#!/usr/bin/env python3
import logging
import os
import urllib.parse
import urllib.request
import tarfile
from tooldog import TMP_DIR
from .utils import *
LOGGER = logging.getLogger(__name__)
[docs]class CodeCollector(object):
"""
Class to download source code from a https://bio.tools entry
"""
ZIP_NAME = "tool.zip"
TAR_NAME = "tool.tar"
TMP_NAME = "tmp"
[docs] def __init__(self, biotool):
"""
:param biotool: Biotool object
:type biotool: :class:`tooldog.biotool_model.Biotool`
"""
self.biotool = biotool
def _make_tar(self, file_path, tarname):
with tarfile.open(tarname, mode='w') as archive:
archive.add(file_path, arcname=self.ZIP_NAME)
def _get_from_repository(self, url):
"""
Get source code from a repository link
:param url: url of the repository
:type url: STRING
"""
# Here we deal with repository, have to use regex to test the url and
# use appropriate strategy to get the code depending the type of repository
if "github.com" in url:
return self._get_from_github(url)
else:
LOGGER.warn("The url ({}) is not a Github url".format(url))
LOGGER.warn("ToolDog only deals with Github repository for the moment...")
def _get_from_github(self, url):
try:
zip_url = os.path.join(url, "archive/master.zip")
response = urllib.request.urlopen(zip_url)
data = response.read()
LOGGER.info('Writing data to zip file...')
zip_path = os.path.join(TMP_DIR, self.ZIP_NAME)
tar_path = os.path.join(TMP_DIR, self.TAR_NAME)
write_to_file(zip_path, data, 'wb')
LOGGER.info('Making tar...')
self._make_tar(zip_path, tar_path)
return tar_path
except:
LOGGER.warn('Something went wrong with the following Github repository: {}'.format(zip_url))
def _get_from_source_code(self, url):
"""
Get source code from a source code link
:param url: url of the source code
:type url: STRING
"""
return None
[docs] def get_source(self):
"""
Retrieve source code of the tool using links provided in https://bio.tools
"""
source_code = None
links = self.biotool.informations.links
for link in links:
link_type = link.type.lower().translate(str.maketrans(' ', '_'))
try:
source_code = getattr(self, '_get_from_{}'.format(link_type))(link.url)
except AttributeError:
LOGGER.warn(link_type + ' link type is not processed yet by ToolDog.')
if source_code is not None:
# For the moment, consider that if a source code has been found,
# we just leave the loop.
break
return source_code
