def get_results(codebase, as_list=False, **kwargs):
"""
Return an ordered mapping of scan results collected from a `codebase`.
if `as_list` consume the "files" iterator in a list sequence.
"""
codebase.add_files_count_to_current_header()
results = dict([('headers', codebase.get_headers()), ])
# add codebase toplevel attributes such as summaries
if codebase.attributes:
results.update(codebase.attributes.to_dict())
files = OutputPlugin.get_files(codebase, **kwargs)
if as_list:
files = list(files)
results['files'] = files
if TRACE:
logger_debug('get_results: files')
files = list(files)
from pprint import pformat
logger_debug(pformat(files))
return results
def write_results(codebase, output_file, pretty=False, **kwargs):
"""
Write headers, files, and other attributes from `codebase` to `output_file`
Enable JSON indentation if `pretty` is True
"""
# Set indentation for JSON output if `pretty` is True
# We use a separate dict for jsonstream kwargs since we are passing
# this function's kwargs as arguments to OutputPlugin.get_files()
if pretty:
jsonstreams_kwargs = dict(indent=2, pretty=True)
else:
jsonstreams_kwargs = dict(indent=None, pretty=False)
# If `output_file` is a path string, open the file at path `output_file` and use it as `output_file`
close_fd = False
if isinstance(output_file, str):
output_file = open(output_file, 'w')
close_fd = True
# Begin wri'w' JSON to `output_file`
with jsonstreams.Stream(
jsonstreams.Type.OBJECT,
fd=output_file,
close_fd=close_fd,
**jsonstreams_kwargs
) as s:
# Write headers
codebase.add_files_count_to_current_header()
codebase_headers = codebase.get_headers()
s.write('headers', codebase_headers)
# Write attributes
if codebase.attributes:
for attribute_key, attribute_value in codebase.attributes.to_dict().items():
s.write(attribute_key, attribute_value)
# Write files
codebase_files = OutputPlugin.get_files(codebase, **kwargs)
# OutputPlugin.get_files() returns a generator, not JSON-serializable
codebase_files = list(codebase_files)
s.write('files', codebase_files)
def build_copyright_paragraphs(codebase, **kwargs):
"""
Yield paragraphs built from the codebase.
The codebase is assumed to contains license and copyright detections.
"""
codebase.add_files_count_to_current_header()
header_para = CopyrightHeaderParagraph(
format=
'https://www.debian.org/doc/packaging-manuals/copyright-format/1.0/',
# TODO: add some details, but not all these
# comment=saneyaml.dump(codebase.get_headers()),
comment=notice,
)
yield header_para
# TODO: create CopyrightLicenseParagraph for common licenses
# TODO: group files that share copyright and license
# TODO: infer files patternsas in decopy
# for now this is dumb and will generate one paragraph per scanned file
for scanned_file in OutputPlugin.get_files(codebase, **kwargs):
if scanned_file['type'] == 'directory':
continue
dfiles = scanned_file['path']
dlicense = build_license(scanned_file)
dcopyright = build_copyright_field(scanned_file)
file_para = CopyrightFilesParagraph.from_dict(
dict(
files=dfiles,
license=dlicense,
copyright=dcopyright,
))
yield file_para