diff --git a/PKG-INFO b/PKG-INFO index 4925bc7..6e8954f 100644 --- a/PKG-INFO +++ b/PKG-INFO @@ -1,10 +1,10 @@ Metadata-Version: 1.0 Name: swh.model -Version: 0.0.13 +Version: 0.0.14 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 Description: UNKNOWN Platform: UNKNOWN diff --git a/swh.model.egg-info/PKG-INFO b/swh.model.egg-info/PKG-INFO index 4925bc7..6e8954f 100644 --- a/swh.model.egg-info/PKG-INFO +++ b/swh.model.egg-info/PKG-INFO @@ -1,10 +1,10 @@ Metadata-Version: 1.0 Name: swh.model -Version: 0.0.13 +Version: 0.0.14 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 Description: UNKNOWN Platform: UNKNOWN diff --git a/swh/model/git.py b/swh/model/git.py index a3503cb..1f95b7b 100644 --- a/swh/model/git.py +++ b/swh/model/git.py @@ -1,575 +1,587 @@ -# Copyright (C) 2015 The Software Heritage developers +# Copyright (C) 2015-2017 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 os +import stat from enum import Enum from swh.model import hashutil, identifiers ROOT_TREE_KEY = b'' class GitType(Enum): BLOB = b'blob' TREE = b'tree' EXEC = b'exec' LINK = b'link' COMM = b'commit' RELE = b'release' REFS = b'ref' class GitPerm(Enum): BLOB = b'100644' TREE = b'40000' EXEC = b'100755' LINK = b'120000' def _compute_directory_git_sha1(hashes): """Compute a directory git sha1 from hashes. Args: hashes: list of tree entries with keys: - sha1_git: the tree entry's sha1 - name: file or subdir's name - perms: the tree entry's sha1 permissions Returns: the binary sha1 of the dictionary's identifier Assumes: Every path exists in hashes. """ directory = { 'entries': [ { 'name': entry['name'], 'perms': int(entry['perms'].value, 8), 'target': entry['sha1_git'], 'type': 'dir' if entry['perms'] == GitPerm.TREE else 'file', } for entry in hashes ] } return hashutil.hash_to_bytes(identifiers.directory_identifier(directory)) def compute_directory_git_sha1(dirpath, hashes): """Compute a directory git sha1 for a dirpath. Args: dirpath: the directory's absolute path hashes: list of tree entries with keys: - sha1_git: the tree entry's sha1 - name: file or subdir's name - perms: the tree entry's sha1 permissions Returns: the binary sha1 of the dictionary's identifier Assumes: Every path exists in hashes. """ return _compute_directory_git_sha1(hashes[dirpath]) def compute_revision_sha1_git(revision): """Compute a revision sha1 git from its dict representation. Args: revision: Additional dictionary information needed to compute a synthetic revision. Following keys are expected: - author - date - committer - committer_date - message - type - directory: binary form of the tree hash Returns: revision sha1 in bytes # FIXME: beware, bytes output from storage api """ return hashutil.hash_to_bytes(identifiers.revision_identifier(revision)) def compute_release_sha1_git(release): """Compute a release sha1 git from its dict representation. Args: release: Additional dictionary information needed to compute a synthetic release. Following keys are expected: - name - message - date - author - revision: binary form of the sha1_git revision targeted by this Returns: release sha1 in bytes """ return hashutil.hash_to_bytes(identifiers.release_identifier(release)) def compute_link_metadata(linkpath): """Given a linkpath, compute the git metadata. Args: linkpath: absolute pathname of the link Returns: Dictionary of values: - data: link's content - length: link's content length - name: basename of the link - perms: git permission for link - type: git type for link - path: absolute path to the link on filesystem """ data = os.readlink(linkpath) link_metadata = hashutil.hash_data(data) link_metadata.update({ 'data': data, 'length': len(data), 'name': os.path.basename(linkpath), 'perms': GitPerm.LINK, 'type': GitType.BLOB, 'path': linkpath }) return link_metadata def compute_blob_metadata(filepath): - """Given a filepath, compute the git metadata. + """Given a filepath resolving to a regular file, compute the metadata. + Other file types (fifo, character or block device, symlink) will + be considered empty regular file. To deal properly with symlinks, + use swh.model.git.compute_link_metadata. Args: - filepath: absolute pathname of the file. + filepath: absolute pathname of the regular file. Returns: Dictionary of values: - name: basename of the file + - length: data length - perms: git permission for file - type: git type for file - path: absolute filepath on filesystem """ - blob_metadata = hashutil.hash_path(filepath) - perms = GitPerm.EXEC if os.access(filepath, os.X_OK) else GitPerm.BLOB + mode = os.lstat(filepath).st_mode + if not stat.S_ISREG(mode): # special (block or character device, fifo) + perms = GitPerm.BLOB + blob_metadata = hashutil.hash_data(b'') + blob_metadata['length'] = 0 + else: + perms = GitPerm.EXEC if os.access(filepath, os.X_OK) else GitPerm.BLOB + blob_metadata = hashutil.hash_path(filepath) + blob_metadata.update({ 'name': os.path.basename(filepath), 'perms': perms, 'type': GitType.BLOB, 'path': filepath }) return blob_metadata def _compute_tree_metadata(dirname, hashes): """Given a dirname, compute the git metadata. Args: dirname: absolute pathname of the directory. hashes: list of tree dirname's entries with keys: - sha1_git: the tree entry's sha1 - name: file or subdir's name - perms: the tree entry's sha1 permissions Returns: Dictionary of values: - sha1_git: tree's sha1 git - name: basename of the directory - perms: git permission for directory - type: git type for directory - path: absolute path to directory on filesystem """ return { 'sha1_git': _compute_directory_git_sha1(hashes), 'name': os.path.basename(dirname), 'perms': GitPerm.TREE, 'type': GitType.TREE, 'path': dirname } def compute_tree_metadata(dirname, ls_hashes): """Given a dirname, compute the git metadata. Args: dirname: absolute pathname of the directory. ls_hashes: dictionary of path, hashes Returns: Dictionary of values: - sha1_git: tree's sha1 git - name: basename of the directory - perms: git permission for directory - type: git type for directory - path: absolute path to directory on filesystem """ return _compute_tree_metadata(dirname, ls_hashes[dirname]) def default_validation_dir(dirpath): """Default validation function. This is the equivalent of the identity function. Args: dirpath: Path to validate Returns: True """ return True def _walk(rootdir, dir_ok_fn=default_validation_dir, remove_empty_folder=False): """Walk the filesystem and yields a 3 tuples (dirpath, dirnames as set of absolute paths, filenames as set of abslute paths) Ignore files which won't pass the dir_ok_fn validation. If remove_empty_folder is True, remove and ignore any encountered empty folder. Args: - rootdir: starting walk root directory path - dir_ok_fn: validation function. if folder encountered are not ok, they are ignored. Default to default_validation_dir which does nothing. - remove_empty_folder: Flag to remove and ignore any encountered empty folders. Yields: 3 tuples dirpath, set of absolute children dirname paths, set of absolute filename paths. """ def basic_gen_dir(rootdir): for dp, dns, fns in os.walk(rootdir, topdown=False): yield (dp, set((os.path.join(dp, dn) for dn in dns)), set((os.path.join(dp, fn) for fn in fns))) if dir_ok_fn == default_validation_dir: if not remove_empty_folder: # os.walk yield from basic_gen_dir(rootdir) else: # os.walk + empty dir cleanup empty_folders = set() for dp, dns, fns in basic_gen_dir(rootdir): if not dns and not fns: empty_folders.add(dp) # need to remove it because folder of empty folder # is an empty folder!!! if os.path.islink(dp): os.remove(dp) else: os.rmdir(dp) parent = os.path.dirname(dp) # edge case about parent containing one empty # folder which become an empty one while not os.listdir(parent): empty_folders.add(parent) if os.path.islink(parent): os.remove(parent) else: os.rmdir(parent) parent = os.path.dirname(parent) continue yield (dp, dns - empty_folders, fns) else: def filtfn(dirnames): return set(filter(dir_ok_fn, dirnames)) gen_dir = ((dp, dns, fns) for dp, dns, fns in basic_gen_dir(rootdir) if dir_ok_fn(dp)) if not remove_empty_folder: # os.walk + filtering for dp, dns, fns in gen_dir: yield (dp, filtfn(dns), fns) else: # os.walk + filtering + empty dir cleanup empty_folders = set() for dp, dns, fns in gen_dir: dps = filtfn(dns) if not dps and not fns: empty_folders.add(dp) # need to remove it because folder of empty folder # is an empty folder!!! if os.path.islink(dp): os.remove(dp) else: os.rmdir(dp) parent = os.path.dirname(dp) # edge case about parent containing one empty # folder which become an empty one while not os.listdir(parent): empty_folders.add(parent) if os.path.islink(parent): os.remove(parent) else: os.rmdir(parent) parent = os.path.dirname(parent) continue yield dp, dps - empty_folders, fns def walk_and_compute_sha1_from_directory(rootdir, dir_ok_fn=default_validation_dir, with_root_tree=True, remove_empty_folder=False): """(Deprecated) TODO migrate the code to compute_hashes_from_directory. Compute git sha1 from directory rootdir. Args: - rootdir: Root directory from which beginning the git hash computation - dir_ok_fn: Filter function to filter directory according to rules defined in the function. By default, all folders are ok. Example override: dir_ok_fn = lambda dirpath: b'svn' not in dirpath - with_root_tree: Determine if we compute the upper root tree's checksums. As a default, we want it. One possible use case where this is not useful is the update (cf. `update_checksums_from`) Returns: Dictionary of entries with keys and as values a list of directory entries. Those are list of dictionary with keys: - 'perms' - 'type' - 'name' - 'sha1_git' - and specifically content: 'sha1', 'sha256', ... Note: One special key is ROOT_TREE_KEY to indicate the upper root of the directory (this is the revision's directory). Raises: Nothing If something is raised, this is a programmatic error. """ ls_hashes = {} all_links = set() if rootdir.endswith(b'/'): rootdir = rootdir.rstrip(b'/') for dirpath, dirnames, filenames in _walk( rootdir, dir_ok_fn, remove_empty_folder): hashes = [] links = (file for file in filenames.union(dirnames) if os.path.islink(file)) for linkpath in links: all_links.add(linkpath) m_hashes = compute_link_metadata(linkpath) hashes.append(m_hashes) for filepath in (file for file in filenames if file not in all_links): m_hashes = compute_blob_metadata(filepath) hashes.append(m_hashes) ls_hashes[dirpath] = hashes dir_hashes = [] for fulldirname in (dir for dir in dirnames if dir not in all_links): tree_hash = _compute_tree_metadata(fulldirname, ls_hashes[fulldirname]) dir_hashes.append(tree_hash) ls_hashes[dirpath].extend(dir_hashes) if with_root_tree: # compute the current directory hashes root_hash = { 'sha1_git': _compute_directory_git_sha1(ls_hashes[rootdir]), 'path': rootdir, 'name': os.path.basename(rootdir), 'perms': GitPerm.TREE, 'type': GitType.TREE } ls_hashes[ROOT_TREE_KEY] = [root_hash] return ls_hashes def compute_hashes_from_directory(rootdir, dir_ok_fn=default_validation_dir, remove_empty_folder=False): """Compute git sha1 from directory rootdir. Args: - rootdir: Root directory from which beginning the git hash computation - dir_ok_fn: Filter function to filter directory according to rules defined in the function. By default, all folders are ok. Example override: dir_ok_fn = lambda dirpath: b'svn' not in dirpath Returns: Dictionary of entries with keys absolute path name. Path-name can be a file/link or directory. The associated value is a dictionary with: - checksums: the dictionary with the hashes for the link/file/dir Those are list of dictionary with keys: - 'perms' - 'type' - 'name' - 'sha1_git' - and specifically content: 'sha1', 'sha256', ... - children: Only for a directory, the set of children paths Note: One special key is the / which indicates the upper root of the directory (this is the revision's directory). Raises: Nothing If something is raised, this is a programmatic error. """ def _get_dict_from_dirpath(_dict, path): """Retrieve the default associated value for key path. """ return _dict.get(path, dict(children=set(), checksums=None)) def _get_dict_from_filepath(_dict, path): """Retrieve the default associated value for key path. """ return _dict.get(path, dict(checksums=None)) ls_hashes = {} all_links = set() if rootdir.endswith(b'/'): rootdir = rootdir.rstrip(b'/') for dirpath, dirnames, filenames in _walk( rootdir, dir_ok_fn, remove_empty_folder): dir_entry = _get_dict_from_dirpath(ls_hashes, dirpath) children = dir_entry['children'] links = (file for file in filenames.union(dirnames) if os.path.islink(file)) for linkpath in links: all_links.add(linkpath) m_hashes = compute_link_metadata(linkpath) d = _get_dict_from_filepath(ls_hashes, linkpath) d['checksums'] = m_hashes ls_hashes[linkpath] = d children.add(linkpath) for filepath in (file for file in filenames if file not in all_links): m_hashes = compute_blob_metadata(filepath) d = _get_dict_from_filepath(ls_hashes, filepath) d['checksums'] = m_hashes ls_hashes[filepath] = d children.add(filepath) for fulldirname in (dir for dir in dirnames if dir not in all_links): d_hashes = _get_dict_from_dirpath(ls_hashes, fulldirname) tree_hash = _compute_tree_metadata( fulldirname, (ls_hashes[p]['checksums'] for p in d_hashes['children']) ) d = _get_dict_from_dirpath(ls_hashes, fulldirname) d['checksums'] = tree_hash ls_hashes[fulldirname] = d children.add(fulldirname) dir_entry['children'] = children ls_hashes[dirpath] = dir_entry # compute the current directory hashes d_hashes = _get_dict_from_dirpath(ls_hashes, rootdir) root_hash = { 'sha1_git': _compute_directory_git_sha1( (ls_hashes[p]['checksums'] for p in d_hashes['children']) ), 'path': rootdir, 'name': os.path.basename(rootdir), 'perms': GitPerm.TREE, 'type': GitType.TREE } d_hashes['checksums'] = root_hash ls_hashes[rootdir] = d_hashes return ls_hashes def children_hashes(children, objects): """Given a collection of children path, yield the corresponding hashes. Args: objects: objects hash as returned by git.compute_hashes_from_directory. children: collection of bytes path Yields: Dictionary hashes """ for p in children: c = objects.get(p) if c: h = c.get('checksums') if h: yield h def objects_per_type(filter_type, objects_per_path): """Given an object dictionary returned by `swh.model.git.compute_hashes_from_directory`, yields corresponding element type's hashes Args: filter_type: one of GitType enum objects_per_path: Yields: Elements of type filter_type's hashes """ for path, obj in objects_per_path.items(): o = obj['checksums'] if o['type'] == filter_type: if 'children' in obj: # for trees if obj['children']: o['children'] = children_hashes(obj['children'], objects_per_path) else: o['children'] = [] yield o diff --git a/swh/model/hashutil.py b/swh/model/hashutil.py index b2558a3..ea28414 100644 --- a/swh/model/hashutil.py +++ b/swh/model/hashutil.py @@ -1,189 +1,201 @@ # 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 binascii import functools import hashlib from io import BytesIO import os # supported hashing algorithms ALGORITHMS = set(['sha1', 'sha256', 'sha1_git']) # should be a multiple of 64 (sha1/sha256's block size) # FWIW coreutils' sha1sum uses 32768 HASH_BLOCK_SIZE = 32768 def _new_git_hash(base_algo, git_type, length): """Initialize a digest object (as returned by python's hashlib) for the requested algorithm, and feed it with the header for a git object of the given type and length. The header for hashing a git object consists of: - The type of the object (encoded in ASCII) - One ASCII space (\x20) - The length of the object (decimal encoded in ASCII) - One NUL byte Args: base_algo: a hashlib-supported algorithm git_type: the type of the git object (supposedly one of 'blob', 'commit', 'tag', 'tree') length: the length of the git object you're encoding Returns: a hashutil.hash object """ h = hashlib.new(base_algo) git_header = '%s %d\0' % (git_type, length) h.update(git_header.encode('ascii')) return h def _new_hash(algo, length=None): """Initialize a digest object (as returned by python's hashlib) for the requested algorithm. See the constant ALGORITHMS for the list of supported algorithms. If a git-specific hashing algorithm is requested (e.g., "sha1_git"), the hashing object will be pre-fed with the needed header; for this to work, length must be given. Args: algo: a hashing algorithm (one of ALGORITHMS) length: the length of the hashed payload (needed for git-specific algorithms) Returns: a hashutil.hash object Raises: ValueError if algo is unknown, or length is missing for a git-specific hash. """ if algo not in ALGORITHMS: raise ValueError('Unexpected hashing algorithm %s, ' 'expected one of %s' % (algo, ', '.join(sorted(ALGORITHMS)))) h = None if algo.endswith('_git'): if length is None: raise ValueError('Missing length for git hashing algorithm') base_algo = algo[:-4] h = _new_git_hash(base_algo, 'blob', length) else: h = hashlib.new(algo) return h def hash_file(fobj, length=None, algorithms=ALGORITHMS, chunk_cb=None): """Hash the contents of the given file object with the given algorithms. Args: fobj: a file-like object length: the length of the contents of the file-like object (for the git-specific algorithms) algorithms: the hashing algorithms used Returns: a dict mapping each algorithm to a bytes digest. Raises: ValueError if algorithms contains an unknown hash algorithm. """ hashes = {algo: _new_hash(algo, length) for algo in algorithms} while True: chunk = fobj.read(HASH_BLOCK_SIZE) if not chunk: break for hash in hashes.values(): hash.update(chunk) if chunk_cb: chunk_cb(chunk) return {algo: hash.digest() for algo, hash in hashes.items()} def hash_path(path, algorithms=ALGORITHMS, chunk_cb=None): """Hash the contents of the file at the given path with the given algorithms. Args: path: the path of the file to hash algorithms: the hashing algorithms used chunk_cb: a callback Returns: a dict mapping each algorithm to a bytes digest. Raises: ValueError if algorithms contains an unknown hash algorithm. OSError on file access error """ length = os.path.getsize(path) with open(path, 'rb') as fobj: hash = hash_file(fobj, length, algorithms, chunk_cb) hash['length'] = length return hash def hash_data(data, algorithms=ALGORITHMS): """Hash the given binary blob with the given algorithms. Args: data: a bytes object algorithms: the hashing algorithms used Returns: a dict mapping each algorithm to a bytes digest Raises: TypeError if data does not support the buffer interface. ValueError if algorithms contains an unknown hash algorithm. """ fobj = BytesIO(data) return hash_file(fobj, len(data), algorithms) def hash_git_data(data, git_type, base_algo='sha1'): """Hash the given data as a git object of type git_type. Args: data: a bytes object git_type: the git object type base_algo: the base hashing algorithm used (default: sha1) Returns: a dict mapping each algorithm to a bytes digest Raises: ValueError if the git_type is unexpected. """ git_object_types = {'blob', 'tree', 'commit', 'tag'} if git_type not in git_object_types: raise ValueError('Unexpected git object type %s, expected one of %s' % (git_type, ', '.join(sorted(git_object_types)))) h = _new_git_hash(base_algo, git_type, len(data)) h.update(data) return h.digest() @functools.lru_cache() def hash_to_hex(hash): """Converts a hash (in hex or bytes form) to its hexadecimal ascii form""" if isinstance(hash, str): return hash return binascii.hexlify(hash).decode('ascii') +@functools.lru_cache() +def hash_to_bytehex(hash): + """Converts a hash to its hexadecimal bytes representation""" + return binascii.hexlify(hash) + + @functools.lru_cache() def hash_to_bytes(hash): """Converts a hash (in hex or bytes form) to its raw bytes form""" if isinstance(hash, bytes): return hash return bytes.fromhex(hash) + + +@functools.lru_cache() +def bytehex_to_hash(hex): + """Converts a hexadecimal bytes representation of a hash to that hash""" + return hash_to_bytes(hex.decode()) diff --git a/swh/model/identifiers.py b/swh/model/identifiers.py index c53513a..4eb2b9d 100644 --- a/swh/model/identifiers.py +++ b/swh/model/identifiers.py @@ -1,487 +1,495 @@ # 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 binascii import datetime from functools import lru_cache from .hashutil import hash_data, hash_git_data @lru_cache() def identifier_to_bytes(identifier): """Convert a text identifier to bytes. Args: identifier: an identifier, either a 40-char hexadecimal string or a bytes object of length 20 Returns: The length 20 bytestring corresponding to the given identifier Raises: ValueError if the identifier is of an unexpected type or length. """ if isinstance(identifier, bytes): if len(identifier) != 20: raise ValueError( 'Wrong length for bytes identifier %s, expected 20' % len(identifier)) return identifier if isinstance(identifier, str): if len(identifier) != 40: raise ValueError( 'Wrong length for str identifier %s, expected 40' % len(identifier)) return bytes.fromhex(identifier) raise ValueError('Wrong type for identifier %s, expected bytes or str' % identifier.__class__.__name__) @lru_cache() def identifier_to_str(identifier): """Convert an identifier to an hexadecimal string. Args: identifier: an identifier, either a 40-char hexadecimal string or a bytes object of length 20 Returns: The length 40 string corresponding to the given identifier, hex encoded Raises: ValueError if the identifier is of an unexpected type or length. """ if isinstance(identifier, str): if len(identifier) != 40: raise ValueError( 'Wrong length for str identifier %s, expected 40' % len(identifier)) return identifier if isinstance(identifier, bytes): if len(identifier) != 20: raise ValueError( 'Wrong length for bytes identifier %s, expected 20' % len(identifier)) return binascii.hexlify(identifier).decode() raise ValueError('Wrong type for identifier %s, expected bytes or str' % identifier.__class__.__name__) def content_identifier(content): """Return the intrinsic identifier for a content. A content's identifier is the sha1, sha1_git and sha256 checksums of its data. Args: content: a content conforming to the Software Heritage schema Returns: A dictionary with all the hashes for the data Raises: KeyError if the content doesn't have a data member. """ hashes = hash_data( content['data'], {'sha1', 'sha1_git', 'sha256'}, ) return hashes def _sort_key(entry): """The sorting key for tree entries""" if entry['type'] == 'dir': return entry['name'] + b'/' else: return entry['name'] @lru_cache() def _perms_to_bytes(perms): """Convert the perms value to its bytes representation""" oc = oct(perms)[2:] return oc.encode('ascii') +def escape_newlines(snippet): + """Escape the newlines present in snippet according to git rules. + + New lines in git manifests are escaped by indenting the next line by one + space.""" + + if b'\n' in snippet: + return b'\n '.join(snippet.split(b'\n')) + else: + return snippet + + def directory_identifier(directory): """Return the intrinsic identifier for a directory. A directory's identifier is the tree sha1 à la git of a directory listing, using the following algorithm, which is equivalent to the git algorithm for trees: 1. Entries of the directory are sorted using the name (or the name with '/' appended for directory entries) as key, in bytes order. 2. For each entry of the directory, the following bytes are output: - the octal representation of the permissions for the entry (stored in the 'perms' member), which is a representation of the entry type: b'100644' (int 33188) for files b'100755' (int 33261) for executable files b'120000' (int 40960) for symbolic links b'40000' (int 16384) for directories b'160000' (int 57344) for references to revisions - an ascii space (b'\x20') - the entry's name (as raw bytes), stored in the 'name' member - a null byte (b'\x00') - the 20 byte long identifier of the object pointed at by the entry, stored in the 'target' member: for files or executable files: their blob sha1_git for symbolic links: the blob sha1_git of a file containing the link destination for directories: their intrinsic identifier for revisions: their intrinsic identifier (Note that there is no separator between entries) """ components = [] for entry in sorted(directory['entries'], key=_sort_key): components.extend([ _perms_to_bytes(entry['perms']), b'\x20', entry['name'], b'\x00', identifier_to_bytes(entry['target']), ]) return identifier_to_str(hash_git_data(b''.join(components), 'tree')) def format_date(date): """Convert a date object into an UTC timestamp encoded as ascii bytes. Git stores timestamps as an integer number of seconds since the UNIX epoch. However, Software Heritage stores timestamps as an integer number of microseconds (postgres type "datetime with timezone"). Therefore, we print timestamps with no microseconds as integers, and timestamps with microseconds as floating point values. We elide the trailing zeroes from microsecond values, to "future-proof" our representation if we ever need more precision in timestamps. """ if not isinstance(date, dict): raise ValueError('format_date only supports dicts, %r received' % date) seconds = date.get('seconds', 0) microseconds = date.get('microseconds', 0) if not microseconds: return str(seconds).encode() else: float_value = ('%d.%06d' % (seconds, microseconds)) return float_value.rstrip('0').encode() @lru_cache() def format_offset(offset, negative_utc=None): """Convert an integer number of minutes into an offset representation. The offset representation is [+-]hhmm where: hh is the number of hours; mm is the number of minutes. A null offset is represented as +0000. """ if offset < 0 or offset == 0 and negative_utc: sign = '-' else: sign = '+' hours = abs(offset) // 60 minutes = abs(offset) % 60 t = '%s%02d%02d' % (sign, hours, minutes) return t.encode() def normalize_timestamp(time_representation): """Normalize a time representation for processing by Software Heritage This function supports a numeric timestamp (representing a number of seconds since the UNIX epoch, 1970-01-01 at 00:00 UTC), a datetime.datetime object (with timezone information), or a normalized Software Heritage time representation (idempotency). Args: time_representation: the representation of a timestamp Returns: a normalized dictionary with three keys - timestamp: a dict with two optional keys: - seconds: the integral number of seconds since the UNIX epoch - microseconds: the integral number of microseconds - offset: the timezone offset as a number of minutes relative to UTC - negative_utc: a boolean representing whether the offset is -0000 when offset = 0. """ if time_representation is None: return None negative_utc = False if isinstance(time_representation, dict): ts = time_representation['timestamp'] if isinstance(ts, dict): seconds = ts.get('seconds', 0) microseconds = ts.get('microseconds', 0) elif isinstance(ts, int): seconds = ts microseconds = 0 else: raise ValueError( 'normalize_timestamp received non-integer timestamp member:' ' %r' % ts) offset = time_representation['offset'] if 'negative_utc' in time_representation: negative_utc = time_representation['negative_utc'] elif isinstance(time_representation, datetime.datetime): seconds = int(time_representation.timestamp()) microseconds = time_representation.microsecond utcoffset = time_representation.utcoffset() if utcoffset is None: raise ValueError( 'normalize_timestamp received datetime without timezone: %s' % time_representation) # utcoffset is an integer number of minutes seconds_offset = utcoffset.total_seconds() offset = int(seconds_offset) // 60 elif isinstance(time_representation, int): seconds = time_representation microseconds = 0 offset = 0 else: raise ValueError( 'normalize_timestamp received non-integer timestamp:' ' %r' % time_representation) return { 'timestamp': { 'seconds': seconds, 'microseconds': microseconds, }, 'offset': offset, 'negative_utc': negative_utc, } def format_author(author): """Format the specification of an author. An author is either a byte string (passed unchanged), or a dict with three keys, fullname, name and email. If the fullname exists, return it; if it doesn't, we construct a fullname using the following heuristics: if the name value is None, we return the email in angle brackets, else, we return the name, a space, and the email in angle brackets. """ if isinstance(author, bytes) or author is None: return author if 'fullname' in author: return author['fullname'] ret = [] if author['name'] is not None: ret.append(author['name']) if author['email'] is not None: ret.append(b''.join([b'<', author['email'], b'>'])) return b' '.join(ret) def format_author_line(header, author, date_offset): """Format a an author line according to git standards. An author line has three components: - a header, describing the type of author (author, committer, tagger) - a name and email, which is an arbitrary bytestring - optionally, a timestamp with UTC offset specification The author line is formatted thus: `header` `name and email`[ `timestamp` `utc_offset`] The timestamp is encoded as a (decimal) number of seconds since the UNIX epoch (1970-01-01 at 00:00 UTC). As an extension to the git format, we support fractional timestamps, using a dot as the separator for the decimal part. The utc offset is a number of minutes encoded as '[+-]HHMM'. Note some tools can pass a negative offset corresponding to the UTC timezone ('-0000'), which is valid and is encoded as such. For convenience, this function returns the whole line with its trailing newline. Args: header: the header of the author line (one of 'author', 'committer', 'tagger') author: an author specification (dict with two bytes values: name and email, or byte value) date_offset: a normalized date/time representation as returned by `normalize_timestamp`. Returns: the newline-terminated byte string containing the author line """ - ret = [header.encode(), b' ', format_author(author)] + ret = [header.encode(), b' ', escape_newlines(format_author(author))] date_offset = normalize_timestamp(date_offset) if date_offset is not None: date_f = format_date(date_offset['timestamp']) offset_f = format_offset(date_offset['offset'], date_offset['negative_utc']) ret.extend([b' ', date_f, b' ', offset_f]) ret.append(b'\n') return b''.join(ret) def revision_identifier(revision): """Return the intrinsic identifier for a revision. The fields used for the revision identifier computation are: - directory - parents - author - author_date - committer - committer_date - metadata -> extra_headers - message A revision's identifier is the 'git'-checksum of a commit manifest constructed as follows (newlines are a single ASCII newline character): ``` tree [for each parent in parents] parent [end for each parents] author committer [for each key, value in extra_headers] [end for each extra_headers] ``` The directory identifier is the ascii representation of its hexadecimal encoding. Author and committer are formatted with the `format_author` function. Dates are formatted with the `format_date_offset` function. Extra headers are an ordered list of [key, value] pairs. Keys are strings and get encoded to utf-8 for identifier computation. Values are either byte strings, unicode strings (that get encoded to utf-8), or integers (that get encoded to their utf-8 decimal representation). Multiline extra header values are escaped by indenting the continuation lines with one ascii space. If the message is None, the manifest ends with the last header. Else, the message is appended to the headers after an empty line. The checksum of the full manifest is computed using the 'commit' git object type. """ components = [ b'tree ', identifier_to_str(revision['directory']).encode(), b'\n', ] for parent in revision['parents']: if parent: components.extend([ b'parent ', identifier_to_str(parent).encode(), b'\n', ]) components.extend([ format_author_line('author', revision['author'], revision['date']), format_author_line('committer', revision['committer'], revision['committer_date']), ]) # Handle extra headers metadata = revision.get('metadata') if not metadata: metadata = {} for key, value in metadata.get('extra_headers', []): # Integer values: decimal representation if isinstance(value, int): value = str(value).encode('utf-8') # Unicode string values: utf-8 encoding if isinstance(value, str): value = value.encode('utf-8') - # multi-line values: indent continuation lines - if b'\n' in value: - value_chunks = value.split(b'\n') - value = b'\n '.join(value_chunks) - # encode the key to utf-8 - components.extend([key.encode('utf-8'), b' ', value, b'\n']) + components.extend([key.encode('utf-8'), b' ', + escape_newlines(value), b'\n']) if revision['message'] is not None: components.extend([b'\n', revision['message']]) commit_raw = b''.join(components) return identifier_to_str(hash_git_data(commit_raw, 'commit')) def target_type_to_git(target_type): """Convert a software heritage target type to a git object type""" return { 'content': b'blob', 'directory': b'tree', 'revision': b'commit', 'release': b'tag', }[target_type] def release_identifier(release): """Return the intrinsic identifier for a release.""" components = [ b'object ', identifier_to_str(release['target']).encode(), b'\n', b'type ', target_type_to_git(release['target_type']), b'\n', b'tag ', release['name'], b'\n', ] if 'author' in release and release['author']: components.append( format_author_line('tagger', release['author'], release['date']) ) if release['message'] is not None: components.extend([b'\n', release['message']]) return identifier_to_str(hash_git_data(b''.join(components), 'tag')) diff --git a/swh/model/tests/test_git.py b/swh/model/tests/test_git.py index b1eac8c..3c233c3 100644 --- a/swh/model/tests/test_git.py +++ b/swh/model/tests/test_git.py @@ -1,687 +1,722 @@ -# Copyright (C) 2015 The Software Heritage developers +# Copyright (C) 2015-2017 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 os import shutil import subprocess import tempfile import unittest from nose.plugins.attrib import attr from nose.tools import istest from swh.model import git class GitHashlib(unittest.TestCase): def setUp(self): self.tree_data = b''.join([b'40000 barfoo\0', bytes.fromhex('c3020f6bf135a38c6df' '3afeb5fb38232c5e07087'), b'100644 blah\0', bytes.fromhex('63756ef0df5e4f10b6efa' '33cfe5c758749615f20'), b'100644 hello\0', bytes.fromhex('907b308167f0880fb2a' '5c0e1614bb0c7620f9dc3')]) self.commit_data = """tree 1c61f7259dcb770f46b194d941df4f08ff0a3970 author Antoine R. Dumont (@ardumont) 1444054085 +0200 committer Antoine R. Dumont (@ardumont) 1444054085 +0200 initial """.encode('utf-8') # NOQA self.tag_data = """object 24d012aaec0bc5a4d2f62c56399053d6cc72a241 type commit tag 0.0.1 tagger Antoine R. Dumont (@ardumont) 1444225145 +0200 blah """.encode('utf-8') # NOQA self.checksums = { 'tree_sha1_git': bytes.fromhex('ac212302c45eada382b27bfda795db' '121dacdb1c'), 'commit_sha1_git': bytes.fromhex('e960570b2e6e2798fa4cfb9af2c399' 'd629189653'), 'tag_sha1_git': bytes.fromhex('bc2b99ba469987bcf1272c189ed534' 'e9e959f120'), } @istest def compute_directory_git_sha1(self): # given dirpath = 'some-dir-path' hashes = { dirpath: [{'perms': git.GitPerm.TREE, 'type': git.GitType.TREE, 'name': b'barfoo', 'sha1_git': bytes.fromhex('c3020f6bf135a38c6df' '3afeb5fb38232c5e07087')}, {'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'name': b'hello', 'sha1_git': bytes.fromhex('907b308167f0880fb2a' '5c0e1614bb0c7620f9dc3')}, {'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'name': b'blah', 'sha1_git': bytes.fromhex('63756ef0df5e4f10b6efa' '33cfe5c758749615f20')}] } # when checksum = git.compute_directory_git_sha1(dirpath, hashes) # then self.assertEqual(checksum, self.checksums['tree_sha1_git']) @istest def compute_revision_sha1_git(self): # given tree_hash = bytes.fromhex('1c61f7259dcb770f46b194d941df4f08ff0a3970') revision = { 'author': { 'name': b'Antoine R. Dumont (@ardumont)', 'email': b'antoine.romain.dumont@gmail.com', }, 'date': { 'timestamp': 1444054085, 'offset': 120, }, 'committer': { 'name': b'Antoine R. Dumont (@ardumont)', 'email': b'antoine.romain.dumont@gmail.com', }, 'committer_date': { 'timestamp': 1444054085, 'offset': 120, }, 'message': b'initial\n', 'type': 'tar', 'directory': tree_hash, 'parents': [], } # when checksum = git.compute_revision_sha1_git(revision) # then self.assertEqual(checksum, self.checksums['commit_sha1_git']) @istest def compute_release_sha1_git(self): # given revision_hash = bytes.fromhex('24d012aaec0bc5a4d2f62c56399053' 'd6cc72a241') release = { 'name': b'0.0.1', 'author': { 'name': b'Antoine R. Dumont (@ardumont)', 'email': b'antoine.romain.dumont@gmail.com', }, 'date': { 'timestamp': 1444225145, 'offset': 120, }, 'message': b'blah\n', 'target_type': 'revision', 'target': revision_hash, } # when checksum = git.compute_release_sha1_git(release) # then self.assertEqual(checksum, self.checksums['tag_sha1_git']) +@attr('fs') +class ComputeBlobMetadata(unittest.TestCase): + @istest + def compute_blob_metadata__special_file_returns_nothing(self): + # prepare + tmp_root_path = tempfile.mkdtemp().encode('utf-8') + name = b'fifo-file' + path = os.path.join(tmp_root_path, name) + + # given + os.mkfifo(path) + + # when + actual_metadata = git.compute_blob_metadata(path) + + # then + expected_metadata = { + 'sha1': b'\xda9\xa3\xee^kK\r2U\xbf\xef\x95`\x18\x90\xaf\xd8\x07\t', + 'sha1_git': b'\xe6\x9d\xe2\x9b\xb2\xd1\xd6CK\x8b)\xaewZ\xd8\xc2' + b'\xe4\x8cS\x91', + 'sha256': b"\xe3\xb0\xc4B\x98\xfc\x1c\x14\x9a\xfb\xf4\xc8\x99o" + b"\xb9$'\xaeA\xe4d\x9b\x93L\xa4\x95\x99\x1bxR\xb8U", + 'perms': git.GitPerm.BLOB, + 'path': path, + 'name': name, + 'type': git.GitType.BLOB, + 'length': 0 + } + + self.assertEquals(actual_metadata, expected_metadata) + + # cleanup + shutil.rmtree(tmp_root_path) + + @attr('fs') class GitHashWalkArborescenceTree: """Root class to ease walk and git hash testing without side-effecty problems. """ def setUp(self): super().setUp() self.tmp_root_path = tempfile.mkdtemp().encode('utf-8') self.maxDiff = None start_path = os.path.dirname(__file__).encode('utf-8') sample_folder = os.path.join(start_path, b'../../../..', b'swh-storage-testdata', b'dir-folders', b'sample-folder.tgz') self.root_path = os.path.join(self.tmp_root_path, b'sample-folder') # uncompress the sample folder subprocess.check_output( ['tar', 'xvf', sample_folder, '-C', self.tmp_root_path]) def tearDown(self): if os.path.exists(self.tmp_root_path): shutil.rmtree(self.tmp_root_path) class GitHashFromScratch(GitHashWalkArborescenceTree, unittest.TestCase): """Test the main `walk_and_compute_sha1_from_directory` algorithm that scans and compute the disk for checksums. """ @istest def walk_and_compute_sha1_from_directory(self): # make a temporary arborescence tree to hash without ignoring anything # same as previous behavior walk0 = git.walk_and_compute_sha1_from_directory(self.tmp_root_path) keys0 = list(walk0.keys()) path_excluded = os.path.join(self.tmp_root_path, b'sample-folder', b'foo') self.assertTrue(path_excluded in keys0) # it is not excluded here # make the same temporary arborescence tree to hash with ignoring one # folder foo walk1 = git.walk_and_compute_sha1_from_directory( self.tmp_root_path, dir_ok_fn=lambda dirpath: b'sample-folder/foo' not in dirpath) keys1 = list(walk1.keys()) self.assertTrue(path_excluded not in keys1) # remove the keys that can't be the same (due to hash definition) # Those are the top level folders keys_diff = [self.tmp_root_path, os.path.join(self.tmp_root_path, b'sample-folder'), git.ROOT_TREE_KEY] for k in keys_diff: self.assertNotEquals(walk0[k], walk1[k]) # The remaining keys (bottom path) should have exactly the same hashes # as before keys = set(keys1) - set(keys_diff) actual_walk1 = {} for k in keys: self.assertEquals(walk0[k], walk1[k]) actual_walk1[k] = walk1[k] expected_checksums = { os.path.join(self.tmp_root_path, b'sample-folder/empty-folder'): [], # noqa os.path.join(self.tmp_root_path, b'sample-folder/bar/barfoo'): [{ # noqa 'type': git.GitType.BLOB, # noqa 'length': 72, 'sha256': b'=\xb5\xae\x16\x80U\xbc\xd9:M\x08(]\xc9\x9f\xfe\xe2\x883\x03\xb2?\xac^\xab\x85\x02s\xa8\xeaUF', # noqa 'name': b'another-quote.org', # noqa 'path': os.path.join(self.tmp_root_path, b'sample-folder/bar/barfoo/another-quote.org'), # noqa 'perms': git.GitPerm.BLOB, # noqa 'sha1': b'\x90\xa6\x13\x8b\xa5\x99\x15&\x1e\x17\x99H8j\xa1\xcc*\xa9"\n', # noqa 'sha1_git': b'\x136\x93\xb1%\xba\xd2\xb4\xac1\x855\xb8I\x01\xeb\xb1\xf6\xb68'}], # noqa os.path.join(self.tmp_root_path, b'sample-folder/bar'): [{ # noqa 'type': git.GitType.TREE, # noqa 'perms': git.GitPerm.TREE, # noqa 'name': b'barfoo', # noqa 'path': os.path.join(self.tmp_root_path, b'sample-folder/bar/barfoo'), # noqa 'sha1_git': b'\xc3\x02\x0fk\xf15\xa3\x8cm\xf3\xaf\xeb_\xb3\x822\xc5\xe0p\x87'}]} # noqa self.assertEquals(actual_walk1, expected_checksums) @istest def walk_and_compute_sha1_from_directory_without_root_tree(self): # compute the full checksums expected_hashes = git.walk_and_compute_sha1_from_directory( self.tmp_root_path) # except for the key on that round actual_hashes = git.walk_and_compute_sha1_from_directory( self.tmp_root_path, with_root_tree=False) # then, removing the root tree hash from the first round del expected_hashes[git.ROOT_TREE_KEY] # should give us the same checksums as the second round self.assertEquals(actual_hashes, expected_hashes) class WithSampleFolderChecksums: def setUp(self): super().setUp() self.rootkey = b'/tmp/tmp7w3oi_j8' self.objects = { b'/tmp/tmp7w3oi_j8': { 'children': {b'/tmp/tmp7w3oi_j8/sample-folder'}, 'checksums': { 'type': git.GitType.TREE, 'name': b'tmp7w3oi_j8', 'sha1_git': b'\xa7A\xfcM\x96\x8c{\x8e<\x94\xff\x86\xe7\x04\x80\xc5\xc7\xe5r\xa9', # noqa 'path': b'/tmp/tmp7w3oi_j8', 'perms': git.GitPerm.TREE }, }, b'/tmp/tmp7w3oi_j8/sample-folder': { 'children': { b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder', b'/tmp/tmp7w3oi_j8/sample-folder/link-to-binary', b'/tmp/tmp7w3oi_j8/sample-folder/link-to-another-quote', b'/tmp/tmp7w3oi_j8/sample-folder/link-to-foo', b'/tmp/tmp7w3oi_j8/sample-folder/some-binary', b'/tmp/tmp7w3oi_j8/sample-folder/bar', b'/tmp/tmp7w3oi_j8/sample-folder/foo', }, 'checksums': { 'type': git.GitType.TREE, 'name': b'sample-folder', 'sha1_git': b'\xe8\xb0\xf1Fj\xf8`\x8c\x8a?\xb9\x87\x9d\xb1r\xb8\x87\xe8\x07Y', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder', 'perms': git.GitPerm.TREE} }, b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder': { 'children': {}, 'checksums': { 'type': git.GitType.TREE, 'name': b'empty-folder', 'sha1_git': b'K\x82]\xc6B\xcbn\xb9\xa0`\xe5K\xf8\xd6\x92\x88\xfb\xeeI\x04', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder', 'perms': git.GitPerm.TREE } }, b'/tmp/tmp7w3oi_j8/sample-folder/link-to-binary': { 'checksums': { 'name': b'link-to-binary', 'sha1': b'\xd0$\x87\x14\x94\x8b:H\xa2T8#*o\x99\xf01\x8fY\xf1', # noqa 'data': b'some-binary', 'sha1_git': b'\xe8kE\xe58\xd9\xb6\x88\x8c\x96\x9c\x89\xfb\xd2*\x85\xaa\x0e\x03f', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/link-to-binary', 'sha256': b'\x14\x12n\x97\xd8?}&\x1cZh\x89\xce\xe76\x19w\x0f\xf0\x9e@\xc5I\x86\x85\xab\xa7E\xbe\x88.\xff', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 11 } }, b'/tmp/tmp7w3oi_j8/sample-folder/link-to-another-quote': { 'checksums': { 'name': b'link-to-another-quote', 'sha1': b'\xcb\xee\xd1^yY\x9c\x90\xdes\x83\xf4 \xfe\xd7\xac\xb4\x8e\xa1q', # noqa 'data': b'bar/barfoo/another-quote.org', 'sha1_git': b'}\\\x08\x11\x1e!\xc8\xa9\xf7\x15@\x93\x99\x98U\x16\x837_\xad', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/link-to-another-quote', # noqa 'sha256': b'\xe6\xe1}\x07\x93\xaau\n\x04@\xeb\x9a\xd5\xb8\x0b%\x80vc~\xf0\xfbh\xf3\xac.Y\xe4\xb9\xac;\xa6', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 28 } }, b'/tmp/tmp7w3oi_j8/sample-folder/link-to-foo': { 'checksums': { 'name': b'link-to-foo', 'sha1': b'\x0b\xee\xc7\xb5\xea?\x0f\xdb\xc9]\r\xd4\x7f<[\xc2u\xda\x8a3', # noqa 'data': b'foo', 'sha1_git': b'\x19\x10(\x15f=#\xf8\xb7ZG\xe7\xa0\x19e\xdc\xdc\x96F\x8c', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/link-to-foo', 'sha256': b',&\xb4kh\xff\xc6\x8f\xf9\x9bE<\x1d0A4\x13B-pd\x83\xbf\xa0\xf9\x8a^\x88bf\xe7\xae', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 3 } }, b'/tmp/tmp7w3oi_j8/sample-folder/some-binary': { 'checksums': { 'name': b'some-binary', 'sha1': b'\x0b\xbc\x12\xd7\xf4\xa2\xa1[\x14=\xa8F\x17\xd9\\\xb2#\xc9\xb2<', # noqa 'sha1_git': b'hv\x95y\xc3\xea\xad\xbeUSy\xb9\xc3S\x8ef(\xba\xe1\xeb', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/some-binary', 'sha256': b'\xba\xc6P\xd3Jv8\xbb\n\xebSBdm$\xe3\xb9\xadkD\xc9\xb3\x83b\x1f\xaaH+\x99\n6}', # noqa 'perms': git.GitPerm.EXEC, 'type': git.GitType.BLOB, 'length': 5} }, b'/tmp/tmp7w3oi_j8/sample-folder/bar': { 'children': {b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo'}, 'checksums': {'type': git.GitType.TREE, 'name': b'bar', 'sha1_git': b'<\x1fW\x83\x94\xf4b?t\xa0\xba\x7f\xe7ar\x9fY\xfcn\xc4', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/bar', 'perms': git.GitPerm.TREE}, }, b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo': { 'children': {b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo/another-quote.org'}, # noqa 'checksums': {'type': git.GitType.TREE, 'name': b'barfoo', 'sha1_git': b'\xc3\x02\x0fk\xf15\xa3\x8cm\xf3\xaf\xeb_\xb3\x822\xc5\xe0p\x87', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo', # noqa 'perms': git.GitPerm.TREE}, }, b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo/another-quote.org': { 'checksums': {'name': b'another-quote.org', 'sha1': b'\x90\xa6\x13\x8b\xa5\x99\x15&\x1e\x17\x99H8j\xa1\xcc*\xa9"\n', # noqa 'sha1_git': b'\x136\x93\xb1%\xba\xd2\xb4\xac1\x855\xb8I\x01\xeb\xb1\xf6\xb68', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo/another-quote.org', # noqa 'sha256': b'=\xb5\xae\x16\x80U\xbc\xd9:M\x08(]\xc9\x9f\xfe\xe2\x883\x03\xb2?\xac^\xab\x85\x02s\xa8\xeaUF', # noqa 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'length': 72} }, b'/tmp/tmp7w3oi_j8/sample-folder/foo': { 'children': { b'/tmp/tmp7w3oi_j8/sample-folder/foo/barfoo', b'/tmp/tmp7w3oi_j8/sample-folder/foo/rel-link-to-barfoo', b'/tmp/tmp7w3oi_j8/sample-folder/foo/quotes.md', }, 'checksums': {'type': git.GitType.TREE, 'name': b'foo', 'sha1_git': b'+A\xc4\x0f\r\x1f\xbf\xfc\xba\x12I}\xb7\x1f\xba\x83\xfc\xca\x96\xe5', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo', 'perms': git.GitPerm.TREE} }, b'/tmp/tmp7w3oi_j8/sample-folder/foo/barfoo': { 'checksums': {'name': b'barfoo', 'sha1': b'\x90W\xeem\x01bPn\x01\xc4\xd9\xd5E\x9az\xdd\x1f\xed\xac7', # noqa 'data': b'bar/barfoo', 'sha1_git': b'\x81\x85\xdf\xb2\xc0\xc2\xc5\x97\xd1ou\xa8\xa0\xc3vhV|=~', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo/barfoo', # noqa 'sha256': b')\xad?W%2\x1b\x94\x032\xc7\x8e@6\x01\xaf\xffa\xda\xea\x85\xe9\xc8\x0bJpc\xb6\x88~\xadh', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 10} }, b'/tmp/tmp7w3oi_j8/sample-folder/foo/rel-link-to-barfoo': { 'checksums': {'name': b'rel-link-to-barfoo', 'sha1': b'\xdcQ"\x1d0\x8f:\xeb\'T\xdbH9\x1b\x85h|(i\xf4', # noqa 'data': b'../bar/barfoo', 'sha1_git': b'\xac\xac2m\xddc\xb0\xbcp\x84\x06Y\xd4\xacCa\x94\x84\xe6\x9f', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo/rel-link-to-barfoo', # noqa 'sha256': b'\x80\x07\xd2\r\xb2\xaf@C_B\xdd\xefK\x8a\xd7k\x80\xad\xbe\xc2k$\x9f\xdf\x04s5?\x8d\x99\xdf\x08', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 13} }, b'/tmp/tmp7w3oi_j8/sample-folder/foo/quotes.md': { 'checksums': {'name': b'quotes.md', 'sha1': b'\x1b\xf0\xbbr\x1a\xc9,\x18\xa1\x9b\x13\xc0\xeb=t\x1c\xbf\xad\xeb\xfc', # noqa 'sha1_git': b'|LW\xba\x9f\xf4\x96\xad\x17\x9b\x8fe\xb1\xd2\x86\xed\xbd\xa3L\x9a', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo/quotes.md', # noqa 'sha256': b'\xca\xca\x94*\xed\xa7\xb3\x08\x85\x9e\xb5o\x90\x9e\xc9m\x07\xa4\x99I\x16\x90\xc4S\xf7;\x98\x00\xa9;\x16Y', # noqa 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'length': 66} }, } class TestObjectsPerType(WithSampleFolderChecksums, unittest.TestCase): @istest def objects_per_type_blob(self): # given expected_blobs = [ { 'name': b'another-quote.org', 'sha1': b'\x90\xa6\x13\x8b\xa5\x99\x15&\x1e\x17\x99H8j\xa1\xcc*\xa9"\n', # noqa 'sha1_git': b'\x136\x93\xb1%\xba\xd2\xb4\xac1\x855\xb8I\x01\xeb\xb1\xf6\xb68', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo/another-quote.org', # noqa 'sha256': b'=\xb5\xae\x16\x80U\xbc\xd9:M\x08(]\xc9\x9f\xfe\xe2\x883\x03\xb2?\xac^\xab\x85\x02s\xa8\xeaUF', # noqa 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'length': 72 }, { 'name': b'link-to-binary', 'sha1': b'\xd0$\x87\x14\x94\x8b:H\xa2T8#*o\x99\xf01\x8fY\xf1', 'data': b'some-binary', 'sha1_git': b'\xe8kE\xe58\xd9\xb6\x88\x8c\x96\x9c\x89\xfb\xd2*\x85\xaa\x0e\x03f', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/link-to-binary', 'sha256': b'\x14\x12n\x97\xd8?}&\x1cZh\x89\xce\xe76\x19w\x0f\xf0\x9e@\xc5I\x86\x85\xab\xa7E\xbe\x88.\xff', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 11 }, { 'name': b'link-to-another-quote', 'sha1': b'\xcb\xee\xd1^yY\x9c\x90\xdes\x83\xf4 \xfe\xd7\xac\xb4\x8e\xa1q', # noqa 'data': b'bar/barfoo/another-quote.org', 'sha1_git': b'}\\\x08\x11\x1e!\xc8\xa9\xf7\x15@\x93\x99\x98U\x16\x837_\xad', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/link-to-another-quote', # noqa 'sha256': b'\xe6\xe1}\x07\x93\xaau\n\x04@\xeb\x9a\xd5\xb8\x0b%\x80vc~\xf0\xfbh\xf3\xac.Y\xe4\xb9\xac;\xa6', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 28 }, { 'name': b'link-to-foo', 'sha1': b'\x0b\xee\xc7\xb5\xea?\x0f\xdb\xc9]\r\xd4\x7f<[\xc2u\xda\x8a3', # noqa 'data': b'foo', 'sha1_git': b'\x19\x10(\x15f=#\xf8\xb7ZG\xe7\xa0\x19e\xdc\xdc\x96F\x8c', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/link-to-foo', 'sha256': b',&\xb4kh\xff\xc6\x8f\xf9\x9bE<\x1d0A4\x13B-pd\x83\xbf\xa0\xf9\x8a^\x88bf\xe7\xae', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 3 }, { 'name': b'some-binary', 'sha1': b'\x0b\xbc\x12\xd7\xf4\xa2\xa1[\x14=\xa8F\x17\xd9\\\xb2#\xc9\xb2<', # noqa 'sha1_git': b'hv\x95y\xc3\xea\xad\xbeUSy\xb9\xc3S\x8ef(\xba\xe1\xeb', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/some-binary', 'sha256': b'\xba\xc6P\xd3Jv8\xbb\n\xebSBdm$\xe3\xb9\xadkD\xc9\xb3\x83b\x1f\xaaH+\x99\n6}', # noqa 'perms': git.GitPerm.EXEC, 'type': git.GitType.BLOB, 'length': 5 }, { 'name': b'barfoo', 'sha1': b'\x90W\xeem\x01bPn\x01\xc4\xd9\xd5E\x9az\xdd\x1f\xed\xac7', # noqa 'data': b'bar/barfoo', 'sha1_git': b'\x81\x85\xdf\xb2\xc0\xc2\xc5\x97\xd1ou\xa8\xa0\xc3vhV|=~', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo/barfoo', 'sha256': b')\xad?W%2\x1b\x94\x032\xc7\x8e@6\x01\xaf\xffa\xda\xea\x85\xe9\xc8\x0bJpc\xb6\x88~\xadh', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 10 }, { 'name': b'rel-link-to-barfoo', 'sha1': b'\xdcQ"\x1d0\x8f:\xeb\'T\xdbH9\x1b\x85h|(i\xf4', 'data': b'../bar/barfoo', 'sha1_git': b'\xac\xac2m\xddc\xb0\xbcp\x84\x06Y\xd4\xacCa\x94\x84\xe6\x9f', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo/rel-link-to-barfoo', # noqa 'sha256': b'\x80\x07\xd2\r\xb2\xaf@C_B\xdd\xefK\x8a\xd7k\x80\xad\xbe\xc2k$\x9f\xdf\x04s5?\x8d\x99\xdf\x08', # noqa 'perms': git.GitPerm.LINK, 'type': git.GitType.BLOB, 'length': 13 }, { 'name': b'quotes.md', 'sha1': b'\x1b\xf0\xbbr\x1a\xc9,\x18\xa1\x9b\x13\xc0\xeb=t\x1c\xbf\xad\xeb\xfc', # noqa 'sha1_git': b'|LW\xba\x9f\xf4\x96\xad\x17\x9b\x8fe\xb1\xd2\x86\xed\xbd\xa3L\x9a', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo/quotes.md', 'sha256': b'\xca\xca\x94*\xed\xa7\xb3\x08\x85\x9e\xb5o\x90\x9e\xc9m\x07\xa4\x99I\x16\x90\xc4S\xf7;\x98\x00\xa9;\x16Y', # noqa 'perms': git.GitPerm.BLOB, 'type': git.GitType.BLOB, 'length': 66 }, ] expected_sha1_blobs = set( ((c['sha1_git'], git.GitType.BLOB) for c in expected_blobs)) # when actual_sha1_blobs = set( ((c['sha1_git'], c['type']) for c in git.objects_per_type(git.GitType.BLOB, self.objects))) # then self.assertEqual(actual_sha1_blobs, expected_sha1_blobs) @istest def objects_per_type_tree(self): def _children_hashes(path, objects=self.objects): return set((c['sha1_git'] for c in git.children_hashes( objects[path]['children'], objects))) expected_trees = [ { 'type': git.GitType.TREE, 'name': b'tmp7w3oi_j8', 'sha1_git': b'\xa7A\xfcM\x96\x8c{\x8e<\x94\xff\x86\xe7\x04\x80\xc5\xc7\xe5r\xa9', # noqa 'path': b'/tmp/tmp7w3oi_j8', 'perms': git.GitPerm.TREE, # we only add children's sha1_git here, in reality, # it's a full dict of hashes. 'children': _children_hashes(b'/tmp/tmp7w3oi_j8') }, { 'type': git.GitType.TREE, 'name': b'sample-folder', 'sha1_git': b'\xe8\xb0\xf1Fj\xf8`\x8c\x8a?\xb9\x87\x9d\xb1r\xb8\x87\xe8\x07Y', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder', 'perms': git.GitPerm.TREE, 'children': _children_hashes( b'/tmp/tmp7w3oi_j8/sample-folder') }, { 'type': git.GitType.TREE, 'name': b'empty-folder', 'sha1_git': b'K\x82]\xc6B\xcbn\xb9\xa0`\xe5K\xf8\xd6\x92\x88\xfb\xeeI\x04', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder', 'perms': git.GitPerm.TREE, 'children': _children_hashes( b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder') }, { 'type': git.GitType.TREE, 'name': b'bar', 'sha1_git': b'<\x1fW\x83\x94\xf4b?t\xa0\xba\x7f\xe7ar\x9fY\xfcn\xc4', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/bar', 'perms': git.GitPerm.TREE, 'children': _children_hashes( b'/tmp/tmp7w3oi_j8/sample-folder/bar') }, { 'type': git.GitType.TREE, 'name': b'barfoo', 'sha1_git': b'\xc3\x02\x0fk\xf15\xa3\x8cm\xf3\xaf\xeb_\xb3\x822\xc5\xe0p\x87', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo', 'perms': git.GitPerm.TREE, 'children': _children_hashes( b'/tmp/tmp7w3oi_j8/sample-folder/bar/barfoo'), }, { 'type': git.GitType.TREE, 'name': b'foo', 'sha1_git': b'+A\xc4\x0f\r\x1f\xbf\xfc\xba\x12I}\xb7\x1f\xba\x83\xfc\xca\x96\xe5', # noqa 'path': b'/tmp/tmp7w3oi_j8/sample-folder/foo', 'perms': git.GitPerm.TREE, 'children': _children_hashes( b'/tmp/tmp7w3oi_j8/sample-folder/foo') }, ] expected_sha1_trees = list( ((c['sha1_git'], git.GitType.TREE, c['children']) for c in expected_trees)) # when actual_sha1_trees = list( ((c['sha1_git'], c['type'], _children_hashes(c['path'])) for c in git.objects_per_type(git.GitType.TREE, self.objects))) self.assertEquals(len(actual_sha1_trees), len(expected_sha1_trees)) for e in actual_sha1_trees: self.assertTrue(e in expected_sha1_trees) class TestComputeHashesFromDirectory(WithSampleFolderChecksums, GitHashWalkArborescenceTree, unittest.TestCase): def __adapt_object_to_rootpath(self, rootpath): def _replace_slash(s, rootpath=self.rootkey, newrootpath=rootpath): return s.replace(rootpath, newrootpath) def _update_children(children): return set((_replace_slash(c) for c in children)) # given expected_objects = {} for path, v in self.objects.items(): p = _replace_slash(path) v['checksums']['path'] = _replace_slash(v['checksums']['path']) v['checksums']['name'] = os.path.basename(v['checksums']['path']) if 'children' in v: v['children'] = _update_children(v['children']) expected_objects[p] = v return expected_objects @istest def compute_hashes_from_directory_default(self): # given expected_objects = self.__adapt_object_to_rootpath(self.tmp_root_path) # when actual_hashes = git.compute_hashes_from_directory(self.tmp_root_path) # then self.assertEquals(actual_hashes, expected_objects) @istest def compute_hashes_from_directory_no_empty_folder(self): # given def _replace_slash(s, rootpath=self.rootkey, newrootpath=self.tmp_root_path): return s.replace(rootpath, newrootpath) expected_objects = self.__adapt_object_to_rootpath(self.tmp_root_path) # when actual_hashes = git.compute_hashes_from_directory( self.tmp_root_path, remove_empty_folder=True) # then # One folder less, so plenty of hashes are different now self.assertNotEquals(actual_hashes, expected_objects) keys = set(actual_hashes.keys()) assert (b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder' in self.objects.keys()) new_empty_folder_path = _replace_slash( b'/tmp/tmp7w3oi_j8/sample-folder/empty-folder') self.assertNotIn(new_empty_folder_path, keys) self.assertEqual(len(keys), len(expected_objects.keys()) - 1) @istest def compute_hashes_from_directory_ignore_some_folder(self): # given def _replace_slash(s, rootpath=self.rootkey, newrootpath=self.tmp_root_path): return s.replace(rootpath, newrootpath) ignore_path = b'/tmp/tmp7w3oi_j8/sample-folder' # when actual_hashes = git.compute_hashes_from_directory( self.tmp_root_path, dir_ok_fn=lambda dirpath: b'sample-folder' not in dirpath) # then # One entry less, so plenty of hashes are different now keys = set(actual_hashes.keys()) assert ignore_path in self.objects.keys() new_ignore_path = _replace_slash(ignore_path) self.assertNotIn(new_ignore_path, keys) # top level directory contains the folder to ignore self.assertEqual(len(keys), 1) diff --git a/swh/model/tests/test_hashutil.py b/swh/model/tests/test_hashutil.py index f795e87..614e7ee 100644 --- a/swh/model/tests/test_hashutil.py +++ b/swh/model/tests/test_hashutil.py @@ -1,112 +1,210 @@ -# Copyright (C) 2015 The Software Heritage developers +# Copyright (C) 2015-2017 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 io import tempfile import unittest from nose.tools import istest from swh.model import hashutil class Hashutil(unittest.TestCase): def setUp(self): self.data = b'1984\n' self.hex_checksums = { 'sha1': '62be35bf00ff0c624f4a621e2ea5595a049e0731', 'sha1_git': '568aaf43d83b2c3df8067f3bedbb97d83260be6d', 'sha256': '26602113b4b9afd9d55466b08580d3c2' '4a9b50ee5b5866c0d91fab0e65907311', } self.checksums = { type: bytes.fromhex(cksum) for type, cksum in self.hex_checksums.items() } self.git_hex_checksums = { 'blob': self.hex_checksums['sha1_git'], 'tree': '5b2e883aa33d2efab98442693ea4dd5f1b8871b0', 'commit': '79e4093542e72f0fcb7cbd75cb7d270f9254aa8f', 'tag': 'd6bf62466f287b4d986c545890716ce058bddf67', } self.git_checksums = { type: bytes.fromhex(cksum) for type, cksum in self.git_hex_checksums.items() } @istest def hash_data(self): checksums = hashutil.hash_data(self.data) self.assertEqual(checksums, self.checksums) @istest def hash_data_unknown_hash(self): with self.assertRaises(ValueError) as cm: hashutil.hash_data(self.data, ['unknown-hash']) self.assertIn('Unexpected hashing algorithm', cm.exception.args[0]) self.assertIn('unknown-hash', cm.exception.args[0]) @istest def hash_git_data(self): checksums = { git_type: hashutil.hash_git_data(self.data, git_type) for git_type in self.git_checksums } self.assertEqual(checksums, self.git_checksums) @istest def hash_git_data_unknown_git_type(self): with self.assertRaises(ValueError) as cm: hashutil.hash_git_data(self.data, 'unknown-git-type') self.assertIn('Unexpected git object type', cm.exception.args[0]) self.assertIn('unknown-git-type', cm.exception.args[0]) @istest def hash_file(self): fobj = io.BytesIO(self.data) checksums = hashutil.hash_file(fobj, length=len(self.data)) self.assertEqual(checksums, self.checksums) @istest def hash_file_missing_length(self): fobj = io.BytesIO(self.data) with self.assertRaises(ValueError) as cm: hashutil.hash_file(fobj, algorithms=['sha1_git']) self.assertIn('Missing length', cm.exception.args[0]) @istest def hash_path(self): with tempfile.NamedTemporaryFile(delete=False) as f: f.write(self.data) f.close() hashes = hashutil.hash_path(f.name) self.checksums['length'] = len(self.data) self.assertEquals(self.checksums, hashes) @istest def hash_to_hex(self): for type in self.checksums: hex = self.hex_checksums[type] hash = self.checksums[type] self.assertEquals(hashutil.hash_to_hex(hex), hex) self.assertEquals(hashutil.hash_to_hex(hash), hex) @istest def hash_to_bytes(self): for type in self.checksums: hex = self.hex_checksums[type] hash = self.checksums[type] self.assertEquals(hashutil.hash_to_bytes(hex), hash) self.assertEquals(hashutil.hash_to_bytes(hash), hash) + + @istest + def hash_to_bytehex(self): + for algo in self.checksums: + self.assertEqual(self.hex_checksums[algo].encode('ascii'), + hashutil.hash_to_bytehex(self.checksums[algo])) + + @istest + def bytehex_to_hash(self): + for algo in self.checksums: + self.assertEqual(self.checksums[algo], + hashutil.bytehex_to_hash( + self.hex_checksums[algo].encode())) + + +class HashlibGit(unittest.TestCase): + + def setUp(self): + self.blob_data = b'42\n' + + self.tree_data = b''.join([b'40000 barfoo\0', + bytes.fromhex('c3020f6bf135a38c6df' + '3afeb5fb38232c5e07087'), + b'100644 blah\0', + bytes.fromhex('63756ef0df5e4f10b6efa' + '33cfe5c758749615f20'), + b'100644 hello\0', + bytes.fromhex('907b308167f0880fb2a' + '5c0e1614bb0c7620f9dc3')]) + + self.commit_data = """tree 1c61f7259dcb770f46b194d941df4f08ff0a3970 +author Antoine R. Dumont (@ardumont) 1444054085 +0200 +committer Antoine R. Dumont (@ardumont) 1444054085 +0200 + +initial +""".encode('utf-8') # NOQA + self.tag_data = """object 24d012aaec0bc5a4d2f62c56399053d6cc72a241 +type commit +tag 0.0.1 +tagger Antoine R. Dumont (@ardumont) 1444225145 +0200 + +blah +""".encode('utf-8') # NOQA + + self.checksums = { + 'blob_sha1_git': bytes.fromhex('d81cc0710eb6cf9efd5b920a8453e1' + 'e07157b6cd'), + 'tree_sha1_git': bytes.fromhex('ac212302c45eada382b27bfda795db' + '121dacdb1c'), + 'commit_sha1_git': bytes.fromhex('e960570b2e6e2798fa4cfb9af2c399' + 'd629189653'), + 'tag_sha1_git': bytes.fromhex('bc2b99ba469987bcf1272c189ed534' + 'e9e959f120'), + } + + @istest + def unknown_header_type(self): + with self.assertRaises(ValueError) as cm: + hashutil.hash_git_data(b'any-data', 'some-unknown-type') + + self.assertIn('Unexpected git object type', cm.exception.args[0]) + + @istest + def hashdata_content(self): + # when + actual_hash = hashutil.hash_git_data(self.blob_data, git_type='blob') + + # then + self.assertEqual(actual_hash, + self.checksums['blob_sha1_git']) + + @istest + def hashdata_tree(self): + # when + actual_hash = hashutil.hash_git_data(self.tree_data, git_type='tree') + + # then + self.assertEqual(actual_hash, + self.checksums['tree_sha1_git']) + + @istest + def hashdata_revision(self): + # when + actual_hash = hashutil.hash_git_data(self.commit_data, + git_type='commit') + + # then + self.assertEqual(actual_hash, + self.checksums['commit_sha1_git']) + + @istest + def hashdata_tag(self): + # when + actual_hash = hashutil.hash_git_data(self.tag_data, git_type='tag') + + # then + self.assertEqual(actual_hash, + self.checksums['tag_sha1_git']) diff --git a/swh/model/tests/test_identifiers.py b/swh/model/tests/test_identifiers.py index 16a34bb..755dba6 100644 --- a/swh/model/tests/test_identifiers.py +++ b/swh/model/tests/test_identifiers.py @@ -1,651 +1,681 @@ -# Copyright (C) 2015 The Software Heritage developers +# Copyright (C) 2015-2017 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 binascii import datetime import unittest from nose.tools import istest from swh.model import hashutil, identifiers class UtilityFunctionsIdentifier(unittest.TestCase): def setUp(self): self.str_id = 'c2e41aae41ac17bd4a650770d6ee77f62e52235b' self.bytes_id = binascii.unhexlify(self.str_id) self.bad_type_id = object() @istest def identifier_to_bytes(self): for id in [self.str_id, self.bytes_id]: self.assertEqual(identifiers.identifier_to_bytes(id), self.bytes_id) # wrong length with self.assertRaises(ValueError) as cm: identifiers.identifier_to_bytes(id[:-2]) self.assertIn('length', str(cm.exception)) with self.assertRaises(ValueError) as cm: identifiers.identifier_to_bytes(self.bad_type_id) self.assertIn('type', str(cm.exception)) @istest def identifier_to_str(self): for id in [self.str_id, self.bytes_id]: self.assertEqual(identifiers.identifier_to_str(id), self.str_id) # wrong length with self.assertRaises(ValueError) as cm: identifiers.identifier_to_str(id[:-2]) self.assertIn('length', str(cm.exception)) with self.assertRaises(ValueError) as cm: identifiers.identifier_to_str(self.bad_type_id) self.assertIn('type', str(cm.exception)) class UtilityFunctionsDateOffset(unittest.TestCase): def setUp(self): self.dates = { b'1448210036': { 'seconds': 1448210036, 'microseconds': 0, }, b'1448210036.002342': { 'seconds': 1448210036, 'microseconds': 2342, }, b'1448210036.12': { 'seconds': 1448210036, 'microseconds': 120000, } } self.broken_dates = [ 1448210036.12, ] self.offsets = { 0: b'+0000', -630: b'-1030', 800: b'+1320', } @istest def format_date(self): for date_repr, date in self.dates.items(): self.assertEqual(identifiers.format_date(date), date_repr) @istest def format_date_fail(self): for date in self.broken_dates: with self.assertRaises(ValueError): identifiers.format_date(date) @istest def format_offset(self): for offset, res in self.offsets.items(): self.assertEqual(identifiers.format_offset(offset), res) class ContentIdentifier(unittest.TestCase): def setUp(self): self.content = { 'status': 'visible', 'length': 5, 'data': b'1984\n', 'ctime': datetime.datetime(2015, 11, 22, 16, 33, 56, tzinfo=datetime.timezone.utc), } self.content_id = hashutil.hash_data(self.content['data']) @istest def content_identifier(self): self.assertEqual(identifiers.content_identifier(self.content), self.content_id) class DirectoryIdentifier(unittest.TestCase): def setUp(self): self.directory = { 'id': 'c2e41aae41ac17bd4a650770d6ee77f62e52235b', 'entries': [ { 'type': 'file', 'perms': 33188, 'name': b'README', 'target': '37ec8ea2110c0b7a32fbb0e872f6e7debbf95e21' }, { 'type': 'file', 'perms': 33188, 'name': b'Rakefile', 'target': '3bb0e8592a41ae3185ee32266c860714980dbed7' }, { 'type': 'dir', 'perms': 16384, 'name': b'app', 'target': '61e6e867f5d7ba3b40540869bc050b0c4fed9e95' }, { 'type': 'file', 'perms': 33188, 'name': b'1.megabyte', 'target': '7c2b2fbdd57d6765cdc9d84c2d7d333f11be7fb3' }, { 'type': 'dir', 'perms': 16384, 'name': b'config', 'target': '591dfe784a2e9ccc63aaba1cb68a765734310d98' }, { 'type': 'dir', 'perms': 16384, 'name': b'public', 'target': '9588bf4522c2b4648bfd1c61d175d1f88c1ad4a5' }, { 'type': 'file', 'perms': 33188, 'name': b'development.sqlite3', 'target': 'e69de29bb2d1d6434b8b29ae775ad8c2e48c5391' }, { 'type': 'dir', 'perms': 16384, 'name': b'doc', 'target': '154705c6aa1c8ead8c99c7915373e3c44012057f' }, { 'type': 'dir', 'perms': 16384, 'name': b'db', 'target': '85f157bdc39356b7bc7de9d0099b4ced8b3b382c' }, { 'type': 'dir', 'perms': 16384, 'name': b'log', 'target': '5e3d3941c51cce73352dff89c805a304ba96fffe' }, { 'type': 'dir', 'perms': 16384, 'name': b'script', 'target': '1b278423caf176da3f3533592012502aa10f566c' }, { 'type': 'dir', 'perms': 16384, 'name': b'test', 'target': '035f0437c080bfd8711670b3e8677e686c69c763' }, { 'type': 'dir', 'perms': 16384, 'name': b'vendor', 'target': '7c0dc9ad978c1af3f9a4ce061e50f5918bd27138' }, { 'type': 'rev', 'perms': 57344, 'name': b'will_paginate', 'target': '3d531e169db92a16a9a8974f0ae6edf52e52659e' } ], } self.empty_directory = { 'id': '4b825dc642cb6eb9a060e54bf8d69288fbee4904', 'entries': [], } @istest def dir_identifier(self): self.assertEqual( identifiers.directory_identifier(self.directory), self.directory['id']) @istest def dir_identifier_empty_directory(self): self.assertEqual( identifiers.directory_identifier(self.empty_directory), self.empty_directory['id']) class RevisionIdentifier(unittest.TestCase): def setUp(self): linus_tz = datetime.timezone(datetime.timedelta(minutes=-420)) gpgsig = b'''\ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.13 (Darwin) iQIcBAABAgAGBQJVJcYsAAoJEBiY3kIkQRNJVAUQAJ8/XQIfMqqC5oYeEFfHOPYZ L7qy46bXHVBa9Qd8zAJ2Dou3IbI2ZoF6/Et89K/UggOycMlt5FKV/9toWyuZv4Po L682wonoxX99qvVTHo6+wtnmYO7+G0f82h+qHMErxjP+I6gzRNBvRr+SfY7VlGdK wikMKOMWC5smrScSHITnOq1Ews5pe3N7qDYMzK0XVZmgDoaem4RSWMJs4My/qVLN e0CqYWq2A22GX7sXl6pjneJYQvcAXUX+CAzp24QnPSb+Q22Guj91TcxLFcHCTDdn qgqMsEyMiisoglwrCbO+D+1xq9mjN9tNFWP66SQ48mrrHYTBV5sz9eJyDfroJaLP CWgbDTgq6GzRMehHT3hXfYS5NNatjnhkNISXR7pnVP/obIi/vpWh5ll6Gd8q26z+ a/O41UzOaLTeNI365MWT4/cnXohVLRG7iVJbAbCxoQmEgsYMRc/pBAzWJtLfcB2G jdTswYL6+MUdL8sB9pZ82D+BP/YAdHe69CyTu1lk9RT2pYtI/kkfjHubXBCYEJSG +VGllBbYG6idQJpyrOYNRJyrDi9yvDJ2W+S0iQrlZrxzGBVGTB/y65S8C+2WTBcE lf1Qb5GDsQrZWgD+jtWTywOYHtCBwyCKSAXxSARMbNPeak9WPlcW/Jmu+fUcMe2x dg1KdHOa34shrKDaOVzW =od6m -----END PGP SIGNATURE-----''' self.revision = { 'id': 'bc0195aad0daa2ad5b0d76cce22b167bc3435590', 'directory': '85a74718d377195e1efd0843ba4f3260bad4fe07', 'parents': ['01e2d0627a9a6edb24c37db45db5ecb31e9de808'], 'author': { 'name': b'Linus Torvalds', 'email': b'torvalds@linux-foundation.org', }, 'date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'committer': { 'name': b'Linus Torvalds', 'email': b'torvalds@linux-foundation.org', }, 'committer_date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'message': b'Linux 4.2-rc2\n', } self.revision_none_metadata = { 'id': 'bc0195aad0daa2ad5b0d76cce22b167bc3435590', 'directory': '85a74718d377195e1efd0843ba4f3260bad4fe07', 'parents': ['01e2d0627a9a6edb24c37db45db5ecb31e9de808'], 'author': { 'name': b'Linus Torvalds', 'email': b'torvalds@linux-foundation.org', }, 'date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'committer': { 'name': b'Linus Torvalds', 'email': b'torvalds@linux-foundation.org', }, 'committer_date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'message': b'Linux 4.2-rc2\n', 'metadata': None, } self.synthetic_revision = { 'id': b'\xb2\xa7\xe1&\x04\x92\xe3D\xfa\xb3\xcb\xf9\x1b\xc1<\x91' b'\xe0T&\xfd', 'author': { 'name': b'Software Heritage', 'email': b'robot@softwareheritage.org', }, 'date': { 'timestamp': {'seconds': 1437047495}, 'offset': 0, 'negative_utc': False, }, 'type': 'tar', 'committer': { 'name': b'Software Heritage', 'email': b'robot@softwareheritage.org', }, 'committer_date': 1437047495, 'synthetic': True, 'parents': [None], 'message': b'synthetic revision message\n', 'directory': b'\xd1\x1f\x00\xa6\xa0\xfe\xa6\x05SA\xd2U\x84\xb5\xa9' b'e\x16\xc0\xd2\xb8', 'metadata': {'original_artifact': [ {'archive_type': 'tar', 'name': 'gcc-5.2.0.tar.bz2', 'sha1_git': '39d281aff934d44b439730057e55b055e206a586', 'sha1': 'fe3f5390949d47054b613edc36c557eb1d51c18e', 'sha256': '5f835b04b5f7dd4f4d2dc96190ec1621b8d89f' '2dc6f638f9f8bc1b1014ba8cad'}]}, } # cat commit.txt | git hash-object -t commit --stdin self.revision_with_extra_headers = { 'id': '010d34f384fa99d047cdd5e2f41e56e5c2feee45', 'directory': '85a74718d377195e1efd0843ba4f3260bad4fe07', 'parents': ['01e2d0627a9a6edb24c37db45db5ecb31e9de808'], 'author': { 'name': b'Linus Torvalds', 'email': b'torvalds@linux-foundation.org', 'fullname': b'Linus Torvalds ', }, 'date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'committer': { 'name': b'Linus Torvalds', 'email': b'torvalds@linux-foundation.org', 'fullname': b'Linus Torvalds ', }, 'committer_date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'message': b'Linux 4.2-rc2\n', 'metadata': { 'extra_headers': [ ['svn-repo-uuid', '046f1af7-66c2-d61b-5410-ce57b7db7bff'], ['svn-revision', 10], ] } } self.revision_with_gpgsig = { 'id': '44cc742a8ca17b9c279be4cc195a93a6ef7a320e', 'directory': 'b134f9b7dc434f593c0bab696345548b37de0558', 'parents': ['689664ae944b4692724f13b709a4e4de28b54e57', 'c888305e1efbaa252d01b4e5e6b778f865a97514'], 'author': { 'name': b'Jiang Xin', 'email': b'worldhello.net@gmail.com', 'fullname': b'Jiang Xin ', }, 'date': { 'timestamp': 1428538899, 'offset': 480, }, 'committer': { 'name': b'Jiang Xin', 'email': b'worldhello.net@gmail.com', }, 'committer_date': { 'timestamp': 1428538899, 'offset': 480, }, 'metadata': { 'extra_headers': [ ['gpgsig', gpgsig], ], }, 'message': b'''Merge branch 'master' of git://github.com/alexhenrie/git-po * 'master' of git://github.com/alexhenrie/git-po: l10n: ca.po: update translation ''' } self.revision_no_message = { 'id': '4cfc623c9238fa92c832beed000ce2d003fd8333', 'directory': 'b134f9b7dc434f593c0bab696345548b37de0558', 'parents': ['689664ae944b4692724f13b709a4e4de28b54e57', 'c888305e1efbaa252d01b4e5e6b778f865a97514'], 'author': { 'name': b'Jiang Xin', 'email': b'worldhello.net@gmail.com', 'fullname': b'Jiang Xin ', }, 'date': { 'timestamp': 1428538899, 'offset': 480, }, 'committer': { 'name': b'Jiang Xin', 'email': b'worldhello.net@gmail.com', }, 'committer_date': { 'timestamp': 1428538899, 'offset': 480, }, 'message': None, } self.revision_empty_message = { 'id': '7442cd78bd3b4966921d6a7f7447417b7acb15eb', 'directory': 'b134f9b7dc434f593c0bab696345548b37de0558', 'parents': ['689664ae944b4692724f13b709a4e4de28b54e57', 'c888305e1efbaa252d01b4e5e6b778f865a97514'], 'author': { 'name': b'Jiang Xin', 'email': b'worldhello.net@gmail.com', 'fullname': b'Jiang Xin ', }, 'date': { 'timestamp': 1428538899, 'offset': 480, }, 'committer': { 'name': b'Jiang Xin', 'email': b'worldhello.net@gmail.com', }, 'committer_date': { 'timestamp': 1428538899, 'offset': 480, }, 'message': b'', } self.revision_only_fullname = { 'id': '010d34f384fa99d047cdd5e2f41e56e5c2feee45', 'directory': '85a74718d377195e1efd0843ba4f3260bad4fe07', 'parents': ['01e2d0627a9a6edb24c37db45db5ecb31e9de808'], 'author': { 'fullname': b'Linus Torvalds ', }, 'date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'committer': { 'fullname': b'Linus Torvalds ', }, 'committer_date': datetime.datetime(2015, 7, 12, 15, 10, 30, tzinfo=linus_tz), 'message': b'Linux 4.2-rc2\n', 'metadata': { 'extra_headers': [ ['svn-repo-uuid', '046f1af7-66c2-d61b-5410-ce57b7db7bff'], ['svn-revision', 10], ] } } @istest def revision_identifier(self): self.assertEqual( identifiers.revision_identifier(self.revision), identifiers.identifier_to_str(self.revision['id']), ) @istest def revision_identifier_none_metadata(self): self.assertEqual( identifiers.revision_identifier(self.revision_none_metadata), identifiers.identifier_to_str(self.revision_none_metadata['id']), ) @istest def revision_identifier_synthetic(self): self.assertEqual( identifiers.revision_identifier(self.synthetic_revision), identifiers.identifier_to_str(self.synthetic_revision['id']), ) @istest def revision_identifier_with_extra_headers(self): self.assertEqual( identifiers.revision_identifier( self.revision_with_extra_headers), identifiers.identifier_to_str( self.revision_with_extra_headers['id']), ) @istest def revision_identifier_with_gpgsig(self): self.assertEqual( identifiers.revision_identifier( self.revision_with_gpgsig), identifiers.identifier_to_str( self.revision_with_gpgsig['id']), ) @istest def revision_identifier_no_message(self): self.assertEqual( identifiers.revision_identifier( self.revision_no_message), identifiers.identifier_to_str( self.revision_no_message['id']), ) @istest def revision_identifier_empty_message(self): self.assertEqual( identifiers.revision_identifier( self.revision_empty_message), identifiers.identifier_to_str( self.revision_empty_message['id']), ) @istest def revision_identifier_only_fullname(self): self.assertEqual( identifiers.revision_identifier( self.revision_only_fullname), identifiers.identifier_to_str( self.revision_only_fullname['id']), ) class ReleaseIdentifier(unittest.TestCase): def setUp(self): linus_tz = datetime.timezone(datetime.timedelta(minutes=-420)) self.release = { 'id': '2b10839e32c4c476e9d94492756bb1a3e1ec4aa8', 'target': b't\x1b"R\xa5\xe1Ml`\xa9\x13\xc7z`\x99\xab\xe7:\x85J', 'target_type': 'revision', 'name': b'v2.6.14', 'author': { 'name': b'Linus Torvalds', 'email': b'torvalds@g5.osdl.org', }, 'date': datetime.datetime(2005, 10, 27, 17, 2, 33, tzinfo=linus_tz), 'message': b'''\ Linux 2.6.14 release -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQBDYWq6F3YsRnbiHLsRAmaeAJ9RCez0y8rOBbhSv344h86l/VVcugCeIhO1 wdLOnvj91G4wxYqrvThthbE= =7VeT -----END PGP SIGNATURE----- ''', 'synthetic': False, } self.release_no_author = { 'id': b'&y\x1a\x8b\xcf\x0em3\xf4:\xefv\x82\xbd\xb5U#mV\xde', 'target': '9ee1c939d1cb936b1f98e8d81aeffab57bae46ab', 'target_type': 'revision', 'name': b'v2.6.12', 'message': b'''\ This is the final 2.6.12 release -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQBCsykyF3YsRnbiHLsRAvPNAJ482tCZwuxp/bJRz7Q98MHlN83TpACdHr37 o6X/3T+vm8K3bf3driRr34c= =sBHn -----END PGP SIGNATURE----- ''', 'synthetic': False, } self.release_no_message = { 'id': 'b6f4f446715f7d9543ef54e41b62982f0db40045', 'target': '9ee1c939d1cb936b1f98e8d81aeffab57bae46ab', 'target_type': 'revision', 'name': b'v2.6.12', 'author': { 'name': b'Linus Torvalds', 'email': b'torvalds@g5.osdl.org', }, 'date': datetime.datetime(2005, 10, 27, 17, 2, 33, tzinfo=linus_tz), 'message': None, } self.release_empty_message = { 'id': '71a0aea72444d396575dc25ac37fec87ee3c6492', 'target': '9ee1c939d1cb936b1f98e8d81aeffab57bae46ab', 'target_type': 'revision', 'name': b'v2.6.12', 'author': { 'name': b'Linus Torvalds', 'email': b'torvalds@g5.osdl.org', }, 'date': datetime.datetime(2005, 10, 27, 17, 2, 33, tzinfo=linus_tz), 'message': b'', } self.release_negative_utc = { 'id': '97c8d2573a001f88e72d75f596cf86b12b82fd01', 'name': b'20081029', 'target': '54e9abca4c77421e2921f5f156c9fe4a9f7441c7', 'target_type': 'revision', 'date': { 'timestamp': {'seconds': 1225281976}, 'offset': 0, 'negative_utc': True, }, 'author': { 'name': b'Otavio Salvador', 'email': b'otavio@debian.org', 'id': 17640, }, 'synthetic': False, 'message': b'tagging version 20081029\n\nr56558\n', } + self.release_newline_in_author = { + 'author': { + 'email': b'esycat@gmail.com', + 'fullname': b'Eugene Janusov\n', + 'name': b'Eugene Janusov\n', + }, + 'date': { + 'negative_utc': None, + 'offset': 600, + 'timestamp': { + 'microseconds': 0, + 'seconds': 1377480558, + }, + }, + 'id': b'\\\x98\xf5Y\xd04\x16-\xe2->\xbe\xb9T3\xe6\xf8\x88R1', + 'message': b'Release of v0.3.2.', + 'name': b'0.3.2', + 'synthetic': False, + 'target': (b'\xc0j\xa3\xd9;x\xa2\x86\\I5\x17' + b'\x000\xf8\xc2\xd79o\xd3'), + 'target_type': 'revision', + } + @istest def release_identifier(self): self.assertEqual( identifiers.release_identifier(self.release), identifiers.identifier_to_str(self.release['id']) ) @istest def release_identifier_no_author(self): self.assertEqual( identifiers.release_identifier(self.release_no_author), identifiers.identifier_to_str(self.release_no_author['id']) ) @istest def release_identifier_no_message(self): self.assertEqual( identifiers.release_identifier(self.release_no_message), identifiers.identifier_to_str(self.release_no_message['id']) ) @istest def release_identifier_empty_message(self): self.assertEqual( identifiers.release_identifier(self.release_empty_message), identifiers.identifier_to_str(self.release_empty_message['id']) ) @istest def release_identifier_negative_utc(self): self.assertEqual( identifiers.release_identifier(self.release_negative_utc), identifiers.identifier_to_str(self.release_negative_utc['id']) ) + + @istest + def release_identifier_newline_in_author(self): + self.assertEqual( + identifiers.release_identifier(self.release_newline_in_author), + identifiers.identifier_to_str(self.release_newline_in_author['id']) + ) diff --git a/version.txt b/version.txt index ec62758..ed4e55c 100644 --- a/version.txt +++ b/version.txt @@ -1 +1 @@ -v0.0.13-0-g58c5a24 \ No newline at end of file +v0.0.14-0-g3e325ca \ No newline at end of file