async def _resolve_path_to_ids(self, path, start_at=None):
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
# parts is list of [path_part_name, is_folder]
parts = [[parse.unquote(x), True] for x in path.strip('/').split('/')]
if not path.endswith('/'):
parts[-1][1] = False
while parts:
current_part = parts.pop(0)
query = "title = '{}' " \
"and trashed = false " \
"and mimeType != 'application/vnd.google-apps.form' " \
"and mimeType != 'application/vnd.google-apps.map' " \
"and mimeType {} '{}'".format(
clean_query(current_part[0]),
'=' if current_part[1] else '!=',
self.FOLDER_MIME_TYPE
)
async with self.request(
'GET',
self.build_url('files', item_id, 'children', q=query, fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
data = await resp.json()
try:
item_id = data['items'][0]['id']
except (KeyError, IndexError):
if parts:
raise exceptions.MetadataError('{} not found'.format(str(path)), code=http.client.NOT_FOUND)
name, ext = os.path.splitext(current_part[0])
if ext not in ('.gdoc', '.gdraw', '.gslides', '.gsheet'):
return ret + [{
'id': None,
'title': current_part[0],
'mimeType': 'folder' if path.endswith('/') else '',
}]
parts.append([name, current_part[1]])
async with self.request(
'GET',
self.build_url('files', item_id, fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
ret.append(await resp.json())
return ret
async def _resolve_path_to_ids(self, path, start_at=None):
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
# parts is list of [path_part_name, is_folder]
parts = [[parse.unquote(x), True] for x in path.strip('/').split('/')]
if not path.endswith('/'):
parts[-1][1] = False
while parts:
current_part = parts.pop(0)
query = "title = '{}' " \
"and trashed = false " \
"and mimeType != 'application/vnd.google-apps.form' " \
"and mimeType {} '{}'".format(
clean_query(current_part[0]),
'=' if current_part[1] else '!=',
self.FOLDER_MIME_TYPE
)
async with self.request(
'GET',
self.build_url('files', item_id, 'children', q=query, fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
data = await resp.json()
try:
item_id = data['items'][0]['id']
except (KeyError, IndexError):
if parts:
raise exceptions.MetadataError('{} not found'.format(str(path)), code=http.client.NOT_FOUND)
name, ext = os.path.splitext(current_part[0])
if ext not in ('.gdoc', '.gdraw', '.gslides', '.gsheet'):
return ret + [{
'id': None,
'title': current_part[0],
'mimeType': 'folder' if path.endswith('/') else '',
}]
parts.append([name, current_part[1]])
async with self.request(
'GET',
self.build_url('files', item_id, fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
ret.append(await resp.json())
return ret
def validate_v1_path(self, path, **kwargs):
if not getattr(self, '_repo', None):
self._repo = yield from self._fetch_repo()
self.default_branch = self._repo['default_branch']
branch_ref = kwargs.get('ref') or kwargs.get(
'branch') or self.default_branch
implicit_folder = path.endswith('/')
url = furl.furl(self.build_repo_url('contents', path))
url.args.update({'ref': branch_ref})
resp = yield from self.make_request('GET',
url.url,
expects=(200, ),
throws=exceptions.MetadataError)
content = yield from resp.json()
explicit_folder = isinstance(content, list)
if implicit_folder != explicit_folder:
raise exceptions.NotFoundError(path)
path = GitHubPath(path)
for part in path.parts:
part._id = (branch_ref, None)
# TODO Validate that filesha is a valid sha
path.parts[-1]._id = (branch_ref, kwargs.get('fileSha'))
return path
def validate_v1_path(self, path, **kwargs):
if not getattr(self, '_repo', None):
self._repo = yield from self._fetch_repo()
self.default_branch = self._repo['default_branch']
branch_ref = kwargs.get('ref') or kwargs.get('branch') or self.default_branch
implicit_folder = path.endswith('/')
url = furl.furl(self.build_repo_url('contents', path))
url.args.update({'ref': branch_ref})
resp = yield from self.make_request(
'GET',
url.url,
expects=(200, 403),
throws=exceptions.MetadataError
)
content = yield from resp.json()
explicit_folder = isinstance(content, list)
if implicit_folder != explicit_folder:
raise exceptions.NotFoundError(path)
path = GitHubPath(path)
for part in path.parts:
part._id = (branch_ref, None)
# TODO Validate that filesha is a valid sha
path.parts[-1]._id = (branch_ref, kwargs.get('fileSha'))
return path
async def validate_v1_path(self, path, **kwargs):
if path == '/':
return GoogleDrivePath('/', _ids=[self.folder['id']], folder=True)
implicit_folder = path.endswith('/')
parts = await self._resolve_path_to_ids(path)
explicit_folder = parts[-1]['mimeType'] == self.FOLDER_MIME_TYPE
if parts[-1]['id'] is None or implicit_folder != explicit_folder:
raise exceptions.NotFoundError(str(path))
names, ids = zip(*[(parse.quote(x['title'], safe=''), x['id']) for x in parts])
return GoogleDrivePath('/'.join(names), _ids=ids, folder='folder' in parts[-1]['mimeType'])
async def validate_v1_path(self, path, **kwargs):
if path == '/':
return GoogleDrivePath('/', _ids=[self.folder['id']], folder=True)
implicit_folder = path.endswith('/')
parts = await self._resolve_path_to_ids(path)
explicit_folder = parts[-1]['mimeType'] == self.FOLDER_MIME_TYPE
if parts[-1]['id'] is None or implicit_folder != explicit_folder:
raise exceptions.NotFoundError(str(path))
names, ids = zip(*[(parse.quote(x['title'], safe=''), x['id']) for x in parts])
return GoogleDrivePath('/'.join(names), _ids=ids, folder='folder' in parts[-1]['mimeType'])
def _resolve_path_to_ids(self, path, start_at=None):
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
parts = [parse.unquote(x) for x in path.strip('/').split('/')]
while parts:
current_part = parts.pop(0)
resp = yield from self.make_request(
'GET',
self.build_url('files',
item_id,
'children',
q="title = '{}'".format(
clean_query(current_part)),
fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
)
try:
item_id = (yield from resp.json())['items'][0]['id']
except (KeyError, IndexError):
if parts:
raise exceptions.MetadataError('{} not found'.format(
str(path)),
code=http.client.NOT_FOUND)
name, ext = os.path.splitext(current_part)
if ext not in ('.gdoc', '.gdraw', '.gslides', '.gsheet'):
return ret + [
{
'id': None,
'title': current_part,
'mimeType': 'folder' if path.endswith('/') else '',
}
]
parts.append(name)
resp = yield from self.make_request(
'GET',
self.build_url('files', item_id, fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
)
ret.append((yield from resp.json()))
return ret
async def _search_tree_for_path(self, path, tree_sha, recursive=True):
"""Search through the given tree for an entity matching the name and type of `path`.
"""
tree = await self._fetch_tree(tree_sha, recursive=True)
if tree['truncated']:
raise GitHubUnsupportedRepoError
implicit_type = 'tree' if path.endswith('/') else 'blob'
for entity in tree['tree']:
if entity['path'] == path.strip('/') and entity['type'] == implicit_type:
return entity
raise exceptions.NotFoundError(str(path))
async def _search_tree_for_path(self, path, tree_sha, recursive=True):
"""Search through the given tree for an entity matching the name and type of `path`.
"""
tree = await self._fetch_tree(tree_sha, recursive=True)
if tree['truncated']:
raise GitHubUnsupportedRepoError
implicit_type = 'tree' if path.endswith('/') else 'blob'
for entity in tree['tree']:
if entity['path'] == path.strip(
'/') and entity['type'] == implicit_type:
return entity
raise exceptions.NotFoundError(str(path))
def _resolve_path_to_ids(self, path, start_at=None):
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
parts = [parse.unquote(x) for x in path.strip('/').split('/')]
while parts:
current_part = parts.pop(0)
resp = yield from self.make_request(
'GET',
self.build_url('files', item_id, 'children', q='title = "{}"'.format(current_part.replace('"', '\\"')), fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
)
try:
item_id = (yield from resp.json())['items'][0]['id']
except (KeyError, IndexError):
if parts:
raise exceptions.MetadataError('{} not found'.format(str(path)), code=http.client.NOT_FOUND)
name, ext = os.path.splitext(current_part)
if ext not in ('.gdoc', '.gsheet'):
return ret + [{
'id': None,
'title': current_part,
'mimeType': 'folder' if path.endswith('/') else '',
}]
parts.append(name)
resp = yield from self.make_request(
'GET',
self.build_url('files', item_id, fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
)
ret.append((yield from resp.json()))
return ret
def _resolve_path_to_ids(self, path, start_at=None):
ret = start_at or [{"title": "", "mimeType": "folder", "id": self.folder["id"]}]
item_id = ret[0]["id"]
parts = [parse.unquote(x) for x in path.strip("/").split("/")]
while parts:
current_part = parts.pop(0)
resp = yield from self.make_request(
"GET",
self.build_url(
"files", item_id, "children", q="title = '{}'".format(clean_query(current_part)), fields="items(id)"
),
expects=(200,),
throws=exceptions.MetadataError,
)
try:
item_id = (yield from resp.json())["items"][0]["id"]
except (KeyError, IndexError):
if parts:
raise exceptions.MetadataError("{} not found".format(str(path)), code=http.client.NOT_FOUND)
name, ext = os.path.splitext(current_part)
if ext not in (".gdoc", ".gdraw", ".gslides", ".gsheet"):
return ret + [
{"id": None, "title": current_part, "mimeType": "folder" if path.endswith("/") else ""}
]
parts.append(name)
resp = yield from self.make_request(
"GET",
self.build_url("files", item_id, fields="id,title,mimeType"),
expects=(200,),
throws=exceptions.MetadataError,
)
ret.append((yield from resp.json()))
return ret
async def _resolve_path_to_ids(self, path, start_at=None):
"""Takes a path and traverses the file tree (ha!) beginning at ``start_at``, looking for
something that matches ``path``. Returns a list of dicts for each part of the path, with
``title``, ``mimeType``, and ``id`` keys.
"""
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
# parts is list of [path_part_name, is_folder]
parts = [[parse.unquote(x), True] for x in path.strip('/').split('/')]
if not path.endswith('/'):
parts[-1][1] = False
while parts:
current_part = parts.pop(0)
query = "title = '{}' " \
"and trashed = false " \
"and mimeType != 'application/vnd.google-apps.form' " \
"and mimeType != 'application/vnd.google-apps.map' " \
"and mimeType {} '{}'".format(
clean_query(current_part[0]),
'=' if current_part[1] else '!=',
self.FOLDER_MIME_TYPE
)
async with self.request(
'GET',
self.build_url('files',
item_id,
'children',
q=query,
fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
data = await resp.json()
try:
item_id = data['items'][0]['id']
except (KeyError, IndexError):
if parts:
# if we can't find an intermediate path part, that's an error
raise exceptions.MetadataError('{} not found'.format(
str(path)),
code=http.client.NOT_FOUND)
# Couldn't find id for last part of path. If path includes Google Doc extension,
# search again without extension (gdrive won't find it if included). Otherwise,
# assume file or folder doesn't yet exist (i.e. id = None)
name, ext = os.path.splitext(current_part[0])
if ext not in ('.gdoc', '.gdraw', '.gslides', '.gsheet'):
return ret + [
{
'id': None,
'title': current_part[0],
'mimeType': 'folder' if path.endswith('/') else '',
}
]
# strip google docs extension and try again
parts.append([name, current_part[1]])
continue
async with self.request(
'GET',
self.build_url('files',
item_id,
fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
ret.append(await resp.json())
return ret
async def _resolve_path_to_ids(self, path, start_at=None):
"""Takes a path and traverses the file tree (ha!) beginning at ``start_at``, looking for
something that matches ``path``. Returns a list of dicts for each part of the path, with
``title``, ``mimeType``, and ``id`` keys.
"""
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
# parts is list of [path_part_name, is_folder]
parts = [[parse.unquote(x), True] for x in path.strip('/').split('/')]
if not path.endswith('/'):
parts[-1][1] = False
while parts:
current_part = parts.pop(0)
part_name, part_is_folder = current_part[0], current_part[1]
name, ext = os.path.splitext(part_name)
if not part_is_folder and ext in ('.gdoc', '.gdraw', '.gslides', '.gsheet'):
gd_ext = drive_utils.get_mimetype_from_ext(ext)
query = "title = '{}' " \
"and trashed = false " \
"and mimeType = '{}'".format(clean_query(name), gd_ext)
else:
query = "title = '{}' " \
"and trashed = false " \
"and mimeType != 'application/vnd.google-apps.form' " \
"and mimeType != 'application/vnd.google-apps.map' " \
"and mimeType {} '{}'".format(
clean_query(part_name),
'=' if part_is_folder else '!=',
self.FOLDER_MIME_TYPE
)
async with self.request(
'GET',
self.build_url('files', item_id, 'children', q=query, fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
data = await resp.json()
try:
item_id = data['items'][0]['id']
except (KeyError, IndexError):
if parts:
# if we can't find an intermediate path part, that's an error
raise exceptions.MetadataError('{} not found'.format(str(path)), code=http.client.NOT_FOUND)
return ret + [{
'id': None,
'title': part_name,
'mimeType': 'folder' if part_is_folder else '',
}]
async with self.request(
'GET',
self.build_url('files', item_id, fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
ret.append(await resp.json())
return ret
async def _resolve_path_to_ids(self, path, start_at=None):
"""Takes a path and traverses the file tree (ha!) beginning at ``start_at``, looking for
something that matches ``path``. Returns a list of dicts for each part of the path, with
``title``, ``mimeType``, and ``id`` keys.
"""
ret = start_at or [{
'title': '',
'mimeType': 'folder',
'id': self.folder['id'],
}]
item_id = ret[0]['id']
# parts is list of [path_part_name, is_folder]
parts = [[parse.unquote(x), True] for x in path.strip('/').split('/')]
if not path.endswith('/'):
parts[-1][1] = False
while parts:
current_part = parts.pop(0)
query = "title = '{}' " \
"and trashed = false " \
"and mimeType != 'application/vnd.google-apps.form' " \
"and mimeType != 'application/vnd.google-apps.map' " \
"and mimeType {} '{}'".format(
clean_query(current_part[0]),
'=' if current_part[1] else '!=',
self.FOLDER_MIME_TYPE
)
async with self.request(
'GET',
self.build_url('files', item_id, 'children', q=query, fields='items(id)'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
data = await resp.json()
try:
item_id = data['items'][0]['id']
except (KeyError, IndexError):
if parts:
# if we can't find an intermediate path part, that's an error
raise exceptions.MetadataError('{} not found'.format(str(path)), code=http.client.NOT_FOUND)
# Couldn't find id for last part of path. If path includes Google Doc extension,
# search again without extension (gdrive won't find it if included). Otherwise,
# assume file or folder doesn't yet exist (i.e. id = None)
name, ext = os.path.splitext(current_part[0])
if ext not in ('.gdoc', '.gdraw', '.gslides', '.gsheet'):
return ret + [{
'id': None,
'title': current_part[0],
'mimeType': 'folder' if path.endswith('/') else '',
}]
# strip google docs extension and try again
parts.append([name, current_part[1]])
continue
async with self.request(
'GET',
self.build_url('files', item_id, fields='id,title,mimeType'),
expects=(200, ),
throws=exceptions.MetadataError,
) as resp:
ret.append(await resp.json())
return ret