diff --git a/swh/web/api/urls.py b/swh/web/api/urls.py
index c42bd375..9d444603 100644
--- a/swh/web/api/urls.py
+++ b/swh/web/api/urls.py
@@ -1,17 +1,18 @@
# Copyright (C) 2017 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU General Public License version 3, or any later version
# See top-level LICENSE file for more information
-import swh.web.api.views.origin # noqa
import swh.web.api.views.content # noqa
+import swh.web.api.views.directory # noqa
+import swh.web.api.views.entity # noqa
+import swh.web.api.views.origin # noqa
import swh.web.api.views.person # noqa
import swh.web.api.views.release # noqa
import swh.web.api.views.revision # noqa
-import swh.web.api.views.directory # noqa
-import swh.web.api.views.entity # noqa
import swh.web.api.views.stat # noqa
+import swh.web.api.views.vault # noqa
from swh.web.api.apiurls import APIUrls
urlpatterns = APIUrls.get_url_patterns()
diff --git a/swh/web/api/views/vault.py b/swh/web/api/views/vault.py
new file mode 100644
index 00000000..a1789ea8
--- /dev/null
+++ b/swh/web/api/views/vault.py
@@ -0,0 +1,138 @@
+# Copyright (C) 2015-2017 The Software Heritage developers
+# See the AUTHORS file at the top-level directory of this distribution
+# License: GNU Affero General Public License version 3, or any later version
+# See top-level LICENSE file for more information
+
+from django.http import HttpResponse
+
+from swh.web.common import service, query
+from swh.web.common.utils import reverse
+from swh.web.api import apidoc as api_doc
+from swh.web.api.apiurls import api_route
+from swh.web.api.views.utils import (
+ api_lookup, doc_exc_id_not_found, doc_exc_bad_id,
+)
+
+
+@api_route('/vault/directory/(?P<dir_id>[a-fA-F0-9]+)/',
+ 'vault-cook-directory')
+@api_doc.route('/vault/directory', tags=['hidden'])
+@api_doc.arg('dir_id',
+ default='d4a96ba891017d0d26c15e509b4e6515e40d75ee',
+ argtype=api_doc.argtypes.sha1_git,
+ argdoc="The directory's sha1 identifier")
+@api_doc.param('email', default=None,
+ argtype=api_doc.argtypes.int,
+ doc="e-mail to notify when the bundle is ready")
+@api_doc.raises(exc=api_doc.excs.badinput, doc=doc_exc_bad_id)
+@api_doc.raises(exc=api_doc.excs.notfound, doc=doc_exc_id_not_found)
+@api_doc.returns(rettype=api_doc.rettypes.dict,
+ retdoc=('dictionary mapping containing the status of '
+ 'the cooking'))
+def api_vault_cook_directory(request, dir_id):
+ """Requests an archive of the directoy identified by dir_id.
+
+ To import the directory in the current directory, use::
+
+ $ tar xvf path/to/directory.tar.gz
+ """
+ email = request.GET.get('email', None)
+ _, obj_id = query.parse_hash_with_algorithms_or_throws(
+ dir_id, ['sha1'], 'Only sha1_git is supported.')
+
+ def _enrich_dir_cook(res):
+ res['fetch_url'] = reverse('vault-fetch-directory',
+ kwargs={'dir_id': dir_id})
+ return res
+
+ return api_lookup(
+ service.vault_cook, 'directory', obj_id, email,
+ notfound_msg="Directory with ID '{}' not found.".format(dir_id),
+ enrich_fn=_enrich_dir_cook)
+
+
+@api_route(r'/vault/directory/(?P<dir_id>[a-fA-F0-9]+)/raw/',
+ 'vault-fetch-directory')
+@api_doc.route('/vault/directory/raw', tags=['hidden'], handle_response=True)
+@api_doc.arg('dir_id',
+ default='d4a96ba891017d0d26c15e509b4e6515e40d75ee',
+ argtype=api_doc.argtypes.sha1_git,
+ argdoc="The directory's sha1 identifier")
+@api_doc.raises(exc=api_doc.excs.badinput, doc=doc_exc_bad_id)
+@api_doc.raises(exc=api_doc.excs.notfound, doc=doc_exc_id_not_found)
+@api_doc.returns(rettype=api_doc.rettypes.octet_stream,
+ retdoc='the cooked directory tarball')
+def api_vault_fetch_directory(request, dir_id):
+ """Fetch the archive of the directoy identified by dir_id."""
+ _, obj_id = query.parse_hash_with_algorithms_or_throws(
+ dir_id, ['sha1'], 'Only sha1_git is supported.')
+ res = api_lookup(
+ service.vault_fetch, 'directory', obj_id,
+ notfound_msg="Directory with ID '{}' not found.".format(dir_id))
+ fname = '{}.tar.gz'.format(dir_id)
+ response = HttpResponse(res, content_type='application/gzip')
+ response['Content-disposition'] = 'attachment; filename={}'.format(fname)
+ return response
+
+
+@api_route(r'/vault/revision_gitfast/(?P<rev_id>[a-fA-F0-9]+)/',
+ 'vault-cook-revision_gitfast')
+@api_doc.route('/vault/revision_gitfast', tags=['hidden'])
+@api_doc.arg('rev_id',
+ default='9174026cfe69d73ef80b27890615f8b2ef5c265a',
+ argtype=api_doc.argtypes.sha1_git,
+ argdoc="The revision's sha1_git identifier")
+@api_doc.param('email', default=None,
+ argtype=api_doc.argtypes.int,
+ doc="e-mail to notify when the bundle is ready")
+@api_doc.raises(exc=api_doc.excs.badinput, doc=doc_exc_bad_id)
+@api_doc.raises(exc=api_doc.excs.notfound, doc=doc_exc_id_not_found)
+@api_doc.returns(rettype=api_doc.rettypes.dict,
+ retdoc='dictionary mapping containing the status of '
+ 'the cooking')
+def api_vault_cook_revision_gitfast(request, rev_id):
+ """Requests an archive of the revision identified by rev_id.
+
+ To import the revision in the current directory, use::
+
+ $ git init
+ $ zcat path/to/revision.gitfast.gz | git fast-import
+ """
+ email = request.GET.get('email', None)
+ _, obj_id = query.parse_hash_with_algorithms_or_throws(
+ rev_id, ['sha1'], 'Only sha1_git is supported.')
+
+ def _enrich_dir_cook(res):
+ res['fetch_url'] = reverse('vault-fetch-revision_gitfast',
+ kwargs={'rev_id': rev_id})
+ return res
+
+ return api_lookup(
+ service.vault_cook, 'revision_gitfast', obj_id, email,
+ notfound_msg="Revision with ID '{}' not found.".format(rev_id),
+ enrich_fn=_enrich_dir_cook)
+
+
+@api_route('/vault/revision_gitfast/(?P<rev_id>[a-fA-F0-9]+)/raw/',
+ 'vault-fetch-revision_gitfast')
+@api_doc.route('/vault/revision_gitfast/raw', tags=['hidden'],
+ handle_response=True)
+@api_doc.arg('rev_id',
+ default='9174026cfe69d73ef80b27890615f8b2ef5c265a',
+ argtype=api_doc.argtypes.sha1_git,
+ argdoc="The revision's sha1_git identifier")
+@api_doc.raises(exc=api_doc.excs.badinput, doc=doc_exc_bad_id)
+@api_doc.raises(exc=api_doc.excs.notfound, doc=doc_exc_id_not_found)
+@api_doc.returns(rettype=api_doc.rettypes.octet_stream,
+ retdoc='the cooked revision git fast-export')
+def api_vault_fetch_revision_gitfast(request, rev_id):
+ """Fetch the archive of the revision identified by rev_id."""
+ _, obj_id = query.parse_hash_with_algorithms_or_throws(
+ rev_id, ['sha1'], 'Only sha1_git is supported.')
+ res = api_lookup(
+ service.vault_fetch, 'revision_gitfast', obj_id,
+ notfound_msg="Revision with ID '{}' not found.".format(rev_id))
+ fname = '{}.gitfast.gz'.format(rev_id)
+ response = HttpResponse(res, content_type='application/gzip')
+ response['Content-disposition'] = 'attachment; filename={}'.format(fname)
+ return response
diff --git a/swh/web/common/service.py b/swh/web/common/service.py
index 300f37da..93bbfbf8 100644
--- a/swh/web/common/service.py
+++ b/swh/web/common/service.py
@@ -1,828 +1,841 @@
# Copyright (C) 2015-2017 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
import os
from collections import defaultdict
from swh.model import hashutil
from swh.web.common import converters
from swh.web.common import query
from swh.web.common.exc import NotFoundExc
from swh.web import config
storage = config.storage()
+vault = config.vault()
MAX_LIMIT = 50 # Top limit the users can ask for
def _first_element(l):
"""Returns the first element in the provided list or None
if it is empty or None"""
return next(iter(l or []), None)
def lookup_multiple_hashes(hashes):
"""Lookup the passed hashes in a single DB connection, using batch
processing.
Args:
An array of {filename: X, sha1: Y}, string X, hex sha1 string Y.
Returns:
The same array with elements updated with elem['found'] = true if
the hash is present in storage, elem['found'] = false if not.
"""
hashlist = [hashutil.hash_to_bytes(elem['sha1']) for elem in hashes]
content_missing = storage.content_missing_per_sha1(hashlist)
missing = [hashutil.hash_to_hex(x) for x in content_missing]
for x in hashes:
x.update({'found': True})
for h in hashes:
if h['sha1'] in missing:
h['found'] = False
return hashes
def lookup_expression(expression, last_sha1, per_page):
"""Lookup expression in raw content.
Args:
expression (str): An expression to lookup through raw indexed
content
last_sha1 (str): Last sha1 seen
per_page (int): Number of results per page
Returns:
List of ctags whose content match the expression
"""
limit = min(per_page, MAX_LIMIT)
ctags = storage.content_ctags_search(expression,
last_sha1=last_sha1,
limit=limit)
for ctag in ctags:
ctag = converters.from_swh(ctag, hashess={'id'})
ctag['sha1'] = ctag['id']
ctag.pop('id')
yield ctag
def lookup_hash(q):
"""Checks if the storage contains a given content checksum
Args: query string of the form <hash_algo:hash>
Returns: Dict with key found containing the hash info if the
hash is present, None if not.
"""
algo, hash = query.parse_hash(q)
found = storage.content_find({algo: hash})
return {'found': found,
'algo': algo}
def search_hash(q):
"""Checks if the storage contains a given content checksum
Args: query string of the form <hash_algo:hash>
Returns: Dict with key found to True or False, according to
whether the checksum is present or not
"""
algo, hash = query.parse_hash(q)
found = storage.content_find({algo: hash})
return {'found': found is not None}
def lookup_content_provenance(q):
"""Return provenance information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
provenance information (dict) list if the content is found.
"""
algo, hash = query.parse_hash(q)
provenances = storage.content_find_provenance({algo: hash})
if not provenances:
return None
return (converters.from_provenance(p) for p in provenances)
def _lookup_content_sha1(q):
"""Given a possible input, query for the content's sha1.
Args:
q: query string of the form <hash_algo:hash>
Returns:
binary sha1 if found or None
"""
algo, hash = query.parse_hash(q)
if algo != 'sha1':
hashes = storage.content_find({algo: hash})
if not hashes:
return None
return hashes['sha1']
return hash
def lookup_content_ctags(q):
"""Return ctags information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
ctags information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
ctags = list(storage.content_ctags_get([sha1]))
if not ctags:
return None
for ctag in ctags:
yield converters.from_swh(ctag, hashess={'id'})
def lookup_content_filetype(q):
"""Return filetype information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
filetype information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
filetype = _first_element(list(storage.content_mimetype_get([sha1])))
if not filetype:
return None
return converters.from_filetype(filetype)
def lookup_content_language(q):
"""Return language information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
language information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
lang = _first_element(list(storage.content_language_get([sha1])))
if not lang:
return None
return converters.from_swh(lang, hashess={'id'})
def lookup_content_license(q):
"""Return license information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
license information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
lang = _first_element(storage.content_fossology_license_get([sha1]))
if not lang:
return None
return converters.from_swh(lang, hashess={'id'})
def lookup_origin(origin):
"""Return information about the origin matching dict origin.
Args:
origin: origin's dict with keys either 'id' or
('type' AND 'url')
Returns:
origin information as dict.
"""
origin_info = storage.origin_get(origin)
if not origin_info:
if 'id' in origin and origin['id']:
msg = 'Origin with id %s not found!' % origin['id']
else:
msg = 'Origin of type %s and url %s not found!' % \
(origin['type'], origin['url'])
raise NotFoundExc(msg)
return converters.from_origin(origin_info)
def lookup_person(person_id):
"""Return information about the person with id person_id.
Args:
person_id as string
Returns:
person information as dict.
"""
person = _first_element(storage.person_get([person_id]))
return converters.from_person(person)
def lookup_directory(sha1_git):
"""Return information about the directory with id sha1_git.
Args:
sha1_git as string
Returns:
directory information as dict.
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
sha1_git,
['sha1'], # HACK: sha1_git really
'Only sha1_git is supported.')
dir = _first_element(storage.directory_get([sha1_git_bin]))
if not dir:
raise NotFoundExc('Directory with sha1_git %s not found' % sha1_git)
directory_entries = storage.directory_ls(sha1_git_bin) or []
return map(converters.from_directory_entry, directory_entries)
def lookup_directory_with_path(directory_sha1_git, path_string):
"""Return directory information for entry with path path_string w.r.t.
root directory pointed by directory_sha1_git
Args:
- directory_sha1_git: sha1_git corresponding to the directory
to which we append paths to (hopefully) find the entry
- the relative path to the entry starting from the directory pointed by
directory_sha1_git
Raises:
NotFoundExc if the directory entry is not found
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
directory_sha1_git,
['sha1'],
'Only sha1_git is supported.')
paths = path_string.strip(os.path.sep).split(os.path.sep)
queried_dir = storage.directory_entry_get_by_path(
sha1_git_bin, list(map(lambda p: p.encode('utf-8'), paths)))
if not queried_dir:
raise NotFoundExc(('Directory entry with path %s from %s not found') %
(path_string, directory_sha1_git))
return converters.from_directory_entry(queried_dir)
def lookup_release(release_sha1_git):
"""Return information about the release with sha1 release_sha1_git.
Args:
release_sha1_git: The release's sha1 as hexadecimal
Returns:
Release information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
release_sha1_git,
['sha1'],
'Only sha1_git is supported.')
res = _first_element(storage.release_get([sha1_git_bin]))
return converters.from_release(res)
def lookup_revision(rev_sha1_git):
"""Return information about the revision with sha1 revision_sha1_git.
Args:
revision_sha1_git: The revision's sha1 as hexadecimal
Returns:
Revision information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
rev_sha1_git,
['sha1'],
'Only sha1_git is supported.')
revision = _first_element(storage.revision_get([sha1_git_bin]))
return converters.from_revision(revision)
def lookup_revision_multiple(sha1_git_list):
"""Return information about the revision with sha1 revision_sha1_git.
Args:
revision_sha1_git: The revision's sha1 as hexadecimal
Returns:
Revision information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
def to_sha1_bin(sha1_hex):
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
sha1_hex,
['sha1'],
'Only sha1_git is supported.')
return sha1_git_bin
sha1_bin_list = (to_sha1_bin(x) for x in sha1_git_list)
revisions = storage.revision_get(sha1_bin_list) or []
return (converters.from_revision(x) for x in revisions)
def lookup_revision_message(rev_sha1_git):
"""Return the raw message of the revision with sha1 revision_sha1_git.
Args:
revision_sha1_git: The revision's sha1 as hexadecimal
Returns:
Decoded revision message as dict {'message': <the_message>}
Raises:
ValueError if the identifier provided is not of sha1 nature.
NotFoundExc if the revision is not found, or if it has no message
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
rev_sha1_git,
['sha1'],
'Only sha1_git is supported.')
revision = _first_element(storage.revision_get([sha1_git_bin]))
if not revision:
raise NotFoundExc('Revision with sha1_git %s not found.'
% rev_sha1_git)
if 'message' not in revision:
raise NotFoundExc('No message for revision with sha1_git %s.'
% rev_sha1_git)
res = {'message': revision['message']}
return res
def lookup_revision_by(origin_id,
branch_name="refs/heads/master",
timestamp=None):
"""Lookup revisions by origin_id, branch_name and timestamp.
If:
- branch_name is not provided, lookup using 'refs/heads/master' as default.
- ts is not provided, use the most recent
Args:
- origin_id: origin of the revision.
- branch_name: revision's branch.
- timestamp: revision's time frame.
Yields:
The revisions matching the criterions.
"""
res = _first_element(storage.revision_get_by(origin_id,
branch_name,
timestamp=timestamp,
limit=1))
return converters.from_revision(res)
def lookup_revision_log(rev_sha1_git, limit):
"""Return information about the revision with sha1 revision_sha1_git.
Args:
revision_sha1_git: The revision's sha1 as hexadecimal
limit: the maximum number of revisions returned
Returns:
Revision information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
rev_sha1_git,
['sha1'],
'Only sha1_git is supported.')
revision_entries = storage.revision_log([sha1_git_bin], limit)
return map(converters.from_revision, revision_entries)
def lookup_revision_log_by(origin_id, branch_name, timestamp, limit):
"""Return information about the revision with sha1 revision_sha1_git.
Args:
origin_id: origin of the revision
branch_name: revision's branch
timestamp: revision's time frame
limit: the maximum number of revisions returned
Returns:
Revision information as dict.
Raises:
NotFoundExc if no revision corresponds to the criterion
NotFoundExc if the corresponding revision has no log
"""
revision_entries = storage.revision_log_by(origin_id,
branch_name,
timestamp,
limit=limit)
if not revision_entries:
return None
return map(converters.from_revision, revision_entries)
def lookup_revision_with_context_by(origin_id, branch_name, ts, sha1_git,
limit=100):
"""Return information about revision sha1_git, limited to the
sub-graph of all transitive parents of sha1_git_root.
sha1_git_root being resolved through the lookup of a revision by origin_id,
branch_name and ts.
In other words, sha1_git is an ancestor of sha1_git_root.
Args:
- origin_id: origin of the revision.
- branch_name: revision's branch.
- timestamp: revision's time frame.
- sha1_git: one of sha1_git_root's ancestors.
- limit: limit the lookup to 100 revisions back.
Returns:
Pair of (root_revision, revision).
Information on sha1_git if it is an ancestor of sha1_git_root
including children leading to sha1_git_root
Raises:
- BadInputExc in case of unknown algo_hash or bad hash.
- NotFoundExc if either revision is not found or if sha1_git is not an
ancestor of sha1_git_root.
"""
rev_root = _first_element(storage.revision_get_by(origin_id,
branch_name,
timestamp=ts,
limit=1))
if not rev_root:
raise NotFoundExc('Revision with (origin_id: %s, branch_name: %s'
', ts: %s) not found.' % (origin_id,
branch_name,
ts))
return (converters.from_revision(rev_root),
lookup_revision_with_context(rev_root, sha1_git, limit))
def lookup_revision_with_context(sha1_git_root, sha1_git, limit=100):
"""Return information about revision sha1_git, limited to the
sub-graph of all transitive parents of sha1_git_root.
In other words, sha1_git is an ancestor of sha1_git_root.
Args:
sha1_git_root: latest revision. The type is either a sha1 (as an hex
string) or a non converted dict.
sha1_git: one of sha1_git_root's ancestors
limit: limit the lookup to 100 revisions back
Returns:
Information on sha1_git if it is an ancestor of sha1_git_root
including children leading to sha1_git_root
Raises:
BadInputExc in case of unknown algo_hash or bad hash
NotFoundExc if either revision is not found or if sha1_git is not an
ancestor of sha1_git_root
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
sha1_git,
['sha1'],
'Only sha1_git is supported.')
revision = _first_element(storage.revision_get([sha1_git_bin]))
if not revision:
raise NotFoundExc('Revision %s not found' % sha1_git)
if isinstance(sha1_git_root, str):
_, sha1_git_root_bin = query.parse_hash_with_algorithms_or_throws(
sha1_git_root,
['sha1'],
'Only sha1_git is supported.')
revision_root = _first_element(storage.revision_get([sha1_git_root_bin])) # noqa
if not revision_root:
raise NotFoundExc('Revision root %s not found' % sha1_git_root)
else:
sha1_git_root_bin = sha1_git_root['id']
revision_log = storage.revision_log([sha1_git_root_bin], limit)
parents = {}
children = defaultdict(list)
for rev in revision_log:
rev_id = rev['id']
parents[rev_id] = []
for parent_id in rev['parents']:
parents[rev_id].append(parent_id)
children[parent_id].append(rev_id)
if revision['id'] not in parents:
raise NotFoundExc('Revision %s is not an ancestor of %s' %
(sha1_git, sha1_git_root))
revision['children'] = children[revision['id']]
return converters.from_revision(revision)
def lookup_directory_with_revision(sha1_git, dir_path=None, with_data=False):
"""Return information on directory pointed by revision with sha1_git.
If dir_path is not provided, display top level directory.
Otherwise, display the directory pointed by dir_path (if it exists).
Args:
sha1_git: revision's hash.
dir_path: optional directory pointed to by that revision.
with_data: boolean that indicates to retrieve the raw data if the path
resolves to a content. Default to False (for the api)
Returns:
Information on the directory pointed to by that revision.
Raises:
BadInputExc in case of unknown algo_hash or bad hash.
NotFoundExc either if the revision is not found or the path referenced
does not exist.
NotImplementedError in case of dir_path exists but do not reference a
type 'dir' or 'file'.
"""
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
sha1_git,
['sha1'],
'Only sha1_git is supported.')
revision = _first_element(storage.revision_get([sha1_git_bin]))
if not revision:
raise NotFoundExc('Revision %s not found' % sha1_git)
dir_sha1_git_bin = revision['directory']
if dir_path:
paths = dir_path.strip(os.path.sep).split(os.path.sep)
entity = storage.directory_entry_get_by_path(
dir_sha1_git_bin, list(map(lambda p: p.encode('utf-8'), paths)))
if not entity:
raise NotFoundExc(
"Directory or File '%s' pointed to by revision %s not found"
% (dir_path, sha1_git))
else:
entity = {'type': 'dir', 'target': dir_sha1_git_bin}
if entity['type'] == 'dir':
directory_entries = storage.directory_ls(entity['target']) or []
return {'type': 'dir',
'path': '.' if not dir_path else dir_path,
'revision': sha1_git,
'content': map(converters.from_directory_entry,
directory_entries)}
elif entity['type'] == 'file': # content
content = storage.content_find({'sha1_git': entity['target']})
if with_data:
c = _first_element(storage.content_get([content['sha1']]))
content['data'] = c['data']
return {'type': 'file',
'path': '.' if not dir_path else dir_path,
'revision': sha1_git,
'content': converters.from_content(content)}
else:
raise NotImplementedError('Entity of type %s not implemented.'
% entity['type'])
def lookup_content(q):
"""Lookup the content designed by q.
Args:
q: The release's sha1 as hexadecimal
"""
algo, hash = query.parse_hash(q)
c = storage.content_find({algo: hash})
return converters.from_content(c)
def lookup_content_raw(q):
"""Lookup the content defined by q.
Args:
q: query string of the form <hash_algo:hash>
Returns:
dict with 'sha1' and 'data' keys.
data representing its raw data decoded.
Raises:
NotFoundExc if the requested content is not found
"""
algo, hash = query.parse_hash(q)
c = storage.content_find({algo: hash})
if not c:
raise NotFoundExc('Content with %s checksum equals to %s not found!' %
(algo, hashutil.hash_to_hex(hash)))
content = _first_element(storage.content_get([c['sha1']]))
return converters.from_content(content)
def stat_counters():
"""Return the stat counters for Software Heritage
Returns:
A dict mapping textual labels to integer values.
"""
return storage.stat_counters()
def _lookup_origin_visits(origin_id, last_visit=None, limit=10):
"""Yields the origin origin_ids' visits.
Args:
origin_id (int): origin to list visits for
last_visit (int): last visit to lookup from
limit (int): Number of elements max to display
Yields:
Dictionaries of origin_visit for that origin
"""
limit = min(limit, MAX_LIMIT)
yield from storage.origin_visit_get(
origin_id, last_visit=last_visit, limit=limit)
def lookup_origin_visits(origin_id, last_visit=None, per_page=10):
"""Yields the origin origin_ids' visits.
Args:
origin_id: origin to list visits for
Yields:
Dictionaries of origin_visit for that origin
"""
visits = _lookup_origin_visits(origin_id, last_visit=last_visit,
limit=per_page)
for visit in visits:
yield converters.from_origin_visit(visit)
def lookup_origin_visit(origin_id, visit_id):
"""Return information about visit visit_id with origin origin_id.
Args:
origin_id: origin concerned by the visit
visit_id: the visit identifier to lookup
Yields:
The dict origin_visit concerned
"""
visit = storage.origin_visit_get_by(origin_id, visit_id)
if not visit:
raise NotFoundExc('Origin with id %s or its visit '
'with id %s not found!' % (origin_id, visit_id))
return converters.from_origin_visit(visit)
def lookup_entity_by_uuid(uuid):
"""Return the entity's hierarchy from its uuid.
Args:
uuid: entity's identifier.
Returns:
List of hierarchy entities from the entity with uuid.
"""
uuid = query.parse_uuid4(uuid)
for entity in storage.entity_get(uuid):
entity = converters.from_swh(entity,
convert={'last_seen', 'uuid'},
convert_fn=lambda x: str(x))
yield entity
def lookup_revision_through(revision, limit=100):
"""Retrieve a revision from the criterion stored in revision dictionary.
Args:
revision: Dictionary of criterion to lookup the revision with.
Here are the supported combination of possible values:
- origin_id, branch_name, ts, sha1_git
- origin_id, branch_name, ts
- sha1_git_root, sha1_git
- sha1_git
Returns:
None if the revision is not found or the actual revision.
"""
if 'origin_id' in revision and \
'branch_name' in revision and \
'ts' in revision and \
'sha1_git' in revision:
return lookup_revision_with_context_by(revision['origin_id'],
revision['branch_name'],
revision['ts'],
revision['sha1_git'],
limit)
if 'origin_id' in revision and \
'branch_name' in revision and \
'ts' in revision:
return lookup_revision_by(revision['origin_id'],
revision['branch_name'],
revision['ts'])
if 'sha1_git_root' in revision and \
'sha1_git' in revision:
return lookup_revision_with_context(revision['sha1_git_root'],
revision['sha1_git'],
limit)
if 'sha1_git' in revision:
return lookup_revision(revision['sha1_git'])
# this should not happen
raise NotImplementedError('Should not happen!')
def lookup_directory_through_revision(revision, path=None,
limit=100, with_data=False):
"""Retrieve the directory information from the revision.
Args:
revision: dictionary of criterion representing a revision to lookup
path: directory's path to lookup.
limit: optional query parameter to limit the revisions log (default to
100). For now, note that this limit could impede the transitivity
conclusion about sha1_git not being an ancestor of.
with_data: indicate to retrieve the content's raw data if path resolves
to a content.
Returns:
The directory pointing to by the revision criterions at path.
"""
rev = lookup_revision_through(revision, limit)
if not rev:
raise NotFoundExc('Revision with criterion %s not found!' % revision)
return (rev['id'],
lookup_directory_with_revision(rev['id'], path, with_data))
+
+
+def vault_cook(obj_type, obj_id, email=None):
+ """Cook a vault bundle.
+ """
+ return vault.cook(obj_type, obj_id, email=email)
+
+
+def vault_fetch(obj_type, obj_id):
+ """Fetch a vault bundle.
+ """
+ return vault.fetch(obj_type, obj_id)
diff --git a/swh/web/config.py b/swh/web/config.py
index 7a534e43..a11eecd4 100644
--- a/swh/web/config.py
+++ b/swh/web/config.py
@@ -1,55 +1,65 @@
# Copyright (C) 2017 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU General Public License version 3, or any later version
# See top-level LICENSE file for more information
from swh.core import config
from swh.storage import get_storage
+from swh.vault.api.client import RemoteVaultClient
DEFAULT_CONFIG = {
'allowed_hosts': ('list', []),
'storage': ('dict', {
'cls': 'remote',
'args': {
'url': 'http://127.0.0.1:5002/',
},
}),
+ 'vault': ('string', 'http://127.0.0.1:5005/'),
'log_dir': ('string', '/tmp/swh/log'),
'debug': ('bool', False),
'host': ('string', '127.0.0.1'), # development property
'port': ('int', 5003), # development property
'secret_key': ('string', 'development key'),
'throttling': ('dict', {
'cache_uri': None, # production: memcached as cache (127.0.0.1:11211)
# development: in-memory cache so None
'scopes': {
'swh_api': {
'limiter_rate': '120/h',
'exempted_networks': ['127.0.0.0/8']
}
}
})
}
swhweb_config = {}
def get_config(config_file='webapp/webapp'):
"""Read the configuration file `config_file`, update the app with
parameters (secret_key, conf) and return the parsed configuration as a
dict. If no configuration file is provided, return a default
configuration."""
if not swhweb_config:
cfg = config.load_named_config(config_file, DEFAULT_CONFIG)
swhweb_config.update(cfg)
config.prepare_folders(swhweb_config, 'log_dir')
swhweb_config['storage'] = get_storage(**swhweb_config['storage'])
+ swhweb_config['vault'] = RemoteVaultClient(swhweb_config['vault'])
return swhweb_config
def storage():
"""Return the current application's SWH storage.
"""
return get_config()['storage']
+
+
+def vault():
+ """Return the current application's SWH vault.
+
+ """
+ return get_config()['vault']