diff --git a/PKG-INFO b/PKG-INFO
index 92654f668ba1da7d7dd0a57ff3bc8312bde790ae..5eff1167ae860b2b06b3a72adab41ae9e12ff384 100644
--- a/PKG-INFO
+++ b/PKG-INFO
@@ -1,14 +1,14 @@
Metadata-Version: 2.1
Name: swh.model
-Version: 0.0.31
+Version: 0.0.32
Summary: Software Heritage data model
Home-page: https://forge.softwareheritage.org/diffusion/DMOD/
Author: Software Heritage developers
Author-email: swh-devel@inria.fr
License: UNKNOWN
-Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest
Project-URL: Funding, https://www.softwareheritage.org/donate
Project-URL: Source, https://forge.softwareheritage.org/source/swh-model
+Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest
Description: swh-model
=========
diff --git a/requirements.txt b/requirements.txt
index 447def3ef89e2d70a432a96982e62af6142e9d71..cd97184a9ab1826ff308c7e3c5806598309a4691 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -3,3 +3,5 @@
# dependency lines, see https://pip.readthedocs.org/en/1.1/requirements.html
vcversioner
Click
+attrs
+hypothesis
diff --git a/swh.model.egg-info/PKG-INFO b/swh.model.egg-info/PKG-INFO
index 92654f668ba1da7d7dd0a57ff3bc8312bde790ae..5eff1167ae860b2b06b3a72adab41ae9e12ff384 100644
--- a/swh.model.egg-info/PKG-INFO
+++ b/swh.model.egg-info/PKG-INFO
@@ -1,14 +1,14 @@
Metadata-Version: 2.1
Name: swh.model
-Version: 0.0.31
+Version: 0.0.32
Summary: Software Heritage data model
Home-page: https://forge.softwareheritage.org/diffusion/DMOD/
Author: Software Heritage developers
Author-email: swh-devel@inria.fr
License: UNKNOWN
-Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest
Project-URL: Funding, https://www.softwareheritage.org/donate
Project-URL: Source, https://forge.softwareheritage.org/source/swh-model
+Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest
Description: swh-model
=========
diff --git a/swh.model.egg-info/SOURCES.txt b/swh.model.egg-info/SOURCES.txt
index 99760038e42434d949912c64782666b9c2fc3bbd..7ca184a9f0997bac93f4d7cec1af327645ca6d11 100644
--- a/swh.model.egg-info/SOURCES.txt
+++ b/swh.model.egg-info/SOURCES.txt
@@ -17,8 +17,10 @@ swh/model/cli.py
swh/model/exceptions.py
swh/model/from_disk.py
swh/model/hashutil.py
+swh/model/hypothesis_strategies.py
swh/model/identifiers.py
swh/model/merkle.py
+swh/model/model.py
swh/model/toposort.py
swh/model/validators.py
swh/model/fields/__init__.py
@@ -30,6 +32,7 @@ swh/model/tests/generate_testdata_from_disk.py
swh/model/tests/test_cli.py
swh/model/tests/test_from_disk.py
swh/model/tests/test_hashutil.py
+swh/model/tests/test_hypothesis_strategies.py
swh/model/tests/test_identifiers.py
swh/model/tests/test_merkle.py
swh/model/tests/test_toposort.py
diff --git a/swh.model.egg-info/requires.txt b/swh.model.egg-info/requires.txt
index 30a1eca133ddf171cece1e272cb7e6c3f9b2d2a2..674d2255919a55a09a4a2f5d3f1d33573eb9da14 100644
--- a/swh.model.egg-info/requires.txt
+++ b/swh.model.egg-info/requires.txt
@@ -1,5 +1,7 @@
vcversioner
Click
+attrs
+hypothesis
[testing]
pytest
diff --git a/swh/model/hypothesis_strategies.py b/swh/model/hypothesis_strategies.py
new file mode 100644
index 0000000000000000000000000000000000000000..e2ca4c60bcdb6cab2c5b3c48683631289e52f465
--- /dev/null
+++ b/swh/model/hypothesis_strategies.py
@@ -0,0 +1,191 @@
+# Copyright (C) 2019 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 hypothesis.strategies import (
+ lists, one_of, composite, builds, integers, sampled_from, binary,
+ dictionaries, none, from_regex, just
+)
+
+
+from .from_disk import DentryPerms
+from .model import (
+ Person, Timestamp, TimestampWithTimezone, Origin, OriginVisit,
+ Snapshot, SnapshotBranch, TargetType, Release, Revision,
+ Directory, DirectoryEntry, Content
+)
+from .identifiers import snapshot_identifier, identifier_to_bytes
+
+
+def sha1_git():
+ return binary(min_size=20, max_size=20)
+
+
+@composite
+def urls(draw):
+ protocol = draw(sampled_from(['git', 'http', 'https', 'deb']))
+ domain = draw(from_regex(r'\A([a-z]([a-z0-9-]*)\.){1,3}[a-z0-9]+\Z'))
+
+ return '%s://%s' % (protocol, domain)
+
+
+def persons():
+ return builds(Person)
+
+
+def timestamps():
+ return builds(
+ Timestamp,
+ seconds=integers(-2**63, 2**63-1),
+ microseconds=integers(0, 1000000))
+
+
+def timestamps_with_timezone():
+ return builds(
+ TimestampWithTimezone,
+ timestamp=timestamps(),
+ offset=integers(-2**16, 2**16-1))
+
+
+def origins():
+ return builds(
+ Origin,
+ type=sampled_from(['git', 'hg', 'svn', 'pypi', 'deb']),
+ url=urls())
+
+
+def origin_visits():
+ return builds(
+ OriginVisit,
+ visit=integers(0, 1000),
+ origin=origins())
+
+
+def releases():
+ return builds(
+ Release,
+ id=sha1_git(),
+ date=timestamps_with_timezone(),
+ author=one_of(none(), persons()),
+ target=one_of(none(), sha1_git()))
+
+
+def revisions():
+ return builds(
+ Revision,
+ id=sha1_git(),
+ date=timestamps_with_timezone(),
+ committer_date=timestamps_with_timezone(),
+ parents=lists(binary()),
+ directory=binary(),
+ metadata=one_of(none(), dictionaries(binary(), binary())))
+
+
+def directory_entries():
+ return builds(
+ DirectoryEntry,
+ target=sha1_git(),
+ perms=sampled_from([perm.value for perm in DentryPerms]))
+
+
+def directories():
+ return builds(
+ Directory,
+ id=sha1_git(),
+ entries=lists(directory_entries()))
+
+
+def contents():
+ def filter_data(content):
+ if content.status != 'visible':
+ content.data = None
+ return content
+
+ return builds(
+ Content,
+ length=integers(0),
+ data=binary(),
+ sha1_git=sha1_git(),
+ ).map(filter_data)
+
+
+def branch_names():
+ return binary()
+
+
+def branch_targets_object():
+ return builds(
+ SnapshotBranch,
+ target=sha1_git(),
+ target_type=sampled_from([
+ TargetType.CONTENT, TargetType.DIRECTORY, TargetType.REVISION,
+ TargetType.RELEASE, TargetType.SNAPSHOT]))
+
+
+def branch_targets_alias():
+ return builds(
+ SnapshotBranch,
+ target_type=just(TargetType.ALIAS))
+
+
+def branch_targets(*, only_objects=False):
+ if only_objects:
+ return branch_targets_object()
+ else:
+ return one_of(branch_targets_alias(), branch_targets_object())
+
+
+@composite
+def snapshots(draw, *, min_size=0, max_size=100, only_objects=False):
+ branches = draw(dictionaries(
+ keys=branch_names(),
+ values=branch_targets(only_objects=only_objects),
+ min_size=min_size,
+ max_size=max_size,
+ ))
+
+ if not only_objects:
+ # Make sure aliases point to actual branches
+ unresolved_aliases = {
+ target.target
+ for target in branches.values()
+ if (target
+ and target.target_type == 'alias'
+ and target.target not in branches)
+ }
+
+ for alias in unresolved_aliases:
+ branches[alias] = draw(branch_targets(only_objects=True))
+
+ while True:
+ try:
+ id_ = snapshot_identifier({
+ 'branches': {
+ name: branch.to_dict()
+ for (name, branch) in branches.items()}})
+ except ValueError as e:
+ print(e.args)
+ for (source, target) in e.args[1]:
+ branches[source] = draw(branch_targets(only_objects=True))
+ else:
+ break
+ return Snapshot(
+ id=identifier_to_bytes(id_),
+ branches=branches)
+
+
+def objects():
+ return one_of(
+ origins().map(lambda x: ('origin', x)),
+ origin_visits().map(lambda x: ('origin_visit', x)),
+ snapshots().map(lambda x: ('snapshot', x)),
+ releases().map(lambda x: ('release', x)),
+ revisions().map(lambda x: ('revision', x)),
+ directories().map(lambda x: ('directory', x)),
+ contents().map(lambda x: ('content', x)),
+ )
+
+
+def object_dicts():
+ return objects().map(lambda x: (x[0], x[1].to_dict()))
diff --git a/swh/model/identifiers.py b/swh/model/identifiers.py
index e7508ebcbbb4fbe05d1a5610725a2412750498ea..486639e63560236c2e2b842623dd6e67f1fd0dc1 100644
--- a/swh/model/identifiers.py
+++ b/swh/model/identifiers.py
@@ -581,6 +581,8 @@ def snapshot_identifier(snapshot, *, ignore_unresolved=False):
if target_id not in snapshot['branches'] or target_id == name:
unresolved.append((name, target_id))
else:
+ print(name)
+ print(target)
target_type = target['target_type'].encode()
target_id = identifier_to_bytes(target['target'])
diff --git a/swh/model/model.py b/swh/model/model.py
new file mode 100644
index 0000000000000000000000000000000000000000..890d1330a4eccc77110c9a115f8137e71e79e83a
--- /dev/null
+++ b/swh/model/model.py
@@ -0,0 +1,205 @@
+# Copyright (C) 2018-2019 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 datetime
+from enum import Enum
+from typing import List, Optional, Dict
+
+import attr
+
+
+# TODO: Limit this to 20 bytes
+Sha1Git = bytes
+
+
+@attr.s
+class Person:
+ name = attr.ib(type=bytes)
+ email = attr.ib(type=bytes)
+ fullname = attr.ib(type=bytes)
+
+
+@attr.s
+class Timestamp:
+ seconds = attr.ib(type=int)
+ microseconds = attr.ib(type=int)
+
+ @seconds.validator
+ def check_seconds(self, attribute, value):
+ """Check that seconds fit in a 64-bits signed integer."""
+ if not (-2**63 <= value < 2**63):
+ raise ValueError('Seconds must be a signed 64-bits integer.')
+
+ @microseconds.validator
+ def check_microseconds(self, attribute, value):
+ """Checks that microseconds are positive and < 1000000."""
+ if not (0 <= value < 10**6):
+ raise ValueError('Microseconds must be in [0, 1000000[.')
+
+
+@attr.s
+class TimestampWithTimezone:
+ timestamp = attr.ib(type=Timestamp)
+ offset = attr.ib(type=int)
+ negative_utc = attr.ib(type=bool)
+
+ def to_dict(self):
+ return attr.asdict(self)
+
+
+@attr.s
+class Origin:
+ type = attr.ib(type=str)
+ url = attr.ib(type=str)
+
+ def to_dict(self):
+ return attr.asdict(self)
+
+
+@attr.s
+class OriginVisit:
+ origin = attr.ib(type=Origin)
+ date = attr.ib(type=datetime.datetime)
+ visit = attr.ib(type=Optional[int])
+ """Should not be set before calling 'origin_visit_add()'."""
+
+ def to_dict(self):
+ ov = attr.asdict(self)
+ ov['origin'] = self.origin.to_dict()
+ ov['date'] = str(self.date)
+ if not ov['visit']:
+ del ov['visit']
+ return ov
+
+
+class TargetType(Enum):
+ CONTENT = 'content'
+ DIRECTORY = 'directory'
+ REVISION = 'revision'
+ RELEASE = 'release'
+ SNAPSHOT = 'snapshot'
+ ALIAS = 'alias'
+
+
+@attr.s
+class SnapshotBranch:
+ target = attr.ib(type=bytes)
+ target_type = attr.ib(type=TargetType)
+
+ @target.validator
+ def check_target(self, attribute, value):
+ if self.target_type != TargetType.ALIAS:
+ if len(value) != 20:
+ raise ValueError('Wrong length for bytes identifier: %d' %
+ len(value))
+
+ def to_dict(self):
+ branch = attr.asdict(self)
+ branch['target_type'] = branch['target_type'].value
+ return branch
+
+
+@attr.s
+class Snapshot:
+ id = attr.ib(type=Sha1Git)
+ branches = attr.ib(type=Dict[bytes, Optional[SnapshotBranch]])
+
+ def to_dict(self):
+ return {
+ 'id': self.id,
+ 'branches': {
+ name: branch.to_dict()
+ for (name, branch) in self.branches.items()
+ }
+ }
+
+
+@attr.s
+class Release:
+ id = attr.ib(type=Sha1Git)
+ name = attr.ib(type=bytes)
+ message = attr.ib(type=bytes)
+ date = attr.ib(type=TimestampWithTimezone)
+ author = attr.ib(type=Optional[Person])
+ target = attr.ib(type=Optional[Sha1Git])
+ target_type = attr.ib(type=TargetType)
+ synthetic = attr.ib(type=bool)
+
+ def to_dict(self):
+ rel = attr.asdict(self)
+ rel['date'] = self.date.to_dict()
+ rel['target_type'] = rel['target_type'].value
+ return rel
+
+
+@attr.s
+class Revision:
+ id = attr.ib(type=Sha1Git)
+ message = attr.ib(type=bytes)
+ author = attr.ib(type=Person)
+ committer = attr.ib(type=Person)
+ date = attr.ib(type=TimestampWithTimezone)
+ committer_date = attr.ib(type=TimestampWithTimezone)
+ parents = attr.ib(type=List[Sha1Git])
+ type = attr.ib(type=str)
+ directory = attr.ib(type=Sha1Git)
+ metadata = attr.ib(type=Optional[dict])
+ synthetic = attr.ib(type=bool)
+
+ def to_dict(self):
+ rev = attr.asdict(self)
+ rev['date'] = self.date.to_dict()
+ rev['committer_date'] = self.committer_date.to_dict()
+ return rev
+
+
+@attr.s
+class DirectoryEntry:
+ name = attr.ib(type=bytes)
+ type = attr.ib(type=str,
+ validator=attr.validators.in_(['file', 'dir', 'rev']))
+ target = attr.ib(type=Sha1Git)
+ perms = attr.ib(type=int)
+ """Usually one of the values of `swh.model.from_disk.DentryPerms`."""
+
+ def to_dict(self):
+ return attr.asdict(self)
+
+
+@attr.s
+class Directory:
+ id = attr.ib(type=Sha1Git)
+ entries = attr.ib(type=List[DirectoryEntry])
+
+ def to_dict(self):
+ dir_ = attr.asdict(self)
+ dir_['entries'] = [entry.to_dict() for entry in self.entries]
+ return dir_
+
+
+@attr.s
+class Content:
+ sha1 = attr.ib(type=bytes)
+ sha1_git = attr.ib(type=Sha1Git)
+ sha256 = attr.ib(type=bytes)
+ blake2s256 = attr.ib(type=bytes)
+
+ data = attr.ib(type=bytes)
+ length = attr.ib(type=int)
+ status = attr.ib(
+ type=str,
+ validator=attr.validators.in_(['visible', 'absent', 'hidden']))
+
+ @length.validator
+ def check_length(self, attribute, value):
+ """Checks the length is positive."""
+ if value < 0:
+ raise ValueError('Length must be positive.')
+
+ def to_dict(self):
+ content = attr.asdict(self)
+ if content['data'] is None:
+ del content['data']
+ return content
diff --git a/swh/model/tests/test_hypothesis_strategies.py b/swh/model/tests/test_hypothesis_strategies.py
new file mode 100644
index 0000000000000000000000000000000000000000..52e1211abcd2124efe7b39265e218b92e9cdc6ec
--- /dev/null
+++ b/swh/model/tests/test_hypothesis_strategies.py
@@ -0,0 +1,38 @@
+# Copyright (C) 2019 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 attr
+from hypothesis import given
+
+from swh.model.hashutil import DEFAULT_ALGORITHMS
+from swh.model.hypothesis_strategies import objects, object_dicts
+
+
+target_types = (
+ 'content', 'directory', 'revision', 'release', 'snapshot', 'alias')
+
+
+@given(objects())
+def test_generation(obj_type_and_obj):
+ (obj_type, object_) = obj_type_and_obj
+ attr.validate(object_)
+
+
+@given(object_dicts())
+def test_dicts_generation(obj_type_and_obj):
+ (obj_type, object_) = obj_type_and_obj
+ assert isinstance(object_, dict)
+ if obj_type == 'content':
+ if object_['status'] == 'visible':
+ assert set(object_) == \
+ set(DEFAULT_ALGORITHMS) | {'length', 'status', 'data'}
+ else:
+ assert set(object_) == \
+ set(DEFAULT_ALGORITHMS) | {'length', 'status'}
+ elif obj_type == 'release':
+ assert object_['target_type'] in target_types
+ elif obj_type == 'snapshot':
+ for branch in object_['branches'].values():
+ assert branch['target_type'] in target_types
diff --git a/version.txt b/version.txt
index 60d42029e96d527ee76fe3586c8cb270128e5a07..808b7d85a5dda3aba60ded8962aad41515e9e0d0 100644
--- a/version.txt
+++ b/version.txt
@@ -1 +1 @@
-v0.0.31-0-g4d40f4d
\ No newline at end of file
+v0.0.32-0-gd1b2156
\ No newline at end of file