def fetch_page(
url=None,
load_complete=lambda browser, url, recursion_level: True,
links=lambda browser, url, recursion_level: (),
keep_only=(),
remove_after=None,
remove_before=None,
remove=(),
remove_javascript=True,
delay=0,
preprocess_browser=lambda browser, url, stage, recursion_level:None,
postprocess_html=lambda root, url, recursion_level: root,
resource_cache={},
output_dir=None,
browser=None,
recursion_level=0
):
output_dir = output_dir or os.getcwdu()
if browser is None:
browser = jsbrowser()
if delay:
time.sleep(delay)
# Load the DOM
if url is not None:
start_time = time.time()
browser.start_load(url)
while not load_complete(browser, url, recursion_level):
browser.run_for_a_time(0.1)
if time.time() - start_time > browser.default_timeout:
raise Timeout('Timed out while waiting for %s to load' % url)
children = links(browser, url, recursion_level)
# Cleanup the DOM
clean_dom(
browser, url, recursion_level, preprocess_browser,
remove_javascript, keep_only, remove_after, remove_before, remove)
# Download resources
download_resources(browser, resource_cache, output_dir)
# Get HTML from the DOM
pages = [save_html(browser, output_dir, postprocess_html, url, recursion_level)]
# Fetch the linked pages
for i, curl in enumerate(children):
odir = os.path.join(output_dir, 'link%d' % (i + 1))
if not os.path.exists(odir):
os.mkdir(odir)
try:
pages.extend(fetch_page(
curl, load_complete=load_complete, links=links, keep_only=keep_only,
remove_after=remove_after, remove_before=remove_before, remove=remove,
preprocess_browser=preprocess_browser, postprocess_html=postprocess_html,
resource_cache=resource_cache, output_dir=odir, browser=browser, delay=delay,
recursion_level=recursion_level+1))
except AbortFetch:
continue
return tuple(pages)
def fetch_page(
url=None,
load_complete=lambda browser, url, recursion_level: True,
links=lambda browser, url, recursion_level: (),
keep_only=(),
remove_after=None,
remove_before=None,
remove=(),
remove_javascript=True,
delay=0,
preprocess_browser=lambda browser, url, stage, recursion_level:None,
postprocess_html=lambda root, url, recursion_level: root,
resource_cache={},
output_dir=None,
browser=None,
recursion_level=0
):
output_dir = output_dir or os.getcwdu()
if browser is None:
browser = jsbrowser()
if delay:
time.sleep(delay)
# Load the DOM
if url is not None:
start_time = time.time()
browser.start_load(url)
while not load_complete(browser, url, recursion_level):
browser.run_for_a_time(0.1)
if time.time() - start_time > browser.default_timeout:
from calibre.web.jsbrowser.browser import Timeout
raise Timeout('Timed out while waiting for %s to load' % url)
children = links(browser, url, recursion_level)
# Cleanup the DOM
clean_dom(
browser, url, recursion_level, preprocess_browser,
remove_javascript, keep_only, remove_after, remove_before, remove)
# Download resources
download_resources(browser, resource_cache, output_dir)
# Get HTML from the DOM
pages = [save_html(browser, output_dir, postprocess_html, url, recursion_level)]
# Fetch the linked pages
for i, curl in enumerate(children):
odir = os.path.join(output_dir, 'link%d' % (i + 1))
if not os.path.exists(odir):
os.mkdir(odir)
try:
pages.extend(fetch_page(
curl, load_complete=load_complete, links=links, keep_only=keep_only,
remove_after=remove_after, remove_before=remove_before, remove=remove,
preprocess_browser=preprocess_browser, postprocess_html=postprocess_html,
resource_cache=resource_cache, output_dir=odir, browser=browser, delay=delay,
recursion_level=recursion_level+1))
except AbortFetch:
continue
return tuple(pages)
download_resources(browser, resource_cache, output_dir)
# Get HTML from the DOM
pages = [save_html(browser, output_dir, postprocess_html, url, recursion_level)]
# Fetch the linked pages
for i, curl in enumerate(children):
odir = os.path.join(output_dir, 'link%d' % (i + 1))
if not os.path.exists(odir):
os.mkdir(odir)
try:
pages.extend(fetch_page(
curl, load_complete=load_complete, links=links, keep_only=keep_only,
remove_after=remove_after, remove_before=remove_before, remove=remove,
preprocess_browser=preprocess_browser, postprocess_html=postprocess_html,
resource_cache=resource_cache, output_dir=odir, browser=browser, delay=delay,
recursion_level=recursion_level+1))
except AbortFetch:
continue
return tuple(pages)
if __name__ == '__main__':
browser = jsbrowser()
fetch_page('http://www.time.com/time/magazine/article/0,9171,2145057,00.html', browser=browser,
links=partial(links_from_selectors, ('.wp-paginate a.page[href]',), 1),
keep_only=('article.post',), remove=('.entry-sharing', '.entry-footer', '.wp-paginate', '.post-rail'))
download_resources(browser, resource_cache, output_dir)
# Get HTML from the DOM
pages = [save_html(browser, output_dir, postprocess_html, url, recursion_level)]
# Fetch the linked pages
for i, curl in enumerate(children):
odir = os.path.join(output_dir, 'link%d' % (i + 1))
if not os.path.exists(odir):
os.mkdir(odir)
try:
pages.extend(fetch_page(
curl, load_complete=load_complete, links=links, keep_only=keep_only,
remove_after=remove_after, remove_before=remove_before, remove=remove,
preprocess_browser=preprocess_browser, postprocess_html=postprocess_html,
resource_cache=resource_cache, output_dir=odir, browser=browser, delay=delay,
recursion_level=recursion_level+1))
except AbortFetch:
continue
return tuple(pages)
if __name__ == '__main__':
browser = jsbrowser()
fetch_page('http://www.time.com/time/magazine/article/0,9171,2145057,00.html', browser=browser,
links=partial(links_from_selectors, ('.wp-paginate a.page[href]',), 1),
keep_only=('article.post',), remove=('.entry-sharing', '.entry-footer', '.wp-paginate', '.post-rail'))
