use consisten double quote docstring formatting

rhodecode/controllers/error.py

@@ -1,108 +1,108 b''
 # -*- coding: utf-8 -*-
 """
     rhodecode.controllers.error
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
     RhodeCode error controller
 
     :created_on: Dec 8, 2010
     :author: marcink
     :copyright: (C) 2010-2012 Marcin Kuzminski <[e-mail]>
     :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 cgi
 import logging
 import paste.fileapp
 
 from pylons import tmpl_context as c, request, config, url
 from pylons.i18n.translation import _
 from pylons.middleware import  media_path
 
 from rhodecode.lib.base import BaseController, render
 
 log = logging.getLogger(__name__)
 
 
 class ErrorController(BaseController):
     """Generates error documents as and when they are required.
 
     The ErrorDocuments middleware forwards to ErrorController when error
     related status codes are returned from the application.
 
     This behavior can be altered by changing the parameters to the
     ErrorDocuments middleware in your config/middleware.py file.
     """
 
     def __before__(self):
         #disable all base actions since we don't need them here
         pass
 
     def document(self):
         resp = request.environ.get('pylons.original_response')
         c.rhodecode_name = config.get('rhodecode_title')
 
         log.debug('### %s ###' % resp.status)
 
         e = request.environ
         c.serv_p = r'%(protocol)s://%(host)s/' \
                                     % {'protocol': e.get('wsgi.url_scheme'),
                                        'host': e.get('HTTP_HOST'), }
 
         c.error_message = cgi.escape(request.GET.get('code', str(resp.status)))
         c.error_explanation = self.get_error_explanation(resp.status_int)
 
         #  redirect to when error with given seconds
         c.redirect_time = 0
         c.redirect_module = _('Home page')
         c.url_redirect = "/"
 
         return render('/errors/error_document.html')
 
     def img(self, id):
         """Serve Pylons' stock images"""
         return self._serve_file(os.path.join(media_path, 'img', id))
 
     def style(self, id):
         """Serve Pylons' stock stylesheets"""
         return self._serve_file(os.path.join(media_path, 'style', id))
 
     def _serve_file(self, path):
         """Call Paste's FileApp (a WSGI application) to serve the file
         at the specified path
         """
         fapp = paste.fileapp.FileApp(path)
         return fapp(request.environ, self.start_response)
 
     def get_error_explanation(self, code):
-        ''' get the error explanations of int codes
-            [400, 401, 403, 404, 500]'''
+        """ get the error explanations of int codes
+            [400, 401, 403, 404, 500]"""
         try:
             code = int(code)
         except Exception:
             code = 500
 
         if code == 400:
             return _('The request could not be understood by the server'
                      ' due to malformed syntax.')
         if code == 401:
             return _('Unauthorized access to resource')
         if code == 403:
             return _("You don't have permission to view this page")
         if code == 404:
             return _('The resource could not be found')
         if code == 500:
             return _('The server encountered an unexpected condition'
                      ' which prevented it from fulfilling the request.')

rhodecode/lib/dbmigrate/migrate/versioning/genmodel.py

@@ -1,284 +1,284 b''
 """
 Code to generate a Python model from a database or differences
 between a model and database.
 
 Some of this is borrowed heavily from the AutoCode project at:
 http://code.google.com/p/sqlautocode/
 """
 
 import sys
 import logging
 
 import sqlalchemy
 
 from rhodecode.lib.dbmigrate import migrate
 from rhodecode.lib.dbmigrate.migrate import changeset
 
 
 log = logging.getLogger(__name__)
 HEADER = """
 ## File autogenerated by genmodel.py
 
 from sqlalchemy import *
 meta = MetaData()
 """
 
 DECLARATIVE_HEADER = """
 ## File autogenerated by genmodel.py
 
 from sqlalchemy import *
 from sqlalchemy.ext import declarative
 
 Base = declarative.declarative_base()
 """
 
 
 class ModelGenerator(object):
     """Various transformations from an A, B diff.
 
     In the implementation, A tends to be called the model and B
     the database (although this is not true of all diffs).
     The diff is directionless, but transformations apply the diff
     in a particular direction, described in the method name.
     """
 
     def __init__(self, diff, engine, declarative=False):
         self.diff = diff
         self.engine = engine
         self.declarative = declarative
 
     def column_repr(self, col):
         kwarg = []
         if col.key != col.name:
             kwarg.append('key')
         if col.primary_key:
             col.primary_key = True  # otherwise it dumps it as 1
             kwarg.append('primary_key')
         if not col.nullable:
             kwarg.append('nullable')
         if col.onupdate:
             kwarg.append('onupdate')
         if col.default:
             if col.primary_key:
                 # I found that PostgreSQL automatically creates a
                 # default value for the sequence, but let's not show
                 # that.
                 pass
             else:
                 kwarg.append('default')
         args = ['%s=%r' % (k, getattr(col, k)) for k in kwarg]
 
         # crs: not sure if this is good idea, but it gets rid of extra
         # u''
         name = col.name.encode('utf8')
 
         type_ = col.type
         for cls in col.type.__class__.__mro__:
             if cls.__module__ == 'sqlalchemy.types' and \
                 not cls.__name__.isupper():
                 if cls is not type_.__class__:
                     type_ = cls()
                 break
 
         type_repr = repr(type_)
         if type_repr.endswith('()'):
             type_repr = type_repr[:-2]
 
         constraints = [repr(cn) for cn in col.constraints]
 
         data = {
             'name': name,
             'commonStuff': ', '.join([type_repr] + constraints + args),
         }
 
         if self.declarative:
             return """%(name)s = Column(%(commonStuff)s)""" % data
         else:
             return """Column(%(name)r, %(commonStuff)s)""" % data
 
     def _getTableDefn(self, table, metaName='meta'):
         out = []
         tableName = table.name
         if self.declarative:
             out.append("class %(table)s(Base):" % {'table': tableName})
             out.append("    __tablename__ = '%(table)s'\n" %
                             {'table': tableName})
             for col in table.columns:
                 out.append("    %s" % self.column_repr(col))
             out.append('\n')
         else:
             out.append("%(table)s = Table('%(table)s', %(meta)s," %
                        {'table': tableName, 'meta': metaName})
             for col in table.columns:
                 out.append("    %s," % self.column_repr(col))
             out.append(")\n")
         return out
 
     def _get_tables(self,missingA=False,missingB=False,modified=False):
         to_process = []
         for bool_,names,metadata in (
             (missingA,self.diff.tables_missing_from_A,self.diff.metadataB),
             (missingB,self.diff.tables_missing_from_B,self.diff.metadataA),
             (modified,self.diff.tables_different,self.diff.metadataA),
                 ):
             if bool_:
                 for name in names:
                     yield metadata.tables.get(name)
 
     def genBDefinition(self):
         """Generates the source code for a definition of B.
 
         Assumes a diff where A is empty.
 
         Was: toPython. Assume database (B) is current and model (A) is empty.
         """
 
         out = []
         if self.declarative:
             out.append(DECLARATIVE_HEADER)
         else:
             out.append(HEADER)
         out.append("")
         for table in self._get_tables(missingA=True):
             out.extend(self._getTableDefn(table))
         return '\n'.join(out)
 
     def genB2AMigration(self, indent='    '):
-        '''Generate a migration from B to A.
+        """Generate a migration from B to A.
 
         Was: toUpgradeDowngradePython
         Assume model (A) is most current and database (B) is out-of-date.
-        '''
+        """
 
         decls = ['from migrate.changeset import schema',
                  'pre_meta = MetaData()',
                  'post_meta = MetaData()',
                 ]
         upgradeCommands = ['pre_meta.bind = migrate_engine',
                            'post_meta.bind = migrate_engine']
         downgradeCommands = list(upgradeCommands)
 
         for tn in self.diff.tables_missing_from_A:
             pre_table = self.diff.metadataB.tables[tn]
             decls.extend(self._getTableDefn(pre_table, metaName='pre_meta'))
             upgradeCommands.append(
                 "pre_meta.tables[%(table)r].drop()" % {'table': tn})
             downgradeCommands.append(
                 "pre_meta.tables[%(table)r].create()" % {'table': tn})
 
         for tn in self.diff.tables_missing_from_B:
             post_table = self.diff.metadataA.tables[tn]
             decls.extend(self._getTableDefn(post_table, metaName='post_meta'))
             upgradeCommands.append(
                 "post_meta.tables[%(table)r].create()" % {'table': tn})
             downgradeCommands.append(
                 "post_meta.tables[%(table)r].drop()" % {'table': tn})
 
         for (tn, td) in self.diff.tables_different.iteritems():
             if td.columns_missing_from_A or td.columns_different:
                 pre_table = self.diff.metadataB.tables[tn]
                 decls.extend(self._getTableDefn(
                     pre_table, metaName='pre_meta'))
             if td.columns_missing_from_B or td.columns_different:
                 post_table = self.diff.metadataA.tables[tn]
                 decls.extend(self._getTableDefn(
                     post_table, metaName='post_meta'))
 
             for col in td.columns_missing_from_A:
                 upgradeCommands.append(
                     'pre_meta.tables[%r].columns[%r].drop()' % (tn, col))
                 downgradeCommands.append(
                     'pre_meta.tables[%r].columns[%r].create()' % (tn, col))
             for col in td.columns_missing_from_B:
                 upgradeCommands.append(
                     'post_meta.tables[%r].columns[%r].create()' % (tn, col))
                 downgradeCommands.append(
                     'post_meta.tables[%r].columns[%r].drop()' % (tn, col))
             for modelCol, databaseCol, modelDecl, databaseDecl in td.columns_different:
                 upgradeCommands.append(
                     'assert False, "Can\'t alter columns: %s:%s=>%s"' % (
                     tn, modelCol.name, databaseCol.name))
                 downgradeCommands.append(
                     'assert False, "Can\'t alter columns: %s:%s=>%s"' % (
                     tn, modelCol.name, databaseCol.name))
 
         return (
             '\n'.join(decls),
             '\n'.join('%s%s' % (indent, line) for line in upgradeCommands),
             '\n'.join('%s%s' % (indent, line) for line in downgradeCommands))
 
     def _db_can_handle_this_change(self,td):
         """Check if the database can handle going from B to A."""
 
         if (td.columns_missing_from_B
             and not td.columns_missing_from_A
             and not td.columns_different):
             # Even sqlite can handle column additions.
             return True
         else:
             return not self.engine.url.drivername.startswith('sqlite')
 
     def runB2A(self):
         """Goes from B to A.
 
         Was: applyModel. Apply model (A) to current database (B).
         """
 
         meta = sqlalchemy.MetaData(self.engine)
 
         for table in self._get_tables(missingA=True):
             table = table.tometadata(meta)
             table.drop()
         for table in self._get_tables(missingB=True):
             table = table.tometadata(meta)
             table.create()
         for modelTable in self._get_tables(modified=True):
             tableName = modelTable.name
             modelTable = modelTable.tometadata(meta)
             dbTable = self.diff.metadataB.tables[tableName]
 
             td = self.diff.tables_different[tableName]
 
             if self._db_can_handle_this_change(td):
 
                 for col in td.columns_missing_from_B:
                     modelTable.columns[col].create()
                 for col in td.columns_missing_from_A:
                     dbTable.columns[col].drop()
                 # XXX handle column changes here.
             else:
                 # Sqlite doesn't support drop column, so you have to
                 # do more: create temp table, copy data to it, drop
                 # old table, create new table, copy data back.
                 #
                 # I wonder if this is guaranteed to be unique?
                 tempName = '_temp_%s' % modelTable.name
 
                 def getCopyStatement():
                     preparer = self.engine.dialect.preparer
                     commonCols = []
                     for modelCol in modelTable.columns:
                         if modelCol.name in dbTable.columns:
                             commonCols.append(modelCol.name)
                     commonColsStr = ', '.join(commonCols)
                     return 'INSERT INTO %s (%s) SELECT %s FROM %s' % \
                         (tableName, commonColsStr, commonColsStr, tempName)
 
                 # Move the data in one transaction, so that we don't
                 # leave the database in a nasty state.
                 connection = self.engine.connect()
                 trans = connection.begin()
                 try:
                     connection.execute(
                         'CREATE TEMPORARY TABLE %s as SELECT * from %s' % \
                             (tempName, modelTable.name))
                     # make sure the drop takes place inside our
                     # transaction with the bind parameter
                     modelTable.drop(bind=connection)
                     modelTable.create(bind=connection)
                     connection.execute(getCopyStatement())
                     connection.execute('DROP TABLE %s' % tempName)
                     trans.commit()
                 except:
                     trans.rollback()
                     raise

rhodecode/lib/dbmigrate/migrate/versioning/schemadiff.py

@@ -1,295 +1,295 b''
 """
    Schema differencing support.
 """
 
 import logging
 import sqlalchemy
 
 from rhodecode.lib.dbmigrate.migrate.changeset import SQLA_06
 from sqlalchemy.types import Float
 
 log = logging.getLogger(__name__)
 
 
 def getDiffOfModelAgainstDatabase(metadata, engine, excludeTables=None):
     """
     Return differences of model against database.
 
     :return: object which will evaluate to :keyword:`True` if there \
       are differences else :keyword:`False`.
     """
     db_metadata = sqlalchemy.MetaData(engine)
     db_metadata.reflect()
 
     # sqlite will include a dynamically generated 'sqlite_sequence' table if
     # there are autoincrement sequences in the database; this should not be
     # compared.
     if engine.dialect.name == 'sqlite':
         if 'sqlite_sequence' in db_metadata.tables:
             db_metadata.remove(db_metadata.tables['sqlite_sequence'])
 
     return SchemaDiff(metadata, db_metadata,
                       labelA='model',
                       labelB='database',
                       excludeTables=excludeTables)
 
 
 def getDiffOfModelAgainstModel(metadataA, metadataB, excludeTables=None):
     """
     Return differences of model against another model.
 
     :return: object which will evaluate to :keyword:`True` if there \
       are differences else :keyword:`False`.
     """
     return SchemaDiff(metadataA, metadataB, excludeTables=excludeTables)
 
 
 class ColDiff(object):
     """
     Container for differences in one :class:`~sqlalchemy.schema.Column`
     between two :class:`~sqlalchemy.schema.Table` instances, ``A``
     and ``B``.
 
     .. attribute:: col_A
 
       The :class:`~sqlalchemy.schema.Column` object for A.
 
     .. attribute:: col_B
 
       The :class:`~sqlalchemy.schema.Column` object for B.
 
     .. attribute:: type_A
 
       The most generic type of the :class:`~sqlalchemy.schema.Column`
       object in A.
 
     .. attribute:: type_B
 
       The most generic type of the :class:`~sqlalchemy.schema.Column`
       object in A.
 
     """
 
     diff = False
 
     def __init__(self,col_A,col_B):
         self.col_A = col_A
         self.col_B = col_B
 
         self.type_A = col_A.type
         self.type_B = col_B.type
 
         self.affinity_A = self.type_A._type_affinity
         self.affinity_B = self.type_B._type_affinity
 
         if self.affinity_A is not self.affinity_B:
             self.diff = True
             return
 
         if isinstance(self.type_A,Float) or isinstance(self.type_B,Float):
             if not (isinstance(self.type_A,Float) and isinstance(self.type_B,Float)):
                 self.diff=True
                 return
 
         for attr in ('precision','scale','length'):
             A = getattr(self.type_A,attr,None)
             B = getattr(self.type_B,attr,None)
             if not (A is None or B is None) and A!=B:
                 self.diff=True
                 return
 
     def __nonzero__(self):
         return self.diff
 
 class TableDiff(object):
     """
     Container for differences in one :class:`~sqlalchemy.schema.Table`
     between two :class:`~sqlalchemy.schema.MetaData` instances, ``A``
     and ``B``.
 
     .. attribute:: columns_missing_from_A
 
       A sequence of column names that were found in B but weren't in
       A.
 
     .. attribute:: columns_missing_from_B
 
       A sequence of column names that were found in A but weren't in
       B.
 
     .. attribute:: columns_different
 
       A dictionary containing information about columns that were
       found to be different.
       It maps column names to a :class:`ColDiff` objects describing the
       differences found.
     """
     __slots__ = (
         'columns_missing_from_A',
         'columns_missing_from_B',
         'columns_different',
         )
 
     def __nonzero__(self):
         return bool(
             self.columns_missing_from_A or
             self.columns_missing_from_B or
             self.columns_different
             )
 
 class SchemaDiff(object):
     """
     Compute the difference between two :class:`~sqlalchemy.schema.MetaData`
     objects.
 
     The string representation of a :class:`SchemaDiff` will summarise
     the changes found between the two
     :class:`~sqlalchemy.schema.MetaData` objects.
 
     The length of a :class:`SchemaDiff` will give the number of
     changes found, enabling it to be used much like a boolean in
     expressions.
 
     :param metadataA:
       First :class:`~sqlalchemy.schema.MetaData` to compare.
 
     :param metadataB:
       Second :class:`~sqlalchemy.schema.MetaData` to compare.
 
     :param labelA:
       The label to use in messages about the first
       :class:`~sqlalchemy.schema.MetaData`.
 
     :param labelB:
       The label to use in messages about the second
       :class:`~sqlalchemy.schema.MetaData`.
 
     :param excludeTables:
       A sequence of table names to exclude.
 
     .. attribute:: tables_missing_from_A
 
       A sequence of table names that were found in B but weren't in
       A.
 
     .. attribute:: tables_missing_from_B
 
       A sequence of table names that were found in A but weren't in
       B.
 
     .. attribute:: tables_different
 
       A dictionary containing information about tables that were found
       to be different.
       It maps table names to a :class:`TableDiff` objects describing the
       differences found.
     """
 
     def __init__(self,
                  metadataA, metadataB,
                  labelA='metadataA',
                  labelB='metadataB',
                  excludeTables=None):
 
         self.metadataA, self.metadataB = metadataA, metadataB
         self.labelA, self.labelB = labelA, labelB
         self.label_width = max(len(labelA),len(labelB))
         excludeTables = set(excludeTables or [])
 
         A_table_names = set(metadataA.tables.keys())
         B_table_names = set(metadataB.tables.keys())
 
         self.tables_missing_from_A = sorted(
             B_table_names - A_table_names - excludeTables
             )
         self.tables_missing_from_B = sorted(
             A_table_names - B_table_names - excludeTables
             )
 
         self.tables_different = {}
         for table_name in A_table_names.intersection(B_table_names):
 
             td = TableDiff()
 
             A_table = metadataA.tables[table_name]
             B_table = metadataB.tables[table_name]
 
             A_column_names = set(A_table.columns.keys())
             B_column_names = set(B_table.columns.keys())
 
             td.columns_missing_from_A = sorted(
                 B_column_names - A_column_names
                 )
 
             td.columns_missing_from_B = sorted(
                 A_column_names - B_column_names
                 )
 
             td.columns_different = {}
 
             for col_name in A_column_names.intersection(B_column_names):
 
                 cd = ColDiff(
                     A_table.columns.get(col_name),
                     B_table.columns.get(col_name)
                     )
 
                 if cd:
                     td.columns_different[col_name]=cd
 
             # XXX - index and constraint differences should
             #       be checked for here
 
             if td:
                 self.tables_different[table_name]=td
 
     def __str__(self):
-        ''' Summarize differences. '''
+        """ Summarize differences. """
         out = []
         column_template ='      %%%is: %%r' % self.label_width
 
         for names,label in (
             (self.tables_missing_from_A,self.labelA),
             (self.tables_missing_from_B,self.labelB),
             ):
             if names:
                 out.append(
                     '  tables missing from %s: %s' % (
                         label,', '.join(sorted(names))
                         )
                     )
 
         for name,td in sorted(self.tables_different.items()):
             out.append(
                '  table with differences: %s' % name
                )
             for names,label in (
                 (td.columns_missing_from_A,self.labelA),
                 (td.columns_missing_from_B,self.labelB),
                 ):
                 if names:
                     out.append(
                         '    %s missing these columns: %s' % (
                             label,', '.join(sorted(names))
                             )
                         )
             for name,cd in td.columns_different.items():
                 out.append('    column with differences: %s' % name)
                 out.append(column_template % (self.labelA,cd.col_A))
                 out.append(column_template % (self.labelB,cd.col_B))
 
         if out:
             out.insert(0, 'Schema diffs:')
             return '\n'.join(out)
         else:
             return 'No schema diffs'
 
     def __len__(self):
         """
         Used in bool evaluation, return of 0 means no diffs.
         """
         return (
             len(self.tables_missing_from_A) +
             len(self.tables_missing_from_B) +
             len(self.tables_different)
             )

rhodecode/lib/vcs/subprocessio.py

@@ -1,415 +1,415 b''
-'''
+"""
 Module provides a class allowing to wrap communication over subprocess.Popen
 input, output, error streams into a meaningfull, non-blocking, concurrent
 stream processor exposing the output data as an iterator fitting to be a
 return value passed by a WSGI applicaiton to a WSGI server per PEP 3333.
 
 Copyright (c) 2011  Daniel Dotsenko <[e-mail]>
 
 This file is part of git_http_backend.py Project.
 
 git_http_backend.py Project is free software: you can redistribute it and/or
 modify it under the terms of the GNU Lesser General Public License as
 published by the Free Software Foundation, either version 2.1 of the License,
 or (at your option) any later version.
 
 git_http_backend.py Project 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 Lesser General Public License for more details.
 
 You should have received a copy of the GNU Lesser General Public License
 along with git_http_backend.py Project.
 If not, see <http://www.gnu.org/licenses/>.
-'''
+"""
 import os
 import subprocess
 from rhodecode.lib.vcs.utils.compat import deque, Event, Thread, _bytes, _bytearray
 
 
 class StreamFeeder(Thread):
     """
     Normal writing into pipe-like is blocking once the buffer is filled.
     This thread allows a thread to seep data from a file-like into a pipe
     without blocking the main thread.
     We close inpipe once the end of the source stream is reached.
     """
     def __init__(self, source):
         super(StreamFeeder, self).__init__()
         self.daemon = True
         filelike = False
         self.bytes = _bytes()
         if type(source) in (type(''), _bytes, _bytearray):  # string-like
             self.bytes = _bytes(source)
         else:  # can be either file pointer or file-like
             if type(source) in (int, long):  # file pointer it is
                 ## converting file descriptor (int) stdin into file-like
                 try:
                     source = os.fdopen(source, 'rb', 16384)
                 except Exception:
                     pass
             # let's see if source is file-like by now
             try:
                 filelike = source.read
             except Exception:
                 pass
         if not filelike and not self.bytes:
             raise TypeError("StreamFeeder's source object must be a readable "
                             "file-like, a file descriptor, or a string-like.")
         self.source = source
         self.readiface, self.writeiface = os.pipe()
 
     def run(self):
         t = self.writeiface
         if self.bytes:
             os.write(t, self.bytes)
         else:
             s = self.source
             b = s.read(4096)
             while b:
                 os.write(t, b)
                 b = s.read(4096)
         os.close(t)
 
     @property
     def output(self):
         return self.readiface
 
 
 class InputStreamChunker(Thread):
     def __init__(self, source, target, buffer_size, chunk_size):
 
         super(InputStreamChunker, self).__init__()
 
         self.daemon = True  # die die die.
 
         self.source = source
         self.target = target
         self.chunk_count_max = int(buffer_size / chunk_size) + 1
         self.chunk_size = chunk_size
 
         self.data_added = Event()
         self.data_added.clear()
 
         self.keep_reading = Event()
         self.keep_reading.set()
 
         self.EOF = Event()
         self.EOF.clear()
 
         self.go = Event()
         self.go.set()
 
     def stop(self):
         self.go.clear()
         self.EOF.set()
         try:
             # this is not proper, but is done to force the reader thread let
             # go of the input because, if successful, .close() will send EOF
             # down the pipe.
             self.source.close()
         except:
             pass
 
     def run(self):
         s = self.source
         t = self.target
         cs = self.chunk_size
         ccm = self.chunk_count_max
         kr = self.keep_reading
         da = self.data_added
         go = self.go
 
         try:
             b = s.read(cs)
         except ValueError:
             b = ''
 
         while b and go.is_set():
             if len(t) > ccm:
                 kr.clear()
                 kr.wait(2)
 #                # this only works on 2.7.x and up
 #                if not kr.wait(10):
 #                    raise Exception("Timed out while waiting for input to be read.")
                 # instead we'll use this
                 if len(t) > ccm + 3:
                     raise IOError("Timed out while waiting for input from subprocess.")
             t.append(b)
             da.set()
             b = s.read(cs)
         self.EOF.set()
         da.set()  # for cases when done but there was no input.
 
 
 class BufferedGenerator():
-    '''
+    """
     Class behaves as a non-blocking, buffered pipe reader.
     Reads chunks of data (through a thread)
     from a blocking pipe, and attaches these to an array (Deque) of chunks.
     Reading is halted in the thread when max chunks is internally buffered.
     The .next() may operate in blocking or non-blocking fashion by yielding
     '' if no data is ready
     to be sent or by not returning until there is some data to send
     When we get EOF from underlying source pipe we raise the marker to raise
     StopIteration after the last chunk of data is yielded.
-    '''
+    """
 
     def __init__(self, source, buffer_size=65536, chunk_size=4096,
                  starting_values=[], bottomless=False):
 
         if bottomless:
             maxlen = int(buffer_size / chunk_size)
         else:
             maxlen = None
 
         self.data = deque(starting_values, maxlen)
 
         self.worker = InputStreamChunker(source, self.data, buffer_size,
                                          chunk_size)
         if starting_values:
             self.worker.data_added.set()
         self.worker.start()
 
     ####################
     # Generator's methods
     ####################
 
     def __iter__(self):
         return self
 
     def next(self):
         while not len(self.data) and not self.worker.EOF.is_set():
             self.worker.data_added.clear()
             self.worker.data_added.wait(0.2)
         if len(self.data):
             self.worker.keep_reading.set()
             return _bytes(self.data.popleft())
         elif self.worker.EOF.is_set():
             raise StopIteration
 
     def throw(self, type, value=None, traceback=None):
         if not self.worker.EOF.is_set():
             raise type(value)
 
     def start(self):
         self.worker.start()
 
     def stop(self):
         self.worker.stop()
 
     def close(self):
         try:
             self.worker.stop()
             self.throw(GeneratorExit)
         except (GeneratorExit, StopIteration):
             pass
 
     def __del__(self):
         self.close()
 
     ####################
     # Threaded reader's infrastructure.
     ####################
     @property
     def input(self):
         return self.worker.w
 
     @property
     def data_added_event(self):
         return self.worker.data_added
 
     @property
     def data_added(self):
         return self.worker.data_added.is_set()
 
     @property
     def reading_paused(self):
         return not self.worker.keep_reading.is_set()
 
     @property
     def done_reading_event(self):
-        '''
+        """
         Done_reding does not mean that the iterator's buffer is empty.
         Iterator might have done reading from underlying source, but the read
         chunks might still be available for serving through .next() method.
 
         @return An Event class instance.
-        '''
+        """
         return self.worker.EOF
 
     @property
     def done_reading(self):
-        '''
+        """
         Done_reding does not mean that the iterator's buffer is empty.
         Iterator might have done reading from underlying source, but the read
         chunks might still be available for serving through .next() method.
 
         @return An Bool value.
-        '''
+        """
         return self.worker.EOF.is_set()
 
     @property
     def length(self):
-        '''
+        """
         returns int.
 
         This is the lenght of the que of chunks, not the length of
         the combined contents in those chunks.
 
         __len__() cannot be meaningfully implemented because this
         reader is just flying throuh a bottomless pit content and
         can only know the lenght of what it already saw.
 
         If __len__() on WSGI server per PEP 3333 returns a value,
         the responce's length will be set to that. In order not to
         confuse WSGI PEP3333 servers, we will not implement __len__
         at all.
-        '''
+        """
         return len(self.data)
 
     def prepend(self, x):
         self.data.appendleft(x)
 
     def append(self, x):
         self.data.append(x)
 
     def extend(self, o):
         self.data.extend(o)
 
     def __getitem__(self, i):
         return self.data[i]
 
 
 class SubprocessIOChunker(object):
-    '''
+    """
     Processor class wrapping handling of subprocess IO.
 
     In a way, this is a "communicate()" replacement with a twist.
 
     - We are multithreaded. Writing in and reading out, err are all sep threads.
     - We support concurrent (in and out) stream processing.
     - The output is not a stream. It's a queue of read string (bytes, not unicode)
       chunks. The object behaves as an iterable. You can "for chunk in obj:" us.
     - We are non-blocking in more respects than communicate()
       (reading from subprocess out pauses when internal buffer is full, but
        does not block the parent calling code. On the flip side, reading from
        slow-yielding subprocess may block the iteration until data shows up. This
        does not block the parallel inpipe reading occurring parallel thread.)
 
     The purpose of the object is to allow us to wrap subprocess interactions into
     and interable that can be passed to a WSGI server as the application's return
     value. Because of stream-processing-ability, WSGI does not have to read ALL
     of the subprocess's output and buffer it, before handing it to WSGI server for
     HTTP response. Instead, the class initializer reads just a bit of the stream
     to figure out if error ocurred or likely to occur and if not, just hands the
     further iteration over subprocess output to the server for completion of HTTP
     response.
 
     The real or perceived subprocess error is trapped and raised as one of
     EnvironmentError family of exceptions
 
     Example usage:
     #    try:
     #        answer = SubprocessIOChunker(
     #            cmd,
     #            input,
     #            buffer_size = 65536,
     #            chunk_size = 4096
     #            )
     #    except (EnvironmentError) as e:
     #        print str(e)
     #        raise e
     #
     #    return answer
 
 
-    '''
+    """
     def __init__(self, cmd, inputstream=None, buffer_size=65536,
                  chunk_size=4096, starting_values=[], **kwargs):
-        '''
+        """
         Initializes SubprocessIOChunker
 
         :param cmd: A Subprocess.Popen style "cmd". Can be string or array of strings
         :param inputstream: (Default: None) A file-like, string, or file pointer.
         :param buffer_size: (Default: 65536) A size of total buffer per stream in bytes.
         :param chunk_size: (Default: 4096) A max size of a chunk. Actual chunk may be smaller.
         :param starting_values: (Default: []) An array of strings to put in front of output que.
-        '''
+        """
 
         if inputstream:
             input_streamer = StreamFeeder(inputstream)
             input_streamer.start()
             inputstream = input_streamer.output
 
         _shell = kwargs.get('shell', True)
         if isinstance(cmd, (list, tuple)):
             cmd = ' '.join(cmd)
 
         kwargs['shell'] = _shell
         _p = subprocess.Popen(cmd,
             bufsize=-1,
             stdin=inputstream,
             stdout=subprocess.PIPE,
             stderr=subprocess.PIPE,
             **kwargs
         )
 
         bg_out = BufferedGenerator(_p.stdout, buffer_size, chunk_size, starting_values)
         bg_err = BufferedGenerator(_p.stderr, 16000, 1, bottomless=True)
 
         while not bg_out.done_reading and not bg_out.reading_paused and not bg_err.length:
             # doing this until we reach either end of file, or end of buffer.
             bg_out.data_added_event.wait(1)
             bg_out.data_added_event.clear()
 
         # at this point it's still ambiguous if we are done reading or just full buffer.
         # Either way, if error (returned by ended process, or implied based on
         # presence of stuff in stderr output) we error out.
         # Else, we are happy.
         _returncode = _p.poll()
         if _returncode or (_returncode == None and bg_err.length):
             try:
                 _p.terminate()
             except:
                 pass
             bg_out.stop()
             bg_err.stop()
             err = '%s' % ''.join(bg_err)
             if err:
                 raise EnvironmentError("Subprocess exited due to an error:\n" + err)
             raise EnvironmentError("Subprocess exited with non 0 ret code:%s" % _returncode)
 
         self.process = _p
         self.output = bg_out
         self.error = bg_err
 
     def __iter__(self):
         return self
 
     def next(self):
         if self.process.poll():
             err = '%s' % ''.join(self.error)
             raise EnvironmentError("Subprocess exited due to an error:\n" + err)
         return self.output.next()
 
     def throw(self, type, value=None, traceback=None):
         if self.output.length or not self.output.done_reading:
             raise type(value)
 
     def close(self):
         try:
             self.process.terminate()
         except:
             pass
         try:
             self.output.close()
         except:
             pass
         try:
             self.error.close()
         except:
             pass
 
     def __del__(self):
         self.close()

rhodecode/tests/vcs/test_git.py

@@ -1,707 +1,707 b''
 from __future__ import with_statement
 
 import os
 import mock
 import datetime
 from rhodecode.lib.vcs.backends.git import GitRepository, GitChangeset
 from rhodecode.lib.vcs.exceptions import RepositoryError, VCSError, NodeDoesNotExistError
 from rhodecode.lib.vcs.nodes import NodeKind, FileNode, DirNode, NodeState
 from rhodecode.lib.vcs.utils.compat import unittest
 from rhodecode.tests.vcs.base import BackendTestMixin
 from rhodecode.tests.vcs.conf import TEST_GIT_REPO, TEST_GIT_REPO_CLONE, get_new_dir
 
 
 class GitRepositoryTest(unittest.TestCase):
 
     def __check_for_existing_repo(self):
         if os.path.exists(TEST_GIT_REPO_CLONE):
             self.fail('Cannot test git clone repo as location %s already '
                       'exists. You should manually remove it first.'
                       % TEST_GIT_REPO_CLONE)
 
     def setUp(self):
         self.repo = GitRepository(TEST_GIT_REPO)
 
     def test_wrong_repo_path(self):
         wrong_repo_path = '/tmp/errorrepo'
         self.assertRaises(RepositoryError, GitRepository, wrong_repo_path)
 
     def test_repo_clone(self):
         self.__check_for_existing_repo()
         repo = GitRepository(TEST_GIT_REPO)
         repo_clone = GitRepository(TEST_GIT_REPO_CLONE,
             src_url=TEST_GIT_REPO, create=True, update_after_clone=True)
         self.assertEqual(len(repo.revisions), len(repo_clone.revisions))
         # Checking hashes of changesets should be enough
         for changeset in repo.get_changesets():
             raw_id = changeset.raw_id
             self.assertEqual(raw_id, repo_clone.get_changeset(raw_id).raw_id)
 
     def test_repo_clone_without_create(self):
         self.assertRaises(RepositoryError, GitRepository,
             TEST_GIT_REPO_CLONE + '_wo_create', src_url=TEST_GIT_REPO)
 
     def test_repo_clone_with_update(self):
         repo = GitRepository(TEST_GIT_REPO)
         clone_path = TEST_GIT_REPO_CLONE + '_with_update'
         repo_clone = GitRepository(clone_path,
             create=True, src_url=TEST_GIT_REPO, update_after_clone=True)
         self.assertEqual(len(repo.revisions), len(repo_clone.revisions))
 
         #check if current workdir was updated
         fpath = os.path.join(clone_path, 'MANIFEST.in')
         self.assertEqual(True, os.path.isfile(fpath),
             'Repo was cloned and updated but file %s could not be found'
             % fpath)
 
     def test_repo_clone_without_update(self):
         repo = GitRepository(TEST_GIT_REPO)
         clone_path = TEST_GIT_REPO_CLONE + '_without_update'
         repo_clone = GitRepository(clone_path,
             create=True, src_url=TEST_GIT_REPO, update_after_clone=False)
         self.assertEqual(len(repo.revisions), len(repo_clone.revisions))
         #check if current workdir was *NOT* updated
         fpath = os.path.join(clone_path, 'MANIFEST.in')
         # Make sure it's not bare repo
         self.assertFalse(repo_clone._repo.bare)
         self.assertEqual(False, os.path.isfile(fpath),
             'Repo was cloned and updated but file %s was found'
             % fpath)
 
     def test_repo_clone_into_bare_repo(self):
         repo = GitRepository(TEST_GIT_REPO)
         clone_path = TEST_GIT_REPO_CLONE + '_bare.git'
         repo_clone = GitRepository(clone_path, create=True,
             src_url=repo.path, bare=True)
         self.assertTrue(repo_clone._repo.bare)
 
     def test_create_repo_is_not_bare_by_default(self):
         repo = GitRepository(get_new_dir('not-bare-by-default'), create=True)
         self.assertFalse(repo._repo.bare)
 
     def test_create_bare_repo(self):
         repo = GitRepository(get_new_dir('bare-repo'), create=True, bare=True)
         self.assertTrue(repo._repo.bare)
 
     def test_revisions(self):
         # there are 112 revisions (by now)
         # so we can assume they would be available from now on
         subset = set([
             'c1214f7e79e02fc37156ff215cd71275450cffc3',
             '38b5fe81f109cb111f549bfe9bb6b267e10bc557',
             'fa6600f6848800641328adbf7811fd2372c02ab2',
             '102607b09cdd60e2793929c4f90478be29f85a17',
             '49d3fd156b6f7db46313fac355dca1a0b94a0017',
             '2d1028c054665b962fa3d307adfc923ddd528038',
             'd7e0d30fbcae12c90680eb095a4f5f02505ce501',
             'ff7ca51e58c505fec0dd2491de52c622bb7a806b',
             'dd80b0f6cf5052f17cc738c2951c4f2070200d7f',
             '8430a588b43b5d6da365400117c89400326e7992',
             'd955cd312c17b02143c04fa1099a352b04368118',
             'f67b87e5c629c2ee0ba58f85197e423ff28d735b',
             'add63e382e4aabc9e1afdc4bdc24506c269b7618',
             'f298fe1189f1b69779a4423f40b48edf92a703fc',
             'bd9b619eb41994cac43d67cf4ccc8399c1125808',
             '6e125e7c890379446e98980d8ed60fba87d0f6d1',
             'd4a54db9f745dfeba6933bf5b1e79e15d0af20bd',
             '0b05e4ed56c802098dfc813cbe779b2f49e92500',
             '191caa5b2c81ed17c0794bf7bb9958f4dcb0b87e',
             '45223f8f114c64bf4d6f853e3c35a369a6305520',
             'ca1eb7957a54bce53b12d1a51b13452f95bc7c7e',
             'f5ea29fc42ef67a2a5a7aecff10e1566699acd68',
             '27d48942240f5b91dfda77accd2caac94708cc7d',
             '622f0eb0bafd619d2560c26f80f09e3b0b0d78af',
             'e686b958768ee96af8029fe19c6050b1a8dd3b2b'])
         self.assertTrue(subset.issubset(set(self.repo.revisions)))
 
 
 
     def test_slicing(self):
         #4 1 5 10 95
         for sfrom, sto, size in [(0, 4, 4), (1, 2, 1), (10, 15, 5),
                                  (10, 20, 10), (5, 100, 95)]:
             revs = list(self.repo[sfrom:sto])
             self.assertEqual(len(revs), size)
             self.assertEqual(revs[0], self.repo.get_changeset(sfrom))
             self.assertEqual(revs[-1], self.repo.get_changeset(sto - 1))
 
 
     def test_branches(self):
         # TODO: Need more tests here
         # Removed (those are 'remotes' branches for cloned repo)
         #self.assertTrue('master' in self.repo.branches)
         #self.assertTrue('gittree' in self.repo.branches)
         #self.assertTrue('web-branch' in self.repo.branches)
         for name, id in self.repo.branches.items():
             self.assertTrue(isinstance(
                 self.repo.get_changeset(id), GitChangeset))
 
     def test_tags(self):
         # TODO: Need more tests here
         self.assertTrue('v0.1.1' in self.repo.tags)
         self.assertTrue('v0.1.2' in self.repo.tags)
         for name, id in self.repo.tags.items():
             self.assertTrue(isinstance(
                 self.repo.get_changeset(id), GitChangeset))
 
     def _test_single_changeset_cache(self, revision):
         chset = self.repo.get_changeset(revision)
         self.assertTrue(revision in self.repo.changesets)
         self.assertTrue(chset is self.repo.changesets[revision])
 
     def test_initial_changeset(self):
         id = self.repo.revisions[0]
         init_chset = self.repo.get_changeset(id)
         self.assertEqual(init_chset.message, 'initial import\n')
         self.assertEqual(init_chset.author,
             'Marcin Kuzminski <[e-mail]>')
         for path in ('vcs/__init__.py',
                      'vcs/backends/BaseRepository.py',
                      'vcs/backends/__init__.py'):
             self.assertTrue(isinstance(init_chset.get_node(path), FileNode))
         for path in ('', 'vcs', 'vcs/backends'):
             self.assertTrue(isinstance(init_chset.get_node(path), DirNode))
 
         self.assertRaises(NodeDoesNotExistError, init_chset.get_node, path='foobar')
 
         node = init_chset.get_node('vcs/')
         self.assertTrue(hasattr(node, 'kind'))
         self.assertEqual(node.kind, NodeKind.DIR)
 
         node = init_chset.get_node('vcs')
         self.assertTrue(hasattr(node, 'kind'))
         self.assertEqual(node.kind, NodeKind.DIR)
 
         node = init_chset.get_node('vcs/__init__.py')
         self.assertTrue(hasattr(node, 'kind'))
         self.assertEqual(node.kind, NodeKind.FILE)
 
     def test_not_existing_changeset(self):
         self.assertRaises(RepositoryError, self.repo.get_changeset,
             'f' * 40)
 
     def test_changeset10(self):
 
         chset10 = self.repo.get_changeset(self.repo.revisions[9])
         README = """===
 VCS
 ===
 
 Various Version Control System management abstraction layer for Python.
 
 Introduction
 ------------
 
 TODO: To be written...
 
 """
         node = chset10.get_node('README.rst')
         self.assertEqual(node.kind, NodeKind.FILE)
         self.assertEqual(node.content, README)
 
 
 class GitChangesetTest(unittest.TestCase):
 
     def setUp(self):
         self.repo = GitRepository(TEST_GIT_REPO)
 
     def test_default_changeset(self):
         tip = self.repo.get_changeset()
         self.assertEqual(tip, self.repo.get_changeset(None))
         self.assertEqual(tip, self.repo.get_changeset('tip'))
 
     def test_root_node(self):
         tip = self.repo.get_changeset()
         self.assertTrue(tip.root is tip.get_node(''))
 
     def test_lazy_fetch(self):
         """
         Test if changeset's nodes expands and are cached as we walk through
         the revision. This test is somewhat hard to write as order of tests
         is a key here. Written by running command after command in a shell.
         """
         hex = '2a13f185e4525f9d4b59882791a2d397b90d5ddc'
         self.assertTrue(hex in self.repo.revisions)
         chset = self.repo.get_changeset(hex)
         self.assertTrue(len(chset.nodes) == 0)
         root = chset.root
         self.assertTrue(len(chset.nodes) == 1)
         self.assertTrue(len(root.nodes) == 8)
         # accessing root.nodes updates chset.nodes
         self.assertTrue(len(chset.nodes) == 9)
 
         docs = root.get_node('docs')
         # we haven't yet accessed anything new as docs dir was already cached
         self.assertTrue(len(chset.nodes) == 9)
         self.assertTrue(len(docs.nodes) == 8)
         # accessing docs.nodes updates chset.nodes
         self.assertTrue(len(chset.nodes) == 17)
 
         self.assertTrue(docs is chset.get_node('docs'))
         self.assertTrue(docs is root.nodes[0])
         self.assertTrue(docs is root.dirs[0])
         self.assertTrue(docs is chset.get_node('docs'))
 
     def test_nodes_with_changeset(self):
         hex = '2a13f185e4525f9d4b59882791a2d397b90d5ddc'
         chset = self.repo.get_changeset(hex)
         root = chset.root
         docs = root.get_node('docs')
         self.assertTrue(docs is chset.get_node('docs'))
         api = docs.get_node('api')
         self.assertTrue(api is chset.get_node('docs/api'))
         index = api.get_node('index.rst')
         self.assertTrue(index is chset.get_node('docs/api/index.rst'))
         self.assertTrue(index is chset.get_node('docs')\
             .get_node('api')\
             .get_node('index.rst'))
 
     def test_branch_and_tags(self):
-        '''
+        """
         rev0 = self.repo.revisions[0]
         chset0 = self.repo.get_changeset(rev0)
         self.assertEqual(chset0.branch, 'master')
         self.assertEqual(chset0.tags, [])
 
         rev10 = self.repo.revisions[10]
         chset10 = self.repo.get_changeset(rev10)
         self.assertEqual(chset10.branch, 'master')
         self.assertEqual(chset10.tags, [])
 
         rev44 = self.repo.revisions[44]
         chset44 = self.repo.get_changeset(rev44)
         self.assertEqual(chset44.branch, 'web-branch')
 
         tip = self.repo.get_changeset('tip')
         self.assertTrue('tip' in tip.tags)
-        '''
+        """
         # Those tests would fail - branches are now going
         # to be changed at main API in order to support git backend
         pass
 
     def _test_slices(self, limit, offset):
         count = self.repo.count()
         changesets = self.repo.get_changesets(limit=limit, offset=offset)
         idx = 0
         for changeset in changesets:
             rev = offset + idx
             idx += 1
             rev_id = self.repo.revisions[rev]
             if idx > limit:
                 self.fail("Exceeded limit already (getting revision %s, "
                     "there are %s total revisions, offset=%s, limit=%s)"
                     % (rev_id, count, offset, limit))
             self.assertEqual(changeset, self.repo.get_changeset(rev_id))
         result = list(self.repo.get_changesets(limit=limit, offset=offset))
         start = offset
         end = limit and offset + limit or None
         sliced = list(self.repo[start:end])
         self.failUnlessEqual(result, sliced,
             msg="Comparison failed for limit=%s, offset=%s"
             "(get_changeset returned: %s and sliced: %s"
             % (limit, offset, result, sliced))
 
     def _test_file_size(self, revision, path, size):
         node = self.repo.get_changeset(revision).get_node(path)
         self.assertTrue(node.is_file())
         self.assertEqual(node.size, size)
 
     def test_file_size(self):
         to_check = (
             ('c1214f7e79e02fc37156ff215cd71275450cffc3',
                 'vcs/backends/BaseRepository.py', 502),
             ('d7e0d30fbcae12c90680eb095a4f5f02505ce501',
                 'vcs/backends/hg.py', 854),
             ('6e125e7c890379446e98980d8ed60fba87d0f6d1',
                 'setup.py', 1068),
 
             ('d955cd312c17b02143c04fa1099a352b04368118',
                 'vcs/backends/base.py', 2921),
             ('ca1eb7957a54bce53b12d1a51b13452f95bc7c7e',
                 'vcs/backends/base.py', 3936),
             ('f50f42baeed5af6518ef4b0cb2f1423f3851a941',
                 'vcs/backends/base.py', 6189),
         )
         for revision, path, size in to_check:
             self._test_file_size(revision, path, size)
 
     def test_file_history(self):
         # we can only check if those revisions are present in the history
         # as we cannot update this test every time file is changed
         files = {
             'setup.py': [
                 '54386793436c938cff89326944d4c2702340037d',
                 '51d254f0ecf5df2ce50c0b115741f4cf13985dab',
                 '998ed409c795fec2012b1c0ca054d99888b22090',
                 '5e0eb4c47f56564395f76333f319d26c79e2fb09',
                 '0115510b70c7229dbc5dc49036b32e7d91d23acd',
                 '7cb3fd1b6d8c20ba89e2264f1c8baebc8a52d36e',
                 '2a13f185e4525f9d4b59882791a2d397b90d5ddc',
                 '191caa5b2c81ed17c0794bf7bb9958f4dcb0b87e',
                 'ff7ca51e58c505fec0dd2491de52c622bb7a806b',
             ],
             'vcs/nodes.py': [
                 '33fa3223355104431402a888fa77a4e9956feb3e',
                 'fa014c12c26d10ba682fadb78f2a11c24c8118e1',
                 'e686b958768ee96af8029fe19c6050b1a8dd3b2b',
                 'ab5721ca0a081f26bf43d9051e615af2cc99952f',
                 'c877b68d18e792a66b7f4c529ea02c8f80801542',
                 '4313566d2e417cb382948f8d9d7c765330356054',
                 '6c2303a793671e807d1cfc70134c9ca0767d98c2',
                 '54386793436c938cff89326944d4c2702340037d',
                 '54000345d2e78b03a99d561399e8e548de3f3203',
                 '1c6b3677b37ea064cb4b51714d8f7498f93f4b2b',
                 '2d03ca750a44440fb5ea8b751176d1f36f8e8f46',
                 '2a08b128c206db48c2f0b8f70df060e6db0ae4f8',
                 '30c26513ff1eb8e5ce0e1c6b477ee5dc50e2f34b',
                 'ac71e9503c2ca95542839af0ce7b64011b72ea7c',
                 '12669288fd13adba2a9b7dd5b870cc23ffab92d2',
                 '5a0c84f3e6fe3473e4c8427199d5a6fc71a9b382',
                 '12f2f5e2b38e6ff3fbdb5d722efed9aa72ecb0d5',
                 '5eab1222a7cd4bfcbabc218ca6d04276d4e27378',
                 'f50f42baeed5af6518ef4b0cb2f1423f3851a941',
                 'd7e390a45f6aa96f04f5e7f583ad4f867431aa25',
                 'f15c21f97864b4f071cddfbf2750ec2e23859414',
                 'e906ef056cf539a4e4e5fc8003eaf7cf14dd8ade',
                 'ea2b108b48aa8f8c9c4a941f66c1a03315ca1c3b',
                 '84dec09632a4458f79f50ddbbd155506c460b4f9',
                 '0115510b70c7229dbc5dc49036b32e7d91d23acd',
                 '2a13f185e4525f9d4b59882791a2d397b90d5ddc',
                 '3bf1c5868e570e39569d094f922d33ced2fa3b2b',
                 'b8d04012574729d2c29886e53b1a43ef16dd00a1',
                 '6970b057cffe4aab0a792aa634c89f4bebf01441',
                 'dd80b0f6cf5052f17cc738c2951c4f2070200d7f',
                 'ff7ca51e58c505fec0dd2491de52c622bb7a806b',
             ],
             'vcs/backends/git.py': [
                 '4cf116ad5a457530381135e2f4c453e68a1b0105',
                 '9a751d84d8e9408e736329767387f41b36935153',
                 'cb681fb539c3faaedbcdf5ca71ca413425c18f01',
                 '428f81bb652bcba8d631bce926e8834ff49bdcc6',
                 '180ab15aebf26f98f714d8c68715e0f05fa6e1c7',
                 '2b8e07312a2e89e92b90426ab97f349f4bce2a3a',
                 '50e08c506174d8645a4bb517dd122ac946a0f3bf',
                 '54000345d2e78b03a99d561399e8e548de3f3203',
             ],
         }
         for path, revs in files.items():
             node = self.repo.get_changeset(revs[0]).get_node(path)
             node_revs = [chset.raw_id for chset in node.history]
             self.assertTrue(set(revs).issubset(set(node_revs)),
                 "We assumed that %s is subset of revisions for which file %s "
                 "has been changed, and history of that node returned: %s"
                 % (revs, path, node_revs))
 
     def test_file_annotate(self):
         files = {
             'vcs/backends/__init__.py': {
                 'c1214f7e79e02fc37156ff215cd71275450cffc3': {
                     'lines_no': 1,
                     'changesets': [
                         'c1214f7e79e02fc37156ff215cd71275450cffc3',
                     ],
                 },
                 '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647': {
                     'lines_no': 21,
                     'changesets': [
                         '49d3fd156b6f7db46313fac355dca1a0b94a0017',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                     ],
                 },
                 'e29b67bd158580fc90fc5e9111240b90e6e86064': {
                     'lines_no': 32,
                     'changesets': [
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '5eab1222a7cd4bfcbabc218ca6d04276d4e27378',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '54000345d2e78b03a99d561399e8e548de3f3203',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '78c3f0c23b7ee935ec276acb8b8212444c33c396',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '2a13f185e4525f9d4b59882791a2d397b90d5ddc',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '78c3f0c23b7ee935ec276acb8b8212444c33c396',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '992f38217b979d0b0987d0bae3cc26dac85d9b19',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                         '16fba1ae9334d79b66d7afed2c2dfbfa2ae53647',
                     ],
                 },
             },
         }
 
         for fname, revision_dict in files.items():
             for rev, data in revision_dict.items():
                 cs = self.repo.get_changeset(rev)
 
                 l1_1 = [x[1] for x in cs.get_file_annotate(fname)]
                 l1_2 = [x[2]().raw_id for x in cs.get_file_annotate(fname)]
                 self.assertEqual(l1_1, l1_2)
                 l1 = l1_1
                 l2 = files[fname][rev]['changesets']
                 self.assertTrue(l1 == l2 , "The lists of revision for %s@rev %s"
                                 "from annotation list should match each other, "
                                 "got \n%s \nvs \n%s " % (fname, rev, l1, l2))
 
     def test_files_state(self):
         """
         Tests state of FileNodes.
         """
         node = self.repo\
             .get_changeset('e6ea6d16e2f26250124a1f4b4fe37a912f9d86a0')\
             .get_node('vcs/utils/diffs.py')
         self.assertTrue(node.state, NodeState.ADDED)
         self.assertTrue(node.added)
         self.assertFalse(node.changed)
         self.assertFalse(node.not_changed)
         self.assertFalse(node.removed)
 
         node = self.repo\
             .get_changeset('33fa3223355104431402a888fa77a4e9956feb3e')\
             .get_node('.hgignore')
         self.assertTrue(node.state, NodeState.CHANGED)
         self.assertFalse(node.added)
         self.assertTrue(node.changed)
         self.assertFalse(node.not_changed)
         self.assertFalse(node.removed)
 
         node = self.repo\
             .get_changeset('e29b67bd158580fc90fc5e9111240b90e6e86064')\
             .get_node('setup.py')
         self.assertTrue(node.state, NodeState.NOT_CHANGED)
         self.assertFalse(node.added)
         self.assertFalse(node.changed)
         self.assertTrue(node.not_changed)
         self.assertFalse(node.removed)
 
         # If node has REMOVED state then trying to fetch it would raise
         # ChangesetError exception
         chset = self.repo.get_changeset(
             'fa6600f6848800641328adbf7811fd2372c02ab2')
         path = 'vcs/backends/BaseRepository.py'
         self.assertRaises(NodeDoesNotExistError, chset.get_node, path)
         # but it would be one of ``removed`` (changeset's attribute)
         self.assertTrue(path in [rf.path for rf in chset.removed])
 
         chset = self.repo.get_changeset(
             '54386793436c938cff89326944d4c2702340037d')
         changed = ['setup.py', 'tests/test_nodes.py', 'vcs/backends/hg.py',
             'vcs/nodes.py']
         self.assertEqual(set(changed), set([f.path for f in chset.changed]))
 
     def test_commit_message_is_unicode(self):
         for cs in self.repo:
             self.assertEqual(type(cs.message), unicode)
 
     def test_changeset_author_is_unicode(self):
         for cs in self.repo:
             self.assertEqual(type(cs.author), unicode)
 
     def test_repo_files_content_is_unicode(self):
         changeset = self.repo.get_changeset()
         for node in changeset.get_node('/'):
             if node.is_file():
                 self.assertEqual(type(node.content), unicode)
 
     def test_wrong_path(self):
         # There is 'setup.py' in the root dir but not there:
         path = 'foo/bar/setup.py'
         tip = self.repo.get_changeset()
         self.assertRaises(VCSError, tip.get_node, path)
 
     def test_author_email(self):
         self.assertEqual('[e-mail]',
           self.repo.get_changeset('c1214f7e79e02fc37156ff215cd71275450cffc3')\
           .author_email)
         self.assertEqual('[e-mail]',
           self.repo.get_changeset('ff7ca51e58c505fec0dd2491de52c622bb7a806b')\
           .author_email)
         self.assertEqual('none@none',
           self.repo.get_changeset('8430a588b43b5d6da365400117c89400326e7992')\
           .author_email)
 
     def test_author_username(self):
         self.assertEqual('Marcin Kuzminski',
           self.repo.get_changeset('c1214f7e79e02fc37156ff215cd71275450cffc3')\
           .author_name)
         self.assertEqual('Lukasz Balcerzak',
           self.repo.get_changeset('ff7ca51e58c505fec0dd2491de52c622bb7a806b')\
           .author_name)
         self.assertEqual('marcink',
           self.repo.get_changeset('8430a588b43b5d6da365400117c89400326e7992')\
           .author_name)
 
 
 class GitSpecificTest(unittest.TestCase):
 
     def test_error_is_raised_for_added_if_diff_name_status_is_wrong(self):
         repo = mock.MagicMock()
         changeset = GitChangeset(repo, 'foobar')
         changeset._diff_name_status = 'foobar'
         with self.assertRaises(VCSError):
             changeset.added
 
     def test_error_is_raised_for_changed_if_diff_name_status_is_wrong(self):
         repo = mock.MagicMock()
         changeset = GitChangeset(repo, 'foobar')
         changeset._diff_name_status = 'foobar'
         with self.assertRaises(VCSError):
             changeset.added
 
     def test_error_is_raised_for_removed_if_diff_name_status_is_wrong(self):
         repo = mock.MagicMock()
         changeset = GitChangeset(repo, 'foobar')
         changeset._diff_name_status = 'foobar'
         with self.assertRaises(VCSError):
             changeset.added
 
 
 class GitSpecificWithRepoTest(BackendTestMixin, unittest.TestCase):
     backend_alias = 'git'
 
     @classmethod
     def _get_commits(cls):
         return [
             {
                 'message': 'Initial',
                 'author': 'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 20),
                 'added': [
                     FileNode('foobar/static/js/admin/base.js', content='base'),
                     FileNode('foobar/static/admin', content='admin',
                         mode=0120000), # this is a link
                     FileNode('foo', content='foo'),
                 ],
             },
             {
                 'message': 'Second',
                 'author': 'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 22),
                 'added': [
                     FileNode('foo2', content='foo2'),
                 ],
             },
         ]
 
     def test_paths_slow_traversing(self):
         cs = self.repo.get_changeset()
         self.assertEqual(cs.get_node('foobar').get_node('static').get_node('js')
             .get_node('admin').get_node('base.js').content, 'base')
 
     def test_paths_fast_traversing(self):
         cs = self.repo.get_changeset()
         self.assertEqual(cs.get_node('foobar/static/js/admin/base.js').content,
             'base')
 
     def test_workdir_get_branch(self):
         self.repo.run_git_command('checkout -b production')
         # Regression test: one of following would fail if we don't check
         # .git/HEAD file
         self.repo.run_git_command('checkout production')
         self.assertEqual(self.repo.workdir.get_branch(), 'production')
         self.repo.run_git_command('checkout master')
         self.assertEqual(self.repo.workdir.get_branch(), 'master')
 
     def test_get_diff_runs_git_command_with_hashes(self):
         self.repo.run_git_command = mock.Mock(return_value=['', ''])
         self.repo.get_diff(0, 1)
         self.repo.run_git_command.assert_called_once_with(
           'diff -U%s --full-index --binary -p -M --abbrev=40 %s %s' %
             (3, self.repo._get_revision(0), self.repo._get_revision(1)))
 
     def test_get_diff_runs_git_command_with_str_hashes(self):
         self.repo.run_git_command = mock.Mock(return_value=['', ''])
         self.repo.get_diff(self.repo.EMPTY_CHANGESET, 1)
         self.repo.run_git_command.assert_called_once_with(
             'show -U%s --full-index --binary -p -M --abbrev=40 %s' %
             (3, self.repo._get_revision(1)))
 
     def test_get_diff_runs_git_command_with_path_if_its_given(self):
         self.repo.run_git_command = mock.Mock(return_value=['', ''])
         self.repo.get_diff(0, 1, 'foo')
         self.repo.run_git_command.assert_called_once_with(
           'diff -U%s --full-index --binary -p -M --abbrev=40 %s %s -- "foo"'
             % (3, self.repo._get_revision(0), self.repo._get_revision(1)))
 
 
 class GitRegressionTest(BackendTestMixin, unittest.TestCase):
     backend_alias = 'git'
 
     @classmethod
     def _get_commits(cls):
         return [
             {
                 'message': 'Initial',
                 'author': 'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 20),
                 'added': [
                     FileNode('bot/__init__.py', content='base'),
                     FileNode('bot/templates/404.html', content='base'),
                     FileNode('bot/templates/500.html', content='base'),
                 ],
             },
             {
                 'message': 'Second',
                 'author': 'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 22),
                 'added': [
                     FileNode('bot/build/migrations/1.py', content='foo2'),
                     FileNode('bot/build/migrations/2.py', content='foo2'),
                     FileNode('bot/build/static/templates/f.html', content='foo2'),
                     FileNode('bot/build/static/templates/f1.html', content='foo2'),
                     FileNode('bot/build/templates/err.html', content='foo2'),
                     FileNode('bot/build/templates/err2.html', content='foo2'),
                 ],
             },
         ]
 
     def test_similar_paths(self):
         cs = self.repo.get_changeset()
         paths = lambda *n:[x.path for x in n]
         self.assertEqual(paths(*cs.get_nodes('bot')), ['bot/build', 'bot/templates', 'bot/__init__.py'])
         self.assertEqual(paths(*cs.get_nodes('bot/build')), ['bot/build/migrations', 'bot/build/static', 'bot/build/templates'])
         self.assertEqual(paths(*cs.get_nodes('bot/build/static')), ['bot/build/static/templates'])
         # this get_nodes below causes troubles !
         self.assertEqual(paths(*cs.get_nodes('bot/build/static/templates')), ['bot/build/static/templates/f.html', 'bot/build/static/templates/f1.html'])
         self.assertEqual(paths(*cs.get_nodes('bot/build/templates')), ['bot/build/templates/err.html', 'bot/build/templates/err2.html'])
         self.assertEqual(paths(*cs.get_nodes('bot/templates/')), ['bot/templates/404.html', 'bot/templates/500.html'])
 
 if __name__ == '__main__':
     unittest.main()