def __init__(self, path_abs, id=None, identifier=None):
path_abs = os.path.normpath(path_abs)
if identifier:
i = identifier
else:
i = Identifier(path=path_abs)
self.identifier = i
self.id = i.id
self.idparts = list(i.parts.values())
self.collection_id = i.collection_id()
self.parent_id = i.parent_id()
self.path_abs = path_abs
self.path = path_abs
self.collection_path = i.collection_path()
self.parent_path = i.parent_path()
self.root = os.path.dirname(self.parent_path)
self.json_path = i.path_abs('json')
self.changelog_path = i.path_abs('changelog')
self.control_path = i.path_abs('control')
self.mets_path = i.path_abs('mets')
self.lock_path = i.path_abs('lock')
self.files_path = i.path_abs('files')
self.path_rel = i.path_rel()
self.json_path_rel = i.path_rel('json')
self.changelog_path_rel = i.path_rel('changelog')
self.control_path_rel = i.path_rel('control')
self.mets_path_rel = i.path_rel('mets')
self.files_path_rel = i.path_rel('files')
def __init__( self, path_abs, id=None, identifier=None ):
path_abs = os.path.normpath(path_abs)
if identifier:
i = identifier
else:
i = Identifier(path=path_abs)
self.identifier = i
self.id = i.id
self.idparts = i.parts.values()
self.collection_id = i.collection_id()
self.parent_id = i.parent_id()
self.path_abs = path_abs
self.path = path_abs
self.collection_path = i.collection_path()
self.parent_path = i.parent_path()
self.root = os.path.dirname(self.parent_path)
self.json_path = i.path_abs('json')
self.changelog_path = i.path_abs('changelog')
self.control_path = i.path_abs('control')
self.mets_path = i.path_abs('mets')
self.lock_path = i.path_abs('lock')
self.files_path = i.path_abs('files')
self.path_rel = i.path_rel()
self.json_path_rel = i.path_rel('json')
self.changelog_path_rel = i.path_rel('changelog')
self.control_path_rel = i.path_rel('control')
self.mets_path_rel = i.path_rel('mets')
self.files_path_rel = i.path_rel('files')
def __init__(self, path_abs, id=None, identifier=None):
"""
>>> c = Collection('/tmp/ddr-testing-123')
>>> c.id
'ddr-testing-123'
>>> c.ead_path_rel
'ead.xml'
>>> c.ead_path
'/tmp/ddr-testing-123/ead.xml'
>>> c.json_path_rel
'collection.json'
>>> c.json_path
'/tmp/ddr-testing-123/collection.json'
"""
path_abs = os.path.normpath(path_abs)
if identifier:
i = identifier
else:
i = Identifier(path=path_abs)
self.identifier = i
self.id = i.id
self.idparts = i.parts.values()
self.path_abs = path_abs
self.path = path_abs
self.root = os.path.split(self.path)[0]
self.json_path = i.path_abs('json')
self.git_path = i.path_abs('git')
self.gitignore_path = i.path_abs('gitignore')
self.annex_path = i.path_abs('annex')
self.changelog_path = i.path_abs('changelog')
self.control_path = i.path_abs('control')
self.ead_path = i.path_abs('ead')
self.lock_path = i.path_abs('lock')
self.files_path = i.path_abs('files')
self.path_rel = i.path_rel()
self.json_path_rel = i.path_rel('json')
self.git_path_rel = i.path_rel('git')
self.gitignore_path_rel = i.path_rel('gitignore')
self.annex_path_rel = i.path_rel('annex')
self.changelog_path_rel = i.path_rel('changelog')
self.control_path_rel = i.path_rel('control')
self.ead_path_rel = i.path_rel('ead')
self.files_path_rel = i.path_rel('files')
self.git_url = '{}:{}'.format(config.GITOLITE, self.id)
def delete(self, document_id, recursive=False):
"""Delete a document and optionally its children.
TODO refactor after upgrading Elasticsearch past 2.4.
delete_by_query was removed sometime during elasticsearch-py 2.*
I think it was added back in a later version so the code stays for now.
For now, instead of deleting based on document_id, we start with
document_id, find all paths beneath it in the filesystem,
and curl DELETE url each individual document from Elasticsearch.
@param document_id:
@param recursive: True or False
"""
logger.debug('delete(%s, %s)' % (document_id, recursive))
oi = Identifier(document_id, config.MEDIA_BASE)
if recursive:
paths = util.find_meta_files(oi.path_abs(),
recursive=recursive,
files_first=1)
else:
paths = [oi.path_abs()]
identifiers = [Identifier(path) for path in paths]
num = len(identifiers)
for n, oi in enumerate(identifiers):
# TODO hard-coded models here!
if oi.model == 'segment':
model = 'entity'
else:
model = oi.model
try:
result = self.es.delete(index=self.index_name(model), id=oi.id)
print(
f'{n}/{num} DELETE {self.index_name(model)} {oi.id} -> {result["result"]}'
)
except docstore.NotFoundError as err:
print(
f'{n}/{num} DELETE {self.index_name(model)} {oi.id} -> 404 Not Found'
)
def signature_abs(obj, basepath):
"""Absolute path to signature image file, if signature_id present.
Expects obj.signature_id to be either a valid file ID
or a special interview signature image
(ex. "denshovh-aart-03", "denshovh-hlarry_g-02")
@returns: str absolute path to signature img, or None
"""
if isinstance(obj, dict):
sid = obj.get('signature_id')
else:
sid = getattr(obj, 'signature_id', None)
# ignore interview signature ID
if sid and INTERVIEW_SIG_REGEX.match(sid):
return None
if sid:
try:
oi = Identifier(sid, basepath)
except:
oi = None
if oi and oi.model == 'file':
return oi.path_abs('access')
return None
def signature_abs(obj, basepath):
"""Absolute path to signature image file, if signature_id present.
Expects obj.signature_id to be either a valid file ID
or a special interview signature image
(ex. "denshovh-aart-03", "denshovh-hlarry_g-02")
@returns: str absolute path to signature img, or None
"""
if isinstance(obj, dict):
sid = obj.get('signature_id')
else:
sid = getattr(obj, 'signature_id', None)
# ignore interview signature ID
if sid and INTERVIEW_SIG_REGEX.match(sid):
return None
if sid:
try:
oi = Identifier(sid, basepath)
except:
oi = None
if oi and oi.model == 'file':
return oi.path_abs('access')
return None