mirror of
https://github.com/ArchiveBox/ArchiveBox
synced 2024-11-10 06:34:16 +00:00
add archive.org submitting support, better parsing
This commit is contained in:
Nick Sweeting
parent
c5395a5d86
commit
99d5150b6e
2 changed files with 157 additions and 61 deletions
213
archive.py
213
archive.py
|
|
@ -12,23 +12,65 @@ import json
|
|||
from datetime import datetime
|
||||
import time
|
||||
|
||||
from subprocess import run, DEVNULL
|
||||
from subprocess import run, PIPE, DEVNULL
|
||||
|
||||
|
||||
### SETTINGS
|
||||
|
||||
INDEX_TEMPLATE = 'index_template.html'
|
||||
|
||||
FETCH_WGET = True
|
||||
FETCH_PDF = True
|
||||
FETCH_SCREENSHOT = True
|
||||
RESOLUTION = '1440,900' # screenshot resolution
|
||||
FETCH_FAVICON = True
|
||||
RESOLUTION = '1440,900'
|
||||
SUBMIT_ARCHIVE_DOT_ORG = True
|
||||
|
||||
CHROME_BINARY = 'google-chrome' # change to chromium browser if using chromium
|
||||
WGET_BINARY = 'wget'
|
||||
|
||||
|
||||
def check_dependencies():
|
||||
for dependency in ('google-chrome', 'wget'):
|
||||
if run(['which', dependency]).returncode:
|
||||
print('[X] Missing dependency: {}'.format(dependency))
|
||||
print('[*] Checking Dependencies:')
|
||||
if FETCH_PDF or FETCH_SCREENSHOT:
|
||||
if run(['which', CHROME_BINARY]).returncode:
|
||||
print('[X] Missing dependency: {}'.format(CHROME_BINARY))
|
||||
print(' See https://github.com/pirate/pocket-archive-stream for help.')
|
||||
raise SystemExit(1)
|
||||
|
||||
# parse chrome --version e.g. Google Chrome 61.0.3114.0 canary / Chromium 59.0.3029.110 built on Ubuntu, running on Ubuntu 16.04
|
||||
result = run([CHROME_BINARY, '--version'], stdout=PIPE)
|
||||
version = result.stdout.decode('utf-8').replace('Google Chrome ', '').replace('Chromium ', '').split(' ', 1)[0].split('.', 1)[0] # TODO: regex might be better
|
||||
if int(version) < 59:
|
||||
print('[X] Chrome version must be 59 or greater for headless PDF and screenshot saving')
|
||||
print(' See https://github.com/pirate/pocket-archive-stream for help.')
|
||||
raise SystemExit(1)
|
||||
|
||||
if FETCH_WGET:
|
||||
if run(['which', 'wget']).returncode:
|
||||
print('[X] Missing dependency: wget')
|
||||
print(' See https://github.com/pirate/pocket-archive-stream for help.')
|
||||
raise SystemExit(1)
|
||||
|
||||
if FETCH_FAVICON or SUBMIT_ARCHIVE_DOT_ORG:
|
||||
if run(['which', 'curl']).returncode:
|
||||
print('[X] Missing dependency: curl')
|
||||
print(' See https://github.com/pirate/pocket-archive-stream for help.')
|
||||
raise SystemExit(1)
|
||||
|
||||
|
||||
### PARSING READER LIST EXPORTS
|
||||
|
||||
def get_link_type(link):
|
||||
if link['base_url'].endswith('.pdf'):
|
||||
return 'PDF'
|
||||
elif link['base_url'].rsplit('.', 1) in ('pdf', 'png', 'jpg', 'jpeg', 'svg', 'bmp', 'gif', 'tiff', 'webp'):
|
||||
return 'image'
|
||||
elif 'wikipedia.org' in link['domain']:
|
||||
return 'wiki'
|
||||
elif 'youtube.com' in link['domain']:
|
||||
return 'youtube'
|
||||
return None
|
||||
|
||||
def parse_pocket_export(html):
|
||||
pattern = re.compile("^\\s*<li><a href=\"(.+)\" time_added=\"(\\d+)\" tags=\"(.*)\">(.+)</a></li>", re.UNICODE) # see sample input in ./example_ril_export.html
|
||||
|
|
@ -37,22 +79,24 @@ def parse_pocket_export(html):
|
|||
if match:
|
||||
fixed_url = match.group(1).replace('http://www.readability.com/read?url=', '') # remove old readability prefixes to get original url
|
||||
without_scheme = fixed_url.replace('http://', '').replace('https://', '')
|
||||
yield {
|
||||
info = {
|
||||
'url': fixed_url,
|
||||
'domain': without_scheme.split('/')[0], # without pathname
|
||||
'base_url': without_scheme.split('?')[0], # without query args
|
||||
'time': datetime.fromtimestamp(int(match.group(2))),
|
||||
'time': datetime.fromtimestamp(int(match.group(2))).strftime('%Y-%m-%d %H:%M'),
|
||||
'timestamp': match.group(2),
|
||||
'tags': match.group(3),
|
||||
'title': match.group(4).replace(' — Readability', '').replace('http://www.readability.com/read?url=', '') or without_scheme,
|
||||
}
|
||||
info['type'] = get_link_type(info)
|
||||
yield info
|
||||
|
||||
def parse_pinboard_export(html):
|
||||
json_content = json.load(html)
|
||||
for line in json_content:
|
||||
if line:
|
||||
erg = line
|
||||
yield {
|
||||
info = {
|
||||
'url': erg['href'],
|
||||
'domain': erg['href'].replace('http://', '').replace('https://', '').split('/')[0],
|
||||
'base_url': erg['href'].replace('https://', '').replace('http://', '').split('?')[0],
|
||||
|
|
@ -61,29 +105,11 @@ def parse_pinboard_export(html):
|
|||
'tags': erg['tags'],
|
||||
'title': erg['description'].replace(' — Readability', ''),
|
||||
}
|
||||
info['type'] = get_link_type(info)
|
||||
yield info
|
||||
|
||||
def dump_index(links, service):
|
||||
with open(INDEX_TEMPLATE, 'r') as f:
|
||||
index_html = f.read()
|
||||
|
||||
link_html = """\
|
||||
<tr>
|
||||
<td>{time}</td>
|
||||
<td><a href="archive/{timestamp}/{base_url}" style="font-size:1.4em;text-decoration:none;color:black;" title="{title}">
|
||||
<img src="archive/{timestamp}/favicon.ico">
|
||||
{title}
|
||||
</td>
|
||||
<td style="text-align:center"><a href="archive/{timestamp}/" title="Files">📂</a></td>
|
||||
<td style="text-align:center"><a href="archive/{timestamp}/output.pdf" title="PDF">📄</a></td>
|
||||
<td style="text-align:center"><a href="archive/{timestamp}/screenshot.png" title="Screenshot">🖼</a></td>
|
||||
<td>🔗 <img src="https://www.google.com/s2/favicons?domain={domain}" height="16px"> <a href="{url}">{url}</a></td>
|
||||
</tr>"""
|
||||
|
||||
with open(''.join((service,'/index.html')), 'w') as f:
|
||||
article_rows = '\n'.join(
|
||||
link_html.format(**link) for link in links
|
||||
)
|
||||
f.write(index_html.format(datetime.now().strftime('%Y-%m-%d %H:%M'), article_rows))
|
||||
### ACHIVING FUNCTIONS
|
||||
|
||||
def fetch_wget(out_dir, link, overwrite=False):
|
||||
# download full site
|
||||
|
|
@ -104,9 +130,9 @@ def fetch_pdf(out_dir, link, overwrite=False):
|
|||
# download PDF
|
||||
if (not os.path.exists('{}/output.pdf'.format(out_dir)) or overwrite) and not link['base_url'].endswith('.pdf'):
|
||||
print(' - Printing PDF')
|
||||
CMD = 'google-chrome --headless --disable-gpu --print-to-pdf'.split(' ')
|
||||
chrome_args = '--headless --disable-gpu --print-to-pdf'.split(' ')
|
||||
try:
|
||||
run([*CMD, link['url']], stdout=DEVNULL, stderr=DEVNULL, cwd=out_dir, timeout=20) # output.pdf
|
||||
run([CHROME_BINARY, *chrome_args, link['url']], stdout=DEVNULL, stderr=DEVNULL, cwd=out_dir, timeout=20) # output.pdf
|
||||
except Exception as e:
|
||||
print(' Exception: {}'.format(e.__class__.__name__))
|
||||
else:
|
||||
|
|
@ -116,14 +142,41 @@ def fetch_screenshot(out_dir, link, overwrite=False):
|
|||
# take screenshot
|
||||
if (not os.path.exists('{}/screenshot.png'.format(out_dir)) or overwrite) and not link['base_url'].endswith('.pdf'):
|
||||
print(' - Snapping Screenshot')
|
||||
CMD = 'google-chrome --headless --disable-gpu --screenshot'.split(' ')
|
||||
chrome_args = '--headless --disable-gpu --screenshot'.split(' ')
|
||||
try:
|
||||
run([*CMD, '--window-size={}'.format(RESOLUTION), link['url']], stdout=DEVNULL, stderr=DEVNULL, cwd=out_dir, timeout=20) # sreenshot.png
|
||||
run([CHROME_BINARY, *chrome_args, '--window-size={}'.format(RESOLUTION), link['url']], stdout=DEVNULL, stderr=DEVNULL, cwd=out_dir, timeout=20) # sreenshot.png
|
||||
except Exception as e:
|
||||
print(' Exception: {}'.format(e.__class__.__name__))
|
||||
else:
|
||||
print(' √ Skipping screenshot')
|
||||
|
||||
def archive_dot_org(out_dir, link, overwrite=False):
|
||||
# submit to archive.org
|
||||
if (not os.path.exists('{}/archive.org.txt'.format(out_dir)) or overwrite):
|
||||
print(' - Submitting to archive.org')
|
||||
submit_url = 'https://web.archive.org/save/{}'.format(link['base_url'])
|
||||
|
||||
success = False
|
||||
try:
|
||||
result = run(['curl', '-I', submit_url], stdout=PIPE, stderr=DEVNULL, cwd=out_dir, timeout=20) # archive.org
|
||||
headers = result.stdout.splitlines()
|
||||
content_location = [h for h in headers if b'Content-Location: ' in h]
|
||||
if content_location:
|
||||
archive_path = content_location[0].split(b'Content-Location: ', 1)[-1].decode('utf-8')
|
||||
saved_url = 'https://web.archive.org{}'.format(archive_path)
|
||||
success = True
|
||||
else:
|
||||
raise Exception('Failed to find Content-Location URL in Archive.org response headers.')
|
||||
except Exception as e:
|
||||
print(' Exception: {}'.format(e.__class__.__name__))
|
||||
|
||||
if success:
|
||||
with open('{}/archive.org.txt'.format(out_dir), 'w') as f:
|
||||
f.write(saved_url)
|
||||
|
||||
else:
|
||||
print(' √ Skipping archive.org')
|
||||
|
||||
def fetch_favicon(out_dir, link, overwrite=False):
|
||||
# download favicon
|
||||
if not os.path.exists('{}/favicon.ico'.format(out_dir)) or overwrite:
|
||||
|
|
@ -139,21 +192,62 @@ def fetch_favicon(out_dir, link, overwrite=False):
|
|||
print(' √ Skipping favicon')
|
||||
|
||||
|
||||
### ORCHESTRATION
|
||||
|
||||
def dump_index(links, service):
|
||||
with open(INDEX_TEMPLATE, 'r') as f:
|
||||
index_html = f.read()
|
||||
|
||||
link_html = """\
|
||||
<tr>
|
||||
<td>{time}</td>
|
||||
<td><a href="archive/{timestamp}/{base_url}" style="font-size:1.4em;text-decoration:none;color:black;" title="{title}">
|
||||
<img src="archive/{timestamp}/favicon.ico">
|
||||
{title} <small style="background-color: #eee;border-radius:4px; float:right">{tags}</small>
|
||||
</td>
|
||||
<td style="text-align:center"><a href="archive/{timestamp}/" title="Files">📂</a></td>
|
||||
<td style="text-align:center"><a href="{pdf_link}" title="PDF">📄</a></td>
|
||||
<td style="text-align:center"><a href="{screenshot_link}" title="Screenshot">🖼</a></td>
|
||||
<td style="text-align:center"><a href="https://web.archive.org/web/{base_url}" title="Archive.org">🏛</a></td>
|
||||
<td>🔗 <img src="https://www.google.com/s2/favicons?domain={domain}" height="16px"> <a href="{url}">{url}</a></td>
|
||||
</tr>"""
|
||||
|
||||
def get_template_vars(link):
|
||||
# since we dont screenshot or PDF links that are images or PDFs, change those links to point to the wget'ed file
|
||||
link_info = {**link}
|
||||
|
||||
if link['type']:
|
||||
link_info.update({'title': '{title} ({type})'.format(**link)})
|
||||
|
||||
if link['type'] in ('PDF', 'image'):
|
||||
link_info.update({
|
||||
'pdf_link': 'archive/{timestamp}/{base_url}'.format(**link),
|
||||
'screenshot_link': 'archive/{timestamp}/{base_url}'.format(**link),
|
||||
})
|
||||
else:
|
||||
link_info.update({
|
||||
'pdf_link': 'archive/{timestamp}/output.pdf'.format(**link),
|
||||
'screenshot_link': 'archive/{timestamp}/screenshot.png'.format(**link)
|
||||
})
|
||||
return link_info
|
||||
|
||||
with open(''.join((service, '/index.html')), 'w') as f:
|
||||
article_rows = '\n'.join(
|
||||
link_html.format(**get_template_vars(link)) for link in links
|
||||
)
|
||||
f.write(index_html.format(datetime.now().strftime('%Y-%m-%d %H:%M'), article_rows))
|
||||
|
||||
def dump_website(link, service, overwrite=False):
|
||||
"""download the DOM, PDF, and a screenshot into a folder named after the link's timestamp"""
|
||||
|
||||
print('[+] [{time}] Archiving "{title}": {url}'.format(**link))
|
||||
print('[+] [{time}] Archiving "{title}": {base_url}'.format(**link))
|
||||
|
||||
out_dir = ''.join((service, '/archive/{timestamp}')).format(**link)
|
||||
if not os.path.exists(out_dir):
|
||||
os.makedirs(out_dir)
|
||||
|
||||
if link['base_url'].endswith('.pdf'):
|
||||
print(' i PDF File')
|
||||
elif 'youtube.com' in link['domain']:
|
||||
print(' i Youtube Video')
|
||||
elif 'wikipedia.org' in link['domain']:
|
||||
print(' i Wikipedia Article')
|
||||
if link['type']:
|
||||
print(' i Type: {}'.format(link['type']))
|
||||
|
||||
if FETCH_WGET:
|
||||
fetch_wget(out_dir, link, overwrite=overwrite)
|
||||
|
|
@ -164,21 +258,22 @@ def dump_website(link, service, overwrite=False):
|
|||
if FETCH_SCREENSHOT:
|
||||
fetch_screenshot(out_dir, link, overwrite=overwrite)
|
||||
|
||||
if SUBMIT_ARCHIVE_DOT_ORG:
|
||||
archive_dot_org(out_dir, link, overwrite=overwrite)
|
||||
|
||||
if FETCH_FAVICON:
|
||||
fetch_favicon(out_dir, link, overwrite=overwrite)
|
||||
|
||||
run(['chmod', '-R', '755', out_dir], timeout=1)
|
||||
|
||||
def create_archive(service_file, service, resume=None):
|
||||
print('[+] [{}] Starting {} archive from {}'.format(datetime.now(), service, service_file))
|
||||
def create_archive(export_file, service, resume=None):
|
||||
print('[+] [{}] Starting {} archive from {}'.format(datetime.now().strftime('%Y-%m-%d %H:%M:%S'), service, export_file))
|
||||
|
||||
if not os.path.exists(service):
|
||||
os.makedirs(service)
|
||||
|
||||
if not os.path.exists(''.join((service,'/archive'))):
|
||||
os.makedirs(''.join((service,'/archive')))
|
||||
if not os.path.exists(''.join((service, '/archive'))):
|
||||
os.makedirs(''.join((service, '/archive')))
|
||||
|
||||
with open(service_file, 'r', encoding='utf-8') as f:
|
||||
with open(export_file, 'r', encoding='utf-8') as f:
|
||||
if service == "pocket":
|
||||
links = parse_pocket_export(f)
|
||||
elif service == "pinboard":
|
||||
|
|
@ -188,17 +283,17 @@ def create_archive(service_file, service, resume=None):
|
|||
links = [link for link in links if link['timestamp'] >= resume]
|
||||
|
||||
if not links:
|
||||
if service == "pocket":
|
||||
print('[X] No links found in {}, is it a getpocket.com/export export?'.format(service_file))
|
||||
elif service == "pinboard":
|
||||
print ('[X] No links found in {}, is it a pinboard.in/export/format:json/ export?'.format(service_file))
|
||||
if service == 'pocket':
|
||||
print('[X] No links found in {}, is it a getpocket.com/export export?'.format(export_file))
|
||||
elif service == 'pinboard':
|
||||
print('[X] No links found in {}, is it a pinboard.in/export/format:json/ export?'.format(export_file))
|
||||
raise SystemExit(1)
|
||||
|
||||
dump_index(links, service)
|
||||
|
||||
run(['chmod', '-R', '755', service], timeout=1)
|
||||
|
||||
print('[*] [{}] Created archive index.'.format(datetime.now()))
|
||||
print('[*] [{}] Created archive index.'.format(datetime.now().strftime('%Y-%m-%d %H:%M:%S')))
|
||||
|
||||
check_dependencies()
|
||||
|
||||
|
|
@ -208,11 +303,11 @@ def create_archive(service_file, service, resume=None):
|
|||
print('[√] [{}] Archive complete.'.format(datetime.now()))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
service_file = 'ril_export.html'
|
||||
argc = len(sys.argv)
|
||||
service_file = sys.argv[1] if argc > 1 else "ril_export.html" # path to export file
|
||||
service = sys.argv[2] if argc > 2 else "pocket" # select service for file format select
|
||||
resume = sys.argv[3] if argc > 3 else None # timestamp to resume dowloading from
|
||||
|
||||
create_archive(service_file, service, resume=resume)
|
||||
if __name__ == '__main__':
|
||||
argc = len(sys.argv)
|
||||
export_file = sys.argv[1] if argc > 1 else "ril_export.html" # path to export file
|
||||
export_type = sys.argv[2] if argc > 2 else "pocket" # select export_type for file format select
|
||||
resume_from = sys.argv[3] if argc > 3 else None # timestamp to resume dowloading from
|
||||
|
||||
create_archive(export_file, export_type, resume=resume_from)
|
||||
|
|
|
|||
|
|
@ -76,11 +76,12 @@
|
|||
<table style="width:100%;height: 90%; overflow-y: scroll;table-layout: fixed">
|
||||
<thead>
|
||||
<tr>
|
||||
<th style="width: 140px;"><img src="https://getpocket.com/favicon.ico" height="12px"> Pocketed Date</th>
|
||||
<th style="width: 120px;"><img src="https://getpocket.com/favicon.ico" height="12px"> Pocketed Date</th>
|
||||
<th style="width: 45vw;">Saved Article</th>
|
||||
<th style="width: 50px">Files</th>
|
||||
<th style="width: 50px">PDF</th>
|
||||
<th style="width: 80px">Screenshot</th>
|
||||
<th style="width: 60px;font-size:0.8em;">Screenshot</th>
|
||||
<th style="width: 50px">A.org</th>
|
||||
<th style="width: 100px;whitespace:nowrap;overflow-x:scroll;display:block">Original URL</th>
|
||||
</tr>
|
||||
</thead>
|
||||
|
|
|
|||
Loading…
Reference in a new issue