def validate_html(directory, args):
"""Validate all the HTML files in a directory."""
# pylint: disable=import-outside-toplevel
from beancount.utils import scrape # To avoid lxml dependency.
files, missing, empty = scrape.validate_local_links_in_dir(directory)
logging.info('%d files processed', len(files))
for target in missing:
logging.error('Missing %s', target)
for target in empty:
logging.error('Empty %s', target)
def do_validate_html(directory, args):
"""Validate all the HTML files under a directory hierarchy.
Args:
directory: A string, the root directory whose contents to validate.
args: A tuple of the rest of arguments.
"""
from beancount.utils import scrape
files, missing, empty = scrape.validate_local_links_in_dir(directory)
logging.info('%d files processed', len(files))
for target in missing:
logging.error('Missing %s', target)
for target in empty:
logging.error('Empty %s', target)
def main():
parser = version.ArgumentParser(description=__doc__)
web_group = web.add_web_arguments(parser)
web_group.set_defaults(port=9475)
group = parser.add_argument_group("Bake process arguments")
group.add_argument('output',
help=('The output directory or archive name. If you '
'specify a filename with a well-known extension,'
'we automatically archive the fetched directory '
'contents to this archive name and delete them.'))
# In order to be able to bake in a reasonable amount of time, we need to
# remove some pages; you can use this switch to do that.
group.add_argument('--render-all-pages',
'--full',
action='store_true',
help=("Don't ignore some of the more numerious pages, "
"like monthly reports."))
opts = parser.parse_args()
# Figure out the archival method.
output_directory, extension = file_utils.path_greedy_split(opts.output)
if extension:
try:
archival_command = ARCHIVERS[extension]
except KeyError as exc:
raise SystemExit(
"ERROR: Unknown archiver type '{}'".format(extension)) from exc
else:
archival_command = None
# Check pre-conditions on input/output filenames.
if not path.exists(opts.filename):
raise SystemExit("ERROR: Missing input file '{}'".format(
opts.filename))
if path.exists(opts.output):
raise SystemExit("ERROR: Output path already exists '{}'".format(
opts.output))
if path.exists(output_directory):
raise SystemExit("ERROR: Output directory already exists '{}'".format(
output_directory))
# Bake to a directory hierarchy of files with local links.
bake_to_directory(opts, output_directory, opts.render_all_pages)
# Verify the bake output files. This is just a sanity checking step.
# You can also use "bean-doctor validate_html <file> to run this manually.
logging.info('Validating HTML output files & links.')
files, missing, empty = scrape.validate_local_links_in_dir(
output_directory)
logging.info('Validation: %d files processed', len(files))
for target in missing:
logging.error("Validation error: Missing '%s'", target)
for target in empty:
logging.error("Validation error: Empty '%s'", target)
# Archive if requested.
if archival_command is not None:
# Normalize the paths and ensure sanity before we start compression.
output_directory = path.abspath(output_directory)
archive_filename = path.abspath(opts.output)
if not path.exists(output_directory):
raise IOError("Directory to archive '{}' does not exist".format(
output_directory))
if path.exists(archive_filename):
raise IOError("Output archive name '{}' already exists".format(
archive_filename))
# Dispatch to a particular compressor.
if isinstance(archival_command, str):
archive(archival_command, output_directory, archive_filename, True)
elif callable(archival_command):
archival_command(output_directory, archive_filename)
# Delete the output directory.
shutil.rmtree(output_directory)
print("Output in '{}'".format(opts.output))