def init(
self,
parsers=None,
repo=None,
params=None,
force=False,
skip_users=None,
from_resources=None,
):
"""Init one or more contributor parsers. Specifically, this is the
action that runs the parser.init() to generate some initial file.
"""
parsers = parsers or []
# Generate a shared repository object
repo = GitHubRepository(repo, skip_users)
# Get resource lookup ids (emails, orcids, logins, most won't be used for init)
resources = self.get_resource_lookups(from_resources, params)
for parser in parsers:
client = get_named_parser(name=parser, repo=repo, params=params)
client.cache = self.cache
client.init(force=force, from_resources=resources)
self.cache.update(client.cache)
# Save the cache
self.save_cache()
def get_resource_lookups(self, from_resources=None, params=None):
"""Based on a name (e.g., GitHub or special case tributors) return
as many lookups of unique ids (email, login, orcid) that
the resource provides
"""
from_resources = from_resources or ["github"]
lookups = {"login": set(), "orcid": set(), "email": set(), "name": set()}
for name in from_resources:
# Special case, tributors is just the entire cache
if name == "tributors":
lookups["login"].update(self.cache)
lookups["name"].update(
{x["name"]: x for _, x in self.cache.items() if "name" in x}
)
lookups["orcid"].update(
{x["orcid"]: x for _, x in self.cache.items() if "orcid" in x}
)
lookups["email"].update(
{x["email"]: x for _, x in self.cache.items() if "email" in x}
)
else:
parser = get_named_parser(name=name, params=params)
lookups["name"].update(parser.name_lookup)
lookups["login"].update(parser.login_lookup)
lookups["email"].update(parser.email_lookup)
lookups["orcid"].update(parser.orcid_lookup)
return lookups
def update(
self,
parsers=None,
repo=None,
params=None,
thresh=1,
skip_users=None,
from_resources=None,
):
"""Update one or more contributor parsers. Specifically, this is the
action that runs the parser.update() after obtaining contributions
from GitHub or a cache.
"""
parsers = parsers or []
# Generate a shared repository object
repo = GitHubRepository(repo, skip_users=skip_users)
bot.debug(f"Found repository {repo}")
# Get resource lookup ids (emails, orcids, logins)
resources = self.get_resource_lookups(from_resources, params)
# Update each metadata file via its parser
for parser in parsers:
client = get_named_parser(name=parser, repo=repo, params=params)
client.cache = self.cache
client.update(thresh=thresh, from_resources=resources)
self.cache.update(client.cache)
# Save the cache
self.save_cache()
def update_resource(self, resources=None, params=None, skip_users=None):
"""Given one or more resource types (an external file or source of
metadata) update the .tributors cache lookup
"""
resources = resources or []
for name in resources:
resource = get_named_parser(name=name, params=params)
resource.cache = self.cache
resource.update_lookup()
# Save the cache
self.save_cache()
def test_parser_allcontrib(tmp_path):
"""test each executor type with the filesystem
"""
from tributors.main.parsers import get_named_parser
# Prepare from resources
from_resources = {"login": set(["vsoch", "manbat", "yarikoptic"])}
# Create a dummy output file
allcontrib = os.path.join(str(tmp_path), ".all-contributorsrc")
params = {"--allcontrib-file": allcontrib}
repo = "singularityhub/sregistry"
# Output file should not exist before inint
assert not os.path.exists(allcontrib)
parser = get_named_parser("allcontrib", repo=repo, params=params)
result = parser.init()
# Trying to write existing file raises system exit
with pytest.raises(SystemExit):
result = parser.init()
assert os.path.exists(allcontrib)
# Ensure the result matches the template
template = {
"projectName": "sregistry",
"projectOwner": "singularityhub",
"repoType": "github",
"repoHost": "https://github.com",
"files": ["README.md"],
"imageSize": 100,
"commit": True,
"commitConvention": "none",
"contributors": [],
"contributorsPerLine": 7,
}
assert template == result
# Test adding contributors, default from GitHub
result = parser.update(from_resources=from_resources, save=False)
assert len(result["contributors"]) == 3
for login in [x["login"] for x in result["contributors"]]:
assert login in from_resources["login"]
# Test adding by email or orcid should not change output
byemail = parser.update(from_resources={"email": set(["*****@*****.**"])},
save=False)
byorcid = parser.update(
from_resources={"orcid": set(["0000-0000-0000-0000"])}, save=False)
assert byemail == byorcid == result
def init(self, parsers=None, repo=None, params=None, force=False):
"""Init one or more contributor parsers. Specifically, this is the
action that runs the parser.init() to generate some initial file.
"""
parsers = parsers or []
for parser in parsers:
client = get_named_parser(parser, repo)
client.init(params=params,
repo=repo,
force=force,
contributors=self.contributors)
# Update contributors caches
self.contributors.update(client.contributors)
self.cache.update(client.cache)
# Save the cache
self.save_cache()
def update(self, parsers=None, repo=None, params=None, thresh=1):
"""Update one or more contributor parsers. Specifically, this is the
action that runs the parser.update() after obtaining contributions
from GitHub or a cache.
"""
parsers = parsers or []
self.orcid_token = get_orcid_token()
for parser in parsers:
client = get_named_parser(parser, repo)
client.orcid_token = self.orcid_token
client.cache = self.cache
client.update(params=params,
repo=repo,
contributors=self.contributors,
thresh=thresh)
# Update contributors caches
self.contributors.update(client.contributors)
self.cache.update(client.cache)
# Save the cache
self.save_cache()
def test_parser_zenodo(tmp_path):
"""test each executor type with the filesystem
"""
from tributors.main.parsers import get_named_parser
# Prepare from resources
from_resources = {"login": set(["vsoch", "manbat", "yarikoptic"])}
# Create a dummy output file
zenodo = os.path.join(str(tmp_path), ".zenodo.json")
params = {"--zenodo-file": zenodo}
repo = "singularityhub/sregistry"
# Output file should not exist before init
assert not os.path.exists(zenodo)
parser = get_named_parser("zenodo", repo=repo, params=params)
result = parser.init(save=False)
assert len(result["creators"]) == 0
# With from resources, should add to creators
result = parser.init(from_resources=from_resources)
assert len(result["creators"]) == 3
# Trying to write existing file raises system exit
with pytest.raises(SystemExit):
result = parser.init()
assert os.path.exists(zenodo)
# Ensure the result matches the template
template = {
"creators": [{
"name": "yarikoptic"
}, {
"name": "manbat"
}, {
"name": "vsoch"
}],
"upload_type":
"software",
"keywords": [
"singularity",
"management",
"containers",
"registry",
"singularity-containers",
"singularityhub",
],
}
assert set(template["keywords"]) == set(result["keywords"])
assert template["upload_type"] == result["upload_type"]
assert len(result["creators"]) == 3
for login in [x["name"] for x in result["creators"]]:
assert login in from_resources["login"]
# Test adding contributors, default from GitHub
parser = get_named_parser("zenodo", repo=repo, params=params)
from_github = parser.update(from_resources=from_resources, save=False)
assert from_github == result
# Test adding by email or orcid should not change output
parser = get_named_parser("zenodo", repo=repo, params=params)
byemail = parser.update(from_resources={"email": set(["*****@*****.**"])},
save=False)
assert byemail == result
# Support adding by orcid, if not known
parser = get_named_parser("zenodo", repo=repo, params=params)
byorcid = parser.update(
from_resources={"orcid": set(["0000-0000-0000-0000"])}, save=False)
assert len(byorcid["creators"]) == 4