|
||||
field_keywords = self.FIELD_KEYWORDS
|
||||
uquery = self.SEARCH_BASE_QUERY.copy()
|
||||
uquery[field_keywords] = query
|
||||
|
||||
def asbytes(x):
|
||||
if isinstance(x, type('')):
|
||||
x = x.encode('utf-8')
|
||||
return x
|
||||
uquery = {asbytes(k):asbytes(v) for k, v in uquery.items()}
|
||||
url = self.SEARCH_BASE_URL + '?' + urlencode(uquery)
|
||||
|
||||
counter = max_results
|
||||
raw = read_url(self.scraper_storage, url, timeout=timeout)
|
||||
if write_html_to is not None:
|
||||
with open(write_html_to, 'w') as f:
|
||||
f.write(raw)
|
||||
doc = html.fromstring(raw)
|
||||
for result in doc.xpath('//div[contains(@class, "s-result-list")]//div[@data-index and @data-asin]'):
|
||||
kformat = ''.join(result.xpath('.//a[contains(text(), "{}")]//text()'.format(self.KINDLE_EDITION)))
|
||||
# Even though we are searching digital-text only Amazon will still
|
||||
# put in results for non Kindle books (author pages). So we need
|
||||
# to explicitly check if the item is a Kindle book and ignore it
|
||||
# if it isn't.
|
||||
if 'kindle' not in kformat.lower():
|
||||
continue
|
||||
asin = result.get('data-asin')
|
||||
if not asin:
|
||||
continue
|
||||
|
||||
cover_url = ''.join(result.xpath('.//img/@src'))
|
||||
title = etree.tostring(result.xpath('.//h2')[0], method='text', encoding='unicode')
|
||||
adiv = result.xpath('.//div[contains(@class, "a-color-secondary")]')[0]
|
||||
aparts = etree.tostring(adiv, method='text', encoding='unicode').split()
|
||||
idx = aparts.index(self.BY)
|
||||
author = ' '.join(aparts[idx+1:]).split('|')[0].strip()
|
||||
price = ''
|
||||
for span in result.xpath('.//span[contains(@class, "a-price")]/span[contains(@class, "a-offscreen")]'):
|
||||
q = ''.join(span.xpath('./text()'))
|
||||
if q:
|
||||
price = q
|
||||
break
|
||||
|
||||
counter -= 1
|
||||
|
||||
s = SearchResult()
|
||||
s.cover_url = cover_url.strip()
|
||||
s.title = title.strip()
|
||||
s.author = author.strip()
|
||||
s.detail_item = asin.strip()
|
||||
s.price = price.strip()
|
||||
s.formats = 'Kindle'
|
||||
|
||||
yield s
|
||||
|
||||
|
||||
def parse_details_amazon(self, idata, search_result):
|
||||
if idata.xpath('boolean(//div[@class="content"]//li/b[contains(text(), "' +
|
||||
self.DRM_SEARCH_TEXT + '")])'):
|
||||
if idata.xpath('boolean(//div[@class="content"]//li[contains(., "' +
|
||||
self.DRM_FREE_TEXT + '") and contains(b, "' +
|
||||
self.DRM_SEARCH_TEXT + '")])'):
|
||||
search_result.drm = SearchResult.DRM_UNLOCKED
|
||||
else:
|
||||
search_result.drm = SearchResult.DRM_UNKNOWN
|
||||
else:
|
||||
search_result.drm = SearchResult.DRM_LOCKED
|
||||
return True
|
||||
|
||||
|
||||
def get_details_amazon(self, search_result, timeout):
|
||||
url = self.DETAILS_URL + search_result.detail_item
|
||||
raw = read_url(self.scraper_storage, url, timeout=timeout)
|
||||
idata = html.fromstring(raw)
|
||||
return parse_details_amazon(self, idata, search_result)
|
||||
|
||||
|
||||
def get_store_link_amazon(self, detail_item):
|
||||
return (self.DETAILS_URL + detail_item) if detail_item else self.STORE_LINK
|
||||
@ -50,6 +50,9 @@ class SearchDialog(QDialog, Ui_Dialog):
|
||||
self.results_view.model().cover_pool.set_thread_count(self.cover_thread_count)
|
||||
self.results_view.model().details_pool.set_thread_count(self.details_thread_count)
|
||||
self.results_view.setCursor(Qt.CursorShape.PointingHandCursor)
|
||||
# needed for live updates of amazon_live.py
|
||||
from calibre.live import start_worker
|
||||
start_worker()
|
||||
|
||||
# Check for results and hung threads.
|
||||
self.checker = QTimer()
|
||||
|
||||
@ -1187,9 +1187,12 @@ class Main(MainWindow, MainWindowMixin, DeviceMixin, EmailMixin, # {{{
|
||||
wait_for_cleanup = cleanup_overseers()
|
||||
from calibre.db.delete_service import shutdown
|
||||
shutdown()
|
||||
from calibre.live import async_stop_worker
|
||||
wait_for_stop = async_stop_worker()
|
||||
time.sleep(2)
|
||||
self.istores.join()
|
||||
wait_for_cleanup()
|
||||
wait_for_stop()
|
||||
return True
|
||||
|
||||
def run_wizard(self, *args):
|
||||
|
||||
@ -58,6 +58,12 @@ def stop_worker(timeout=2*DEFAULT_TIMEOUT):
|
||||
w.join(timeout)
|
||||
|
||||
|
||||
def async_stop_worker():
|
||||
t = Thread(name='StopLiveDownloadWorker', target=stop_worker, daemon=True)
|
||||
t.start()
|
||||
return t.join
|
||||
|
||||
|
||||
def report_failure(full_name):
|
||||
print(f'Failed to download live module {full_name}', file=sys.stderr)
|
||||
import traceback
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user