def test_balsa_pipeline1(ind, topurl, outd, clonedir):
list(initiate_dataset(
template="balsa",
dataset_name='dataladtest-WG33',
path=outd,
data_fields=['dataset_id'])({'dataset_id': 'WG33'}))
with chpwd(outd):
pipeline = ofpipeline('WG33', url=topurl)
out = run_pipeline(pipeline)
eq_(len(out), 1)
repo = AnnexRepo(outd, create=False) # to be used in the checks
# Inspect the tree -- that we have all the branches
branches = {'master', 'incoming', 'incoming-processed', 'git-annex'}
eq_(set(repo.get_branches()), branches)
# since now we base incoming on master -- and there were nothing custom
# in master after incoming-processed, both branches should be the same
eq_(repo.get_hexsha('master'), repo.get_hexsha('incoming-processed'))
# but that one is different from incoming
assert_not_equal(repo.get_hexsha('incoming'), repo.get_hexsha('incoming-processed'))
commits = {b: list(repo.get_branch_commits(b)) for b in branches}
# all commits out there -- init ds + init crawler + 1*(incoming, processed)
# The number of commits in master differs based on the create variant used
# (the one DataLad's master makes only one commit).
ncommits_master = len(commits["master"])
assert_in(ncommits_master, [4, 5])
# incoming branches from master but lacks one merge commit.
eq_(len(commits['incoming']), ncommits_master - 1)
# incoming-processed is on master.
eq_(len(commits['incoming-processed']), ncommits_master)
with chpwd(outd):
eq_(set(glob('*')), {'dir1', 'file1.nii'})
all_files = sorted(find_files('.'))
fpath = opj(outd, 'file1.nii')
ok_file_has_content(fpath, "content of file1.nii")
ok_file_under_git(fpath, annexed=True)
fpath2 = opj(outd, 'dir1', 'file2.nii')
ok_file_has_content(fpath2, "content of file2.nii")
ok_file_under_git(fpath2, annexed=True)
target_files = {
'./.datalad/crawl/crawl.cfg',
'./.datalad/crawl/statuses/incoming.json',
'./.datalad/meta/balsa.json',
'./.datalad/config',
'./file1.nii', './dir1/file2.nii',
}
eq_(set(all_files), target_files)
def test_balsa_pipeline1(ind, topurl, outd, clonedir):
list(
initiate_dataset(template="balsa",
dataset_name='dataladtest-WG33',
path=outd,
data_fields=['dataset_id'])({
'dataset_id': 'WG33'
}))
with chpwd(outd):
pipeline = ofpipeline('WG33', url=topurl)
out = run_pipeline(pipeline)
eq_(len(out), 1)
repo = AnnexRepo(outd, create=False) # to be used in the checks
# Inspect the tree -- that we have all the branches
branches = {'master', 'incoming', 'incoming-processed', 'git-annex'}
eq_(set(repo.get_branches()), branches)
assert_not_equal(repo.get_hexsha('master'),
repo.get_hexsha('incoming-processed'))
# and that one is different from incoming
assert_not_equal(repo.get_hexsha('incoming'),
repo.get_hexsha('incoming-processed'))
commits = {b: list(repo.get_branch_commits(b)) for b in branches}
eq_(len(commits['incoming']), 1)
eq_(len(commits['incoming-processed']), 2)
eq_(
len(commits['master']), 6
) # all commits out there -- init ds + init crawler + 1*(incoming, processed, merge)
with chpwd(outd):
eq_(set(glob('*')), {'dir1', 'file1.nii'})
all_files = sorted(find_files('.'))
fpath = opj(outd, 'file1.nii')
ok_file_has_content(fpath, "content of file1.nii")
ok_file_under_git(fpath, annexed=True)
fpath2 = opj(outd, 'dir1', 'file2.nii')
ok_file_has_content(fpath2, "content of file2.nii")
ok_file_under_git(fpath2, annexed=True)
target_files = {
'./.datalad/crawl/crawl.cfg',
'./.datalad/crawl/statuses/incoming.json',
'./.datalad/meta/balsa.json',
'./.datalad/config',
'./file1.nii',
'./dir1/file2.nii',
}
eq_(set(all_files), target_files)
def test_publish_recursive(origin, src_path, dst_path, sub1_pub, sub2_pub):
# prepare src
source = install(src_path, source=origin, recursive=True)[0]
# create plain git at target:
target = GitRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# subdatasets have no remote yet, so recursive publishing should fail:
with assert_raises(ValueError) as cm:
publish(dataset=source, to="target", recursive=True)
assert_in("No sibling 'target' found.", str(cm.exception))
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, 'subm 2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# publish recursively
with swallow_logs(new_level=logging.DEBUG) as cml:
res = publish(dataset=source, to="target", recursive=True)
assert_not_in('forced update', cml.out,
"we probably haven't merged git-annex before pushing")
# testing result list
# (Note: Dataset lacks __eq__ for now. Should this be based on path only?)
assert_is_instance(res, tuple)
assert_is_instance(res[0], list)
assert_is_instance(res[1], list)
eq_(res[1], []) # nothing failed/was skipped
for item in res[0]:
assert_is_instance(item, Dataset)
eq_({res[0][0].path, res[0][1].path, res[0][2].path},
{src_path, sub1.path, sub2.path})
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
eq_(list(target.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
eq_(list(sub1_target.get_branch_commits("master")),
list(sub1.get_branch_commits("master")))
eq_(list(sub1_target.get_branch_commits("git-annex")),
list(sub1.get_branch_commits("git-annex")))
eq_(list(sub2_target.get_branch_commits("master")),
list(sub2.get_branch_commits("master")))
eq_(list(sub2_target.get_branch_commits("git-annex")),
list(sub2.get_branch_commits("git-annex")))
# test for publishing with --since. By default since no changes, only current pushed
res_ = publish(dataset=source, recursive=True)
# only current one would get pushed
eq_(set(r.path for r in res_[0]), {src_path})
# all get pushed
res_ = publish(dataset=source, recursive=True, since='HEAD^')
eq_(set(r.path for r in res_[0]), {src_path, sub1.path, sub2.path})
# Let's now update one subm
with open(opj(sub2.path, "file.txt"), 'w') as f:
f.write('')
sub2.add('file.txt')
sub2.commit("")
# TODO: Doesn't work: https://github.com/datalad/datalad/issues/636
#source.save("changed sub2", auto_add_changes=True)
source.repo.commit("", options=['-a'])
res_ = publish(dataset=source, recursive=True)
# only updated ones were published
eq_(set(r.path for r in res_[0]), {src_path, sub2.path})
def test_publish_with_data(origin, src_path, dst_path, sub1_pub, sub2_pub):
# prepare src
source = install(src_path, source=origin, recursive=True)[0]
source.repo.get('test-annex.dat')
# create plain git at target:
target = AnnexRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = GitRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, 'subm 2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# TMP: Insert the fetch to prevent GitPython to fail after the push,
# because it cannot resolve the SHA of the old commit of the remote,
# that git reports back after the push.
# TODO: Figure out, when to fetch things in general; Alternatively:
# Is there an option for push, that prevents GitPython from failing?
source.repo.fetch("target")
res = publish(dataset=source, to="target", path=['test-annex.dat'])
eq_(res, ([source, 'test-annex.dat'], []))
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
# TODO: last commit in git-annex branch differs. Probably fine,
# but figure out, when exactly to expect this for proper testing:
eq_(
list(target.get_branch_commits("git-annex"))[1:],
list(source.repo.get_branch_commits("git-annex"))[1:])
# we need compare target/master:
target.checkout("master")
eq_(target.file_has_content(['test-annex.dat']), [True])
source.repo.fetch("target")
res = publish(dataset=source, to="target", path=['.'])
eq_(res, ([source, 'test-annex.dat'], []))
source.repo.fetch("target")
import glob
res = publish(dataset=source,
to="target",
path=glob.glob1(source.path, '*'))
# Note: This leads to recursive publishing, since expansion of '*'
# contains the submodules themselves in this setup
# collect result paths:
result_paths = []
for item in res[0]:
if isinstance(item, Dataset):
result_paths.append(item.path)
else:
result_paths.append(item)
eq_(
{
source.path,
opj(source.path, "subm 1"),
opj(source.path, "subm 2"), 'test-annex.dat'
}, set(result_paths))
def test_publish_with_data(origin, src_path, dst_path, sub1_pub, sub2_pub, dst_clone_path):
# prepare src
source = install(src_path, source=origin, recursive=True)
source.repo.get('test-annex.dat')
# create plain git at target:
target = AnnexRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = GitRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, 'subm 2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# TMP: Insert the fetch to prevent GitPython to fail after the push,
# because it cannot resolve the SHA of the old commit of the remote,
# that git reports back after the push.
# TODO: Figure out, when to fetch things in general; Alternatively:
# Is there an option for push, that prevents GitPython from failing?
source.repo.fetch("target")
res = publish(dataset=source, to="target", path=['test-annex.dat'])
# first it would publish data and then push
eq_(res, (['test-annex.dat', source], []))
# XXX master was not checked out in dst!
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
# TODO: last commit in git-annex branch differs. Probably fine,
# but figure out, when exactly to expect this for proper testing:
# yoh: they differ because local annex records information about now
# file being available in that remote, and remote one does it via a call in
# the hook I guess. So they both get the same information but in two
# different commits. I do not observe such behavior of remote having git-annex
# automagically updated in older clones
# which do not have post-receive hook on remote side
eq_(list(target.get_branch_commits("git-annex"))[1:],
list(source.repo.get_branch_commits("git-annex"))[1:])
# we need compare target/master:
target.checkout("master")
ok_(target.file_has_content('test-annex.dat'))
# make sure that whatever we published is actually consumable
dst_clone = install(
dst_clone_path, source=dst_path,
result_xfm='datasets', return_type='item-or-list')
nok_(dst_clone.repo.file_has_content('test-annex.dat'))
res = dst_clone.get('test-annex.dat')
ok_(dst_clone.repo.file_has_content('test-annex.dat'))
source.repo.fetch("target")
res = publish(dataset=source, to="target", path=['.'])
# there is nothing to publish on 2nd attempt
#eq_(res, ([source, 'test-annex.dat'], []))
eq_(res, ([], []))
source.repo.fetch("target")
import glob
res = publish(dataset=source, to="target", path=glob.glob1(source.path, '*'))
# Note: This leads to recursive publishing, since expansion of '*'
# contains the submodules themselves in this setup
# collect result paths:
result_paths = []
for item in res[0]:
result_paths.append(item.path if isinstance(item, Dataset) else item)
# only the subdatasets, targets are plain git repos, hence
# no file content is pushed, all content in super was pushed
# before
eq_({sub1.path, sub2.path},
set(result_paths))
# if we publish again -- nothing to be published
eq_(source.publish(to="target"), ([], []))
# if we drop a file and publish again -- dataset should be published
# since git-annex branch was updated
source.drop('test-annex.dat')
eq_(source.publish(to="target"), ([source], []))
eq_(source.publish(to="target"), ([], [])) # and empty again if we try again
def test_publish_recursive(pristine_origin, origin_path, src_path, dst_path, sub1_pub, sub2_pub):
# we will be publishing back to origin, so to not alter testrepo
# we will first clone it
origin = install(origin_path, source=pristine_origin, recursive=True)
# prepare src
source = install(src_path, source=origin_path, recursive=True)
# create plain git at target:
target = GitRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# subdatasets have no remote yet, so recursive publishing should fail:
with assert_raises(ValueError) as cm:
publish(dataset=source, to="target", recursive=True)
assert_in("Unknown target sibling 'target'", exc_str(cm.exception))
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
# we will be testing presence of the file content, so let's make it progress
sub2_target.config.set('receive.denyCurrentBranch', 'updateInstead', where='local')
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, 'subm 2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# publish recursively
with swallow_logs(new_level=logging.DEBUG) as cml:
res = publish(dataset=source, to="target", recursive=True)
assert_not_in(
'forced update', cml.out,
"we probably haven't merged git-annex before pushing"
)
# testing result list
# (Note: Dataset lacks __eq__ for now. Should this be based on path only?)
assert_is_instance(res, tuple)
assert_is_instance(res[0], list)
assert_is_instance(res[1], list)
eq_(res[1], []) # nothing failed/was skipped
for item in res[0]:
assert_is_instance(item, Dataset)
eq_({res[0][0].path, res[0][1].path, res[0][2].path},
{src_path, sub1.path, sub2.path})
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
eq_(list(target.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
eq_(list(sub1_target.get_branch_commits("master")),
list(sub1.get_branch_commits("master")))
eq_(list(sub1_target.get_branch_commits("git-annex")),
list(sub1.get_branch_commits("git-annex")))
eq_(list(sub2_target.get_branch_commits("master")),
list(sub2.get_branch_commits("master")))
eq_(list(sub2_target.get_branch_commits("git-annex")),
list(sub2.get_branch_commits("git-annex")))
# we are tracking origin but origin has different git-annex, since we
# cloned from it, so it is not aware of our git-annex
neq_(list(origin.repo.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
# So if we first publish to it recursively, we would update
# all sub-datasets since git-annex branch would need to be pushed
res_ = publish(dataset=source, recursive=True)
eq_(set(r.path for r in res_[0]),
set(opj(*([source.path] + x)) for x in ([], ['subm 1'], ['subm 2'])))
# and now should carry the same state for git-annex
eq_(list(origin.repo.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
# test for publishing with --since. By default since no changes, nothing pushed
res_ = publish(dataset=source, recursive=True)
eq_(set(r.path for r in res_[0]), set())
# still nothing gets pushed, because origin is up to date
res_ = publish(dataset=source, recursive=True, since='HEAD^')
eq_(set(r.path for r in res_[0]), set([]))
# and we should not fail if we run it from within the dataset
with chpwd(source.path):
res_ = publish(recursive=True, since='HEAD^')
eq_(set(r.path for r in res_[0]), set([]))
# Let's now update one subm
with open(opj(sub2.path, "file.txt"), 'w') as f:
f.write('')
# add to subdataset, does not alter super dataset!
# MIH: use `to_git` because original test author used
# and explicit `GitRepo.add` -- keeping this for now
Dataset(sub2.path).add('file.txt', to_git=True)
# Let's now update one subm
create_tree(sub2.path, {'file.dat': 'content'})
# add to subdataset, without reflecting the change in its super(s)
Dataset(sub2.path).add('file.dat')
# note: will publish to origin here since that is what it tracks
res_published, res_skipped = publish(dataset=source, recursive=True)
# only updates published, i.e. just the subdataset, super wasn't altered
# nothing copied!
eq_(res_published, [Dataset(sub2.path)])
eq_(res_skipped, [])
# since published to origin -- destination should not get that file
nok_(lexists(opj(sub2_target.path, 'file.dat')))
res_published, res_skipped = publish(dataset=source, to='target', recursive=True)
eq_(res_published, [Dataset(sub2.path)])
# Note: with updateInstead only in target2 and not saving change in
# super-dataset we would have made remote dataset, if we had entire
# hierarchy, to be somewhat inconsistent.
# But here, since target datasets are independent -- it is ok
# and the file itself was not transferred but now exists
ok_(lexists(opj(sub2_target.path, 'file.dat')))
nok_(sub2_target.file_has_content('file.dat'))
# but now we can redo publish recursively, at least stating to consider
# explicitly to copy .
res_published, res_skipped = publish(
'.',
dataset=source, to='target',
recursive=True
)
ok_(sub2_target.file_has_content('file.dat'))
eq_(res_published, ['file.dat']) # note that this report makes little sense without path to the repository
def test_publish_with_data(origin, src_path, dst_path, sub1_pub, sub2_pub,
dst_clone_path):
# prepare src
source = install(src_path, source=origin, recursive=True)
source.repo.get('test-annex.dat')
# create plain git at target:
target = AnnexRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# now, set up targets for the submodules:
# the need to be annexes, because we want to be able to copy data to them
# further down
sub1_target = AnnexRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, '2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
res = publish(dataset=source,
to="target",
path=['test-annex.dat'],
result_xfm='paths')
# first it would publish data and then push
# TODO order is not fixed (yet)
#eq_(res, [opj(source.path, 'test-annex.dat'), source.path])
eq_(set(res), set([opj(source.path, 'test-annex.dat'), source.path]))
# XXX master was not checked out in dst!
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
# TODO: last commit in git-annex branch differs. Probably fine,
# but figure out, when exactly to expect this for proper testing:
# yoh: they differ because local annex records information about now
# file being available in that remote, and remote one does it via a call in
# the hook I guess. So they both get the same information but in two
# different commits. I do not observe such behavior of remote having git-annex
# automagically updated in older clones
# which do not have post-receive hook on remote side
eq_(
list(target.get_branch_commits("git-annex"))[1:],
list(source.repo.get_branch_commits("git-annex"))[1:])
# we need compare target/master:
target.checkout("master")
ok_(target.file_has_content('test-annex.dat'))
# make sure that whatever we published is actually consumable
dst_clone = install(dst_clone_path,
source=dst_path,
result_xfm='datasets',
return_type='item-or-list')
nok_(dst_clone.repo.file_has_content('test-annex.dat'))
res = dst_clone.get('test-annex.dat')
ok_(dst_clone.repo.file_has_content('test-annex.dat'))
res = publish(dataset=source, to="target", path=['.'])
# there is nothing to publish on 2nd attempt
#eq_(res, ([source, 'test-annex.dat'], []))
assert_result_count(res, 1, status='notneeded')
import glob
res = publish(dataset=source,
to="target",
path=glob.glob1(source.path, '*'))
# Note: This leads to recursive publishing, since expansion of '*'
# contains the submodules themselves in this setup
# only the subdatasets, targets are plain git repos, hence
# no file content is pushed, all content in super was pushed
# before
assert_result_count(res, 3)
assert_result_count(res, 1, status='ok', path=sub1.path)
assert_result_count(res, 1, status='ok', path=sub2.path)
assert_result_count(res, 1, status='notneeded', path=source.path)
# if we publish again -- nothing to be published
res = source.publish(to="target")
assert_result_count(res, 1, status='notneeded', path=source.path)
# if we drop a file and publish again -- dataset should be published
# since git-annex branch was updated
source.drop('test-annex.dat')
res = source.publish(to="target")
assert_result_count(res, 1, status='ok', path=source.path)
# and empty again if we try again
res = source.publish(to="target")
assert_result_count(res, 1, status='notneeded', path=source.path)
def test_publish_recursive(pristine_origin, origin_path, src_path, dst_path,
sub1_pub, sub2_pub):
# we will be publishing back to origin, so to not alter testrepo
# we will first clone it
origin = install(origin_path, source=pristine_origin, recursive=True)
# prepare src
source = install(src_path, source=origin.path, recursive=True)
# we will be trying to push into this later on, need to give permissions...
origin_sub2 = Dataset(opj(origin_path, '2'))
origin_sub2.config.set('receive.denyCurrentBranch',
'updateInstead',
where='local')
## TODO this manual fixup is needed due to gh-1548 -- needs proper solution
#os.remove(opj(origin_sub2.path, '.git'))
#os.rename(opj(origin_path, '.git', 'modules', '2'), opj(origin_sub2.path, '.git'))
# create plain git at target:
target = GitRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# subdatasets have no remote yet, so recursive publishing should fail:
res = publish(dataset=source,
to="target",
recursive=True,
on_failure='ignore')
assert_result_count(res, 3)
assert_result_count(res, 1, status='ok', type='dataset', path=source.path)
assert_result_count(res,
2,
status='error',
message=("Unknown target sibling '%s' for publication",
'target'))
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
# we will be testing presence of the file content, so let's make it progress
sub2_target.config.set('receive.denyCurrentBranch',
'updateInstead',
where='local')
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, '2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# publish recursively
with swallow_logs(new_level=logging.DEBUG) as cml:
res = publish(dataset=source, to="target", recursive=True)
assert_not_in('forced update', cml.out,
"we probably haven't merged git-annex before pushing")
# testing result list
# base dataset was already published above, notneeded again
assert_status(('ok', 'notneeded'), res) # nothing failed
assert_result_count(res, 3, type='dataset')
eq_({r['path'] for r in res}, {src_path, sub1.path, sub2.path})
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
eq_(list(target.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
eq_(list(sub1_target.get_branch_commits("master")),
list(sub1.get_branch_commits("master")))
eq_(list(sub1_target.get_branch_commits("git-annex")),
list(sub1.get_branch_commits("git-annex")))
eq_(list(sub2_target.get_branch_commits("master")),
list(sub2.get_branch_commits("master")))
eq_(list(sub2_target.get_branch_commits("git-annex")),
list(sub2.get_branch_commits("git-annex")))
# we are tracking origin but origin has different git-annex, since we
# cloned from it, so it is not aware of our git-annex
neq_(list(origin.repo.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
# So if we first publish to it recursively, we would update
# all sub-datasets since git-annex branch would need to be pushed
res_ = publish(dataset=source, recursive=True)
assert_result_count(res_, 1, status='ok', path=source.path)
assert_result_count(res_, 1, status='ok', path=sub1.path)
assert_result_count(res_, 1, status='ok', path=sub2.path)
# and now should carry the same state for git-annex
eq_(list(origin.repo.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
# test for publishing with --since. By default since no changes, nothing pushed
res_ = publish(dataset=source, recursive=True)
assert_result_count(res_, 3, status='notneeded', type='dataset')
# still nothing gets pushed, because origin is up to date
res_ = publish(dataset=source, recursive=True, since='HEAD^')
assert_result_count(res_, 3, status='notneeded', type='dataset')
# and we should not fail if we run it from within the dataset
with chpwd(source.path):
res_ = publish(recursive=True, since='HEAD^')
assert_result_count(res_, 3, status='notneeded', type='dataset')
# Let's now update one subm
with open(opj(sub2.path, "file.txt"), 'w') as f:
f.write('')
# add to subdataset, does not alter super dataset!
# MIH: use `to_git` because original test author used
# and explicit `GitRepo.add` -- keeping this for now
Dataset(sub2.path).add('file.txt', to_git=True)
# Let's now update one subm
create_tree(sub2.path, {'file.dat': 'content'})
# add to subdataset, without reflecting the change in its super(s)
Dataset(sub2.path).add('file.dat')
# note: will publish to origin here since that is what it tracks
res_ = publish(dataset=source, recursive=True, on_failure='ignore')
## only updates published, i.e. just the subdataset, super wasn't altered
## nothing copied!
assert_status(('ok', 'notneeded'), res_)
assert_result_count(res_, 1, status='ok', path=sub2.path, type='dataset')
assert_result_count(res_, 0, path=opj(sub2.path, 'file.dat'), type='file')
# since published to origin -- destination should not get that file
nok_(lexists(opj(sub2_target.path, 'file.dat')))
res_ = publish(dataset=source, to='target', recursive=True)
assert_status(('ok', 'notneeded'), res_)
assert_result_count(res_, 1, status='ok', path=sub2.path, type='dataset')
assert_result_count(res_, 0, path=opj(sub2.path, 'file.dat'), type='file')
# Note: with updateInstead only in target2 and not saving change in
# super-dataset we would have made remote dataset, if we had entire
# hierarchy, to be somewhat inconsistent.
# But here, since target datasets are independent -- it is ok
# and the file itself was transferred
ok_(lexists(opj(sub2_target.path, 'file.dat')))
nok_(sub2_target.file_has_content('file.dat'))
## but now we can redo publish recursively, with explicitly requested data transfer
res_ = publish(dataset=source,
to='target',
recursive=True,
transfer_data='all')
ok_(sub2_target.file_has_content('file.dat'))
assert_result_count(res_, 1, status='ok', path=opj(sub2.path, 'file.dat'))
# Let's save those present changes and publish while implying "since last
# merge point"
source.save(message="Changes in subm2")
# and test if it could deduce the remote/branch to push to
source.config.set('branch.master.remote', 'target', where='local')
with chpwd(source.path):
res_ = publish(since='', recursive=True)
# TODO: somehow test that there were no even attempt to diff within "subm 1"
# since if `--since=''` worked correctly, nothing has changed there and it
# should have not been even touched
assert_status(('ok', 'notneeded'), res_)
assert_result_count(res_, 1, status='ok', path=source.path, type='dataset')
def test_publish_with_data(origin, src_path, dst_path, sub1_pub, sub2_pub, dst_clone_path):
# prepare src
source = install(src_path, source=origin, recursive=True)
source.repo.get('test-annex.dat')
# create plain git at target:
target = AnnexRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# now, set up targets for the submodules:
# the need to be annexes, because we want to be able to copy data to them
# further down
sub1_target = AnnexRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, '2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
res = publish(dataset=source, to="target", path=['test-annex.dat'], result_xfm='paths')
# first it would publish data and then push
# TODO order is not fixed (yet)
#eq_(res, [opj(source.path, 'test-annex.dat'), source.path])
eq_(set(res), set([opj(source.path, 'test-annex.dat'), source.path]))
# XXX master was not checked out in dst!
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
# TODO: last commit in git-annex branch differs. Probably fine,
# but figure out, when exactly to expect this for proper testing:
# yoh: they differ because local annex records information about now
# file being available in that remote, and remote one does it via a call in
# the hook I guess. So they both get the same information but in two
# different commits. I do not observe such behavior of remote having git-annex
# automagically updated in older clones
# which do not have post-receive hook on remote side
eq_(list(target.get_branch_commits("git-annex"))[1:],
list(source.repo.get_branch_commits("git-annex"))[1:])
# we need compare target/master:
target.checkout("master")
ok_(target.file_has_content('test-annex.dat'))
# make sure that whatever we published is actually consumable
dst_clone = install(
dst_clone_path, source=dst_path,
result_xfm='datasets', return_type='item-or-list')
nok_(dst_clone.repo.file_has_content('test-annex.dat'))
res = dst_clone.get('test-annex.dat')
ok_(dst_clone.repo.file_has_content('test-annex.dat'))
res = publish(dataset=source, to="target", path=['.'])
# there is nothing to publish on 2nd attempt
#eq_(res, ([source, 'test-annex.dat'], []))
assert_result_count(res, 1, status='notneeded')
import glob
res = publish(dataset=source, to="target", path=glob.glob1(source.path, '*'))
# Note: This leads to recursive publishing, since expansion of '*'
# contains the submodules themselves in this setup
# only the subdatasets, targets are plain git repos, hence
# no file content is pushed, all content in super was pushed
# before
assert_result_count(res, 3)
assert_result_count(res, 1, status='ok', path=sub1.path)
assert_result_count(res, 1, status='ok', path=sub2.path)
assert_result_count(res, 1, status='notneeded', path=source.path)
# if we publish again -- nothing to be published
res = source.publish(to="target")
assert_result_count(res, 1, status='notneeded', path=source.path)
# if we drop a file and publish again -- dataset should be published
# since git-annex branch was updated
source.drop('test-annex.dat')
res = source.publish(to="target")
assert_result_count(res, 1, status='ok', path=source.path)
# and empty again if we try again
res = source.publish(to="target")
assert_result_count(res, 1, status='notneeded', path=source.path)
def test_publish_recursive(pristine_origin, origin_path, src_path, dst_path, sub1_pub, sub2_pub):
# we will be publishing back to origin, so to not alter testrepo
# we will first clone it
origin = install(origin_path, source=pristine_origin, recursive=True)
# prepare src
source = install(src_path, source=origin.path, recursive=True)
# we will be trying to push into this later on, need to give permissions...
origin_sub2 = Dataset(opj(origin_path, '2'))
origin_sub2.config.set(
'receive.denyCurrentBranch', 'updateInstead', where='local')
## TODO this manual fixup is needed due to gh-1548 -- needs proper solution
#os.remove(opj(origin_sub2.path, '.git'))
#os.rename(opj(origin_path, '.git', 'modules', '2'), opj(origin_sub2.path, '.git'))
# create plain git at target:
target = GitRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# subdatasets have no remote yet, so recursive publishing should fail:
res = publish(dataset=source, to="target", recursive=True, on_failure='ignore')
assert_result_count(res, 3)
assert_result_count(
res, 1, status='ok', type='dataset', path=source.path)
assert_result_count(
res, 2, status='error',
message=("Unknown target sibling '%s' for publication", 'target'))
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
# we will be testing presence of the file content, so let's make it progress
sub2_target.config.set('receive.denyCurrentBranch', 'updateInstead', where='local')
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, '2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# publish recursively
with swallow_logs(new_level=logging.DEBUG) as cml:
res = publish(dataset=source, to="target", recursive=True)
assert_not_in(
'forced update', cml.out,
"we probably haven't merged git-annex before pushing"
)
# testing result list
# base dataset was already published above, notneeded again
assert_status(('ok', 'notneeded'), res) # nothing failed
assert_result_count(
res, 3, type='dataset')
eq_({r['path'] for r in res},
{src_path, sub1.path, sub2.path})
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
eq_(list(target.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
eq_(list(sub1_target.get_branch_commits("master")),
list(sub1.get_branch_commits("master")))
eq_(list(sub1_target.get_branch_commits("git-annex")),
list(sub1.get_branch_commits("git-annex")))
eq_(list(sub2_target.get_branch_commits("master")),
list(sub2.get_branch_commits("master")))
eq_(list(sub2_target.get_branch_commits("git-annex")),
list(sub2.get_branch_commits("git-annex")))
# we are tracking origin but origin has different git-annex, since we
# cloned from it, so it is not aware of our git-annex
neq_(list(origin.repo.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
# So if we first publish to it recursively, we would update
# all sub-datasets since git-annex branch would need to be pushed
res_ = publish(dataset=source, recursive=True)
assert_result_count(res_, 1, status='ok', path=source.path)
assert_result_count(res_, 1, status='ok', path=sub1.path)
assert_result_count(res_, 1, status='ok', path=sub2.path)
# and now should carry the same state for git-annex
eq_(list(origin.repo.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
# test for publishing with --since. By default since no changes, nothing pushed
res_ = publish(dataset=source, recursive=True)
assert_result_count(
res_, 3, status='notneeded', type='dataset')
# still nothing gets pushed, because origin is up to date
res_ = publish(dataset=source, recursive=True, since='HEAD^')
assert_result_count(
res_, 3, status='notneeded', type='dataset')
# and we should not fail if we run it from within the dataset
with chpwd(source.path):
res_ = publish(recursive=True, since='HEAD^')
assert_result_count(
res_, 3, status='notneeded', type='dataset')
# Let's now update one subm
with open(opj(sub2.path, "file.txt"), 'w') as f:
f.write('')
# add to subdataset, does not alter super dataset!
# MIH: use `to_git` because original test author used
# and explicit `GitRepo.add` -- keeping this for now
Dataset(sub2.path).add('file.txt', to_git=True)
# Let's now update one subm
create_tree(sub2.path, {'file.dat': 'content'})
# add to subdataset, without reflecting the change in its super(s)
Dataset(sub2.path).add('file.dat')
# note: will publish to origin here since that is what it tracks
res_ = publish(dataset=source, recursive=True, on_failure='ignore')
## only updates published, i.e. just the subdataset, super wasn't altered
## nothing copied!
assert_status(('ok', 'notneeded'), res_)
assert_result_count(res_, 1, status='ok', path=sub2.path, type='dataset')
assert_result_count(res_, 0, path=opj(sub2.path, 'file.dat'), type='file')
# since published to origin -- destination should not get that file
nok_(lexists(opj(sub2_target.path, 'file.dat')))
res_ = publish(dataset=source, to='target', recursive=True)
assert_status(('ok', 'notneeded'), res_)
assert_result_count(res_, 1, status='ok', path=sub2.path, type='dataset')
assert_result_count(res_, 0, path=opj(sub2.path, 'file.dat'), type='file')
# Note: with updateInstead only in target2 and not saving change in
# super-dataset we would have made remote dataset, if we had entire
# hierarchy, to be somewhat inconsistent.
# But here, since target datasets are independent -- it is ok
# and the file itself was transferred
ok_(lexists(opj(sub2_target.path, 'file.dat')))
nok_(sub2_target.file_has_content('file.dat'))
## but now we can redo publish recursively, with explicitly requested data transfer
res_ = publish(
dataset=source, to='target',
recursive=True,
transfer_data='all'
)
ok_(sub2_target.file_has_content('file.dat'))
assert_result_count(
res_, 1, status='ok', path=opj(sub2.path, 'file.dat'))
# Let's save those present changes and publish while implying "since last
# merge point"
source.save(message="Changes in subm2")
# and test if it could deduce the remote/branch to push to
source.config.set('branch.master.remote', 'target', where='local')
with chpwd(source.path):
res_ = publish(since='', recursive=True)
# TODO: somehow test that there were no even attempt to diff within "subm 1"
# since if `--since=''` worked correctly, nothing has changed there and it
# should have not been even touched
assert_status(('ok', 'notneeded'), res_)
assert_result_count(res_, 1, status='ok', path=source.path, type='dataset')
def test_openfmri_pipeline2(ind, topurl, outd):
# no versioned files -- should still work! ;)
list(
initiate_dataset(template="openfmri",
dataset_name='dataladtest-ds666',
path=outd,
data_fields=['dataset'])({
'dataset': 'ds666'
}))
with chpwd(outd):
pipeline = ofpipeline('ds666', versioned_urls=False, topurl=topurl)
out = run_pipeline(pipeline)
eq_(len(out), 1)
repo = AnnexRepo(outd, create=False) # to be used in the checks
# Inspect the tree -- that we have all the branches
branches = {'master', 'incoming', 'incoming-processed', 'git-annex'}
eq_(set(repo.get_branches()), branches)
# We do not have custom changes in master yet, so it just follows incoming-processed atm
# eq_(repo.get_hexsha('master'), repo.get_hexsha('incoming-processed'))
# Since we did initiate_dataset -- now we have separate master!
assert_not_equal(repo.get_hexsha('master'),
repo.get_hexsha('incoming-processed'))
# and that one is different from incoming
assert_not_equal(repo.get_hexsha('incoming'),
repo.get_hexsha('incoming-processed'))
# actually the tree should look quite neat with 1.0.0 tag having 1 parent in incoming
# 1.0.1 having 1.0.0 and the 2nd commit in incoming as parents
commits = {b: list(repo.get_branch_commits(b)) for b in branches}
commits_hexsha = {
b: list(repo.get_branch_commits(b, value='hexsha'))
for b in branches
}
commits_l = {
b: list(repo.get_branch_commits(b, limit='left-only'))
for b in branches
}
# all commits out there:
# backend set, dataset init, crawler, init, incoming (shares with master -1),
# (2 or 3 commits, depending on create variant)
# incoming-processed, merge, aggregate metadata:
ncommits_master = len(commits['master'])
assert_in(ncommits_master, [5, 6])
assert_in(len(commits_l['master']), [4, 5])
eq_(len(commits['incoming']), ncommits_master - 2)
eq_(len(commits_l['incoming']), ncommits_master - 2)
eq_(len(commits['incoming-processed']), ncommits_master - 1)
eq_(len(commits_l['incoming-processed']), ncommits_master - 2)
# rerun pipeline -- make sure we are on the same in all branches!
with chpwd(outd):
out = run_pipeline(pipeline)
eq_(len(out), 1)
commits_hexsha_ = {
b: list(repo.get_branch_commits(b, value='hexsha'))
for b in branches
}
eq_(commits_hexsha, commits_hexsha_) # i.e. nothing new
eq_(out[0]['datalad_stats'], ActivityStats(files=2, skipped=2, urls=2))
eq_(out[0]['datalad_stats'], out[0]['datalad_stats'].get_total())
os.rename(opj(ind, 'ds666', 'ds666_R2.0.0.tar.gz'),
opj(ind, 'ds666', 'ds666.tar.gz'))
with chpwd(outd):
out = run_pipeline(pipeline)
eq_(len(out), 1)
eq_(out[0]['datalad_stats'], ActivityStats()) # was committed
stats_total = out[0]['datalad_stats'].get_total()
stats_total.downloaded_size = 0
eq_(
stats_total,
ActivityStats(files=4,
overwritten=1,
skipped=1,
downloaded=1,
merges=[['incoming', 'incoming-processed']],
versions=['1.0.0'],
renamed=1,
urls=2,
add_annex=2))
# in reality there is also 1.0.0+1 tag since file changed but no version suffix
eq_(repo.get_tags(output='name'), ['1.0.0', '1.0.0+1'])
check_dropall_get(repo)
def test_openfmri_pipeline1(ind, topurl, outd, clonedir):
index_html = opj(ind, 'ds666', 'index.html')
list(
initiate_dataset(template="openfmri",
dataset_name='dataladtest-ds666',
path=outd,
data_fields=['dataset'])({
'dataset': 'ds666'
}))
repo = AnnexRepo(outd, create=False) # to be used in the checks
# Since datalad 0.11.2 all .metadata/objects go under annex.
# Here we have a test where we force drop all annexed content,
# to mitigate that let's place all metadata under git
dotdatalad_attributes_file = opj('.datalad', '.gitattributes')
repo.set_gitattributes([('metadata/objects/**', {
'annex.largefiles': '(nothing)'
})], dotdatalad_attributes_file)
# --amend so we do not cause change in # of commits below
repo.commit("gitattributes",
files=dotdatalad_attributes_file,
options=['--amend'])
with chpwd(outd):
pipeline = ofpipeline('ds666', versioned_urls=False, topurl=topurl)
out = run_pipeline(pipeline)
eq_(len(out), 1)
# Inspect the tree -- that we have all the branches
branches = {'master', 'incoming', 'incoming-processed', 'git-annex'}
eq_(set(repo.get_branches()), branches)
# We do not have custom changes in master yet, so it just follows incoming-processed atm
# eq_(repo.get_hexsha('master'), repo.get_hexsha('incoming-processed'))
# Since we did initiate_dataset -- now we have separate master!
assert_not_equal(repo.get_hexsha('master'),
repo.get_hexsha('incoming-processed'))
# and that one is different from incoming
assert_not_equal(repo.get_hexsha('incoming'),
repo.get_hexsha('incoming-processed'))
t1w_fpath_nover = opj(outd, 'sub1', 'anat', 'sub-1_T1w.dat')
ok_file_has_content(t1w_fpath_nover, "mighty load in old format")
#
# And now versioned files were specified!
#
add_to_index(index_html, content=_versioned_files)
with chpwd(outd):
pipeline = ofpipeline('ds666', versioned_urls=False, topurl=topurl)
out = run_pipeline(pipeline)
eq_(len(out), 1)
ok_(
not exists(t1w_fpath_nover),
"%s file should no longer be there if unversioned files get removed correctly"
% t1w_fpath_nover)
repo = AnnexRepo(outd, create=False) # to be used in the checks
# Inspect the tree -- that we have all the branches
branches = {'master', 'incoming', 'incoming-processed', 'git-annex'}
eq_(set(repo.get_branches()), branches)
# We do not have custom changes in master yet, so it just follows incoming-processed atm
# eq_(repo.get_hexsha('master'), repo.get_hexsha('incoming-processed'))
# Since we did initiate_dataset -- now we have separate master!
assert_not_equal(repo.get_hexsha('master'),
repo.get_hexsha('incoming-processed'))
# and that one is different from incoming
assert_not_equal(repo.get_hexsha('incoming'),
repo.get_hexsha('incoming-processed'))
# actually the tree should look quite neat with 1.0.0 tag having 1 parent in incoming
# 1.0.1 having 1.0.0 and the 2nd commit in incoming as parents
commits = {b: list(repo.get_branch_commits(b)) for b in branches}
commits_hexsha = {
b: list(repo.get_branch_commits(b, value='hexsha'))
for b in branches
}
commits_l = {
b: list(repo.get_branch_commits(b, limit='left-only'))
for b in branches
}
# all commits out there:
# backend set, dataset init, crawler init
# (2 or 3 commits, depending on create variant)
# + 3*(incoming, processed, merge)
# + 3*aggregate-metadata update
# - 1 since now that incoming starts with master, there is one less merge
# In --incremental mode there is a side effect of absent now
# 2*remove of obsolete metadata object files,
# see https://github.com/datalad/datalad/issues/2772
ncommits_master = len(commits['master'])
assert_in(ncommits_master, [13, 14])
assert_in(len(commits_l['master']), [8, 9])
eq_(len(commits['incoming']), ncommits_master - 8)
eq_(len(commits_l['incoming']), ncommits_master - 8)
eq_(len(commits['incoming-processed']), ncommits_master - 5)
eq_(len(commits_l['incoming-processed']), ncommits_master - 8)
# Check tags for the versions
eq_(out[0]['datalad_stats'].get_total().versions, ['1.0.0', '1.0.1'])
# +1 because original "release" was assumed to be 1.0.0
repo_tags = repo.get_tags()
eq_(repo.get_tags(output='name'), ['1.0.0', '1.0.0+1', '1.0.1'])
# Ben: The tagged ones currently are the ones with the message
# '[DATALAD] dataset aggregate metadata update\n':
eq_(repo_tags[0]['hexsha'],
commits_l['master'][4].hexsha) # next to the last one
eq_(repo_tags[-1]['hexsha'], commits_l['master'][0].hexsha) # the last one
def hexsha(l):
return l.__class__(x.hexsha for x in l)
# Verify that we have desired tree of merges
eq_(hexsha(commits_l['incoming-processed'][0].parents),
(commits_l['incoming-processed'][1].hexsha,
commits_l['incoming'][0].hexsha))
eq_(
hexsha(commits_l['incoming-processed'][2].parents),
(
commits_l['incoming-processed'][3].hexsha, # also in master
commits_l['incoming'][2].hexsha,
))
# ben: The following two comparisons are targeting these commits:
# commit "Merge branch 'incoming-processed'\n" in commits_l['master'],
# parents are:
# commit "[DATALAD] dataset aggregate metadata update\n" in commits_l['master'] and
# commit "[DATALAD] Added files from extracted archives\n\nFiles processed: 6\n renamed: 2\n +annex: 3\nBranches merged: incoming->incoming-processed\n" in commits_l['incoming-processed']
eq_(hexsha(commits_l['master'][1].parents),
(commits_l['master'][2].hexsha,
commits_l['incoming-processed'][0].hexsha))
eq_(hexsha(commits_l['master'][3].parents),
(commits_l['master'][4].hexsha,
commits_l['incoming-processed'][1].hexsha))
with chpwd(outd):
eq_(set(glob('*')), {'changelog.txt', 'sub-1'})
all_files = sorted(find_files('.'))
t1w_fpath = opj(outd, 'sub-1', 'anat', 'sub-1_T1w.dat')
ok_file_has_content(t1w_fpath, "mighty load 1.0.1")
ok_file_under_git(opj(outd, 'changelog.txt'), annexed=False)
ok_file_under_git(t1w_fpath, annexed=True)
try:
# this is the new way
from datalad.metadata.metadata import get_ds_aggregate_db_locations
ds = Dataset('.')
dbloc, objbase = get_ds_aggregate_db_locations(ds)
dbloc = op.relpath(dbloc, start=ds.path)
except ImportError:
# this stopped working in early 2019 versions of datalad
from datalad.metadata.metadata import agginfo_relpath
dbloc = agginfo_relpath
target_files = {
'./.datalad/config',
'./.datalad/crawl/crawl.cfg',
# no more!
# './.datalad/config.ttl', './.datalad/datalad.ttl',
'./.datalad/crawl/statuses/incoming.json',
'./.datalad/crawl/versions/incoming.json',
'./changelog.txt',
'./sub-1/anat/sub-1_T1w.dat',
'./sub-1/beh/responses.tsv',
'./' + dbloc,
}
target_incoming_files = {
'.gitattributes', # we marked default backend right in the incoming
# we now base 'incoming' on master branch, so we get all those as well
'.datalad/.gitattributes',
'.datalad/config',
'.datalad/crawl/crawl.cfg',
'changelog.txt',
'ds666.tar.gz',
'ds666-beh_R1.0.1.tar.gz',
'ds666_R1.0.0.tar.gz',
'ds666_R1.0.1.tar.gz',
'ds666_R2.0.0.tar.gz',
'.datalad/crawl/statuses/incoming.json',
'.datalad/crawl/versions/incoming.json'
}
# Ben: metadata object files may differ in their names containing some checksum-ish shit ...
# TODO: Check how those names are constructed and may be at least count the number of created object files in addition to that comparison
eq_(
set([
f for f in all_files
if not f.startswith('./.datalad/metadata/objects/')
]), target_files)
# check that -beh was committed in 2nd commit in incoming, not the first one
assert_not_in('ds666-beh_R1.0.1.tar.gz',
repo.get_files(commits_l['incoming'][-1].hexsha))
assert_in('ds666-beh_R1.0.1.tar.gz',
repo.get_files(commits_l['incoming'][0].hexsha))
# rerun pipeline -- make sure we are on the same in all branches!
with chpwd(outd):
out = run_pipeline(pipeline)
eq_(len(out), 1)
commits_hexsha_ = {
b: list(repo.get_branch_commits(b, value='hexsha'))
for b in branches
}
eq_(commits_hexsha, commits_hexsha_) # i.e. nothing new
# actually we do manage to add_git 1 (README) since it is generated committed directly to git
# BUT now fixed -- if not committed (was the same), should be marked as skipped
# Nothing was committed so stats leaked all the way up
eq_(out[0]['datalad_stats'], ActivityStats(files=5, skipped=5, urls=5))
eq_(out[0]['datalad_stats'], out[0]['datalad_stats'].get_total())
# rerun pipeline when new content is available
# add new revision, rerun pipeline and check that stuff was processed/added correctly
add_to_index(
index_html,
content=
'<a href="ds666_R2.0.0.tar.gz">Raw data on AWS version 2.0.0</a>')
with chpwd(outd):
out = run_pipeline(pipeline)
all_files_updated = sorted(find_files('.'))
eq_(len(out), 1)
assert_not_equal(out[0]['datalad_stats'].get_total(), ActivityStats())
# there is no overlays ATM, so behav would be gone since no 2.0.0 for it!
target_files.remove('./sub-1/beh/responses.tsv')
# Ben: metadata object files may differ in their names containing some checksum-ish shit ...
# TODO: Check how those names are constructed and may be at least count the number of created object files in addition to that comparison
eq_(
set([
f for f in all_files_updated
if not f.startswith('./.datalad/metadata/objects/')
]), target_files)
# new instance so it re-reads git stuff etc
# repo = AnnexRepo(outd, create=False) # to be used in the checks
commits_ = {b: list(repo.get_branch_commits(b)) for b in branches}
commits_hexsha_ = {
b: list(repo.get_branch_commits(b, value='hexsha'))
for b in branches
}
commits_l_ = {
b: list(repo.get_branch_commits(b, limit='left-only'))
for b in branches
}
assert_not_equal(commits_hexsha, commits_hexsha_)
eq_(out[0]['datalad_stats'],
ActivityStats()) # commit happened so stats were consumed
# numbers seems to be right
total_stats = out[0]['datalad_stats'].get_total()
# but for some reason downloaded_size fluctuates.... why? probably archiving...?
total_stats.downloaded_size = 0
eq_(
total_stats,
ActivityStats(
files=8,
skipped=5,
downloaded=1,
renamed=1,
urls=6,
add_annex=2, # add_git=1, # README
versions=['2.0.0'],
merges=[['incoming', 'incoming-processed']]))
check_dropall_get(repo)
# Let's see if pipeline would remove files we stopped tracking
remove_from_index(index_html, '<a href=.ds666_R1.0.0[^<]*</a>')
with chpwd(outd):
with swallow_logs(new_level=logging.WARNING) as cml:
out = run_pipeline(pipeline)
# since files get removed in incoming, but repreprocessed completely
# incomming-processed and merged into master -- new commits will come
# They shouldn't have any difference but still should be new commits
assert_in("There is already a tag 2.0.0 in the repository",
cml.out)
eq_(len(out), 1)
incoming_files = repo.get_files('incoming')
target_incoming_files.remove('ds666_R1.0.0.tar.gz')
eq_(set(incoming_files), target_incoming_files)
commits_hexsha_removed = {
b: list(repo.get_branch_commits(b, value='hexsha'))
for b in branches
}
# our 'statuses' database should have recorded the change thus got a diff
# which propagated through all branches
for b in 'master', 'incoming-processed':
# with non persistent DB we had no changes
# eq_(repo.repo.branches[b].commit.diff(commits_hexsha_[b][0]), [])
eq_(repo.repo.branches[b].commit.diff(commits_hexsha_[b][0])[0].a_path,
'.datalad/crawl/statuses/incoming.json')
dincoming = repo.repo.branches['incoming'].commit.diff(
commits_hexsha_['incoming'][0])
eq_(len(dincoming),
2) # 2 diff objects -- 1 file removed, 1 statuses updated
eq_(set([d.a_path for d in dincoming]),
{'.datalad/crawl/statuses/incoming.json', 'ds666_R1.0.0.tar.gz'})
# since it seems to diff "from current to the specified", it will be listed as new_file
assert any(d.new_file for d in dincoming)
eq_(out[0]['datalad_stats'].get_total().removed, 1)
assert_not_equal(commits_hexsha_, commits_hexsha_removed)
# we will check if a clone would be crawling just as good
from datalad.api import crawl
# make a brand new clone
GitRepo.clone(outd, clonedir)
def _pipeline(*args, **kwargs):
"""Helper to mock openfmri.pipeline invocation so it looks at our 'server'"""
kwargs = updated(kwargs, {'topurl': topurl, 'versioned_urls': False})
return ofpipeline(*args, **kwargs)
with chpwd(clonedir), patch.object(openfmri, 'pipeline', _pipeline):
output, stats = crawl(
) # we should be able to recrawl without doing anything
ok_(stats, ActivityStats(files=6, skipped=6, urls=5))
def test_publish_recursive(origin, src_path, dst_path, sub1_pub, sub2_pub):
# prepare src
source = install(src_path, source=origin, recursive=True)[0]
# create plain git at target:
target = GitRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# subdatasets have no remote yet, so recursive publishing should fail:
with assert_raises(ValueError) as cm:
publish(dataset=source, to="target", recursive=True)
assert_in("No sibling 'target' found", exc_str(cm.exception))
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = AnnexRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, 'subm 2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# publish recursively
with swallow_logs(new_level=logging.DEBUG) as cml:
res = publish(dataset=source, to="target", recursive=True)
assert_not_in(
'forced update', cml.out,
"we probably haven't merged git-annex before pushing"
)
# testing result list
# (Note: Dataset lacks __eq__ for now. Should this be based on path only?)
assert_is_instance(res, tuple)
assert_is_instance(res[0], list)
assert_is_instance(res[1], list)
eq_(res[1], []) # nothing failed/was skipped
for item in res[0]:
assert_is_instance(item, Dataset)
eq_({res[0][0].path, res[0][1].path, res[0][2].path},
{src_path, sub1.path, sub2.path})
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
eq_(list(target.get_branch_commits("git-annex")),
list(source.repo.get_branch_commits("git-annex")))
eq_(list(sub1_target.get_branch_commits("master")),
list(sub1.get_branch_commits("master")))
eq_(list(sub1_target.get_branch_commits("git-annex")),
list(sub1.get_branch_commits("git-annex")))
eq_(list(sub2_target.get_branch_commits("master")),
list(sub2.get_branch_commits("master")))
eq_(list(sub2_target.get_branch_commits("git-annex")),
list(sub2.get_branch_commits("git-annex")))
# test for publishing with --since. By default since no changes, only current pushed
res_ = publish(dataset=source, recursive=True)
# only current one would get pushed
eq_(set(r.path for r in res_[0]), {src_path})
# all get pushed
res_ = publish(dataset=source, recursive=True, since='HEAD^')
eq_(set(r.path for r in res_[0]), {src_path, sub1.path, sub2.path})
# Let's now update one subm
with open(opj(sub2.path, "file.txt"), 'w') as f:
f.write('')
sub2.add('file.txt')
sub2.commit("")
# TODO: Doesn't work: https://github.com/datalad/datalad/issues/636
#source.save("changed sub2", all_changes=True)
source.repo.commit("", options=['-a'])
res_ = publish(dataset=source, recursive=True)
# only updated ones were published
eq_(set(r.path for r in res_[0]), {src_path, sub2.path})
def test_publish_with_data(origin, src_path, dst_path, sub1_pub, sub2_pub):
# prepare src
source = install(src_path, source=origin, recursive=True)[0]
source.repo.get('test-annex.dat')
# create plain git at target:
target = AnnexRepo(dst_path, create=True)
target.checkout("TMP", ["-b"])
source.repo.add_remote("target", dst_path)
# now, set up targets for the submodules:
sub1_target = GitRepo(sub1_pub, create=True)
sub1_target.checkout("TMP", ["-b"])
sub2_target = GitRepo(sub2_pub, create=True)
sub2_target.checkout("TMP", ["-b"])
sub1 = GitRepo(opj(src_path, 'subm 1'), create=False)
sub2 = GitRepo(opj(src_path, 'subm 2'), create=False)
sub1.add_remote("target", sub1_pub)
sub2.add_remote("target", sub2_pub)
# TMP: Insert the fetch to prevent GitPython to fail after the push,
# because it cannot resolve the SHA of the old commit of the remote,
# that git reports back after the push.
# TODO: Figure out, when to fetch things in general; Alternatively:
# Is there an option for push, that prevents GitPython from failing?
source.repo.fetch("target")
res = publish(dataset=source, to="target", path=['test-annex.dat'])
eq_(res, ([source, 'test-annex.dat'], []))
eq_(list(target.get_branch_commits("master")),
list(source.repo.get_branch_commits("master")))
# TODO: last commit in git-annex branch differs. Probably fine,
# but figure out, when exactly to expect this for proper testing:
eq_(list(target.get_branch_commits("git-annex"))[1:],
list(source.repo.get_branch_commits("git-annex"))[1:])
# we need compare target/master:
target.checkout("master")
eq_(target.file_has_content(['test-annex.dat']), [True])
source.repo.fetch("target")
res = publish(dataset=source, to="target", path=['.'])
eq_(res, ([source, 'test-annex.dat'], []))
source.repo.fetch("target")
import glob
res = publish(dataset=source, to="target", path=glob.glob1(source.path, '*'))
# Note: This leads to recursive publishing, since expansion of '*'
# contains the submodules themselves in this setup
# collect result paths:
result_paths = []
for item in res[0]:
if isinstance(item, Dataset):
result_paths.append(item.path)
else:
result_paths.append(item)
eq_({source.path, opj(source.path, "subm 1"),
opj(source.path, "subm 2"), 'test-annex.dat'},
set(result_paths))
