)

from rhodecode.lib.vcs.nodes import (

    FileNode, DirNode, NodeKind, RootNode, RemovedFileNode, SubModuleNode,

    ChangedFileNodesGenerator, AddedFileNodesGenerator, RemovedFileNodesGenerator

)

from rhodecode.lib.vcs.utils import (

    safe_unicode, safe_str, safe_int, date_fromtimestamp

)

from rhodecode.lib.vcs.utils.lazy import LazyProperty

class GitChangeset(BaseChangeset):

    """

    Represents state of the repository at single revision.

    """

    def __init__(self, repository, revision):

        self._stat_modes = {}

        self.repository = repository

        try:

            commit = self.repository._repo[revision]

            if isinstance(commit, objects.Tag):

                revision = commit.object[1]

                commit = self.repository._repo.get_object(commit.object[1])

        except KeyError:

            raise RepositoryError("Cannot get object with id %s" % revision)

        self.raw_id = revision

        self.id = self.raw_id

        self.short_id = self.raw_id[:12]

        self._commit = commit

        self._tree_id = commit.tree

        self._committer_property = 'committer'

        self._author_property = 'author'

        self._date_property = 'commit_time'

        self._date_tz_property = 'commit_timezone'

        self.revision = repository.revisions.index(revision)

        self.nodes = {}

        self._paths = {}

    @LazyProperty

    def message(self):

        return safe_unicode(self._commit.message)

    @LazyProperty

    def committer(self):

        return safe_unicode(getattr(self._commit, self._committer_property))

    @LazyProperty

    def author(self):

        return safe_unicode(getattr(self._commit, self._author_property))

    @LazyProperty

    def date(self):

        return date_fromtimestamp(getattr(self._commit, self._date_property),

                                  getattr(self._commit, self._date_tz_property))

    @LazyProperty

    def _timestamp(self):

        return getattr(self._commit, self._date_property)

    @LazyProperty

    def status(self):

        """

        Returns modified, added, removed, deleted files for current changeset

        """

        return self.changed, self.added, self.removed

    @LazyProperty

    def tags(self):

        _tags = []

        for tname, tsha in self.repository.tags.iteritems():

            if tsha == self.raw_id:

                _tags.append(tname)

        return _tags

    @LazyProperty

    def branch(self):

        heads = self.repository._heads(reverse=False)

        ref = heads.get(self.raw_id)

        if ref:

            return safe_unicode(ref)

    def _fix_path(self, path):

        """

        Paths are stored without trailing slash so we need to get rid off it if

        needed.

        """

        if path.endswith('/'):

            path = path.rstrip('/')

        return path

    def _get_id_for_path(self, path):

        # FIXME: Please, spare a couple of minutes and make those codes cleaner;

        if not path in self._paths:

            path = path.strip('/')

            # set root tree

            tree = self.repository._repo[self._tree_id]

            if path == '':

                self._paths[''] = tree.id

                return tree.id

            splitted = path.split('/')

            dirs, name = splitted[:-1], splitted[-1]

            curdir = ''

            # initially extract things from root dir

            for item, stat, id in tree.iteritems():

                if curdir:

                    name = '/'.join((curdir, item))

                else:

                    name = item

                self._paths[name] = id

                self._stat_modes[name] = stat

            for dir in dirs:

                if curdir:

                    curdir = '/'.join((curdir, dir))

                else:

                    curdir = dir

                dir_id = None

                for item, stat, id in tree.iteritems():

                    if dir == item:

                        dir_id = id

                if dir_id:

                    # Update tree

                    tree = self.repository._repo[dir_id]

                    if not isinstance(tree, objects.Tree):

                        raise ChangesetError('%s is not a directory' % curdir)

                else:

                    raise ChangesetError('%s have not been found' % curdir)

                # cache all items from the given traversed tree

                for item, stat, id in tree.iteritems():

                    if curdir:

                        name = '/'.join((curdir, item))

                    else:

                        name = item

                    self._paths[name] = id

                    self._stat_modes[name] = stat

            if not path in self._paths:

                raise NodeDoesNotExistError("There is no file nor directory "

                    "at the given path '%s' at revision %s"

        return self._paths[path]

    def _get_kind(self, path):

        obj = self.repository._repo[self._get_id_for_path(path)]

        if isinstance(obj, objects.Blob):

            return NodeKind.FILE

        elif isinstance(obj, objects.Tree):

            return NodeKind.DIR

    def _get_filectx(self, path):

        path = self._fix_path(path)

        if self._get_kind(path) != NodeKind.FILE:

            raise ChangesetError("File does not exist for revision %s at "

                " '%s'" % (self.raw_id, path))

        return path

    def _get_file_nodes(self):

        return chain(*(t[2] for t in self.walk()))

    @LazyProperty

    def parents(self):

        """

        Returns list of parents changesets.

        """

        return [self.repository.get_changeset(parent)

                for parent in self._commit.parents]

    @LazyProperty

    def children(self):

        """

        Returns list of children changesets.

        """

        rev_filter = _git_path = settings.GIT_REV_FILTER

        so, se = self.repository.run_git_command(

            "rev-list %s --children | grep '^%s'" % (rev_filter, self.raw_id)

        )

        children = []

        for l in so.splitlines():

            childs = l.split(' ')[1:]

            children.extend(childs)

        return [self.repository.get_changeset(cs) for cs in children]

    def next(self, branch=None):

        if branch and self.branch != branch:

            raise VCSError('Branch option used on changeset not belonging '

                           'to that branch')

        def _next(changeset, branch):

            try:

                next_ = changeset.revision + 1

                next_rev = changeset.repository.revisions[next_]

            except IndexError:

                raise ChangesetDoesNotExistError

            cs = changeset.repository.get_changeset(next_rev)

            if branch and branch != cs.branch:

                return _next(cs, branch)

            return cs

        return _next(self, branch)

    def prev(self, branch=None):

        if branch and self.branch != branch:

            raise VCSError('Branch option used on changeset not belonging '

                           'to that branch')

        def _prev(changeset, branch):

            try:

                prev_ = changeset.revision - 1

                if prev_ < 0:

                    raise IndexError

                prev_rev = changeset.repository.revisions[prev_]

            except IndexError:

                raise ChangesetDoesNotExistError

            cs = changeset.repository.get_changeset(prev_rev)

            if branch and branch != cs.branch:

                return _prev(cs, branch)

            return cs

        return _prev(self, branch)

    def diff(self, ignore_whitespace=True, context=3):

        rev1 = self.parents[0] if self.parents else self.repository.EMPTY_CHANGESET

        rev2 = self

        return ''.join(self.repository.get_diff(rev1, rev2,

                                    ignore_whitespace=ignore_whitespace,

                                    context=context))

    def get_file_mode(self, path):

        """

        Returns stat mode of the file at the given ``path``.

        """

        # ensure path is traversed

        self._get_id_for_path(path)

        return self._stat_modes[path]

    def get_file_content(self, path):

        """

        Returns content of the file at given ``path``.

        """

        id = self._get_id_for_path(path)

        blob = self.repository._repo[id]

        return blob.as_pretty_string()

    def get_file_size(self, path):

        """

        Returns size of the file at given ``path``.

        """

        id = self._get_id_for_path(path)

        blob = self.repository._repo[id]

        return blob.raw_length()

    def get_file_changeset(self, path):

        """

        Returns last commit of the file at the given ``path``.

        """

        return self.get_file_history(path, limit=1)[0]

    def get_file_history(self, path, limit=None):

        """

        Returns history of file as reversed list of ``Changeset`` objects for

        which file at given ``path`` has been modified.

        TODO: This function now uses os underlying 'git' and 'grep' commands

        which is generally not good. Should be replaced with algorithm

        iterating commits.

        """

        self._get_filectx(path)

        cs_id = safe_str(self.id)

        f_path = safe_str(path)

        if limit:

            cmd = 'log -n %s --pretty="format: %%H" -s -p %s -- "%s"' % (

                      safe_int(limit, 0), cs_id, f_path

                   )

        else:

            cmd = 'log --pretty="format: %%H" -s -p %s -- "%s"' % (

                      cs_id, f_path

                   )

        so, se = self.repository.run_git_command(cmd)

        ids = re.findall(r'[0-9a-fA-F]{40}', so)

        return [self.repository.get_changeset(id) for id in ids]

    def get_file_history_2(self, path):

        """

        Returns history of file as reversed list of ``Changeset`` objects for

        which file at given ``path`` has been modified.

        """

        self._get_filectx(path)

        from dulwich.walk import Walker

        include = [self.id]

        walker = Walker(self.repository._repo.object_store, include,

                        paths=[path], max_entries=1)

        return [self.repository.get_changeset(sha)

                for sha in (x.commit.id for x in walker)]

    def get_file_annotate(self, path):

        """

        Returns a generator of four element tuples with

            lineno, sha, changeset lazy loader and line

        TODO: This function now uses os underlying 'git' command which is

        generally not good. Should be replaced with algorithm iterating

        commits.

        """

        cmd = 'blame -l --root -r %s -- "%s"' % (self.id, path)

        # -l     ==> outputs long shas (and we need all 40 characters)

        # --root ==> doesn't put '^' character for bounderies

        # -r sha ==> blames for the given revision

        so, se = self.repository.run_git_command(cmd)

        for i, blame_line in enumerate(so.split('\n')[:-1]):

            ln_no = i + 1

            sha, line = re.split(r' ', blame_line, 1)

            yield (ln_no, sha, lambda: self.repository.get_changeset(sha), line)

    def fill_archive(self, stream=None, kind='tgz', prefix=None,

                     subrepos=False):

        """

        Fills up given stream.

        :param stream: file like object.

        :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``.

            Default: ``tgz``.

        :param prefix: name of root directory in archive.

            Default is repository name and changeset's raw_id joined with dash

            (``repo-tip.<KIND>``).

import time

import datetime

import posixpath

from dulwich import objects

from dulwich.repo import Repo

from rhodecode.lib.vcs.backends.base import BaseInMemoryChangeset

from rhodecode.lib.vcs.exceptions import RepositoryError

from rhodecode.lib.vcs.utils import safe_str

class GitInMemoryChangeset(BaseInMemoryChangeset):

    def commit(self, message, author, parents=None, branch=None, date=None,

               **kwargs):

        """

        Performs in-memory commit (doesn't check workdir in any way) and

        returns newly created ``Changeset``. Updates repository's

        ``revisions``.

        :param message: message of the commit

        :param author: full username, i.e. "Joe Doe <joe.doe@example.com>"

        :param parents: single parent or sequence of parents from which commit

          would be derieved

        :param date: ``datetime.datetime`` instance. Defaults to

          ``datetime.datetime.now()``.

        :param branch: branch name, as string. If none given, default backend's

          branch would be used.

        :raises ``CommitError``: if any error occurs while committing

        """

        self.check_integrity(parents)

        from .repository import GitRepository

        if branch is None:

            branch = GitRepository.DEFAULT_BRANCH_NAME

        repo = self.repository._repo

        object_store = repo.object_store

        ENCODING = "UTF-8"

        DIRMOD = 040000

        # Create tree and populates it with blobs

        commit_tree = self.parents[0] and repo[self.parents[0]._commit.tree] or\

            objects.Tree()

        for node in self.added + self.changed:

            # Compute subdirs if needed

            dirpath, nodename = posixpath.split(node.path)

            parent = commit_tree

            ancestors = [('', parent)]

            # Tries to dig for the deepest existing tree

            while dirnames:

                curdir = dirnames.pop(0)

                try:

                    dir_id = parent[curdir][1]

                except KeyError:

                    # put curdir back into dirnames and stops

                    dirnames.insert(0, curdir)

                    break

                else:

                    # If found, updates parent

                    parent = self.repository._repo[dir_id]

                    ancestors.append((curdir, parent))

            # Now parent is deepest existing tree and we need to create subtrees

            # for dirnames (in reverse order) [this only applies for nodes from added]

            new_trees = []

            if not node.is_binary:

                content = node.content.encode(ENCODING)

            else:

                content = node.content

            blob = objects.Blob.from_string(content)

            node_path = node.name.encode(ENCODING)

            if dirnames:

                # If there are trees which should be created we need to build

                # them now (in reverse order)

                reversed_dirnames = list(reversed(dirnames))

                curtree = objects.Tree()

                curtree[node_path] = node.mode, blob.id

                new_trees.append(curtree)

                for dirname in reversed_dirnames[:-1]:

                    newtree = objects.Tree()

                    #newtree.add(DIRMOD, dirname, curtree.id)

                    newtree[dirname] = DIRMOD, curtree.id

                    new_trees.append(newtree)

                    curtree = newtree

                parent[reversed_dirnames[-1]] = DIRMOD, curtree.id

            else:

                parent.add(name=node_path, mode=node.mode, hexsha=blob.id)

            new_trees.append(parent)

            # Update ancestors

            for parent, tree, path in reversed([(a[1], b[1], b[0]) for a, b in

                zip(ancestors, ancestors[1:])]):

                parent[path] = DIRMOD, tree.id

                object_store.add_object(tree)

            object_store.add_object(blob)

            for tree in new_trees:

                object_store.add_object(tree)

        for node in self.removed:

            paths = node.path.split('/')

            tree = commit_tree

            trees = [tree]

            # Traverse deep into the forest...

            for path in paths:

                try:

                    obj = self.repository._repo[tree[path][1]]

                    if isinstance(obj, objects.Tree):

                        trees.append(obj)

                        tree = obj

                except KeyError:

                    break

            # Cut down the blob and all rotten trees on the way back...

            for path, tree in reversed(zip(paths, trees)):

                del tree[path]

                if tree:

                    # This tree still has elements - don't remove it or any

                    # of it's parents

                    break

        object_store.add_object(commit_tree)

        # Create commit

        commit = objects.Commit()

        commit.tree = commit_tree.id

        commit.parents = [p._commit.id for p in self.parents if p]

        commit.author = commit.committer = safe_str(author)

        commit.encoding = ENCODING

        commit.message = safe_str(message)

        # Compute date

        if date is None:

            date = time.time()

        elif isinstance(date, datetime.datetime):

            date = time.mktime(date.timetuple())

        author_time = kwargs.pop('author_time', date)

        commit.commit_time = int(date)

        commit.author_time = int(author_time)

        tz = time.timezone

        author_tz = kwargs.pop('author_timezone', tz)

# -*- coding: utf-8 -*-

"""

    vcs.nodes

    ~~~~~~~~~

    Module holding everything related to vcs nodes.

    :created_on: Apr 8, 2010

    :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.

"""

import os

import stat

import posixpath

import mimetypes

from rhodecode.lib.vcs.backends.base import EmptyChangeset

from rhodecode.lib.vcs.exceptions import NodeError, RemovedFileNodeError

from rhodecode.lib.vcs.utils.lazy import LazyProperty

class NodeKind:

    SUBMODULE = -1

    DIR = 1

    FILE = 2

class NodeState:

    ADDED = u'added'

    CHANGED = u'changed'

    NOT_CHANGED = u'not changed'

    REMOVED = u'removed'

class NodeGeneratorBase(object):

    """

    Base class for removed added and changed filenodes, it's a lazy generator

    class that will create filenodes only on iteration or call

    The len method doesn't need to create filenodes at all

    """

    def __init__(self, current_paths, cs):

        self.cs = cs

        self.current_paths = current_paths

    def __call__(self):

        return [n for n in self]

    def __getslice__(self, i, j):

        for p in self.current_paths[i:j]:

            yield self.cs.get_node(p)

    def __len__(self):

        return len(self.current_paths)

    def __iter__(self):

        for p in self.current_paths:

            yield self.cs.get_node(p)

class AddedFileNodesGenerator(NodeGeneratorBase):

    """

    Class holding Added files for current changeset

    """

    pass

class ChangedFileNodesGenerator(NodeGeneratorBase):

    """

    Class holding Changed files for current changeset

    """

    pass

class RemovedFileNodesGenerator(NodeGeneratorBase):

    """

    Class holding removed files for current changeset

    """

    def __iter__(self):

        for p in self.current_paths:

            yield RemovedFileNode(path=p)

    def __getslice__(self, i, j):

        for p in self.current_paths[i:j]:

            yield RemovedFileNode(path=p)

class Node(object):

    """

    Simplest class representing file or directory on repository.  SCM backends

    should use ``FileNode`` and ``DirNode`` subclasses rather than ``Node``

    directly.

    Node's ``path`` cannot start with slash as we operate on *relative* paths

    only. Moreover, every single node is identified by the ``path`` attribute,

    so it cannot end with slash, too. Otherwise, path could lead to mistakes.

    """

    def __init__(self, path, kind):

        if path.startswith('/'):

            raise NodeError("Cannot initialize Node objects with slash at "

                "the beginning as only relative paths are supported")

        if path == '' and kind != NodeKind.DIR:

            raise NodeError("Only DirNode and its subclasses may be "

                            "initialized with empty path")

        self.kind = kind

        #self.dirs, self.files = [], []

        if self.is_root() and not self.is_dir():

            raise NodeError("Root node cannot be FILE kind")

    @LazyProperty

    def parent(self):

        parent_path = self.get_parent_path()

        if parent_path:

            if self.changeset:

                return self.changeset.get_node(parent_path)

            return DirNode(parent_path)

        return None

    @LazyProperty

    def unicode_path(self):

        return safe_unicode(self.path)

    @LazyProperty

    def name(self):

        """

        Returns name of the node so if its path

        then only last part is returned.

        """

        return safe_unicode(self.path.rstrip('/').split('/')[-1])

    def _get_kind(self):

        return self._kind

    def _set_kind(self, kind):

        if hasattr(self, '_kind'):

            raise NodeError("Cannot change node's kind")

        else:

            self._kind = kind

            # Post setter check (path's trailing slash)

            if self.path.endswith('/'):

                raise NodeError("Node's path cannot end with slash")

    kind = property(_get_kind, _set_kind)

    def __cmp__(self, other):

        """

        Comparator using name of the node, needed for quick list sorting.

        """

        kind_cmp = cmp(self.kind, other.kind)

        if kind_cmp:

            return kind_cmp

        return cmp(self.name, other.name)

    def __eq__(self, other):

        for attr in ['name', 'path', 'kind']:

            if getattr(self, attr) != getattr(other, attr):

                return False

        if self.is_file():

            if self.content != other.content:

                return False

        else:

            # For DirNode's check without entering each dir

            self_nodes_paths = list(sorted(n.path for n in self.nodes))

            other_nodes_paths = list(sorted(n.path for n in self.nodes))

            if self_nodes_paths != other_nodes_paths:

                return False

        return True

    def __nq__(self, other):

        return not self.__eq__(other)

    def __repr__(self):

        return '<%s %r>' % (self.__class__.__name__, self.path)

    def __str__(self):

        return self.__repr__()

    def __unicode__(self):

        return self.name

    def get_parent_path(self):

        """

        Returns node's parent path or empty string if node is root.

        """

        if self.is_root():

            return ''

        return posixpath.dirname(self.path.rstrip('/')) + '/'

    def is_file(self):

        """

        Returns ``True`` if node's kind is ``NodeKind.FILE``, ``False``

        otherwise.

        """

        return self.kind == NodeKind.FILE

    def is_dir(self):

        """

        """

        if scm_type == 'hg':

            from rhodecode.lib.vcs.backends.hg import \

                MercurialInMemoryChangeset as IMC

        elif scm_type == 'git':

            from rhodecode.lib.vcs.backends.git import \

                GitInMemoryChangeset as IMC

        return IMC

    def pull_changes(self, repo, username):

        dbrepo = self.__get_repo(repo)

        clone_uri = dbrepo.clone_uri

        if not clone_uri:

            raise Exception("This repository doesn't have a clone uri")

        repo = dbrepo.scm_instance

        repo_name = dbrepo.repo_name

        try:

            if repo.alias == 'git':

                repo.fetch(clone_uri)

                # git doesn't really have something like post-fetch action

                # we fake that now. #TODO: extract fetched revisions somehow

                # here

                self._handle_push(repo,

                                  username=username,

                                  action='push_remote',

                                  repo_name=repo_name,

                                  revisions=[])

            else:

                self._handle_rc_scm_extras(username, dbrepo.repo_name,

                                           repo.alias, action='push_remote')

                repo.pull(clone_uri)

            self.mark_for_invalidation(repo_name)

        except Exception:

            log.error(traceback.format_exc())

            raise

    def commit_change(self, repo, repo_name, cs, user, author, message,

                      content, f_path):

        """

        Commits changes

        :param repo: SCM instance

        """

        user = self._get_user(user)

        IMC = self._get_IMC_module(repo.alias)

        # decoding here will force that we have proper encoded values

        # in any other case this will throw exceptions and deny commit

        content = safe_str(content)

        path = safe_str(f_path)

        # message and author needs to be unicode

        # proper backend should then translate that into required type

        message = safe_unicode(message)

        author = safe_unicode(author)

        imc = IMC(repo)

        imc.change(FileNode(path, content, mode=cs.get_file_mode(f_path)))

        tip = imc.commit(message=message,

                       author=author,

                       parents=[cs], branch=cs.branch)

        self.mark_for_invalidation(repo_name)

        self._handle_push(repo,

                          username=user.username,

                          action='push_local',

                          repo_name=repo_name,

                          revisions=[tip.raw_id])

        return tip

    def create_nodes(self, user, repo, message, nodes, parent_cs=None,

                     author=None, trigger_push_hook=True):

        """

        Commits given multiple nodes into repo

        :param user: RhodeCode User object or user_id, the commiter

        :param repo: RhodeCode Repository object

        :param message: commit message

        :param nodes: mapping {filename:{'content':content},...}

        :param parent_cs: parent changeset, can be empty than it's initial commit

        :param author: author of commit, cna be different that commiter only for git

        :param trigger_push_hook: trigger push hooks

        :returns: new commited changeset

        """

        user = self._get_user(user)

        scm_instance = repo.scm_instance_no_cache()

        processed_nodes = []

        for f_path in nodes:

            if f_path.startswith('/') or f_path.startswith('.') or '../' in f_path:

                raise NonRelativePathError('%s is not an relative path' % f_path)

            if f_path:

                f_path = os.path.normpath(f_path)

            f_path = safe_str(f_path)

            # decoding here will force that we have proper encoded values

            # in any other case this will throw exceptions and deny commit

            if isinstance(content, (basestring,)):

                content = safe_str(content)

            elif isinstance(content, (file, cStringIO.OutputType,)):

                content = content.read()

            else:

                raise Exception('Content is of unrecognized type %s' % (

                    type(content)

                ))

            processed_nodes.append((f_path, content))

        message = safe_unicode(message)

        commiter = user.full_contact

        author = safe_unicode(author) if author else commiter

        IMC = self._get_IMC_module(scm_instance.alias)

        imc = IMC(scm_instance)

        if not parent_cs:

            parent_cs = EmptyChangeset(alias=scm_instance.alias)

        if isinstance(parent_cs, EmptyChangeset):

            # EmptyChangeset means we we're editing empty repository

            parents = None

        else:

            parents = [parent_cs]

        # add multiple nodes

        for path, content in processed_nodes:

            imc.add(FileNode(path, content=content))

        tip = imc.commit(message=message,

                         author=author,

                         parents=parents,

                         branch=parent_cs.branch)

        self.mark_for_invalidation(repo.repo_name)

        if trigger_push_hook:

            self._handle_push(scm_instance,

                              username=user.username,

                              action='push_local',

                              repo_name=repo.repo_name,

                              revisions=[tip.raw_id])

        return tip

    def get_nodes(self, repo_name, revision, root_path='/', flat=True):

        """

        recursive walk in root dir and return a set of all path in that dir

        based on repository walk function

        :param repo_name: name of repository

        :param revision: revision for which to list nodes

        :param root_path: root path to list

        :param flat: return as a list, if False returns a dict with decription

        """

        _files = list()

        _dirs = list()

        try:

            _repo = self.__get_repo(repo_name)

            changeset = _repo.scm_instance.get_changeset(revision)

            root_path = root_path.lstrip('/')

            for topnode, dirs, files in changeset.walk(root_path):

                for f in files:

                    _files.append(f.path if flat else {"name": f.path,

                                                       "type": "file"})

                for d in dirs:

                    _dirs.append(d.path if flat else {"name": d.path,

                                                      "type": "dir"})

        except RepositoryError: