diff --git a/swh/loader/core/converters.py b/swh/loader/core/converters.py index c6ebb86..148cc97 100644 --- a/swh/loader/core/converters.py +++ b/swh/loader/core/converters.py @@ -1,229 +1,228 @@ # Copyright (C) 2015-2016 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 """Convert objects to dictionaries suitable for swh.storage""" import datetime import os from swh.model.hashutil import hash_to_hex from swh.model import git def to_datetime(ts): """Convert a timestamp to utc datetime. """ return datetime.datetime.utcfromtimestamp(ts).replace( tzinfo=datetime.timezone.utc) def format_to_minutes(offset_str): """Convert a git string timezone format string (e.g +0200, -0310) to minutes. Args: offset_str: a string representing an offset. Returns: A positive or negative number of minutes of such input """ sign = offset_str[0] hours = int(offset_str[1:3]) minutes = int(offset_str[3:]) + (hours * 60) return minutes if sign == '+' else -1 * minutes def blob_to_content(obj, log=None, max_content_size=None, origin_id=None): """Convert obj to a swh storage content. Note: - If obj represents a link, the length and data are already provided so we use them directly. - 'data' is returned only if max_content_size is not reached. Returns: obj converted to content as a dictionary. """ filepath = obj['path'] if 'length' in obj: # link already has it size = obj['length'] else: size = os.lstat(filepath).st_size ret = { 'sha1': obj['sha1'], 'sha256': obj['sha256'], 'sha1_git': obj['sha1_git'], 'length': size, 'perms': obj['perms'].value, 'type': obj['type'].value, } if max_content_size and size > max_content_size: if log: log.info('Skipping content %s, too large (%s > %s)' % (hash_to_hex(obj['sha1_git']), size, max_content_size)) ret.update({'status': 'absent', 'reason': 'Content too large', 'origin': origin_id}) return ret if 'data' in obj: # link already has it data = obj['data'] else: data = open(filepath, 'rb').read() ret.update({ 'data': data, 'status': 'visible' }) return ret # Map of type to swh types _entry_type_map = { git.GitType.TREE: 'dir', git.GitType.BLOB: 'file', git.GitType.COMM: 'rev', } def tree_to_directory(tree, log=None): """Format a tree as a directory """ entries = [] for entry in tree['children']: entries.append({ 'type': _entry_type_map[entry['type']], 'perms': int(entry['perms'].value), 'name': entry['name'], 'target': entry['sha1_git'] }) return { 'id': tree['sha1_git'], 'entries': entries } -def commit_to_revision(commit, objects, log=None): +def commit_to_revision(commit, log=None): """Format a commit as a revision. """ - upper_directory = objects[git.ROOT_TREE_KEY][0] return { 'date': { 'timestamp': commit['author_date'], 'offset': format_to_minutes(commit['author_offset']), }, 'committer_date': { 'timestamp': commit['committer_date'], 'offset': format_to_minutes(commit['committer_offset']), }, 'type': commit['type'], - 'directory': upper_directory['sha1_git'], + 'directory': commit['directory'], 'message': commit['message'].encode('utf-8'), 'author': { 'name': commit['author_name'].encode('utf-8'), 'email': commit['author_email'].encode('utf-8'), }, 'committer': { 'name': commit['committer_name'].encode('utf-8'), 'email': commit['committer_email'].encode('utf-8'), }, 'synthetic': True, 'metadata': commit['metadata'], 'parents': [], } def annotated_tag_to_release(release, log=None): """Format a swh release. """ return { 'target': release['target'], 'target_type': release['target_type'], 'name': release['name'].encode('utf-8'), 'message': release['comment'].encode('utf-8'), 'date': { 'timestamp': release['date'], 'offset': format_to_minutes(release['offset']), }, 'author': { 'name': release['author_name'].encode('utf-8'), 'email': release['author_email'].encode('utf-8'), }, 'synthetic': True, } def ref_to_occurrence(ref): """Format a reference as an occurrence""" occ = ref.copy() if 'branch' in ref: branch = ref['branch'] if isinstance(branch, str): occ['branch'] = branch.encode('utf-8') else: occ['branch'] = branch return occ def shallow_blob(obj): """Convert a full swh content/blob to just what's needed by swh-storage for filtering. Returns: A shallow copy of a full swh content/blob object. """ return { 'sha1': obj['sha1'], 'sha256': obj['sha256'], 'sha1_git': obj['sha1_git'], 'length': obj['length'] } def shallow_tree(tree): """Convert a full swh directory/tree to just what's needed by swh-storage for filtering. Returns: A shallow copy of a full swh directory/tree object. """ return tree['sha1_git'] def shallow_commit(commit): """Convert a full swh revision/commit to just what's needed by swh-storage for filtering. Returns: A shallow copy of a full swh revision/commit object. """ return commit['id'] def shallow_tag(tag): """Convert a full swh release/tag to just what's needed by swh-storage for filtering. Returns: A shallow copy of a full swh release/tag object. """ return tag['id'] diff --git a/swh/loader/core/tests/test_converters.py b/swh/loader/core/tests/test_converters.py index 67c337b..5fc428e 100644 --- a/swh/loader/core/tests/test_converters.py +++ b/swh/loader/core/tests/test_converters.py @@ -1,408 +1,405 @@ # Copyright (C) 2015 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 os import shutil import tempfile import unittest from nose.tools import istest from swh.loader.core import converters from swh.model import git def tmpfile_with_content(fromdir, contentfile): """Create a temporary file with content contentfile in directory fromdir. """ tmpfilepath = tempfile.mktemp( suffix='.swh', prefix='tmp-file-for-test', dir=fromdir) with open(tmpfilepath, 'wb') as f: f.write(contentfile) return tmpfilepath class TestConverters(unittest.TestCase): @classmethod def setUpClass(cls): super().setUpClass() cls.tmpdir = tempfile.mkdtemp(prefix='test-swh-loader-dir.') @classmethod def tearDownClass(cls): shutil.rmtree(cls.tmpdir) super().tearDownClass() @istest def format_to_minutes(self): self.assertEquals(converters.format_to_minutes('+0100'), 60) self.assertEquals(converters.format_to_minutes('-0200'), -120) self.assertEquals(converters.format_to_minutes('+1250'), 12*60+50) self.assertEquals(converters.format_to_minutes('+0000'), 0) self.assertEquals(converters.format_to_minutes('-0000'), 0) @istest def annotated_tag_to_release(self): # given release = { 'id': '123', 'target': '456', 'target_type': 'revision', 'name': 'some-release', 'comment': 'some-comment-on-release', 'date': 1444054085, 'offset': '-0300', 'author_name': 'someone', 'author_email': 'someone@whatelse.eu', } expected_release = { 'target': '456', 'target_type': 'revision', 'name': b'some-release', 'message': b'some-comment-on-release', 'date': { 'timestamp': 1444054085, 'offset': -180 }, 'author': { 'name': b'someone', 'email': b'someone@whatelse.eu', }, 'synthetic': True, } # when actual_release = converters.annotated_tag_to_release(release) # then self.assertDictEqual(actual_release, expected_release) @istest def blob_to_content_visible_data(self): # given contentfile = b'temp file for testing blob to content conversion' tmpfilepath = tmpfile_with_content(self.tmpdir, contentfile) obj = { 'path': tmpfilepath, 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', } expected_blob = { 'data': contentfile, 'length': len(contentfile), 'status': 'visible', 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', 'perms': git.GitPerm.BLOB.value, 'type': git.GitType.BLOB.value, } # when actual_blob = converters.blob_to_content(obj) # then self.assertEqual(actual_blob, expected_blob) @istest def blob_to_content_link(self): # given contentfile = b'temp file for testing blob to content conversion' tmpfilepath = tmpfile_with_content(self.tmpdir, contentfile) tmplinkpath = tempfile.mktemp(dir=self.tmpdir) os.symlink(tmpfilepath, tmplinkpath) obj = { 'path': tmplinkpath, 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', } expected_blob = { 'data': contentfile, 'length': len(tmpfilepath), 'status': 'visible', 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', 'perms': git.GitPerm.BLOB.value, 'type': git.GitType.BLOB.value, } # when actual_blob = converters.blob_to_content(obj) # then self.assertEqual(actual_blob, expected_blob) @istest def blob_to_content_link_with_data_length_populated(self): # given tmplinkpath = tempfile.mktemp(dir=self.tmpdir) obj = { 'length': 10, # wrong for test purposes 'data': 'something wrong', # again for test purposes 'path': tmplinkpath, 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', } expected_blob = { 'length': 10, 'data': 'something wrong', 'status': 'visible', 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', 'perms': git.GitPerm.BLOB.value, 'type': git.GitType.BLOB.value, } # when actual_blob = converters.blob_to_content(obj) # then self.assertEqual(actual_blob, expected_blob) @istest def blob_to_content2_absent_data(self): # given contentfile = b'temp file for testing blob to content conversion' tmpfilepath = tmpfile_with_content(self.tmpdir, contentfile) obj = { 'path': tmpfilepath, 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', } expected_blob = { 'length': len(contentfile), 'status': 'absent', 'sha1': 'some-sha1', 'sha256': 'some-sha256', 'sha1_git': 'some-sha1git', 'perms': git.GitPerm.BLOB.value, 'type': git.GitType.BLOB.value, 'reason': 'Content too large', 'origin': 190 } # when actual_blob = converters.blob_to_content(obj, None, max_content_size=10, origin_id=190) # then self.assertEqual(actual_blob, expected_blob) @istest def tree_to_directory_no_entries(self): # given tree = { 'path': 'foo', 'sha1_git': b'tree_sha1_git', 'children': [{'type': git.GitType.TREE, 'perms': git.GitPerm.TREE, 'name': 'bar', 'sha1_git': b'sha1-target'}, {'type': git.GitType.BLOB, 'perms': git.GitPerm.BLOB, 'name': 'file-foo', 'sha1_git': b'file-foo-sha1-target'}] } expected_directory = { 'id': b'tree_sha1_git', 'entries': [{'type': 'dir', 'perms': int(git.GitPerm.TREE.value), 'name': 'bar', 'target': b'sha1-target'}, {'type': 'file', 'perms': int(git.GitPerm.BLOB.value), 'name': 'file-foo', 'target': b'file-foo-sha1-target'}] } # when actual_directory = converters.tree_to_directory(tree) # then self.assertEqual(actual_directory, expected_directory) @istest def commit_to_revision(self): # given commit = { 'sha1_git': 'commit-git-sha1', + 'directory': 'targeted-tree-sha1', 'author_date': 1444054085, 'author_offset': '+0000', 'committer_date': 1444054085, 'committer_offset': '-0000', 'type': 'tar', 'message': 'synthetic-message-input', 'author_name': 'author-name', 'author_email': 'author-email', 'committer_name': 'committer-name', 'committer_email': 'committer-email', 'metadata': {'checksums': {'sha1': b'sha1-as-bytes'}}, 'directory': 'targeted-tree-sha1', } - objects = { - git.ROOT_TREE_KEY: [{'sha1_git': 'targeted-tree-sha1'}] - } - expected_revision = { 'date': { 'timestamp': 1444054085, 'offset': 0, }, 'committer_date': { 'timestamp': 1444054085, 'offset': 0, }, 'type': 'tar', 'directory': 'targeted-tree-sha1', 'message': b'synthetic-message-input', 'author': { 'name': b'author-name', 'email': b'author-email', }, 'committer': { 'name': b'committer-name', 'email': b'committer-email', }, 'synthetic': True, 'metadata': {'checksums': {'sha1': b'sha1-as-bytes'}}, 'parents': [], } # when - actual_revision = converters.commit_to_revision(commit, objects) + actual_revision = converters.commit_to_revision(commit) # then self.assertEquals(actual_revision, expected_revision) @istest def ref_to_occurrence_1(self): # when actual_occ = converters.ref_to_occurrence({ 'id': 'some-id', 'branch': 'some/branch' }) # then self.assertEquals(actual_occ, { 'id': 'some-id', 'branch': b'some/branch' }) @istest def ref_to_occurrence_2(self): # when actual_occ = converters.ref_to_occurrence({ 'id': 'some-id', 'branch': b'some/branch' }) # then self.assertEquals(actual_occ, { 'id': 'some-id', 'branch': b'some/branch' }) @istest def shallow_blob(self): # when actual_blob = converters.shallow_blob({ 'length': 1451, 'sha1_git': b'\xd1\xdd\x9a@\xeb\xf6!\x99\xd4[S\x05\xa8Y\xa3\x80\xa7\xb1;\x9c', 'name': b'LDPCL', 'type': b'blob', 'sha256': b'\xe6it!\x99\xb37UT\x8f\x0e\x8f\xd7o\x92"\xce\xa3\x1d\xd2\xe5D>M\xaaj/\x03\x138\xad\x1b', # noqa 'perms': b'100644', 'sha1': b'.\x18Y\xd6M\x8c\x9a\xa4\xe1\xf1\xc7\x95\x082\xcf\xc9\xd8\nV)', 'path': b'/tmp/tmp.c86tq5o9.swh.loader/pkg-doc-linux/copyrights/non-free/LDPCL' # noqa }) # then self.assertEqual(actual_blob, { 'sha1': b'.\x18Y\xd6M\x8c\x9a\xa4\xe1\xf1\xc7\x95\x082\xcf\xc9\xd8\nV)', 'sha1_git': b'\xd1\xdd\x9a@\xeb\xf6!\x99\xd4[S\x05\xa8Y\xa3\x80\xa7\xb1;\x9c', 'sha256': b'\xe6it!\x99\xb37UT\x8f\x0e\x8f\xd7o\x92"\xce\xa3\x1d\xd2\xe5D>M\xaaj/\x03\x138\xad\x1b', # noqa 'length': 1451, }) @istest def shallow_tree(self): # when actual_shallow_tree = converters.shallow_tree({ 'length': 1451, 'sha1_git': b'tree-id', 'type': b'tree', 'sha256': b'\xe6it!\x99\xb37UT\x8f\x0e\x8f\xd7o\x92"\xce\xa3\x1d\xd2\xe5D>M\xaaj/\x03\x138\xad\x1b', # noqa 'perms': b'100644', 'sha1': b'.\x18Y\xd6M\x8c\x9a\xa4\xe1\xf1\xc7\x95\x082\xcf\xc9\xd8\nV)', }) # then self.assertEqual(actual_shallow_tree, b'tree-id') @istest def shallow_commit(self): # when actual_shallow_commit = converters.shallow_commit({ 'sha1_git': b'\xd1\xdd\x9a@\xeb\xf6!\x99\xd4[S\x05\xa8Y\xa3\x80\xa7\xb1;\x9c', 'type': b'commit', 'id': b'let-me-see-some-id', }) # then self.assertEqual(actual_shallow_commit, b'let-me-see-some-id') @istest def shallow_tag(self): # when actual_shallow_tag = converters.shallow_tag({ 'sha1': b'\xd1\xdd\x9a@\xeb\xf6!\x99\xd4[S\x05\xa8Y\xa3\x80\xa7\xb1;\x9c', 'type': b'tag', 'id': b'this-is-not-the-id-you-are-looking-for', }) # then self.assertEqual(actual_shallow_tag, b'this-is-not-the-id-you-are-looking-for') # noqa