wyag/libwyag/objects.py

import zlib
import hashlib
from abc import ABC, abstractmethod
from libwyag import repository


class GitObject:

    fmt: bytes = b""
    repo = None

    def __init__(self, repo, data=None):
        self.repo = repo
        if data is not None:
            self.deserialize(data)

    @abstractmethod
    def serialize(self):
        """
        This function MUST be implemented by subclasses.
        It must read the object's contents from self.data, a byte string, and do
        whatever it takes to convert it into a meaningful representation.  
        What exactly that means depend on each subclass.
        """
        return NotImplemented

    @abstractmethod
    def deserialize(self, data):
        return NotImplemented


class GitBlob(GitObject):
    fmt = b"blob"

    def serialize(self):
        return self.blobdata

    def deserialize(self, data):
        self.blobdata = data


class GitCommit(GitObject):
    pass


class GitTree(GitObject):
    pass


class GitTag(GitObject):
    pass


def read_object(repo, sha):
    """Read object object_id from Git repository repo.  Return a
    GitObject whose exact type depends on the object."""

    path = repository.repo_file(repo, "objects", sha[0:2], sha[2:])

    with open(path, "rb") as f:
        raw = zlib.decompress(f.read())

        # Read object type
        x = raw.find(b" ")
        fmt = raw[0:x]

        # Read and validate object size
        y = raw.find(b"\x00", x)
        size = int(raw[x:y].decode("ascii"))
        if size != len(raw) - y - 1:
            raise Exception("Malformed object {0}: bad length".format(sha))

        # Pick constructor
        if fmt == b"commit":
            c = GitCommit
        elif fmt == b"tree":
            c = GitTree
        elif fmt == b"tag":
            c = GitTag
        elif fmt == b"blob":
            c = GitBlob
        else:
            raise Exception(f"Unknown type {fmt.decode('ascii')} for object {sha}")

        # Call constructor and return object
        return c(repo, raw[y + 1 :])


def find_object(repo, name: str, fmt=None, follow=True):
    """ Name resolution function """
    pass


def write_object(obj: GitObject, actually_write=True):
    # Serialize object data
    data = obj.serialize()
    # Add header
    result = obj.fmt + b" " + str(len(data)).encode() + b"\x00" + data
    # Compute hash
    sha = hashlib.sha1(result).hexdigest()

    if actually_write:
        # Compute path
        path = repository.repo_file(
            obj.repo, "objects", sha[0:2], sha[2:], mkdir=actually_write
        )

        with open(path, "wb") as f:
            # Compress and write
            f.write(zlib.compress(result))

    return sha


def hash_object(fd, fmt, repo=None):
    data = fd.read()

    # Choose constructor depending on
    # object type found in header.
    if fmt == b"commit":
        obj = GitCommit(repo, data)
    elif fmt == b"tree":
        obj = GitTree(repo, data)
    elif fmt == b"tag":
        obj = GitTag(repo, data)
    elif fmt == b"blob":
        obj = GitBlob(repo, data)
    else:
        raise Exception("Unknown type %s!" % fmt)

    return write_object(obj, repo)