Files controller for RhodeCode

    :created_on: Apr 21, 2010

    :author: marcink

    :copyright: (C) 2009-2011 Marcin Kuzminski <marcin@python-works.com>

    :license: GPLv3, see COPYING for more details.

"""

# This program is free software: you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation, either version 3 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program.  If not, see <http://www.gnu.org/licenses/>.

import os

import logging

import rhodecode.lib.helpers as h

from pylons import request, response, session, tmpl_context as c, url

from pylons.i18n.translation import _

from pylons.controllers.util import redirect

from rhodecode.lib.auth import LoginRequired, HasRepoPermissionAnyDecorator

from rhodecode.lib.base import BaseRepoController, render

from rhodecode.lib.utils import EmptyChangeset

from rhodecode.model.repo import RepoModel

from vcs.backends import ARCHIVE_SPECS

from vcs.exceptions import RepositoryError, ChangesetDoesNotExistError, \

    EmptyRepositoryError, ImproperArchiveTypeError, VCSError

from vcs.nodes import FileNode, NodeKind

from vcs.utils import diffs as differ

log = logging.getLogger(__name__)

class FilesController(BaseRepoController):

    @LoginRequired()

    @HasRepoPermissionAnyDecorator('repository.read', 'repository.write',

                                   'repository.admin')

    def __before__(self):

        super(FilesController, self).__before__()

        c.cut_off_limit = self.cut_off_limit

    def __get_cs_or_redirect(self, rev, repo_name):

        """

        Safe way to get changeset if error occur it redirects to tip with

        proper message

        :param rev: revision to fetch

        :param repo_name: repo name to redirect after

        """

        try:

            return c.rhodecode_repo.get_changeset(rev)

        except EmptyRepositoryError, e:

            h.flash(_('There are no files yet'), category='warning')

            redirect(h.url('summary_home', repo_name=repo_name))

        except RepositoryError, e:

            h.flash(str(e), category='warning')

            redirect(h.url('files_home', repo_name=repo_name, revision='tip'))

    def __get_filenode_or_redirect(self, repo_name, cs, path):

        """

        Returns file_node, if error occurs or given path is directory,

        it'll redirect to top level path

        :param repo_name: repo_name

        :param cs: given changeset

        :param path: path to lookup

        """

        try:

            file_node = cs.get_node(path)

            if file_node.is_dir():

                raise RepositoryError('given path is a directory')

        except RepositoryError, e:

            h.flash(str(e), category='warning')

            redirect(h.url('files_home', repo_name=repo_name,

                           revision=cs.raw_id))

        return file_node

    def index(self, repo_name, revision, f_path):

        #reditect to given revision from form if given

        post_revision = request.POST.get('at_rev', None)

        if post_revision:

            redirect(url('files_home', repo_name=c.repo_name,

                         revision=cs.raw_id, f_path=f_path))

        c.changeset = self.__get_cs_or_redirect(revision, repo_name)

        c.branch = request.GET.get('branch', None)

        c.f_path = f_path

        cur_rev = c.changeset.revision

        #prev link

        try:

            prev_rev = c.rhodecode_repo.get_changeset(cur_rev).prev(c.branch)

            c.url_prev = url('files_home', repo_name=c.repo_name,

                         revision=prev_rev.raw_id, f_path=f_path)

            if c.branch:

                c.url_prev += '?branch=%s' % c.branch

        except (ChangesetDoesNotExistError, VCSError):

            c.url_prev = '#'

        #next link

        try:

            next_rev = c.rhodecode_repo.get_changeset(cur_rev).next(c.branch)

            c.url_next = url('files_home', repo_name=c.repo_name,

                     revision=next_rev.raw_id, f_path=f_path)

            if c.branch:

                c.url_next += '?branch=%s' % c.branch

        except (ChangesetDoesNotExistError, VCSError):

            c.url_next = '#'

        #files or dirs

        try:

            c.files_list = c.changeset.get_node(f_path)

            if c.files_list.is_file():

                c.file_history = self._get_node_history(c.changeset, f_path)

            else:

                c.file_history = []

        except RepositoryError, e:

            h.flash(str(e), category='warning')

            redirect(h.url('files_home', repo_name=repo_name,

                           revision=revision))

        return render('files/files.html')

    def rawfile(self, repo_name, revision, f_path):

        cs = self.__get_cs_or_redirect(revision, repo_name)

        file_node = self.__get_filenode_or_redirect(repo_name, cs, f_path)

        response.content_disposition = 'attachment; filename=%s' % \

            f_path.split(os.sep)[-1].encode('utf8', 'replace')

        response.content_type = file_node.mimetype

        return file_node.content

    def raw(self, repo_name, revision, f_path):

        cs = self.__get_cs_or_redirect(revision, repo_name)

        file_node = self.__get_filenode_or_redirect(repo_name, cs, f_path)

        response.content_type = 'text/plain'

        return file_node.content

    def annotate(self, repo_name, revision, f_path):

        c.cs = self.__get_cs_or_redirect(revision, repo_name)

        c.file = self.__get_filenode_or_redirect(repo_name, c.cs, f_path)

        c.file_history = self._get_node_history(c.cs, f_path)

        c.f_path = f_path

        return render('files/files_annotate.html')

    def archivefile(self, repo_name, fname):

        fileformat = None

        revision = None

        ext = None

        for a_type, ext_data in ARCHIVE_SPECS.items():

            archive_spec = fname.split(ext_data[1])

            if len(archive_spec) == 2 and archive_spec[1] == '':

                fileformat = a_type or ext_data[1]

                revision = archive_spec[0]

                ext = ext_data[1]

        try:

            dbrepo = RepoModel().get_by_repo_name(repo_name)

            if dbrepo.enable_downloads is False:

                return _('downloads disabled')

            cs = c.rhodecode_repo.get_changeset(revision)

            content_type = ARCHIVE_SPECS[fileformat][0]

        except ChangesetDoesNotExistError:

            return _('Unknown revision %s') % revision

        except EmptyRepositoryError:

            return _('Empty repository')

        except (ImproperArchiveTypeError, KeyError):

            return _('Unknown archive type')

        response.content_type = content_type

        response.content_disposition = 'attachment; filename=%s-%s%s' \

            % (repo_name, revision, ext)

        return cs.get_chunked_archive(stream=None, kind=fileformat)

    def diff(self, repo_name, f_path):

        diff1 = request.GET.get('diff1')

        diff2 = request.GET.get('diff2')

        c.action = request.GET.get('diff')

        c.no_changes = diff1 == diff2

        c.f_path = f_path

        try:

            if diff1 not in ['', None, 'None', '0' * 12, '0' * 40]:

                c.changeset_1 = c.rhodecode_repo.get_changeset(diff1)

                node1 = c.changeset_1.get_node(f_path)

            else:

                node1 = FileNode('.', '', changeset=c.changeset_1)

            if diff2 not in ['', None, 'None', '0' * 12, '0' * 40]:

                c.changeset_2 = c.rhodecode_repo.get_changeset(diff2)

                node2 = c.changeset_2.get_node(f_path)

            else:

                node2 = FileNode('.', '', changeset=c.changeset_2)

        except RepositoryError:

            return redirect(url('files_home',

                                repo_name=c.repo_name, f_path=f_path))

        if c.action == 'download':

            diff = differ.DiffProcessor(differ.get_gitdiff(node1, node2),

                                        format='gitdiff')

            diff_name = '%s_vs_%s.diff' % (diff1, diff2)

            response.content_type = 'text/plain'

            response.content_disposition = 'attachment; filename=%s' \

                                                    % diff_name

            return diff.raw_diff()

        elif c.action == 'raw':

            diff = differ.DiffProcessor(differ.get_gitdiff(node1, node2),

                                        format='gitdiff')

            response.content_type = 'text/plain'

            return diff.raw_diff()

        elif c.action == 'diff':

            if node1.is_binary or node2.is_binary:

                c.cur_diff = _('Binary file')

            elif node1.size > self.cut_off_limit or \

                    node2.size > self.cut_off_limit:

                c.cur_diff = _('Diff is too big to display')

            else:

                diff = differ.DiffProcessor(differ.get_gitdiff(node1, node2),

                                        format='gitdiff')

                c.cur_diff = diff.as_html()

        else:

            #default option

            if node1.is_binary or node2.is_binary:

                c.cur_diff = _('Binary file')

            elif node1.size > self.cut_off_limit or \

                    node2.size > self.cut_off_limit:

                c.cur_diff = _('Diff is too big to display')

            else:

                diff = differ.DiffProcessor(differ.get_gitdiff(node1, node2),

                                        format='gitdiff')

                c.cur_diff = diff.as_html()

        if not c.cur_diff:

            c.no_changes = True

        return render('files/file_diff.html')

    def _get_node_history(self, cs, f_path):

        changesets = cs.get_file_history(f_path)

        hist_l = []

        changesets_group = ([], _("Changesets"))

        branches_group = ([], _("Branches"))

        tags_group = ([], _("Tags"))

        for chs in changesets:

            n_desc = 'r%s:%s' % (chs.revision, chs.short_id)

            changesets_group[0].append((chs.raw_id, n_desc,))

        hist_l.append(changesets_group)

        for name, chs in c.rhodecode_repo.branches.items():

            #chs = chs.split(':')[-1]

            branches_group[0].append((chs, name),)

        hist_l.append(branches_group)

        for name, chs in c.rhodecode_repo.tags.items():

            #chs = chs.split(':')[-1]

            tags_group[0].append((chs, name),)

        hist_l.append(tags_group)

        return hist_l

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

"""

    rhodecode.lib.utils

    ~~~~~~~~~~~~~~~~~~~

    Utilities library for RhodeCode

    :created_on: Apr 18, 2010

    :author: marcink

    :copyright: (C) 2009-2011 Marcin Kuzminski <marcin@python-works.com>

    :license: GPLv3, see COPYING for more details.

"""

# This program is free software: you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation, either version 3 of the License, or

# (at your option) any later version.

#

# This program is distributed in the hope that it will be useful,

# but WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

# GNU General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program.  If not, see <http://www.gnu.org/licenses/>.

import os

import logging

import datetime

import traceback

import paste

import beaker

from paste.script.command import Command, BadCommand

from UserDict import DictMixin

from mercurial import ui, config, hg

from mercurial.error import RepoError

from webhelpers.text import collapse, remove_formatting, strip_tags

from vcs.backends.base import BaseChangeset

from vcs.utils.lazy import LazyProperty

from rhodecode.model import meta

from rhodecode.model.caching_query import FromCache

from rhodecode.model.db import Repository, User, RhodeCodeUi, UserLog, Group

from rhodecode.model.repo import RepoModel

from rhodecode.model.user import UserModel

log = logging.getLogger(__name__)

def recursive_replace(str, replace=' '):

    """Recursive replace of given sign to just one instance

    :param str: given string

    :param replace: char to find and replace multiple instances

    Examples::

    >>> recursive_replace("Mighty---Mighty-Bo--sstones",'-')

    'Mighty-Mighty-Bo-sstones'

    """

    if str.find(replace * 2) == -1:

        return str

    else:

        str = str.replace(replace * 2, replace)

        return recursive_replace(str, replace)

def repo_name_slug(value):

    """Return slug of name of repository

    This function is called on each creation/modification

    of repository to prevent bad names in repo

    """

    slug = remove_formatting(value)

    slug = strip_tags(slug)

    for c in """=[]\;'"<>,/~!@#$%^&*()+{}|: """:

        slug = slug.replace(c, '-')

    slug = recursive_replace(slug, '-')

    slug = collapse(slug, '-')

    return slug

def get_repo_slug(request):

    return request.environ['pylons.routes_dict'].get('repo_name')

def action_logger(user, action, repo, ipaddr='', sa=None):

    """

    Action logger for various actions made by users

    :param user: user that made this action, can be a unique username string or

        object containing user_id attribute

    :param action: action to log, should be on of predefined unique actions for

        easy translations

    :param repo: string name of repository or object containing repo_id,

        that action was made on

    :param ipaddr: optional ip address from what the action was made

    :param sa: optional sqlalchemy session

    """

    if not sa:

        sa = meta.Session()

    try:

        um = UserModel()

        if hasattr(user, 'user_id'):

            user_obj = user

        elif isinstance(user, basestring):

            user_obj = um.get_by_username(user, cache=False)

        else:

            raise Exception('You have to provide user object or username')

        rm = RepoModel()

        if hasattr(repo, 'repo_id'):

            repo_obj = rm.get(repo.repo_id, cache=False)

            repo_name = repo_obj.repo_name

        elif  isinstance(repo, basestring):

            repo_name = repo.lstrip('/')

            repo_obj = rm.get_by_repo_name(repo_name, cache=False)

        else:

            raise Exception('You have to provide repository to action logger')

        user_log = UserLog()

        user_log.user_id = user_obj.user_id

        user_log.action = action

        user_log.repository_id = repo_obj.repo_id

        user_log.repository_name = repo_name

        user_log.action_date = datetime.datetime.now()

        user_log.user_ip = ipaddr

        sa.add(user_log)

        sa.commit()

        log.info('Adding user %s, action %s on %s', user_obj, action, repo)

    except:

        log.error(traceback.format_exc())

        sa.rollback()

def get_repos(path, recursive=False):

    """

    Scans given path for repos and return (name,(type,path)) tuple

    :param path: path to scann for repositories

    :param recursive: recursive search and return names with subdirs in front

    """

    from vcs.utils.helpers import get_scm

    from vcs.exceptions import VCSError

    if path.endswith(os.sep):

        #remove ending slash for better results

        path = path[:-1]

    def _get_repos(p):

        if not os.access(p, os.W_OK):

            return

        for dirpath in os.listdir(p):

            print dirpath

            if os.path.isfile(os.path.join(p, dirpath)):

                continue

            cur_path = os.path.join(p, dirpath)

            try:

                scm_info = get_scm(cur_path)

                yield scm_info[1].split(path)[-1].lstrip(os.sep), scm_info

            except VCSError:

                if not recursive:

                    continue

                #check if this dir containts other repos for recursive scan

                rec_path = os.path.join(p, dirpath)

                if os.path.isdir(rec_path):

                    for inner_scm in _get_repos(rec_path):

                        yield inner_scm

    return _get_repos(path)

def check_repo_fast(repo_name, base_path):

    """

    Check given path for existence of directory

    :param repo_name:

    :param base_path:

    :return False: if this directory is present

    """

    return True

def check_repo(repo_name, base_path, verify=True):

    repo_path = os.path.join(base_path, repo_name)

    try:

        if not check_repo_fast(repo_name, base_path):

            return False

        r = hg.repository(ui.ui(), repo_path)

        if verify:

            hg.verify(r)

        #here we hnow that repo exists it was verified

        log.info('%s repo is already created', repo_name)

        return False

    except RepoError:

        #it means that there is no valid repo there...

        log.info('%s repo is free for creation', repo_name)

        return True

def ask_ok(prompt, retries=4, complaint='Yes or no, please!'):

    while True:

        ok = raw_input(prompt)

        retries = retries - 1

        print complaint

#propagated from mercurial documentation

ui_sections = ['alias', 'auth',

                'decode/encode', 'defaults',

                'diff', 'email',

                'extensions', 'format',

                'merge-patterns', 'merge-tools',

                'hooks', 'http_proxy',

                'smtp', 'patch',

                'paths', 'profiling',

                'server', 'trusted',

                'ui', 'web', ]

def make_ui(read_from='file', path=None, checkpaths=True):

    """A function that will read python rc files or database

    and make an mercurial ui object from read options

    :param path: path to mercurial config file

    :param checkpaths: check the path

    :param read_from: read from 'file' or 'db'

    """

    baseui = ui.ui()

    #clean the baseui object

    baseui._ocfg = config.config()

    baseui._ucfg = config.config()

    baseui._tcfg = config.config()

    if read_from == 'file':

        if not os.path.isfile(path):

            log.warning('Unable to read config file %s' % path)

            return False

        log.debug('reading hgrc from %s', path)

        cfg = config.config()

        cfg.read(path)

        for section in ui_sections:

            for k, v in cfg.items(section):

                log.debug('settings ui from file[%s]%s:%s', section, k, v)

                baseui.setconfig(section, k, v)

    elif read_from == 'db':

        sa = meta.Session()

        ret = sa.query(RhodeCodeUi)\

            .options(FromCache("sql_cache_short",

                               "get_hg_ui_settings")).all()

        hg_ui = ret

        for ui_ in hg_ui:

            if ui_.ui_active:

                log.debug('settings ui from db[%s]%s:%s', ui_.ui_section,

                          ui_.ui_key, ui_.ui_value)

                baseui.setconfig(ui_.ui_section, ui_.ui_key, ui_.ui_value)

        meta.Session.remove()

    return baseui

def set_rhodecode_config(config):

    """Updates pylons config with new settings from database

    :param config:

    """

    from rhodecode.model.settings import SettingsModel

    hgsettings = SettingsModel().get_app_settings()

    for k, v in hgsettings.items():

        config[k] = v

def invalidate_cache(cache_key, *args):

    """Puts cache invalidation task into db for

    further global cache invalidation

    """

    from rhodecode.model.scm import ScmModel

    if cache_key.startswith('get_repo_cached_'):

        name = cache_key.split('get_repo_cached_')[-1]

        ScmModel().mark_for_invalidation(name)

class EmptyChangeset(BaseChangeset):

    """

    An dummy empty changeset. It's possible to pass hash when creating

    an EmptyChangeset

    """

        self._empty_cs = cs

        self.revision = -1

        self.message = ''

        self.author = ''

        self.date = ''

    @LazyProperty

    def raw_id(self):

        """Returns raw string identifying this changeset, useful for web

        representation.

        """

        return self._empty_cs

    @LazyProperty

    def short_id(self):

        return self.raw_id[:12]

    def get_file_changeset(self, path):

        return self

    def get_file_content(self, path):

        return u''

    def get_file_size(self, path):

        return 0

def map_groups(groups):

    """Checks for groups existence, and creates groups structures.

    It returns last group in structure

    :param groups: list of groups structure

    """

    sa = meta.Session()

    parent = None

    group = None

    for lvl, group_name in enumerate(groups[:-1]):

        group = sa.query(Group).filter(Group.group_name == group_name).scalar()

        if group is None:

            group = Group(group_name, parent)

            sa.add(group)

            sa.commit()

        parent = group

    return group

def repo2db_mapper(initial_repo_list, remove_obsolete=False):

    """maps all repos given in initial_repo_list, non existing repositories

    are created, if remove_obsolete is True it also check for db entries

    that are not in initial_repo_list and removes them.

    :param initial_repo_list: list of repositories found by scanning methods

    :param remove_obsolete: check for obsolete entries in database

    """

    sa = meta.Session()

    rm = RepoModel()

    user = sa.query(User).filter(User.admin == True).first()

    added = []

    for name, repo in initial_repo_list.items():

        group = map_groups(name.split('/'))

        if not rm.get_by_repo_name(name, cache=False):

            log.info('repository %s not found creating default', name)

            added.append(name)

            form_data = {

                         'repo_name':name,

                         'repo_type':repo.alias,

                         'description':repo.description \

                            if repo.description != 'unknown' else \

                                        '%s repository' % name,

                         'private':False,

                         'group_id':getattr(group, 'group_id', None)

                         }

            rm.create(form_data, user, just_db=True)

    removed = []

    if remove_obsolete:

        #remove from database those repositories that are not in the filesystem

        for repo in sa.query(Repository).all():

            if repo.repo_name not in initial_repo_list.keys():

                removed.append(repo.repo_name)

                sa.delete(repo)

                sa.commit()

    return added, removed

class OrderedDict(dict, DictMixin):

    def __init__(self, *args, **kwds):

        if len(args) > 1:

            raise TypeError('expected at most 1 arguments, got %d' % len(args))

        try:

            self.__end

        except AttributeError:

            self.clear()

        self.update(*args, **kwds)

    def clear(self):

        self.__end = end = []

        end += [None, end, end]         # sentinel node for doubly linked list

        self.__map = {}                 # key --> [key, prev, next]

        dict.clear(self)

    def __setitem__(self, key, value):

        if key not in self:

            end = self.__end

            curr = end[1]

            curr[2] = end[1] = self.__map[key] = [key, curr, end]

        dict.__setitem__(self, key, value)

    def __delitem__(self, key):

        dict.__delitem__(self, key)

        key, prev, next = self.__map.pop(key)

        prev[2] = next

        next[1] = prev

    def __iter__(self):

        end = self.__end

        curr = end[2]

        while curr is not end:

            yield curr[0]

            curr = curr[2]

    def __reversed__(self):

        end = self.__end

        curr = end[1]

        while curr is not end:

            yield curr[0]

            curr = curr[1]

    def popitem(self, last=True):

        if not self:

            raise KeyError('dictionary is empty')

        if last:

            key = reversed(self).next()

        else:

            key = iter(self).next()

        value = self.pop(key)

        return key, value

    def __reduce__(self):

        items = [[k, self[k]] for k in self]

        tmp = self.__map, self.__end

        del self.__map, self.__end

        inst_dict = vars(self).copy()

        self.__map, self.__end = tmp

        if inst_dict:

            return (self.__class__, (items,), inst_dict)

        return self.__class__, (items,)

    def keys(self):

        return list(self)

    setdefault = DictMixin.setdefault

    update = DictMixin.update

    pop = DictMixin.pop

    values = DictMixin.values

    items = DictMixin.items

    iterkeys = DictMixin.iterkeys

    itervalues = DictMixin.itervalues

    iteritems = DictMixin.iteritems

    def __repr__(self):

        if not self:

            return '%s()' % (self.__class__.__name__,)

        return '%s(%r)' % (self.__class__.__name__, self.items())

    def copy(self):

        return self.__class__(self)

    @classmethod

    def fromkeys(cls, iterable, value=None):

        d = cls()

        for key in iterable:

            d[key] = value

        return d

    def __eq__(self, other):

        if isinstance(other, OrderedDict):

            return len(self) == len(other) and self.items() == other.items()

        return dict.__eq__(self, other)

    def __ne__(self, other):

        return not self == other

#set cache regions for beaker so celery can utilise it

def add_cache(settings):

    cache_settings = {'regions':None}

    for key in settings.keys():

        for prefix in ['beaker.cache.', 'cache.']:

            if key.startswith(prefix):

                name = key.split(prefix)[1].strip()

                cache_settings[name] = settings[key].strip()

    if cache_settings['regions']:

        for region in cache_settings['regions'].split(','):

            region = region.strip()

            region_settings = {}

            for key, value in cache_settings.items():

                if key.startswith(region):

                    region_settings[key.split('.')[1]] = value

            region_settings['expire'] = int(region_settings.get('expire',

                                                                60))

            region_settings.setdefault('lock_dir',

                                       cache_settings.get('lock_dir'))

            region_settings.setdefault('data_dir',

                                       cache_settings.get('data_dir'))

            if 'type' not in region_settings:

                region_settings['type'] = cache_settings.get('type',

                                                             'memory')

            beaker.cache.cache_regions[region] = region_settings

def get_current_revision():

    """Returns tuple of (number, id) from repository containing this package

    or None if repository could not be found.

    """

    try:

        from vcs import get_repo

        from vcs.utils.helpers import get_scm

        from vcs.exceptions import RepositoryError, VCSError

        repopath = os.path.join(os.path.dirname(__file__), '..', '..')

        scm = get_scm(repopath)[0]

        repo = get_repo(path=repopath, alias=scm)

        tip = repo.get_changeset()

        return (tip.revision, tip.short_id)

    except (ImportError, RepositoryError, VCSError), err:

        logging.debug("Cannot retrieve rhodecode's revision. Original error "

                      "was: %s" % err)

        return None

# TEST FUNCTIONS AND CREATORS

def create_test_index(repo_location, full_index):

    """Makes default test index

    :param repo_location:

    :param full_index:

    """

    from rhodecode.lib.indexers.daemon import WhooshIndexingDaemon