def get(self, key_path, filename, progress_cb):
# Note, that we need the path with hash dirs, since we don't have access
# to annexremote.dirhash from within IO classes
url = self.base_url + "/annex/objects/" + str(key_path)
from datalad.support.network import download_url
download_url(url, filename, overwrite=True)
def test_download_url(toppath, topurl):
furl = "%sfile.dat" % topurl
# fails if URL is dysfunctional
assert_raises(DownloadError, download_url, furl + 'magic', toppath)
# working download
tfpath = opj(toppath, "file-downloaded.dat")
download_url(furl, tfpath)
ok_file_has_content(tfpath, 'abc')
# fails if destfile exists
assert_raises(DownloadError, download_url, furl, tfpath)
# works when forced
download_url(furl, tfpath, overwrite=True)
def postclonecfg_ria(ds, props):
"""Configure a dataset freshly cloned from a RIA store"""
repo = ds.repo
# RIA uses hashdir mixed, copying data to it via git-annex (if cloned via
# ssh) would make it see a bare repo and establish a hashdir lower annex
# object tree.
# Moreover, we want the ORA remote to receive all data for the store, so its
# objects could be moved into archives (the main point of a RIA store).
RIA_REMOTE_NAME = 'origin' # don't hardcode everywhere
ds.config.set(
'remote.{}.annex-ignore'.format(RIA_REMOTE_NAME), 'true',
where='local')
# chances are that if this dataset came from a RIA store, its subdatasets
# may live there too. Place a subdataset source candidate config that makes
# get probe this RIA store when obtaining subdatasets
ds.config.set(
# we use the label 'origin' for this candidate in order to not have to
# generate a complicated name from the actual source specification.
# we pick a cost of 200 to sort it before datalad's default candidates
# for non-RIA URLs, because they prioritize hierarchical layouts that
# cannot be found in a RIA store
'datalad.get.subdataset-source-candidate-200origin',
# use the entire original URL, up to the fragment + plus dataset ID
# placeholder, this should make things work with any store setup we
# support (paths, ports, ...)
props['source'].split('#', maxsplit=1)[0] + '#{id}',
where='local')
# setup publication dependency, if a corresponding special remote exists
# and was enabled (there could be RIA stores that actually only have repos)
# make this function be a generator
ora_remotes = [s for s in ds.siblings('query', result_renderer='disabled')
if s.get('annex-externaltype') == 'ora']
if not ora_remotes and any(
r.get('externaltype') == 'ora'
for r in (repo.get_special_remotes().values()
if hasattr(repo, 'get_special_remotes')
else [])):
# no ORA remote autoenabled, but configuration known about at least one.
# Let's check origin's config for datalad.ora-remote.uuid as stored by
# create-sibling-ria and enable try enabling that one.
lgr.debug("Found no autoenabled ORA special remote. Trying to look it "
"up in source config ...")
# First figure whether we cloned via SSH, HTTP or local path and then
# get that config file the same way:
config_content = None
scheme = props['giturl'].split(':', 1)[0]
if scheme in ['http', 'https']:
try:
config_content = download_url(
"{}{}config".format(
props['giturl'],
'/' if not props['giturl'].endswith('/') else ''))
except DownloadError as e:
lgr.debug("Failed to get config file from source:\n%s",
exc_str(e))
elif scheme == 'ssh':
# TODO: switch the following to proper command abstraction:
# SSHRemoteIO ignores the path part ATM. No remote CWD! (To be
# changed with command abstractions). So we need to get that part to
# have a valid path to origin's config file:
cfg_path = PurePosixPath(URL(props['giturl']).path) / 'config'
op = SSHRemoteIO(props['giturl'])
try:
config_content = op.read_file(cfg_path)
except RIARemoteError as e:
lgr.debug("Failed to get config file from source: %s",
exc_str(e))
elif scheme == 'file':
# TODO: switch the following to proper command abstraction:
op = LocalIO()
cfg_path = Path(URL(props['giturl']).localpath) / 'config'
try:
config_content = op.read_file(cfg_path)
except (RIARemoteError, OSError) as e:
lgr.debug("Failed to get config file from source: %s",
exc_str(e))
else:
lgr.debug("Unknown URL-Scheme %s in %s. Can handle SSH, HTTP or "
"FILE scheme URLs.", scheme, props['source'])
# 3. And read it
org_uuid = None
if config_content:
# TODO: We might be able to spare the saving to a file.
# "git config -f -" is not explicitly documented but happens
# to work and would read from stdin. Make sure we know this
# works for required git versions and on all platforms.
with make_tempfile(content=config_content) as cfg_file:
runner = GitWitlessRunner()
try:
result = runner.run(
['git', 'config', '-f', cfg_file,
'datalad.ora-remote.uuid'],
protocol=StdOutCapture
)
org_uuid = result['stdout'].strip()
except CommandError as e:
# doesn't contain what we are looking for
lgr.debug("Found no UUID for ORA special remote at "
"'%s' (%s)", RIA_REMOTE_NAME, exc_str(e))
# Now, enable it. If annex-init didn't fail to enable it as stored, we
# wouldn't end up here, so enable with store URL as suggested by the URL
# we cloned from.
if org_uuid:
srs = repo.get_special_remotes()
if org_uuid in srs.keys():
# TODO: - Double-check autoenable value and only do this when
# true?
# - What if still fails? -> Annex shouldn't change config
# in that case
# we only need the store:
new_url = props['source'].split('#')[0]
try:
repo.enable_remote(srs[org_uuid]['name'],
options=['url={}'.format(new_url)]
)
lgr.info("Reconfigured %s for %s",
srs[org_uuid]['name'], new_url)
# update ora_remotes for considering publication dependency
# below
ora_remotes = [s for s in
ds.siblings('query',
result_renderer='disabled')
if s.get('annex-externaltype', None) ==
'ora']
except CommandError as e:
lgr.debug("Failed to reconfigure ORA special remote: %s",
exc_str(e))
else:
lgr.debug("Unknown ORA special remote uuid at '%s': %s",
RIA_REMOTE_NAME, org_uuid)
if ora_remotes:
if len(ora_remotes) == 1:
yield from ds.siblings('configure',
name=RIA_REMOTE_NAME,
publish_depends=ora_remotes[0]['name'],
result_filter=None,
result_renderer='disabled')
else:
lgr.warning("Found multiple ORA remotes. Couldn't decide which "
"publishing to 'origin' should depend on: %s. Consider "
"running 'datalad siblings configure -s origin "
"--publish-depends ORAREMOTENAME' to set publication "
"dependency manually.",
[r['name'] for r in ora_remotes])
#!/usr/bin/env python3
import os.path as op
import sys
import xml.dom.minidom
import datalad.api as dl
from datalad.support.network import download_url
ds = dl.Dataset(op.dirname(op.dirname(op.realpath(__file__))))
if 'datalad' not in ds.repo.get_remotes():
from datalad.customremotes.base import init_datalad_remote
init_datalad_remote(ds.repo, 'datalad', autoenable=True)
# doc = xml.dom.minidom.parse('/tmp/outi-7T.xml')
topurl = 'https://db.humanconnectome.org/data/archive/projects/HCP_Resources/resources/7T_Movies/'
doc = xml.dom.minidom.parseString(download_url(topurl))
files = [{f: e.getAttribute(f)
for f in ('ID', 'URI', 'digest', 'name')}
for e in doc.getElementsByTagName("cat:entry")]
# from pprint import pprint
# pprint(files)
added = list(
ds.addurls(files, topurl + 'files/{URI}', '{URI}', fast=False, save=False))
print(f"Processed {len(added)} entries")