def main():
""" Main function to start the confluence-dumper. """
# Configure console for unicode output via stdout/stderr
sys.stdout = codecs.getwriter('utf-8')(sys.stdout)
sys.stderr = codecs.getwriter('utf-8')(sys.stderr)
# Welcome output
print_welcome_output()
# Delete old export
if os.path.exists(settings.EXPORT_FOLDER):
shutil.rmtree(settings.EXPORT_FOLDER)
os.makedirs(settings.EXPORT_FOLDER)
# Read HTML template
template_file = open(settings.TEMPLATE_FILE)
html_template = template_file.read()
# Fetch all spaces if spaces were not configured via settings
if len(settings.SPACES_TO_EXPORT) > 0:
spaces_to_export = settings.SPACES_TO_EXPORT
else:
spaces_to_export = []
page_url = '%s/rest/api/space?limit=25' % settings.CONFLUENCE_BASE_URL
while page_url:
response = utils.http_get(
page_url,
auth=settings.HTTP_AUTHENTICATION,
headers=settings.HTTP_CUSTOM_HEADERS,
verify_peer_certificate=settings.VERIFY_PEER_CERTIFICATE,
proxies=settings.HTTP_PROXIES)
for space in response['results']:
spaces_to_export.append(space['key'])
if 'next' in response['_links'].keys():
page_url = response['_links']['next']
page_url = '%s%s' % (settings.CONFLUENCE_BASE_URL, page_url)
else:
page_url = None
print('Exporting %d space(s): %s\n' %
(len(spaces_to_export), ', '.join(spaces_to_export)))
# Export spaces
space_counter = 0
duplicate_space_names = {}
space_matching = {}
for space in spaces_to_export:
space_counter += 1
# Create folders for this space
space_folder_name = provide_unique_file_name(duplicate_space_names,
space_matching,
space,
is_folder=True)
space_folder = '%s/%s' % (settings.EXPORT_FOLDER, space_folder_name)
try:
os.makedirs(space_folder)
download_folder = '%s/%s' % (space_folder,
settings.DOWNLOAD_SUB_FOLDER)
os.makedirs(download_folder)
space_url = '%s/rest/api/space/%s?expand=homepage' % (
settings.CONFLUENCE_BASE_URL, space)
response = utils.http_get(
space_url,
auth=settings.HTTP_AUTHENTICATION,
headers=settings.HTTP_CUSTOM_HEADERS,
verify_peer_certificate=settings.VERIFY_PEER_CERTIFICATE,
proxies=settings.HTTP_PROXIES)
space_name = response['name']
print('SPACE (%d/%d): %s (%s)' %
(space_counter, len(spaces_to_export), space_name, space))
space_page_id = settings.from_page_id or response['homepage']['id']
path_collection = fetch_page_recursively(space_page_id,
space_folder,
download_folder,
html_template)
if path_collection:
# Create index file for this space
space_index_path = '%s/index.html' % space_folder
space_index_title = 'Index of Space %s (%s)' % (space_name,
space)
space_index_content = create_html_index(path_collection)
utils.write_html_2_file(space_index_path, space_index_title,
space_index_content, html_template)
except utils.ConfluenceException as e:
error_print('ERROR: %s' % e)
except OSError:
print(
'WARNING: The space %s has been exported already. Maybe you mentioned it twice in the settings'
% space)
# Finished output
print_finished_output()
def fetch_page_recursively(page_id,
folder_path,
download_folder,
html_template,
depth=0,
page_duplicate_file_names=None,
page_file_matching=None,
attachment_duplicate_file_names=None,
attachment_file_matching=None):
""" Fetches a Confluence page and its child pages (with referenced downloads).
:param page_id: Confluence page id.
:param folder_path: Folder to place downloaded pages in.
:param download_folder: Folder to place downloaded files in.
:param html_template: HTML template used to export Confluence pages.
:param depth: (optional) Hierarchy depth of the handled Confluence page.
:param page_duplicate_file_names: A dict in the structure {'<sanitized page filename>': amount of duplicates}
:param page_file_matching: A dict in the structure {'<page title>': '<used offline filename>'}
:param attachment_duplicate_file_names: A dict in the structure {'<sanitized attachment filename>': amount of \
duplicates}
:param attachment_file_matching: A dict in the structure {'<attachment title>': '<used offline filename>'}
:returns: Information about downloaded files (pages, attachments, images, ...) as a dict (None for exceptions)
"""
if not page_duplicate_file_names:
page_duplicate_file_names = {}
if not page_file_matching:
page_file_matching = {}
if not attachment_duplicate_file_names:
attachment_duplicate_file_names = {}
if not attachment_file_matching:
attachment_file_matching = {}
page_url = '%s/rest/api/content/%s?expand=children.page,children.attachment,body.view.value' \
% (settings.CONFLUENCE_BASE_URL, page_id)
try:
response = utils.http_get(
page_url,
auth=settings.HTTP_AUTHENTICATION,
headers=settings.HTTP_CUSTOM_HEADERS,
verify_peer_certificate=settings.VERIFY_PEER_CERTIFICATE,
proxies=settings.HTTP_PROXIES)
page_content = response['body']['view']['value']
page_title = response['title']
print('%sPAGE: %s (%s)' % ('\t' * (depth + 1), page_title, page_id))
# Construct unique file name
file_name = provide_unique_file_name(page_duplicate_file_names,
page_file_matching,
str(page_id),
explicit_file_extension='html')
# Remember this file and all children
path_collection = {
'file_path': file_name,
'page_title': page_title,
'child_pages': [],
'child_attachments': []
}
# Download attachments of this page
# TODO: Outsource/Abstract the following two while loops because of much duplicate code.
page_url = '%s/rest/api/content/%s/child/attachment?limit=25' % (
settings.CONFLUENCE_BASE_URL, page_id)
counter = 0
while page_url:
response = utils.http_get(
page_url,
auth=settings.HTTP_AUTHENTICATION,
headers=settings.HTTP_CUSTOM_HEADERS,
verify_peer_certificate=settings.VERIFY_PEER_CERTIFICATE,
proxies=settings.HTTP_PROXIES)
counter += len(response['results'])
for attachment in response['results']:
download_url = attachment['_links']['download']
attachment_id = attachment['id'][3:]
attachment_info = download_attachment(
download_url,
download_folder,
attachment_id,
attachment_duplicate_file_names,
attachment_file_matching,
depth=depth + 1)
path_collection['child_attachments'].append(attachment_info)
if 'next' in response['_links'].keys():
page_url = response['_links']['next']
page_url = '%s%s' % (settings.CONFLUENCE_BASE_URL, page_url)
else:
page_url = None
# Export HTML file
page_content = handle_html_references(page_content,
page_duplicate_file_names,
page_file_matching,
depth=depth + 1)
file_path = '%s/%s' % (folder_path, file_name)
page_content += create_html_attachment_index(
path_collection['child_attachments'])
utils.write_html_2_file(file_path, page_title, page_content,
html_template)
# Save another file with page id which forwards to the original one
id_file_path = '%s/%s.html' % (folder_path, page_id)
id_file_page_title = 'Forward to page %s' % page_title
original_file_link = utils.encode_url(
utils.sanitize_for_filename(file_name))
# id_file_page_content = settings.HTML_FORWARD_MESSAGE % (original_file_link, page_title)
# id_file_forward_header = '<meta http-equiv="refresh" content="0; url=%s" />' % original_file_link
# utils.write_html_2_file(id_file_path, id_file_page_title, id_file_page_content, html_template,
# additional_headers=[id_file_forward_header])
# Iterate through all child pages
page_url = '%s/rest/api/content/%s/child/page?limit=25' % (
settings.CONFLUENCE_BASE_URL, page_id)
counter = 0
while page_url:
response = utils.http_get(
page_url,
auth=settings.HTTP_AUTHENTICATION,
headers=settings.HTTP_CUSTOM_HEADERS,
verify_peer_certificate=settings.VERIFY_PEER_CERTIFICATE,
proxies=settings.HTTP_PROXIES)
counter += len(response['results'])
for child_page in response['results']:
paths = fetch_page_recursively(
child_page['id'],
folder_path,
download_folder,
html_template,
depth=depth + 1,
page_duplicate_file_names=page_duplicate_file_names,
page_file_matching=page_file_matching)
if paths:
path_collection['child_pages'].append(paths)
if 'next' in response['_links'].keys():
page_url = response['_links']['next']
page_url = '%s%s' % (settings.CONFLUENCE_BASE_URL, page_url)
else:
page_url = None
return path_collection
except utils.ConfluenceException as e:
error_print('%sERROR: %s' % ('\t' * (depth + 1), e))
return None
