SimulationTool.py

##############################################################################
#
# Copyright (c) 2002, 2005 Nexedi SARL and Contributors. All Rights Reserved.
#                    Jean-Paul Smets-Solanes <jp@nexedi.com>
#                    Romain Courteaud <romain@nexedi.com>
#
# WARNING: This program as such is intended to be used by professional
# programmers who take the whole responsability of assessing all potential
# consequences resulting from its eventual inadequacies and bugs
# End users who are looking for a ready-to-use solution with commercial
# garantees and support are strongly adviced to contract a Free Software
# Service Company
#
# 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 2
# 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, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
#
##############################################################################

from Products.CMFCore.utils import getToolByName

from AccessControl import ClassSecurityInfo
from Globals import InitializeClass, DTMLFile
from Products.ERP5Type import Permissions
from Products.ERP5Type.Tool.BaseTool import BaseTool

from Products.ERP5 import _dtmldir

from zLOG import LOG

from Products.ERP5.Capacity.GLPK import solve
from Numeric import zeros, resize
from DateTime import DateTime

from Products.ERP5 import DeliverySolver
from Products.ERP5 import TargetSolver
from Products.PythonScripts.Utility import allow_class

from Products.ZSQLCatalog.SQLCatalog import Query, ComplexQuery

from Shared.DC.ZRDB.Results import Results
from Products.ERP5Type.Utils import mergeZRDBResults

class SimulationTool(BaseTool):
    """
    The SimulationTool implements the ERP5
    simulation algorithmics.


    Examples of applications:

    -

    -
    ERP5 main purpose:

    -

    -

    """
    id = 'portal_simulation'
    meta_type = 'ERP5 Simulation Tool'
    portal_type = 'Simulation Tool'
    allowed_types = ( 'ERP5 Applied Rule', )

    # Declarative Security
    security = ClassSecurityInfo()

    #
    #   ZMI methods
    #
    security.declareProtected( Permissions.ManagePortal, 'manage_overview' )
    manage_overview = DTMLFile( 'explainSimulationTool', _dtmldir )

    def filtered_meta_types(self, user=None):
      # Filters the list of available meta types.
      all = SimulationTool.inheritedAttribute('filtered_meta_types')(self)
      meta_types = []
      for meta_type in self.all_meta_types():
        if meta_type['name'] in self.allowed_types:
          meta_types.append(meta_type)
      return meta_types

    def tpValues(self) :
      """ show the content in the left pane of the ZMI """
      return self.objectValues()

    security.declarePrivate('manage_afterAdd')
    def manage_afterAdd(self, item, container) :
      """Init permissions right after creation.

      Permissions in simulation tool are simple:
       o Each member can access and create some content.
       o Only manager can view, because simulation can be seen as
         sensitive information.
      """
      item.manage_permission(Permissions.AddPortalContent,
            ['Member', 'Author', 'Manager'])
      item.manage_permission(Permissions.AccessContentsInformation,
            ['Member', 'Auditor', 'Manager'])
      item.manage_permission(Permissions.View,
            ['Manager',])
      BaseTool.inheritedAttribute('manage_afterAdd')(self, item, container)

    def solveDelivery(self, delivery, dsolver_name, tsolver_name,
                                     additional_parameters=None,**kw):
      """
        Solve a delivery by calling DeliverySolver and TargetSolver
      """
      self.solveMovementOrDelivery(delivery, dsolver_name, tsolver_name,
          delivery=1,additional_parameters=additional_parameters,**kw)

    def solveMovement(self, movement, dsolver_name, tsolver_name,
                                       additional_parameters=None,**kw):
      """
        Solve a movement by calling DeliverySolver and TargetSolver
      """
      return self.solveMovementOrDelivery(movement, dsolver_name, tsolver_name,
          movement=1,additional_parameters=additional_parameters,**kw)

    def solveMovementOrDelivery(self, obj, dsolver_name, tsolver_name,
                                          movement=0,delivery=0,
                                          additional_parameters=None,**kw):
      """
        Solve a delivery by calling DeliverySolver and TargetSolver
      """
      for solver_name, solver_module in [(dsolver_name, DeliverySolver),\
                                         (tsolver_name, TargetSolver)]:

        if solver_name is not None:
          solver_file_path = "%s.%s" % (solver_module.__name__,
                                        solver_name)
          __import__(solver_file_path)
          solver_file = getattr(solver_module, solver_name)
          solver_class = getattr(solver_file, solver_name)
          solver = solver_class(additional_parameters=additional_parameters,**kw)

          if movement:
            return solver.solveMovement(obj)
          if delivery:
            return solver.solveDelivery(obj)

    #######################################################
    # Stock Management

    def _generatePropertyUidList(self, prop, as_text=0):
      """
      converts relative_url or text (single element or list or dict)
        to an object usable by buildSQLQuery

      as_text == 0: tries to lookup an uid from the relative_url
      as_text == 1: directly passes the argument as text
      """
      if prop is None :
        return []
      category_tool = getToolByName(self, 'portal_categories')
      property_uid_list = []
      if isinstance(prop, str):
        if not as_text:
          prop_value = category_tool.getCategoryValue(prop)
          if prop_value is None:
            raise ValueError, 'Category %s does not exists' % prop
          property_uid_list.append(prop_value.getUid())
        else:
          property_uid_list.append(prop)
      elif isinstance(prop, (list, tuple)):
        for property_item in prop :
          if not as_text:
            prop_value = category_tool.getCategoryValue(property_item)
            if prop_value is None:
              raise ValueError, 'Category %s does not exists' % property_item
            property_uid_list.append(prop_value.getUid())
          else:
            property_uid_list.append(property_item)
      elif isinstance(prop, dict):
        tmp_uid_list = []
        if isinstance(prop['query'], str):
          prop['query'] = [prop['query']]
        for property_item in prop['query'] :
          if not as_text:
            prop_value = category_tool.getCategoryValue(property_item)
            if prop_value is None:
              raise ValueError, 'Category %s does not exists' % property_item
            tmp_uid_list.append(prop_value.getUid())
          else:
            tmp_uid_list.append(property_item)
        if tmp_uid_list:
          property_uid_list = {}
          property_uid_list['operator'] = prop['operator']
          property_uid_list['query'] = tmp_uid_list
      return property_uid_list

    def _getSimulationStateQuery(self, **kw):
      simulation_state_dict = self._getSimulationStateDict(**kw)
      return self._buildSimulationStateQuery(simulation_state_dict=simulation_state_dict)
      
    def _buildSimulationStateQuery(self, simulation_state_dict, table='stock'):
      input_simulation_state = simulation_state_dict.get(
                                 'input_simulation_state')
      output_simulation_state = simulation_state_dict.get(
                                 'output_simulation_state')
      simulation_state = simulation_state_dict.get('simulation_state')
      if simulation_state is not None:
        simulation_query = Query(operator='IN',
                                 **{'%s.simulation_state' % (table, ):
                                    simulation_state})
      elif input_simulation_state is not None:
        input_quantity_query = Query(**{'%s.quantity' % (table, ): '>0'})
        input_simulation_query = Query(operator='IN',
                                       **{'%s.simulation_state' % (table, ):
                                          input_simulation_state})
        simulation_query = ComplexQuery(input_quantity_query,
                                        input_simulation_query,
                                        operator='AND')
        if output_simulation_state is not None:
          output_quantity_query = Query(**{'%s.quantity' % (table, ): '<0'})
          output_simulation_query = Query(operator='IN',
                                          **{'%s.simulation_state' % (table, ):
                                             output_simulation_state})
          output_query = ComplexQuery(output_quantity_query,
                                      output_simulation_query,
                                      operator='AND')
          simulation_query = ComplexQuery(simulation_query, output_query,
                                          operator='OR')
      else:
        simulation_query = '1'
      return simulation_query

    def _getSimulationStateDict(self, simulation_state=None, omit_transit=0,
                                input_simulation_state=None,
                                output_simulation_state=None,
                                transit_simulation_state=None,
                                strict_simulation_state=None):
      """
      This method is used in order to give what should be
      the input_simulation_state or output_simulation_state
      depending on many parameters
      """
      string_or_list = (str, list, tuple)
      # Simulation States
      # If strict_simulation_state is set, we directly put it into the dictionary
      simulation_dict = {}
      if strict_simulation_state:
        if isinstance(simulation_state, string_or_list)\
                and simulation_state:
           simulation_query = Query(
                   **{'stock.simulation_state': simulation_state})
      else:
        # first, we evaluate simulation_state
        sql_kw = {}
        if simulation_state and isinstance(simulation_state, string_or_list):
          if isinstance(simulation_state, str):
            sql_kw['input_simulation_state'] = [simulation_state]
            sql_kw['output_simulation_state'] = [simulation_state]
          else:
            sql_kw['input_simulation_state'] = simulation_state
            sql_kw['output_simulation_state'] = simulation_state
        # then, if omit_transit == 1, we evaluate (simulation_state -
        # transit_simulation_state) for input_simulation_state
        if omit_transit:
          if isinstance(simulation_state, string_or_list)\
                and simulation_state:
            if isinstance(transit_simulation_state, string_or_list)\
                  and transit_simulation_state:
              # when we know both are usable, we try to calculate
              # (simulation_state - transit_simulation_state)
              if isinstance(simulation_state, str):
                simulation_state = [simulation_state]
              if isinstance(transit_simulation_state, str) :
                transit_simulation_state = [transit_simulation_state]
              delivered_simulation_state_list = []
              for state in simulation_state :
                if state not in transit_simulation_state :
                  delivered_simulation_state_list.append(state)
              sql_kw['input_simulation_state'] = delivered_simulation_state_list

        # alternatively, the user can directly define input_simulation_state
        # and output_simulation_state
        if input_simulation_state and isinstance(input_simulation_state,
                                                  string_or_list):
          if isinstance(input_simulation_state, str):
            input_simulation_state = [input_simulation_state]
          sql_kw['input_simulation_state'] = input_simulation_state
        if output_simulation_state and isinstance(output_simulation_state,
                                                  string_or_list):
          if isinstance(output_simulation_state, str):
            output_simulation_state = [output_simulation_state]
          sql_kw['output_simulation_state'] = output_simulation_state
        # XXX In this case, we must not set sql_kw[input_simumlation_state] before
        input_simulation_state = None
        output_simulation_state = None
        if sql_kw.has_key('input_simulation_state'):
          input_simulation_state = sql_kw.get('input_simulation_state')
        if sql_kw.has_key('output_simulation_state'):
          output_simulation_state = sql_kw.get('output_simulation_state')
        if input_simulation_state is not None \
           or output_simulation_state is not None:
          sql_kw.pop('input_simulation_state',None)
          sql_kw.pop('output_simulation_state',None)
        if input_simulation_state is not None:
          if output_simulation_state is not None:
            if input_simulation_state == output_simulation_state:
              simulation_dict['simulation_state'] = input_simulation_state
            else:
              simulation_dict['input_simulation_state'] = input_simulation_state
              simulation_dict['output_simulation_state'] = output_simulation_state
          else:
            simulation_dict['input_simulation_state'] = input_simulation_state
        elif output_simulation_state is not None:
          simulation_dict['simulation_state'] = output_simulation_state
      return simulation_dict

    def _getOmitQuery(self, query_table=None, omit_input=0, omit_output=0, **kw):
      omit_dict = self._getOmitDict(omit_input=omit_input, omit_output=omit_output)
      return self._buildOmitQuery(query_table=query_table, omit_dict=omit_dict)
      
    def _buildOmitQuery(self, query_table, omit_dict):
      """
      Build a specific query in order to take:
      - negatives quantity values if omit_input
      - postives quantity values if omit_output
      """
      omit_query = None
      omit_input = omit_dict.get('input', False)
      omit_output = omit_dict.get('output', False)
      if omit_input or omit_output:
        # Make sure to check some conditions
        condition_expression = \
          "%(query_table)s.node_uid <> %(query_table)s.mirror_node_uid \
         OR %(query_table)s.section_uid <> %(query_table)s.mirror_section_uid \
         OR %(query_table)s.mirror_node_uid IS NULL \
         OR %(query_table)s.mirror_section_uid IS NULL \
         OR %(query_table)s.payment_uid IS NOT NULL \
           " % {'query_table': query_table}
        if omit_input:
          quantity_query = Query(**{'%s.quantity' % query_table: '<0'})
          omit_query = ComplexQuery(quantity_query, condition_expression,
                                    operator='AND')
        if omit_output:
          quantity_query = Query(**{'%s.quantity' % query_table: '>0'})
          if omit_query is None:
            omit_query = ComplexQuery(quantity_query, condition_expression,
                                      operator='AND')
          else:
            output_query = ComplexQuery(quantity_query, condition_expression,
                                        operator='AND')
            omit_query = ComplexQuery(omit_query, output_query, operator='AND')

      return omit_query

    def _getOmitDict(self, omit_input=False, omit_output=False):
      return {'input': omit_input, 'output': omit_output}

    def _generateSQLKeywordDict(self, table='stock', **kw):
        sql_kw, new_kw = self._generateKeywordDict(**kw)
        return self._generateSQLKeywordDictFromKeywordDict(table=table,
                 sql_kw=sql_kw, new_kw=new_kw)

    def _generateSQLKeywordDictFromKeywordDict(self, table='stock', sql_kw={},
                                               new_kw={}):
        ctool = getToolByName(self, 'portal_catalog')
        sql_kw = sql_kw.copy()
        new_kw = new_kw.copy()
        # Some columns cannot be found automatically, prepend table name to
        # avoid ambiguities.
        # Group-by expression
        group_by = new_kw.pop('group_by', [])
        if len(group_by):
          new_kw['group_by_expression'] = ', '.join(['%s.%s' % (table, x) \
                                                     for x in group_by])
        # Column values
        column_value_dict = new_kw.pop('column_value_dict', {})
        for key, value in column_value_dict.iteritems():
          new_kw['%s.%s' % (table, key)] = value
        # Related keys
        # First, the passthrough (acts as default values)
        for key, value in new_kw.pop('related_key_dict_passthrough', {})\
            .iteritems():
          new_kw[key] = value
        # Second, calculated values
        for key, value in new_kw.pop('related_key_dict', {}).iteritems():
          new_kw['%s_%s' % (table, key)] = value
        # Simulation states matched with input and output omission
        def joinQueriesIfNeeded(query_a, query_b, operator):
          if None not in (query_a, query_b):
            return ComplexQuery(query_a, query_b, operator=operator)
          elif query_a is not None:
            return query_a
          return query_b
        def getSimulationQuery(simulation_dict, omit_dict):
          simulation_query = self._buildSimulationStateQuery(
                               simulation_state_dict=simulation_dict,
                               table=table)
          omit_query = self._buildOmitQuery(query_table=table,
                                            omit_dict=omit_dict)
          return joinQueriesIfNeeded(query_a=simulation_query,
                                     query_b=omit_query, operator='AND')
        regular_query = getSimulationQuery(new_kw.pop('simulation_dict', {}),
                                           new_kw.pop('omit_dict', {}))
        reserved_kw = new_kw.pop('reserved_kw', None)
        if reserved_kw is not None:
          reserved_query = getSimulationQuery(
            reserved_kw.pop('simulation_dict', {}),
            reserved_kw.pop('omit_dict', {}))
          simulation_query = joinQueriesIfNeeded(query_a=regular_query,
                                                 query_b=reserved_query,
                                                 operator='OR')
        else:
          simulation_query = regular_query
        if simulation_query is not None:
          new_kw['query'] = simulation_query

        # Sort on
        if 'sort_on' in new_kw:
          table_column_list = ctool.getSQLCatalog()._getCatalogSchema(
                                                              table=table)
          sort_on = new_kw['sort_on']
          new_sort_on = []
          for column_id, sort_direction in sort_on:
            if column_id in table_column_list:
              column_id = '%s.%s' % (table, column_id)
            new_sort_on.append((column_id, sort_direction))
          new_kw['sort_on'] = tuple(new_sort_on)
        sql_kw.update(ctool.buildSQLQuery(**new_kw))
        return sql_kw

    def _generateKeywordDict(self,
        # dates
        from_date=None, to_date=None, at_date=None,
        omit_mirror_date=1,
        # instances
        resource=None, node=None, payment=None,
        section=None, mirror_section=None, item=None,
        # used for tracking
        input=0, output=0,
        # categories
        resource_category=None, node_category=None, payment_category=None,
        section_category=None, mirror_section_category=None,
        # categories with strict membership
        resource_category_strict_membership=None,
        node_category_strict_membership=None,
        payment_category_strict_membership=None,
        section_category_strict_membership=None,
        mirror_section_category_strict_membership=None,
        # simulation_state
        strict_simulation_state=0,
        simulation_state=None, transit_simulation_state = None, omit_transit=0,
        input_simulation_state=None, output_simulation_state=None,
        reserved_kw=None,
        # variations
        variation_text=None, sub_variation_text=None,
        variation_category=None,
        # uids
        resource_uid=None, node_uid=None, section_uid=None, payment_uid=None,
        mirror_node_uid=None, mirror_section_uid=None,
        # omit input and output
        omit_input=0,
        omit_output=0,
        # group by
        group_by_node=0,
        group_by_mirror_node=0,
        group_by_section=0,
        group_by_mirror_section=0,
        group_by_payment=0,
        group_by_sub_variation=0,
        group_by_variation=0,
        group_by_movement=0,
        group_by_resource=0,
        group_by_date=0,
        # sort_on
        sort_on=None,
        # keywords for related keys
        **kw):
      """
      Generates keywords and calls buildSQLQuery

      - omit_mirror_date: normally, date's parameters are only based on date
        column. If 0, it also used the mirror_date column.
      """
      new_kw = {}
      sql_kw = {}

      # input and output are used by getTrackingList
      sql_kw['input'] = input
      sql_kw['output'] = output
      # Add sort_on parameter if defined
      if sort_on is not None:
        new_kw['sort_on'] = sort_on

      class DictMixIn(dict):
        def set(dictionary, key, value):
          result = not(not(value))
          if result:
            dictionary[key] = value
          return result

        def setUIDList(dictionary, key, value, as_text=0):
          uid_list = self._generatePropertyUidList(value, as_text=as_text)
          return dictionary.set(key, uid_list)

      column_value_dict = DictMixIn()

      if omit_mirror_date:
        date_dict = {'query':[], 'operator':'and'}
        if from_date :
          date_dict['query'].append(from_date)
          date_dict['range'] = 'min'
          if to_date :
            date_dict['query'].append(to_date)
            date_dict['range'] = 'minmax'
          elif at_date :
            date_dict['query'].append(at_date)
            date_dict['range'] = 'minngt'
        elif to_date :
          date_dict['query'].append(to_date)
          date_dict['range'] = 'max'
        elif at_date :
          date_dict['query'].append(at_date)
          date_dict['range'] = 'ngt'
        column_value_dict['date'] = date_dict
      else:
        column_value_dict['date'] = {'query': [to_date], 'range': 'ngt'}
        column_value_dict['mirror_date'] = {'query': [from_date], 'range': 'nlt'}

      column_value_dict.set('resource_uid', resource_uid)
      column_value_dict.set('payment_uid', payment_uid)
      if column_value_dict.set('section_uid', section_uid):
        sql_kw['section_filtered'] = 1
      column_value_dict.set('node_uid', node_uid)
      column_value_dict.set('mirror_node_uid', mirror_node_uid)
      column_value_dict.set('mirror_section_uid', mirror_section_uid)
      column_value_dict.setUIDList('resource_uid', resource)
      column_value_dict.setUIDList('aggregate_uid', item)
      column_value_dict.setUIDList('node_uid', node)
      column_value_dict.setUIDList('payment_uid', payment)
      if column_value_dict.setUIDList('section_uid', section):
        sql_kw['section_filtered'] = 1
      column_value_dict.setUIDList('mirror_section_uid', mirror_section)
      column_value_dict.setUIDList('variation_text', variation_text,
                                   as_text=1)
      column_value_dict.setUIDList('sub_variation_text', sub_variation_text,
                                   as_text=1)
      new_kw['column_value_dict'] = column_value_dict.copy()

      related_key_dict = DictMixIn()
      # category membership
      related_key_dict.setUIDList('resource_category_uid', resource_category)
      related_key_dict.setUIDList('node_category_uid', node_category)
      related_key_dict.setUIDList('payment_category_uid', payment_category)
      if related_key_dict.setUIDList('section_category_uid',
                                     section_category):
        sql_kw['section_filtered'] = 1
      related_key_dict.setUIDList('mirror_section_category_uid',
                                  mirror_section_category)
      # category strict membership
      related_key_dict.setUIDList('resource_category_strict_membership_uid',
                                  resource_category_strict_membership)
      related_key_dict.setUIDList('node_category_strict_membership_uid',
                                  node_category_strict_membership)
      related_key_dict.setUIDList('payment_category_strict_membership_uid',
                                  payment_category_strict_membership)
      if related_key_dict.setUIDList('section_category_strict_membership_uid',
                                     section_category_strict_membership):
        sql_kw['section_filtered'] = 1
      related_key_dict.setUIDList(
        'mirror_section_category_strict_membership_uid',
        mirror_section_category_strict_membership)
      
      new_kw['related_key_dict'] = related_key_dict.copy()
      new_kw['related_key_dict_passthrough'] = kw

      #variation_category_uid_list = self._generatePropertyUidList(variation_category)
      #if len(variation_category_uid_list) :
      #  new_kw['variationCategory'] = variation_category_uid_list
      
      simulation_dict =  self._getSimulationStateDict(
                                simulation_state=simulation_state, 
                                omit_transit=omit_transit,
                                input_simulation_state=input_simulation_state,
                                output_simulation_state=output_simulation_state,
                                transit_simulation_state=transit_simulation_state,
                                strict_simulation_state=strict_simulation_state)
      new_kw['simulation_dict'] = simulation_dict
      omit_dict = self._getOmitDict(omit_input=omit_input,
                                    omit_output=omit_output)
      new_kw['omit_dict'] = omit_dict
      if reserved_kw is not None:
        if not isinstance(reserved_kw, dict):
          # Not a dict when taken from URL, so, cast is needed 
          # to make pop method available
          reserved_kw = dict(reserved_kw)
        new_reserved_kw = {}
        reserved_omit_input = reserved_kw.pop('omit_input',0)
        reserved_omit_output = reserved_kw.pop('omit_output',0)
        new_reserved_kw['omit_dict'] = self._getOmitDict(
                                         omit_input=reserved_omit_input,
                                         omit_output=reserved_omit_output)
        new_reserved_kw['simulation_dict'] = self._getSimulationStateDict(**reserved_kw)
        new_kw['reserved_kw'] = new_reserved_kw
      # It is necessary to use here another SQL query (or at least a subquery)
      # to get _DISTINCT_ uid from predicate_category table.
      # Otherwise, by using a where_expression, cells which fit conditions
      # more than one time are counted more than one time, and the resulting
      # inventory is false
      # XXX Perhaps is there a better solution
      add_kw = {}
      if variation_category is not None and variation_category:
        where_expression = self.getPortalObject().portal_categories\
          .buildSQLSelector(
            category_list = variation_category,
            query_table = 'predicate_category')
        if where_expression != '':
          add_kw['where_expression'] = where_expression
          add_kw['predicate_category.uid'] = '!=NULL'
          add_kw['group_by_expression'] = 'uid'
          add_query = self.portal_catalog(**add_kw)
          uid_list = []
          for line in add_query:
            uid_list.append(line.uid)
          new_kw['where_expression'] = '( %s )' % ' OR '.join(
                      ['catalog.uid=%s' % uid for uid in uid_list])

      # build the group by expression
      group_by_expression_list = []
      if group_by_node:
        group_by_expression_list.append('node_uid')
      if group_by_mirror_node:
        group_by_expression_list.append('mirror_node_uid')
      if group_by_section:
        group_by_expression_list.append('section_uid')
      if group_by_mirror_section:
        group_by_expression_list.append('mirror_section_uid')
      if group_by_payment:
        group_by_expression_list.append('payment_uid')
      if group_by_sub_variation:
        group_by_expression_list.append('sub_variation_text')
      if group_by_variation:
        group_by_expression_list.append('variation_text')
      if group_by_movement:
        group_by_expression_list.append('uid')
      if group_by_resource:
        group_by_expression_list.append('resource_uid')
      if group_by_date:
        group_by_expression_list.append('date')
      if group_by_expression_list:
        new_kw['group_by'] = group_by_expression_list
      return sql_kw, new_kw 

    #######################################################
    # Inventory management
    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventory')
    def getInventory(self, src__=0, simulation_period='', **kw):
      """
      Returns an inventory of a single or multiple resources on a single or
      multiple nodes as a single float value

      from_date (>=) - only take rows which date is >= from_date

      to_date   (<)  - only take rows which date is < to_date

      at_date   (<=) - only take rows which date is <= at_date

      resource (only in generic API in simulation)

      node           -  only take rows in stock table which node_uid is
                        equivalent to node

      payment        -  only take rows in stock table which payment_uid is
                        equivalent to payment

      section        -  only take rows in stock table which section_uid is
                        equivalent to section

      mirror_section -  only take rows in stock table which mirror_section_uid is
                        mirror_section

      resource_category  -  only take rows in stock table which
                        resource_uid is member of resource_category

      node_category   - only take rows in stock table which node_uid is
                        member of section_category

      payment_category   -  only take rows in stock table which payment_uid
                            is in section_category

      section_category -  only take rows in stock table which section_uid is
                          member of section_category

      mirror_section_category - only take rows in stock table which 
                                mirror_section_uid is member of
				mirror_section_category

      node_filter     - only take rows in stock table which node_uid
                        matches node_filter

      payment_filter  - only take rows in stock table which payment_uid
                        matches payment_filter

      section_filter  - only take rows in stock table which section_uid
                        matches section_filter

      mirror_section_filter - only take rows in stock table which
                              mirror_section_uid matches mirror_section_filter

      variation_text -  only take rows in stock table with specified
                        variation_text.
                        This needs to be extended with some kind of
                        variation_category ?
                        XXX this way of implementing variation selection is far
                        from perfect

      sub_variation_text - only take rows in stock table with specified
                        variation_text

      variation_category - variation or list of possible variations (it is not
                        a cross-search ; SQL query uses OR)

      simulation_state - only take rows with specified simulation_state

      transit_simulation_state - specifies which states are transit states

      omit_transit   -  do not evaluate transit_simulation_state

      input_simulation_state - only take rows with specified simulation_state
                        and quantity > 0

      output_simulation_state - only take rows with specified simulation_state
                        and quantity < 0

      ignore_variation -  do not take into account variation in inventory
                        calculation (useless on getInventory, but useful on
                        getInventoryList)

      standardise    -  provide a standard quantity rather than an SKU (XXX
                        not implemented yet)

      omit_simulation - doesn't take into account simulation movements

      omit_input     -  doesn't take into account movement with quantity < 0

      omit_output    -  doesn't take into account movement with quantity > 0

      selection_domain, selection_report - see ListBox

      group_by_variation - (useless on getInventory, but useful on
                        getInventoryList)

      group_by_node  -  (useless on getInventory, but useful on
                        getInventoryList)

      group_by_mirror_node - (useless on getInventory, but useful on
                        getInventoryList)

      group_by_sub_variation - (useless on getInventory, but useful on
                        getInventoryList)

      group_by_movement - (useless on getInventory, but useful on
                        getInventoryList)

      precision - the precision used to round quantities and prices.

      **kw           -  if we want extended selection with more keywords (but
                        bad performance) check what we can do with
                        buildSQLQuery

      NOTE: we may want to define a parameter so that we can select the kind of
      inventory statistics we want to display (ex. sum, average, cost, etc.)
      """
      # JPS: this is a hint for implementation of xxx_filter arguments
      # node_uid_count = portal_catalog.countResults(**node_filter)
      # if node_uid_count not too big:
      #   node_uid_list = cache(portal_catalog(**node_filter))
      #   pass this list to ZSQL method
      # else:
      #   build a table in MySQL
      #   and join that table with the stock table
      method = getattr(self,'get%sInventoryList' % simulation_period)
      result = method(inventory_list=0, ignore_group_by=1, src__=src__, **kw)
      if src__:
        return result

      total_result = 0.0
      if len(result) > 0:
        if len(result) != 1:
          raise ValueError, 'Sorry we must have only one'
        inventory = result[0].total_quantity
        if inventory is not None:
          total_result = inventory

      return total_result

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getCurrentInventory')
    def getCurrentInventory(self, **kw):
      """
      Returns current inventory
      """
      return self.getInventory(simulation_period='Current', **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getAvailableInventory')
    def getAvailableInventory(self, **kw):
      """
      Returns available inventory
      (current inventory - reserved_inventory)
      """
      return self.getInventory(simulation_period='Available', **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getFutureInventory')
    def getFutureInventory(self, **kw):
      """
      Returns future inventory
      """
      return self.getInventory(simulation_period='Future', **kw)
    
    def _getDefaultGroupByParameters(self, ignore_group_by=0, 
        group_by_node=0, group_by_mirror_node=0,
        group_by_section=0, group_by_mirror_section=0,
        group_by_payment=0,
        group_by_variation=0, group_by_sub_variation=0,
        group_by_movement=0, group_by_date=0,
        group_by_resource=None,
        movement_list_mode=0,
        **ignored):
      """
      Set defaults group_by parameters

      If ignore_group_by is true, this function returns an empty dict.

      If any group-by is provided, automatically group by resource aswell
      unless group_by_resource is explicitely set to false.
      If no group by is provided, use the default group by: movement, node and
      resource, unless movement_list_mode is true, in that case, group by
      movement, node, resource and date (this is historically the default in
      getMovementHistoryList), section, mirror_section and payment (this is to
      make sure two lines will appear when we are, for instance both source and
      destination, implementation might not be optimal, because it uses lots of
      group by statements in SQL).
      """
      new_group_by_dict = {}
      if not ignore_group_by:
        if group_by_node or group_by_mirror_node or group_by_section or \
           group_by_mirror_section or group_by_payment or \
           group_by_sub_variation or group_by_variation or \
           group_by_movement or group_by_date:
          if group_by_resource is None:
            group_by_resource = 1
          new_group_by_dict['group_by_resource'] = group_by_resource
        elif group_by_resource is None:
          new_group_by_dict['group_by_movement'] = 1
          new_group_by_dict['group_by_node'] = 1
          new_group_by_dict['group_by_resource'] = 1
          if movement_list_mode:
            new_group_by_dict['group_by_date'] = 1
            new_group_by_dict['group_by_mirror_node'] = 1
            new_group_by_dict['group_by_section'] = 1
            new_group_by_dict['group_by_mirror_section'] = 1
            new_group_by_dict['group_by_payment'] = 1
      return new_group_by_dict

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventoryList')
    def getInventoryList(self, src__=0, optimisation__=True,
                         ignore_variation=0, standardise=0,
                         omit_simulation=0, 
                         default_stock_table='stock',
                         selection_domain=None, selection_report=None,
                         statistic=0, inventory_list=1, 
                         precision=None, connection_id=None,
                         quantity_unit=None, **kw):
      """
        Returns a list of inventories for a single or multiple
        resources on a single or multiple nodes, grouped by resource,
        node, section, etc. Every line defines an inventory value for
        a given group of resource, node, section.
        NOTE: we may want to define a parameter so that we can select
        the kind of inventory statistics we want to display (ex. sum,
        average, cost, etc.)

        Optimisation queries.
        Optimisation of a stock lookup is done to avoid a table scan
        of all lines corresponding to a given node, section or payment,
        because they grow with time and query time should not.
        First query: Fetch fitting full inventory dates.
          For each node, section or payment, find the first anterior full
          inventory.
        Second query: Fetch full inventory amounts.
          Fetch values of inventory identified in the first query.
        Tird query: Classic stock table read.
          Fetch all rows in stock table which are posterior to the inventory.
        Final result
          Add results of the second and third queries, and return it.

        Missing optimisations:
         - In a getInventory case where everything is specified for the
           resource, it's not required for the inventory to be full, it
           just need to be done for the right resource.
           If the resource isn't completely defined, we require inventory
           to be full, which is implemented.
         - Querying multiple nodes/categories/payments in one call prevents
           from using optimisation, it should be equivalent to multiple calls
           on individual nodes/categories/payments.
         - 
      """
      if src__:
        sql_source_list = []
      # If no group at all, give a default sort group by
      kw.update(self._getDefaultGroupByParameters(**kw))
      sql_kw, new_kw = self._generateKeywordDict(**kw)
      stock_sql_kw = self._generateSQLKeywordDictFromKeywordDict(
                       table=default_stock_table, sql_kw=sql_kw, new_kw=new_kw)
      Resource_zGetFullInventoryDate = \
        getattr(self, 'Resource_zGetFullInventoryDate', None)
      EQUAL_DATE_TABLE_ID = 'inventory_stock'
      GREATER_THAN_DATE_TABLE_ID = 'stock'
      optimisation_success = optimisation__ and ('from_date' not in kw) and \
                             Resource_zGetFullInventoryDate is not None and \
                             (GREATER_THAN_DATE_TABLE_ID == default_stock_table)
      # Generate first query parameter dict
      if optimisation_success:
        def getFirstQueryParameterDict(query_generator_kw):
          optimisation_success = True
          AVAILABLE_CRITERIONS_IN_INVENTORY_TABLE = ['node_uid',
                                                     'section_uid',
                                                     'payment_uid']
          group_by_list = query_generator_kw.get('group_by', [])
          column_value_dict = query_generator_kw.get('column_value_dict', {})
          new_group_by_list = []
          new_column_value_dict = {}
          for criterion_id in AVAILABLE_CRITERIONS_IN_INVENTORY_TABLE:
            criterion_value_list = column_value_dict.get(criterion_id, [])
            if not isinstance(criterion_value_list, (list, tuple)):
              criterion_value_list = [criterion_value_list]
            if len(criterion_value_list) > 0:
              if len(criterion_value_list) > 1:
                # Impossible to optimise if there is more than one possible
                # value per criterion.
                optimisation_success = False
                break
              new_column_value_dict[criterion_id] = criterion_value_list
              new_group_by_list.append(criterion_id)
            elif criterion_id in group_by_list:
              new_group_by_list.append(criterion_id)
          group_by_expression = ', '.join(new_group_by_list)
          column_id_list = new_column_value_dict.keys()
          column_value_list_list = new_column_value_dict.values()
          date_value_list = column_value_dict.get('date', {}).get('query', [])
          where_expression = None
          if len(date_value_list) == 1:
            date = date_value_list[0]
            if isinstance(date, DateTime):
              date = date.toZone('UTC').ISO()
              
            # build a query for date to take range into account
            date_query_kw = {"inventory.date" : date,
                             "operator" : column_value_dict.get('date', {}).get('operator', []),
                             "range"  : column_value_dict.get('date', {}).get('range', []),
                             }
            date_query = Query(**date_query_kw)
            date_query_result = date_query()
            if date_query_result['where_expression'] not in ('',None):
              where_expression = date_query_result['where_expression']
          elif len(date_value_list) > 1:
            # When more than one date is provided, we must not optimise.
            # Also, as we should never end up here (the only currently known
            # case where there are 2 dates is when a from_date is provided
            # along with either an at_date or a to_date, and we disable
            # optimisation when from_date is given), emit a log.
            # This can happen if there are more date parameters than mentioned
            # above.
            LOG('SimulationTool', 100, 'There is more than one date condition'
              ' so optimisation got disabled. The result of this call will be'
              ' correct but it requires investigation as some cases might'
              ' have gone unnoticed and produced wrong results.')
            optimisation_success = False
          return {'group_by_expression': group_by_expression,
                  'column_id_list': column_id_list,
                  'column_value_list_list': column_value_list_list,
                  'where_expression' : where_expression,}, optimisation_success
        first_query_param_dict, optimisation_success = getFirstQueryParameterDict(new_kw)
        if optimisation_success:
          if len(first_query_param_dict['column_id_list']):
            inventory_date_line_list = self.Resource_zGetFullInventoryDate(
                                         **first_query_param_dict)
            if src__:
              sql_source_list.append(
                self.Resource_zGetFullInventoryDate(src__=src__,
                  **first_query_param_dict))
            # Check that all expected uids have been found, otherwise a full
            # inventory of a node/section/payment might be missing.
            if len(inventory_date_line_list) >= max([len(x) for x in \
               first_query_param_dict['column_value_list_list']]):
              # Generate a where expression which filters on dates retrieved
              # in the first query to be used in the second query.
              # Also, generate a where expression to use in the third query,
              # since it is based on the same data.
              # XXX: uggly duplicated query generation code
              # XXX: duplicates SQL variable formatting present in
              #      ERP5Type/patches/sqlvar.py about datetime SQL columns.
              # Note: This code can generate queries like:
              #  date = 2000/01/01 and date >= 2001/01/01
              #  When latest full inventory is at 2000/01/01 and given
              #  from_date is 2001/01/01.
              #  It is not a serious problem since MySQL detects incompatible
              #  conditions and immediately returns (with 0 rows).
              
              # get search key definitions from portal_catalog
              ctool = getToolByName(self, 'portal_catalog')
              portal_catalog = ctool.getSQLCatalog()
              keyword_search_keys = list(portal_catalog.sql_catalog_keyword_search_keys)    
              datetime_search_keys = list(portal_catalog.sql_catalog_datetime_search_keys)
              full_text_search_keys = list(portal_catalog.sql_catalog_full_text_search_keys)
              search_key_mapping = dict(key_alias_dict = None,
                                        keyword_search_keys = keyword_search_keys,
                                        datetime_search_keys = datetime_search_keys,
                                        full_text_search_keys = full_text_search_keys)
              equal_date_query_list = []
              greater_than_date_query_list = []
              for inventory_date_line_dict in \
                  inventory_date_line_list.dictionaries():
                date = inventory_date_line_dict['date']
                non_date_value_dict = dict([(k, v) for k, v \
                  in inventory_date_line_dict.iteritems() if k != 'date'])
                equal_date_query_list.append(
                  ComplexQuery(
                    ComplexQuery(operator='AND',
                      *[Query(**{'%s.%s' % (EQUAL_DATE_TABLE_ID, k): v}) \
                        for k, v in non_date_value_dict.iteritems()]),
                    Query(**{'%s.date' % (EQUAL_DATE_TABLE_ID, ): date}),
                    operator='AND'))
                greater_than_date_query_list.append(
                  ComplexQuery(
                    ComplexQuery(operator='AND',
                      *[Query(**{'%s.%s' % (GREATER_THAN_DATE_TABLE_ID, k): \
                                 v}) \
                        for k, v in non_date_value_dict.iteritems()]),
                    # 'Use explicitly Universal' otherwise DateTime 
                    # search key will convert it to UTC one more time
                    Query(**{'%s.date' % (GREATER_THAN_DATE_TABLE_ID, ): \
                             '>%s Universal' % (date.ISO(), )}), 
                    operator='AND'))
              assert len(equal_date_query_list) == \
                     len(greater_than_date_query_list)
              assert len(equal_date_query_list) > 0
              equal_date_query = ComplexQuery(operator='OR',
                *equal_date_query_list).asSQLExpression(**search_key_mapping)\
                ['where_expression']
              greater_than_date_query = ComplexQuery(operator='OR',
                *greater_than_date_query_list).asSQLExpression(**search_key_mapping)\
                ['where_expression']
              inventory_stock_sql_kw = \
                self._generateSQLKeywordDictFromKeywordDict(
                  table=EQUAL_DATE_TABLE_ID, sql_kw=sql_kw, new_kw=new_kw)
              inventory_stock_where_query = \
                inventory_stock_sql_kw.get('where_expression', '(1)')
              assert isinstance(inventory_stock_where_query, basestring) \
                     and len(inventory_stock_where_query)
              inventory_stock_sql_kw['where_expression'] = '(%s) AND (%s)' % \
                (inventory_stock_where_query, equal_date_query)
              where_query = stock_sql_kw.get('where_expression', '(1)')
              assert isinstance(where_query, basestring) and len(where_query)
              stock_sql_kw['where_expression'] = '(%s) AND (%s)' % \
                (where_query, greater_than_date_query)
              # Get initial inventory amount
              initial_inventory_line_list = self.Resource_zGetInventoryList(
                stock_table_id=EQUAL_DATE_TABLE_ID,
                src__=src__, ignore_variation=ignore_variation,
                standardise=standardise, omit_simulation=omit_simulation,
                selection_domain=selection_domain,
                selection_report=selection_report, precision=precision,
                inventory_list=inventory_list,
                statistic=statistic, quantity_unit=quantity_unit,
                **inventory_stock_sql_kw)
              # Get delta inventory
              delta_inventory_line_list = self.Resource_zGetInventoryList(
                stock_table_id=GREATER_THAN_DATE_TABLE_ID,
                src__=src__, ignore_variation=ignore_variation,
                standardise=standardise, omit_simulation=omit_simulation,
                selection_domain=selection_domain,
                selection_report=selection_report, precision=precision,
                inventory_list=inventory_list,
                statistic=statistic, quantity_unit=quantity_unit,
                **stock_sql_kw)
              # Match & add initial and delta inventories
              if src__:
                sql_source_list.extend((initial_inventory_line_list,
                                        delta_inventory_line_list))
              else:
                if 'group_by' in new_kw:
                  group_by_id_list = []
                  group_by_id_list_append = group_by_id_list.append
                  for group_by_id in new_kw['group_by']:
                    if group_by_id == 'uid':
                      group_by_id_list_append('stock_uid')
                    else:
                      group_by_id_list_append(group_by_id)
                  def getInventoryListKey(line):
                    """
                      Generate a key based on values used in SQL group_by
                    """
                    return tuple([line[x] for x in group_by_id_list])
                else:
                  def getInventoryListKey(line):
                    """
                      No group by criterion, regroup everything.
                    """
                    return 'dummy_key'
                result_column_id_dict = {}
                result_column_id_dict['inventory'] = None
                result_column_id_dict['total_quantity'] = None
                result_column_id_dict['total_price'] = None
                if quantity_unit:
                    result_column_id_dict['converted_quantity'] = None
                def addLineValues(line_a=None, line_b=None):
                  """
                    Addition columns of 2 lines and return a line with same
                    schema. If one of the parameters is None, returns the
                    other parameters.

                    Arythmetic modifications on additions:
                      None + x = x
                      None + None = None
                  """
                  if line_a is None:
                    return line_b
                  if line_b is None:
                    return line_a
                  result = {}
                  for key, value in line_a.iteritems():
                    if key in result_column_id_dict:
                      value_b = line_b[key]
                      if None not in (value, value_b):
                        result[key] = value + value_b
                      elif value is not None:
                        result[key] = value
                      else:
                        result[key] = value_b
                    elif line_a[key] == line_b[key]:
                      result[key] = line_a[key]
                    elif key not in ('date', 'stock_uid', 'path'):
                      LOG('InventoryTool.getInventoryList.addLineValues', 0,
                          'missmatch for %s column: %s and %s' % \
                          (key, line_a[key], line_b[key]))
                  return result
                inventory_list_dict = {}
                for line_list in (initial_inventory_line_list,
                                  delta_inventory_line_list):
                  for line in line_list.dictionaries():
                    line_key = getInventoryListKey(line)
                    line_a = inventory_list_dict.get(line_key)
                    inventory_list_dict[line_key] = addLineValues(line_a,
                                                                  line)
                ## XXX: Returns a dict instead of an <r> instance
                ## As long as they are accessed like dicts it's ok, but...
                #result = inventory_list_dict.values()
                sorted_inventory_list = inventory_list_dict.values()
                sort_on = new_kw.get('sort_on', tuple())
                if len(sort_on) != 0:
                  def cmp_inventory_line(line_a, line_b):
                    """
                      Compare 2 inventory lines and sort them according to
                      sort_on parameter.
                    """
                    result = 0
                    for key, sort_direction in sort_on:
                      if not(key in line_a and key in line_b):
                        raise Exception, "Impossible to sort result since " \
                          "columns sort happens on are not available in " \
                          "result."
                      result = cmp(line_a[key], line_b[key])
                      if result != 0:
                        if len(sort_direction[0]) and \
                           sort_direction[0].upper() != 'A':
                          # Default sort is ascending, if a sort is given and
                          # it does not start with an 'A' then reverse sort.
                          # Tedious syntax checking is MySQL's job, and
                          # happened when queries were executed.
                          result *= -1
                        break
                    return result
                  sorted_inventory_list.sort(cmp_inventory_line)
                result = Results((delta_inventory_line_list.\
                                    _searchable_result_columns(),
                                 tuple(sorted_inventory_list)))
            else:
              # Not all required full inventories are found
              optimisation_success = False
          else:
            # Not enough criterions to trigger optimisation
            optimisation_success = False
      if not optimisation_success:
        result = self.Resource_zGetInventoryList(
                    stock_table_id=default_stock_table,
                    src__=src__, ignore_variation=ignore_variation,
                    standardise=standardise, omit_simulation=omit_simulation,
                    selection_domain=selection_domain,
                    selection_report=selection_report, precision=precision,
                    inventory_list=inventory_list, connection_id=connection_id,
                    statistic=statistic, quantity_unit=quantity_unit,
                    **stock_sql_kw)
        if src__:
          sql_source_list.append(result)
      if src__:
        result = ';\n-- NEXT QUERY\n'.join(sql_source_list)
      return result

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getConvertedInventoryList')
    def getConvertedInventoryList(self, metric_type, quantity_unit=1,
                                  simulation_period='', **kw):
      """
      Return list of inventory with a 'converted_quantity' additional column,
      which contains the sum of measurements for the specified metric type,
      expressed in the 'quantity_unit' unit.

      metric_type   - category relative url
      quantity_unit - int, float or category relative url
      """
      getCategory = self.getPortalObject().portal_categories.getCategoryValue

      kw['metric_type_uid'] = Query(
        metric_type_uid=getCategory(metric_type, 'metric_type').getUid(),
        table_alias_list=(("measure", "measure"),))

      if isinstance(quantity_unit, str):
        quantity_unit = float(getCategory(quantity_unit, 'quantity_unit')
                              .getProperty('quantity'))

      method = getattr(self,'get%sInventoryList' % simulation_period)
      return method(quantity_unit=quantity_unit, **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getAllInventoryList')
    def getAllInventoryList(self, src__=0, **kw):
      """
      Returns list of inventory, for all periods.
      Performs 1 SQL request for each simulation state, and merge the results.
      Rename relevant columns with a '${simulation}_' prefix
      (ex: 'total_price' -> 'current_total_price').
      """
      columns = ('total_quantity', 'total_price', 'converted_quantity')

      # Guess the columns to use to identify each row, by looking at the GROUP
      # clause. Note that the call to 'mergeZRDBResults' will crash if the GROUP
      # clause contains a column not requested in the SELECT clause.
      kw.update(self._getDefaultGroupByParameters(**kw), ignore_group_by=1)
      group_by_list = self._generateKeywordDict(**kw)[1].get('group_by', ())

      results = []
      edit_result = {}
      get_false_value = lambda row, column_name: row.get(column_name) or 0

      for simulation in 'current', 'available', 'future':
        method = getattr(self, 'get%sInventoryList' % simulation.capitalize())
        rename = {'inventory': None} # inventory column is deprecated
        for column in columns:
          rename[column] = new_name = '%s_%s' % (simulation, column)
          edit_result[new_name] = get_false_value
        results += (method(src__=src__, **kw), rename),

      if src__:
        return ';\n-- NEXT QUERY\n'.join(r[0] for r in results)
      return mergeZRDBResults(results, group_by_list, edit_result)


    security.declareProtected(Permissions.AccessContentsInformation,
                              'getCurrentInventoryList')
    def getCurrentInventoryList(self, omit_transit=1, 
                                transit_simulation_state=None, **kw):
      """
        Returns list of current inventory grouped by section or site
      """
      kw['simulation_state'] = self.getPortalCurrentInventoryStateList() + \
                               self.getPortalTransitInventoryStateList()
      if transit_simulation_state is None:
        transit_simulation_state = self.getPortalTransitInventoryStateList()

      return self.getInventoryList(
                            omit_transit=omit_transit,
                            transit_simulation_state=transit_simulation_state,
                            **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getAvailableInventoryList')
    def getAvailableInventoryList(self, omit_transit=1, transit_simulation_state=None, **kw):
      """
        Returns list of current inventory grouped by section or site
      """
      if transit_simulation_state is None:
        transit_simulation_state = self.getPortalTransitInventoryStateList()
      kw['simulation_state'] = self.getPortalCurrentInventoryStateList() + \
                               self.getPortalTransitInventoryStateList()
      reserved_kw = {'simulation_state': self.getPortalReservedInventoryStateList(),
                     'transit_simulation_state': transit_simulation_state,
                     'omit_input': 1}
      return self.getInventoryList(reserved_kw=reserved_kw, omit_transit=omit_transit,
                     transit_simulation_state=transit_simulation_state, **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getFutureInventoryList')
    def getFutureInventoryList(self, **kw):
      """
        Returns list of future inventory grouped by section or site
      """
      kw['simulation_state'] = tuple(
                 list(self.getPortalFutureInventoryStateList()) + \
                 list(self.getPortalTransitInventoryStateList()) + \
                 list(self.getPortalReservedInventoryStateList()) + \
                 list(self.getPortalCurrentInventoryStateList()))
      return self.getInventoryList(**kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventoryStat')
    def getInventoryStat(self, simulation_period='', **kw):
      """
      getInventoryStat is the pending to getInventoryList in order to
      provide statistics on getInventoryList lines in ListBox such as:
      total of inventories, number of variations, number of different
      nodes, etc.
      """
      kw['group_by_variation'] = 0
      method = getattr(self,'get%sInventoryList' % simulation_period)
      return method(statistic=1, inventory_list=0, 
                                   ignore_group_by=1, **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getCurrentInventoryStat')
    def getCurrentInventoryStat(self, **kw):
      """
      Returns statistics of current inventory grouped by section or site
      """
      return self.getInventoryStat(simulation_period='Current', **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getAvailableInventoryStat')
    def getAvailableInventoryStat(self, **kw):
      """
      Returns statistics of available inventory grouped by section or site
      """
      return self.getInventoryStat(simulation_period='Available', **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getFutureInventoryStat')
    def getFutureInventoryStat(self, **kw):
      """
      Returns statistics of future inventory grouped by section or site
      """
      return self.getInventoryStat(simulation_period='Future', **kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventoryChart')
    def getInventoryChart(self, src__=0, **kw):
      """
      Returns a list of couples derived from getInventoryList in order
      to feed a chart renderer. Each couple consist of a label
      (node, section, payment, combination of node & section, etc.)
      and an inventory value.

      Mostly useful for charts in ERP5 forms.
      """
      result = self.getInventoryList(src__=src__, **kw)
      if src__ :
        return result

      return map(lambda r: (r.node_title, r.total_quantity), result)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getCurrentInventoryChart')
    def getCurrentInventoryChart(self, **kw):
      """
      Returns list of current inventory grouped by section or site
      """
      kw['simulation_state'] = self.getPortalCurrentInventoryStateList()
      return self.getInventoryChart(**kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getFutureInventoryChart')
    def getFutureInventoryChart(self, **kw):
      """
      Returns list of future inventory grouped by section or site
      """
      kw['simulation_state'] = tuple(
                      list(self.getPortalFutureInventoryStateList()) + \
                      list(self.getPortalReservedInventoryStateList()) + \
                      list(self.getPortalCurrentInventoryStateList()))
      return self.getInventoryChart(**kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventoryAssetPrice')
    def getInventoryAssetPrice(self, src__=0, 
                               simulation_period='', **kw):
      """
      Same thing as getInventory but returns an asset
      price rather than an inventory.
      """
      method = getattr(self,'get%sInventoryList' % simulation_period)
      result = method( src__=src__, inventory_list=0, ignore_group_by=1, **kw)
      if src__ :
        return result

      total_result = 0.0
      if len(result) > 0:
        for result_line in result:
          if result_line.total_price is not None:
            total_result += result_line.total_price

      return total_result

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getCurrentInventoryAssetPrice')
    def getCurrentInventoryAssetPrice(self, **kw):
      """
      Returns list of current inventory grouped by section or site
      """
      kw['simulation_state'] = self.getPortalCurrentInventoryStateList()
      return self.getInventoryAssetPrice(simulation_period='Current',**kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getAvailableInventoryAssetPrice')
    def getAvailableInventoryAssetPrice(self, **kw):
      """
      Returns list of available inventory grouped by section or site
      (current inventory - deliverable)
      """
      kw['simulation_state'] = tuple(
                    list(self.getPortalReservedInventoryStateList()) + \
                    list(self.getPortalCurrentInventoryStateList()))
      return self.getInventoryAssetPrice(**kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getFutureInventoryAssetPrice')
    def getFutureInventoryAssetPrice(self, **kw):
      """
      Returns list of future inventory grouped by section or site
      """
      kw['simulation_state'] = tuple(
               list(self.getPortalFutureInventoryStateList()) + \
               list(self.getPortalReservedInventoryStateList()) + \
               list(self.getPortalCurrentInventoryStateList()))
      return self.getInventoryAssetPrice(**kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventoryHistoryList')
    def getInventoryHistoryList(self, src__=0, ignore_variation=0,
                                standardise=0, omit_simulation=0, omit_input=0,
                                omit_output=0, selection_domain=None,
                                selection_report=None, precision=None, **kw):
      """
      Returns a time based serie of inventory values
      for a single or a group of resource, node, section, etc. This is useful
      to list the evolution with time of inventory values (quantity, asset price).

      TODO:
        - make sure getInventoryHistoryList can return
	  cumulative values calculated by SQL (JPS)
      """
      sql_kw = self._generateSQLKeywordDict(**kw)
      return self.Resource_getInventoryHistoryList(
                      src__=src__, ignore_variation=ignore_variation,
                      standardise=standardise, omit_simulation=omit_simulation,
                      omit_input=omit_input, omit_output=omit_output,
                      selection_domain=selection_domain,
                      selection_report=selection_report, precision=precision,
                      **sql_kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getInventoryHistoryChart')
    def getInventoryHistoryChart(self, src__=0, ignore_variation=0,
                                 standardise=0, omit_simulation=0,
                                 omit_input=0, omit_output=0,
                                 selection_domain=None,
                                 selection_report=None, precision=None, **kw):
      """
      getInventoryHistoryChart is the pensing to getInventoryHistoryList
      to ease the rendering of time based graphs which show the evolution
      of one ore more inventories. Each item in the serie consists of
      time, value and "colour" (multiple graphs can be drawn for example
      for each variation of a resource)
      """
      sql_kw = self._generateSQLKeywordDict(**kw)

      return self.Resource_getInventoryHistoryChart(
                    src__=src__, ignore_variation=ignore_variation,
                    standardise=standardise, omit_simulation=omit_simulation,
                    omit_input=omit_input, omit_output=omit_output,
                    selection_domain=selection_domain,
                    selection_report=selection_report, precision=precision,
                    **sql_kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getMovementHistoryList')
    def getMovementHistoryList(self, src__=0, ignore_variation=0,
                               standardise=0, omit_simulation=0,
                               omit_input=0, omit_output=0,
                               selection_domain=None, selection_report=None,
                               initial_running_total_quantity=0,
                               initial_running_total_price=0, precision=None,
                               **kw):
      """Returns a list of movements which modify the inventory
      for a single or a group of resource, node, section, etc.
      A running total quantity and a running total price are available on
      brains. The initial values can be passed, in case you want to have an
      "initial summary line".
      """
      kw['movement_list_mode'] = 1
      kw.update(self._getDefaultGroupByParameters(**kw))
      sql_kw = self._generateSQLKeywordDict(**kw)

      return self.Resource_zGetMovementHistoryList(
                         src__=src__, ignore_variation=ignore_variation,
                         standardise=standardise,
                         omit_simulation=omit_simulation,
                         omit_input=omit_input, omit_output=omit_output,
                         selection_domain=selection_domain,
                         selection_report=selection_report,
                         initial_running_total_quantity=
                                  initial_running_total_quantity,
                         initial_running_total_price=
                                  initial_running_total_price,
                         precision=precision, **sql_kw)

    security.declareProtected(Permissions.AccessContentsInformation,
                              'getMovementHistoryStat')
    def getMovementHistoryStat(self, src__=0, ignore_variation=0,
                               standardise=0, omit_simulation=0, omit_input=0,
                               omit_output=0, selection_domain=None,
                               selection_report=None, precision=None, **kw):
      """
      getMovementHistoryStat is the pending to getMovementHistoryList
      for ListBox stat
      """
      sql_kw = self._generateSQLKeywordDict(**kw)
      return self.Resource_zGetInventory(src__=src__,
          ignore_variation=ignore_variation, standardise=standardise,
          omit_simulation=omit_simulation, omit_input=omit_input,
          omit_output=omit_output, selection_domain=selection_domain,
          selection_report=selection_report, precision=precision, **sql_kw)

    security.declareProtected(Permissions.AccessContentsInformation, 
                              'getNextNegativeInventoryDate')
    def getNextNegativeInventoryDate(self, src__=0, **kw):
      """
      Returns statistics of inventory grouped by section or site
      """
      #sql_kw = self._generateSQLKeywordDict(order_by_expression='stock.date', **kw)
      #sql_kw['group_by_expression'] = 'stock.uid'
      #sql_kw['order_by_expression'] = 'stock.date'

      result = self.getInventoryList(src__=src__,
          sort_on = (('date', 'ascending'),), group_by_movement=1, **kw)
      if src__ :
        return result

      total_inventory = 0.
      for inventory in result:
        if inventory['inventory'] is not None:
          total_inventory += inventory['inventory']
          if total_inventory < 0:
            return inventory['date']

      return None

    #######################################################
    # Traceability management
    security.declareProtected(Permissions.AccessContentsInformation, 'getTrackingList')
    def getTrackingList(self, src__=0,
        selection_domain=None, selection_report=None,
        strict_simulation_state=1, history=0, **kw) :
      """
      Returns a list of items in the form

        uid (of item)
        date
        node_uid
        section_uid
        resource_uid
        variation_text
        delivery_uid

      If at_date is provided, returns the a list which answers
      to the question "where are those items at this date" or
      "which are those items which are there a this date"

      If at_date is not provided, returns a history of positions
      which answers the question "where have those items been
      between this time and this time". This will be handled by
      something like getTrackingHistoryList

      This method is only suitable for singleton items (an item which can
      only be at a single place at a given time). Such items include
      containers, serial numbers (ex. for engine), rolls with subrolls,

      This method is not suitable for batches (ex. a coloring batch).
      For such items, standard getInventoryList method is appropriate

      Parameters are the same as for getInventory.

      Default sort orders is based on dates, reverse.


      from_date (>=) -

      to_date   (<)  -

      at_date   (<=) - only take rows which date is <= at_date

      history (boolean) - keep history variations

      resource (only in generic API in simulation)

      node        -  only take rows in stock table which node_uid is equivalent to node

      section        -  only take rows in stock table which section_uid is equivalent to section

      resource_category        -  only take rows in stock table which resource_uid is in resource_category

      node_category        -  only take rows in stock table which node_uid is in section_category

      section_category        -  only take rows in stock table which section_uid is in section_category

      variation_text - only take rows in stock table with specified variation_text
                       this needs to be extended with some kind of variation_category ?
                       XXX this way of implementing variation selection is far from perfect

      variation_category - variation or list of possible variations

      simulation_state - only take rows with specified simulation_state

      selection_domain, selection_report - see ListBox

      **kw  - if we want extended selection with more keywords (but bad performance)
              check what we can do with buildSQLQuery

      Extra parameters for getTrackingList :

      item

      input - if set, answers to the question "which are those items which have been
              delivered for the first time after from_date". Cannot be used with output

      output - if set, answers to the question "which are those items which have been
               delivered for the last time before at_date or to_date". Cannot be used with input

      """
      new_kw = self._generateSQLKeywordDict(table='item',strict_simulation_state=strict_simulation_state,**kw)
      for key in ('at_date', 'to_date'):
        value = kw.get(key, None)
        if value is not None:
          if isinstance(value, DateTime):
            value = value.toZone('UTC').ISO()
          value = '%s' % (value, )
        # Do not remove dates in new_kw, they are required in 
        # order to do a "select item left join item on date"
        new_kw[key] = value

      # Extra parameters for the SQL Method
      new_kw['join_on_item'] = not history and (new_kw.get('at_date') or \
                               new_kw.get('to_date') or \
                               new_kw.get('input') or \
                               new_kw.get('output'))
      new_kw['date_condition_in_join'] = not (new_kw.get('input') or new_kw.get('output'))

      # Pass simulation state to request
      if kw.has_key('item.simulation_state'):
          new_kw['simulation_state_list'] = kw['item.simulation_state']
      else:
          new_kw['simulation_state_list'] =  None

      return self.Resource_zGetTrackingList(src__=src__,
                                            selection_domain=selection_domain,
                                            selection_report=selection_report,
                                            **new_kw)

    security.declareProtected(Permissions.AccessContentsInformation, 'getCurrentTrackingList')
    def getCurrentTrackingList(self, **kw):
      """
      Returns list of current inventory grouped by section or site
      """
      kw['item.simulation_state'] = self.getPortalCurrentInventoryStateList()
      return self.getTrackingList(**kw)
    
    security.declareProtected(Permissions.AccessContentsInformation, 'getCurrentTrackingHistoryList')
    def getCurrentTrackingHistoryList(self, **kw):
      """
      Returns list of current inventory grouped by section or site
      """
      kw['item.simulation_state'] = self.getPortalCurrentInventoryStateList()
      return self.getTrackingHistoryList(**kw)
    
    security.declareProtected(Permissions.AccessContentsInformation, 'getTrackingHistoryList')
    def getTrackingHistoryList(self, **kw):
      """
      Returns history list of inventory grouped by section or site
      """
      kw['history'] = 1
      return self.getTrackingList(**kw)

    security.declareProtected(Permissions.AccessContentsInformation, 'getFutureTrackingList')
    def getFutureTrackingList(self, **kw):
      """
      Returns list of future inventory grouped by section or site
      """
      kw['item.simulation_state'] = tuple(list(self.getPortalFutureInventoryStateList())
          + list(self.getPortalReservedInventoryStateList()) + list(self.getPortalCurrentInventoryStateList()))
      return self.getTrackingList(**kw)

    #######################################################
    # Capacity Management
    security.declareProtected( Permissions.ModifyPortalContent, 'updateCapacity' )
    def updateCapacity(self, node):
      capacity_item_list = []
      for o in node.contentValues():
        if o.isCapacity():
          # Do whatever is needed
          capacity_item_list += o.asCapacityItemList()
          pass
      # Do whatever with capacity_item_list
      # and store the resulting new capacity in node
      node._capacity_item_list = capacity_item_list

    security.declareProtected( Permissions.ModifyPortalContent, 'isMovementInsideCapacity' )
    def isMovementInsideCapacity(self, movement):
      """
        Purpose: provide answer to customer for the question "can you do it ?"

        movement:
          date
          source destination (2 nodes)
          source_section ...
      """
      # Get nodes and dat
      source_node = movement.getSourceValue()
      destination_node = movement.getDestinationValue()
      start_date = movement.getStartDate()
      stop_date = movement.getStopDate()
      # Return result
      return self.isNodeInsideCapacity(source_node, start_date, additional_movement=movement, sign=1) and self.isNodeInsideCapacity(destination_node, stop_date, additional_movement=movement, sign=-1)

    security.declareProtected( Permissions.ModifyPortalContent, 'isNodeInsideCapacity' )
    def isNodeInsideCapacity(self, node, date, simulation_state=None, additional_movement=None, sign=1):
      """
        Purpose: decide if a node is consistent with its capacity definitions
        at a certain date (ie. considreing the stock / production history
      """
      # First get the current inventory situation for this node
      inventory_list = node.getInventoryList(XXXXX)
      # Add additional movement
      if additional_movement:
          inventory_list = inventory_list + sign * additional_movement # needs to be implemented
      # Return answer
      return self.isAmountListInsideCapacity(node, inventory_list)

    security.declareProtected( Permissions.ModifyPortalContent, 'isAmountListInsideCapacity' )
    def isAmountListInsideCapacity(self, node, amount_list,
         resource_aggregation_base_category=None, resource_aggregation_depth=None):
      """
        Purpose: decide if a list of amounts is consistent with the capacity of a node

        If any resource in amount_list is missing in the capacity of the node, resource
        aggregation is performed, based on resource_aggregation_base_category. If the
        base category is not specified, it is an error (should guess instead?). The resource
        aggregation is done at the level of resource_aggregation_depth in the tree
        of categories. If resource_aggregation_depth is not specified, it's an error.

        Assumptions: amount_list is an association list, like ((R1 V1) (R2 V2)).
                     node has an attribute '_capacity_item_list' which is a list of association lists.
                     resource_aggregation_base_category is a Base Category object or a list of Base
                     Category objects or None.
                     resource_aggregation_depth is a strictly positive integer or None.
      """
      # Make a copy of the attribute _capacity_item_list, because it may be necessary
      # to modify it for resource aggregation.
      capacity_item_list = node._capacity_item_list[:]

      # Make a mapping between resources and its indices.
      resource_map = {}
      index = 0
      for alist in capacity_item_list:
        for pair in alist:
          resource = pair[0]
#          LOG('isAmountListInsideCapacity', 0,
#              "resource is %s" % repr(resource))
          if resource not in resource_map:
            resource_map[resource] = index
            index += 1

      # Build a point from the amount list.
      point = zeros(index, 'd') # Fill up zeros for safety.
      mask_map = {}     # This is used to skip items in amount_list.
      for amount in amount_list:
        if amount[0] in mask_map:
          continue
        # This will fail, if amount_list has any different resource from the capacity.
        # If it has any different point, then we should ......
        #
        # There would be two possible different solutions:
        # 1) If a missing resource is a meta-resource of resources supported by the capacity,
        #    it is possible to add the resource into the capacity by aggregation.
        # 2) If a missing resource has a meta-resource as a parent and the capacity supports
        #    the meta-resource directly or indirectly (`indirectly' means `by aggregation'),
        #    it is possible to convert the missing resource into the meta-resource.
        #
        # However, another way has been implemented here. This does the following, if the resource
        # is not present in the capacity:
        # 1) If the value is zero, just ignore the resource, because zero is always acceptable.
        # 2) Attempt to aggregate resources both of the capacity and of the amount list. This aggregation
        #    is performed at the depth of 'resource_aggregation_depth' under the base category
        #    'resource_aggregation_base_category'.
        #
        resource = amount[0]
        if resource in resource_map:
          point[resource_map[amount[0]]] = amount[1]
        else:
          if amount[1] == 0:
            # If the value is zero, no need to consider.
            pass
          elif resource_aggregation_base_category is None or resource_aggregation_depth is None:
            # XXX use an appropriate error class
            # XXX should guess a base category instead of emitting an exception
            raise RuntimeError, "The resource '%s' is not found in the capacity, and the argument 'resource_aggregation_base_category' or the argument 'resource_aggregation_depth' is not specified" % resource
          else:
            # It is necessary to aggregate resources, to guess the capacity of this resource.

            def getAggregationResourceUrl(url, depth):
              # Return a partial url of the argument 'url'.
              # If 'url' is '/foo/bar/baz' and 'depth' is 2, return '/foo/bar'.
              pos = 0
              for i in range(resource_aggregation_depth):
                pos = url.find('/', pos+1)
                if pos < 0:
                  break
              if pos < 0:
                return None
              pos = url.find('/', pos+1)
              if pos < 0:
                pos = len(url)
              return url[:pos]

            def getAggregatedResourceList(aggregation_url, category, resource_list):
              # Return a list of resources which should be aggregated. 'aggregation_url' is used
              # for a top url of those resources. 'category' is a base category for the aggregation.
              aggregated_resource_list = []
              for resource in resource_list:
                for url in resource.getCategoryMembershipList(category, base=1):
                  if url.startswith(aggregation_url):
                    aggregated_resource_list.append(resource)
              return aggregated_resource_list

            def getAggregatedItemList(item_list, resource_list, aggregation_resource):
              # Return a list of association lists, which is a result of an aggregation.
              # 'resource_list' is a list of resources which should be aggregated.
              # 'aggregation_resource' is a category object which is a new resource created by
              # this aggregation.
              # 'item_list' is a list of association lists.
              new_item_list = []
              for alist in item_list:
                new_val = 0
                new_alist = []
                # If a resource is not a aggregated, then add it to the new alist as it is.
                # Otherwise, aggregate it to a single value.
                for pair in alist:
                  if pair[0] in resource_list:
                    new_val += pair[1]
                  else:
                    new_alist.append(pair)
                # If it is zero, ignore this alist, as it is nonsense.
                if new_val != 0:
                  new_alist.append([aggregation_resource, new_val])
                  new_item_list.append(new_alist)
              return new_item_list

            # Convert this to a string if necessary, for convenience.
            if not isinstance(resource_aggregation_base_category, (tuple, list)):
              resource_aggregation_base_category = (resource_aggregation_base_category,)

            done = 0
#            LOG('isAmountListInsideCapacity', 0,
#                "resource_aggregation_base_category is %s" % repr(resource_aggregation_base_category))
            for category in resource_aggregation_base_category:
              for resource_url in resource.getCategoryMembershipList(category, base=1):
                aggregation_url = getAggregationResourceUrl(resource_url,
                                                            resource_aggregation_depth)
                if aggregation_url is None:
                  continue
                aggregated_resource_list = getAggregatedResourceList (aggregation_url,
                                                                      category,
                                                                      resource_map.keys())
                # If any, do the aggregation.
                if len(aggregated_resource_list) > 0:
                  aggregation_resource = self.portal_categories.resolveCategory(aggregation_url)
                  # Add the resource to the mapping.
 #                 LOG('aggregation_resource', 0, str(aggregation_resource))
                  resource_map[aggregation_resource] = index
                  index += 1
                  # Add the resource to the point.
                  point = resize(point, (index,))
                  val = 0
                  for aggregated_amount in amount_list:
                    for url in aggregated_amount[0].getCategoryMembershipList(category, base=1):
                      if url.startswith(aggregation_url):
                        val += aggregated_amount[1]
                        mask_map[aggregated_amount[0]] = None
                        break
                  point[index-1] = val
                  # Add capacity definitions of the resource into the capacity.
                  capacity_item_list += getAggregatedItemList(capacity_item_list,
                                                              aggregated_resource_list,
                                                              aggregation_resource)
                  done = 1
                  break
              if done:
                break
            if not done:
              raise RuntimeError, "Aggregation failed"

      # Build a matrix from the capacity item list.
#      LOG('resource_map', 0, str(resource_map))
      matrix = zeros((len(capacity_item_list)+1, index), 'd')
      for index in range(len(capacity_item_list)):
        for pair in capacity_item_list[index]:
          matrix[index,resource_map[pair[0]]] = pair[1]

#      LOG('isAmountListInsideCapacity', 0,
#          "matrix = %s, point = %s, capacity_item_list = %s" % (str(matrix), str(point), str(capacity_item_list)))
      return solve(matrix, point)


    # Asset Price Calculation
    def updateAssetPrice(self, resource, variation_text, section_category, node_category,
                         strict_membership=0, simulation_state=None):
      if simulation_state is None:
        simulation_state = self.getPortalCurrentInventoryStateList()
      category_tool = getToolByName(self, 'portal_categories')
      section_value = category_tool.resolveCategory(section_category)
      node_value = category_tool.resolveCategory(node_category)
      # Initialize price
      current_asset_price = 0.0 # Missing: initial inventory price !!!
      current_inventory = 0.0
      # Parse each movement
      brain_list = self.Resource_zGetMovementHistoryList(resource=[resource],
                             variation_text=variation_text,
                             section_category=section_category,
                             node_category=node_category,
                             strict_membership=strict_membership,
                             simulation_state=simulation_state) # strict_membership not taken into account
                             # We select movements related to certain nodes (ex. Stock) and sections (ex.Coramy Group)
      result = []
      for b in brain_list:
        m = b.getObject()
        if m is not None:
          previous_inventory = current_inventory
          inventory_quantity = b.quantity # We should use the aggregated quantity provided by Resource_zGetMovementHistoryList
          quantity = m.getQuantity() # The movement quantity is important to determine the meaning of source and destination
          # Maybe we should take care of target qty in delired deliveries
          if quantity is None:
            quantity = 0.0
          if m.getSourceValue() is None:
            # This is a production movement or an inventory movement
            # Use Industrial Price
            current_inventory += inventory_quantity # Update inventory
            if m.getPortalType() in ('Inventory Line', 'Inventory Cell'): # XX should be replaced by isInventory ???
              asset_price = m.getPrice()
              if asset_price in (0.0, None):
                asset_price = current_asset_price # Use current price if no price defined
            else: # this is a production
              asset_price = m.getIndustrialPrice()
              if asset_price is None: asset_price = current_asset_price  # Use current price if no price defined
            result.append((m.getRelativeUrl(), m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                          m.getQuantity(), 'Production or Inventory', 'Price: %s' % asset_price
                        ))
          elif m.getDestinationValue() is None:
            # This is a consumption movement or an inventory movement
            current_inventory += inventory_quantity # Update inventory
            asset_price = current_asset_price
            result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                          m.getQuantity(), 'Consumption or Inventory', 'Price: %s' % asset_price
                        ))
          elif m.getSourceValue().isAcquiredMemberOf(node_category) and m.getDestinationValue().isAcquiredMemberOf(node_category):
            # This is an internal movement
            current_inventory += inventory_quantity # Update inventory
            asset_price = current_asset_price
            result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                          m.getQuantity(), 'Internal', 'Price: %s' % asset_price
                        ))
          elif m.getSourceValue().isAcquiredMemberOf(node_category) and quantity < 0:
            # This is a physically inbound movement - try to use commercial price
            if m.getSourceSectionValue() is None:
              # No meaning
              current_inventory += inventory_quantity # Update inventory
              asset_price = current_asset_price
              result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Error', 'Price: %s' % asset_price
                          ))
            elif m.getDestinationSectionValue() is None:
              # No meaning
              current_inventory += inventory_quantity # Update inventory
              asset_price = current_asset_price
              result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Error', 'Price: %s' % asset_price
                          ))
            elif m.getDestinationSectionValue().isAcquiredMemberOf(section_category):
              current_inventory += inventory_quantity # Update inventory
              if m.getDestinationValue().isAcquiredMemberOf('site/Piquage'):
                # Production
                asset_price = m.getIndustrialPrice()
                if asset_price is None: asset_price = current_asset_price  # Use current price if no price defined
                result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                              m.getQuantity(), 'Production', 'Price: %s' % asset_price
                            ))
              else:
                # Inbound from same section
                asset_price = current_asset_price
                result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                              m.getQuantity(), 'Inbound same section', 'Price: %s' % asset_price
                            ))
            else:
              current_inventory += inventory_quantity # Update inventory
              asset_price = m.getPrice()
              result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Inbound different section', 'Price: %s' % asset_price
                          ))
          elif m.getDestinationValue().isAcquiredMemberOf(node_category) and quantity > 0:
            # This is a physically inbound movement - try to use commercial price
            if m.getSourceSectionValue() is None:
              # No meaning
              current_inventory += inventory_quantity # Update inventory
              asset_price = current_asset_price
              result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Error', 'Price: %s' % asset_price
                          ))
            elif m.getDestinationSectionValue() is None:
              # No meaning
              current_inventory += inventory_quantity # Update inventory
              asset_price = current_asset_price
              result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Error', 'Price: %s' % asset_price
                          ))
            elif m.getSourceSectionValue().isAcquiredMemberOf(section_category):
              current_inventory += inventory_quantity # Update inventory
              if m.getSourceValue().isAcquiredMemberOf('site/Piquage'):
                # Production
                asset_price = m.getIndustrialPrice()
                if asset_price is None: asset_price = current_asset_price  # Use current price if no price defined
                result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                              m.getQuantity(), 'Production', 'Price: %s' % asset_price
                            ))
              else:
                # Inbound from same section
                asset_price = current_asset_price
                result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Inbound same section', 'Price: %s' % asset_price
                          ))
            else:
              current_inventory += inventory_quantity # Update inventory
              asset_price = m.getPrice()
              result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Inbound different section', 'Price: %s' % asset_price
                          ))
          else:
            # Outbound movement
            current_inventory += inventory_quantity # Update inventory
            asset_price = current_asset_price
            result.append((m.getRelativeUrl(),m.getStartDate(), m.getSource(), m.getSourceSection(), m.getDestination(), m.getDestinationSection(),
                            m.getQuantity(), 'Outbound', 'Price: %s' % asset_price
                          ))

          # Update asset_price
          if current_inventory > 0:
            if inventory_quantity is not None:
              # Update price with an average of incoming goods and current goods
              current_asset_price = ( current_asset_price * previous_inventory + asset_price * inventory_quantity ) / float(current_inventory)
          else:
            # New price is the price of incoming goods - negative stock has no meaning for asset calculation
            current_asset_price = asset_price

          result.append(('###New Asset Price', current_asset_price, 'New Inventory', current_inventory))

          # Update Asset Price on the right side
          if m.getSourceSectionValue() is not None and m.getSourceSectionValue().isAcquiredMemberOf(section_category):
            # for each movement, source section is member of one and one only accounting category
            # therefore there is only one and one only source asset price
            m._setSourceAssetPrice(current_asset_price)
            #quantity = m.getInventoriatedQuantity()
            #if quantity:
            #  #total_asset_price = - current_asset_price * quantity
            #  #m.Movement_zSetSourceTotalAssetPrice(uid=m.getUid(), total_asset_price = total_asset_price)
            #  m._setSourceAssetPrice(current_asset_price)
          if m.getDestinationSectionValue() is not None and m.getDestinationSectionValue().isMemberOf(section_category):
            # for each movement, destination section is member of one and one only accounting category
            # therefore there is only one and one only destination asset price
            m._setDestinationAssetPrice(current_asset_price)
            #quantity = m.getInventoriatedQuantity()
            #if quantity:
            #  total_asset_price = current_asset_price * quantity
            #  m.Movement_zSetDestinationTotalAssetPrice(uid=m.getUid(), total_asset_price = total_asset_price)
          # Global reindexing required afterwards in any case: so let us do it now
          # Until we get faster methods (->reindexObject())
          #m.immediateReindexObject()
          m.reindexObject()
          #m.activate(priority=7).immediateReindexObject() # Too slow

      return result

    # Used for mergeDeliveryList.
    class MergeDeliveryListError(Exception): pass

    security.declareProtected( Permissions.ModifyPortalContent, 'mergeDeliveryList' )
    def mergeDeliveryList(self, delivery_list):
      """
        Merge multiple deliveries into one delivery.
        All delivery lines are merged into the first one.
        The first one is therefore called main_delivery here.
        The others are cancelled.
        Return the main delivery.
      """
      # Sanity checks.
      if len(delivery_list) == 0:
        raise self.MergeDeliveryListError, "No delivery is passed"
      elif len(delivery_list) == 1:
        raise self.MergeDeliveryListError, "Only one delivery is passed"

      main_delivery = delivery_list[0]
      delivery_list = delivery_list[1:]

      # Another sanity check. It is necessary for them to be identical in some attributes.
      for delivery in delivery_list:
        for attr in ('portal_type', 'simulation_state',
                     'source', 'destination',
                     'source_section', 'destination_section',
                     'source_decision', 'destination_decision',
                     'source_administration', 'destination_administration',
                     'source_payment', 'destination_payment'):
          main_value = main_delivery.getProperty(attr)
          value = delivery.getProperty(attr)
          if  main_value != value:
            raise self.MergeDeliveryListError, \
              "%s is not the same between %s and %s (%s and %s)" % (attr, delivery.getId(), main_delivery.getId(), value, main_value)

      # One more sanity check. Check if discounts are the same, if any.
      main_discount_list = main_delivery.contentValues(filter = {'portal_type': self.getPortalDiscountTypeList()})
      for delivery in delivery_list:
        discount_list = delivery.contentValues(filter = {'portal_type': self.getPortalDiscountTypeList()})
        if len(main_discount_list) != len(discount_list):
          raise self.MergeDeliveryListError, "Discount is not the same between %s and %s" % (delivery.getId(), main_delivery.getId())
        for discount in discount_list:
          for main_discount in main_discount_list:
            if discount.getDiscount() == main_discount.getDiscount() \
               and discount.getDiscountRatio() == main_discount.getDiscountRatio() \
               and discount.getDiscountType() == main_discount.getDiscountType() \
               and discount.getImmediateDiscount() == main_discount.getImmediateDiscount():
              break
          else:
            raise self.MergeDeliveryListError, "Discount is not the same between %s and %s" % (delivery.getId(), main_delivery.getId())

      # One more sanity check. Check if payment conditions are the same, if any.
      main_payment_condition_list = main_delivery.contentValues(filter = {'portal_type': self.getPortalPaymentConditionTypeList()})
      for delivery in delivery_list:
        payment_condition_list = delivery.contentValues(filter = {'portal_type': self.getPortalPaymentConditionTypeList()})
        if len(main_payment_condition_list) != len(payment_condition_list):
          raise self.MergeDeliveryListError, "Payment Condition is not the same between %s and %s" % (delivery.getId(), main_delivery.getId())
        for condition in payment_condition_list:
          for main_condition in main_payment_condition_list:
            if condition.getPaymentMode() == main_condition.getPaymentMode() \
               and condition.getPaymentAdditionalTerm() == main_condition.getPaymentAdditionalTerm() \
               and condition.getPaymentAmount() == main_condition.getPaymentAmount() \
               and condition.getPaymentEndOfMonth() == main_condition.getPaymentEndOfMonth() \
               and condition.getPaymentRatio() == main_condition.getPaymentRatio() \
               and condition.getPaymentTerm() == main_condition.getPaymentTerm():
              break
          else:
            raise self.MergeDeliveryListError, "Payment Condition is not the same between %s and %s" % (delivery.getId(), main_delivery.getId())

      # Make sure that all activities are flushed, to get simulation movements from delivery cells.
      for delivery in delivery_list:
        for order in delivery.getCausalityValueList(portal_type = self.getPortalOrderTypeList()):
          for applied_rule in order.getCausalityRelatedValueList(portal_type = 'Applied Rule'):
            applied_rule.flushActivity(invoke = 1)
        for causality_related_delivery in delivery.getCausalityValueList(portal_type = self.getPortalDeliveryTypeList()):
          for applied_rule in causality_related_delivery.getCausalityRelatedValueList(portal_type = 'Applied Rule'):
            applied_rule.flushActivity(invoke = 1)

      # Get a list of simulated movements and invoice movements.
      main_simulated_movement_list = main_delivery.getSimulatedMovementList()
      main_invoice_movement_list = main_delivery.getInvoiceMovementList()
      simulated_movement_list = main_simulated_movement_list[:]
      invoice_movement_list = main_invoice_movement_list[:]
      for delivery in delivery_list:
        simulated_movement_list.extend(delivery.getSimulatedMovementList())
        invoice_movement_list.extend(delivery.getInvoiceMovementList())

      #for movement in simulated_movement_list + invoice_movement_list:
      #  parent = movement.aq_parent
      #  LOG('mergeDeliveryList', 0, 'movement = %s, parent = %s, movement.getPortalType() = %s, parent.getPortalType() = %s' % (repr(movement), repr(parent), repr(movement.getPortalType()), repr(parent.getPortalType())))

      LOG('mergeDeliveryList', 0, 'simulated_movement_list = %s, invoice_movement_list = %s' % (str(simulated_movement_list), str(invoice_movement_list)))
      for main_movement_list, movement_list in \
        ((main_simulated_movement_list, simulated_movement_list),
         (main_invoice_movement_list, invoice_movement_list)):
        root_group = self.collectMovement(movement_list,
                                          check_order = 0,
                                          check_path = 0,
                                          check_date = 0,
                                          check_criterion = 1,
                                          check_resource = 1,
                                          check_base_variant = 1,
                                          check_variant = 1)
        for criterion_group in root_group.group_list:
          LOG('mergeDeliveryList dump tree', 0, 'criterion = %s, movement_list = %s, group_list = %s' % (repr(criterion_group.criterion), repr(criterion_group.movement_list), repr(criterion_group.group_list)))
          for resource_group in criterion_group.group_list:
            LOG('mergeDeliveryList dump tree', 0, 'resource = %s, movement_list = %s, group_list = %s' % (repr(resource_group.resource), repr(resource_group.movement_list), repr(resource_group.group_list)))
            for base_variant_group in resource_group.group_list:
              LOG('mergeDeliveryList dump tree', 0, 'base_category_list = %s, movement_list = %s, group_list = %s' % (repr(base_variant_group.base_category_list), repr(base_variant_group.movement_list), repr(base_variant_group.group_list)))
              for variant_group in base_variant_group.group_list:
                LOG('mergeDeliveryList dump tree', 0, 'category_list = %s, movement_list = %s, group_list = %s' % (repr(variant_group.category_list), repr(variant_group.movement_list), repr(variant_group.group_list)))

        for criterion_group in root_group.group_list:
          for resource_group in criterion_group.group_list:
            for base_variant_group in resource_group.group_list:
              # Get a list of categories.
              category_dict = {}
              for variant_group in base_variant_group.group_list:
                for category in variant_group.category_list:
                  category_dict[category] = 1
              category_list = category_dict.keys()

              # Try to find a delivery line.
              delivery_line = None
              for movement in base_variant_group.movement_list:
                if movement in main_movement_list:
                  if movement.aq_parent.getPortalType() in self.getPortalSimulatedMovementTypeList() \
                    or movement.aq_parent.getPortalType() in self.getPortalInvoiceMovementTypeList():
                    delivery_line = movement.aq_parent
                  else:
                    delivery_line = movement
                  LOG('mergeDeliveryList', 0, 'delivery_line %s is found: criterion = %s, resource = %s, base_category_list = %s' % (repr(delivery_line), repr(criterion_group.criterion), repr(resource_group.resource), repr(base_variant_group.base_category_list)))
                  break

              if delivery_line is None:
                # Not found. So create a new delivery line.
                movement = base_variant_group.movement_list[0]
                if movement.aq_parent.getPortalType() in self.getPortalSimulatedMovementTypeList() \
                  or movement.aq_parent.getPortalType() in self.getPortalInvoiceMovementTypeList():
                  delivery_line_type = movement.aq_parent.getPortalType()
                else:
                  delivery_line_type = movement.getPortalType()
                delivery_line = main_delivery.newContent(portal_type = delivery_line_type,
                                                         resource = resource_group.resource)
                LOG('mergeDeliveryList', 0, 'New delivery_line %s is created: criterion = %s, resource = %s, base_category_list = %s' % (repr(delivery_line), repr(criterion_group.criterion), repr(resource_group.resource), repr(base_variant_group.base_category_list)))

              # Update the base categories and categories.
              #LOG('mergeDeliveryList', 0, 'base_category_list = %s, category_list = %s' % (repr(base_category_list), repr(category_list)))
              delivery_line.setVariationBaseCategoryList(base_variant_group.base_category_list)
              delivery_line.setVariationCategoryList(category_list)

              object_to_update = None
              for variant_group in base_variant_group.group_list:
                if len(variant_group.category_list) == 0:
                  object_to_update = delivery_line
                else:
                  for delivery_cell in delivery_line.contentValues():
                    predicate_value_list = delivery_cell.getPredicateValueList()
                    LOG('mergeDeliveryList', 0, 'delivery_cell = %s, predicate_value_list = %s, variant_group.category_list = %s' % (repr(delivery_cell), repr(predicate_value_list), repr(variant_group.category_list)))
                    if len(predicate_value_list) == len(variant_group.category_list):
                      for category in variant_group.category_list:
                        if category not in predicate_value_list:
                          break
                      else:
                        object_to_update = delivery_cell
                        break

                #LOG('mergeDeliveryList', 0, 'object_to_update = %s' % repr(object_to_update))
                if object_to_update is not None:
                  cell_price = object_to_update.getPrice() or 0.0
                  cell_quantity = object_to_update.getQuantity() or 0.0
                  cell_target_quantity = object_to_update.getNetConvertedTargetQuantity() or 0.0 # XXX What to do ?
                  cell_total_price = cell_target_quantity * cell_price
                  cell_category_list = list(object_to_update.getCategoryList())

                  for movement in variant_group.movement_list:
                    if movement in main_movement_list:
                      continue
                    LOG('mergeDeliveryList', 0, 'movement = %s' % repr(movement))
                    cell_quantity += movement.getQuantity()
                    cell_target_quantity += movement.getNetConvertedTargetQuantity()
                    try:
                      # XXX WARNING - ADD PRICED QUANTITY
                      cell_price = movement.getPrice()
                      cell_total_price += movement.getNetConvertedTargetQuantity() * cell_price
                    except TypeError:
                      cell_total_price = None
                    for category in movement.getCategoryList():
                      if category not in cell_category_list:
                        cell_category_list.append(category)
                    # Make sure that simulation movements point to an appropriate delivery line or
                    # delivery cell.
                    if hasattr(movement, 'getDeliveryRelatedValueList'):
                      for simulation_movement in \
                        movement.getDeliveryRelatedValueList(portal_type = 'Simulation Movement'):
                        simulation_movement.setDeliveryValue(object_to_update)
                        #simulation_movement.reindexObject()
                    if hasattr(movement, 'getOrderRelatedValueList'):
                      for simulation_movement in \
                        movement.getOrderRelatedValueList(portal_type = 'Simulation Movement'):
                        simulation_movement.setOrderValue(object_to_update)
                        #simulation_movement.reindexObject()

                  if cell_target_quantity != 0 and cell_total_price is not None:
                    average_price = cell_total_price / cell_target_quantity
                  else:
                    average_price = 0

                  LOG('mergeDeliveryList', 0, 'object_to_update = %s, cell_category_list = %s, cell_target_quantity = %s, cell_quantity = %s, average_price = %s' % (repr(object_to_update), repr(cell_category_list), repr(cell_target_quantity), repr(cell_quantity), repr(average_price)))
                  object_to_update.setCategoryList(cell_category_list)
                  if object_to_update.getPortalType() in self.getPortalSimulatedMovementTypeList():
                    object_to_update.edit(target_quantity = cell_target_quantity,
                                          quantity = cell_quantity,
                                          price = average_price,
                                          )
                  elif object_to_update.getPortalType() in self.getPortalInvoiceMovementTypeList():
                    # Invoices do not have target quantities, and the price never change.
                    object_to_update.edit(quantity = cell_quantity,
                                          price = cell_price,
                                          )
                  else:
                    raise self.MergeDeliveryListError, "Unknown portal type %s" % str(object_to_update.getPortalType())
                  #object_to_update.immediateReindexObject()
                else:
                  raise self.MergeDeliveryListError, "No object to update"

      # Merge containers. Just copy them from other deliveries into the main.
      for delivery in delivery_list:
        container_id_list = delivery.contentIds(filter = {'portal_type': self.getPortalContainerTypeList()})
        if len(container_id_list) > 0:
          copy_data = delivery.manage_copyObjects(ids = container_id_list)
          new_id_list = main_delivery.manage_pasteObjects(copy_data)

      # Unify the list of causality.
      causality_list = main_delivery.getCausalityValueList()
      for delivery in delivery_list:
        for causality in delivery.getCausalityValueList():
          if causality not in causality_list:
            causality_list.append(causality)
      LOG("mergeDeliveryList", 0, "causality_list = %s" % str(causality_list))
      main_delivery.setCausalityValueList(causality_list)

      # Cancel deliveries.
      for delivery in delivery_list:
        LOG("mergeDeliveryList", 0, "cancelling %s" % repr(delivery))
        delivery.cancel()

      # Reindex the main delivery.
      main_delivery.reindexObject()

      return main_delivery

    #######################################################
    # Sequence
    security.declareProtected(Permissions.AccessContentsInformation, 
                              'getSequence')
    def getSequence(self, **kw):
      """
      getSequence is take the same parameters as Sequence constructor,
      and return a Sequence.
      """
      return Sequence(**kw)

    #######################################################
    # Time Management
    security.declareProtected(Permissions.AccessContentsInformation, 
                              'getAvailableTime')
    def getAvailableTime(self, from_date=None, to_date=None, 
                         portal_type=[], node=[], 
                         resource=[], src__=0, **kw):
      """
      Calculate available time for a node
      Returns an inventory of a single or multiple resources on a single
      node as a single float value

      from_date (>=) - only take rows which mirror_date is >= from_date

      to_date   (<)  - only take rows which date is < to_date

      node           - only take rows in stock table which node_uid is
                       equivalent to node

      resource       - only take rows in stock table which resource_uid is
                       equivalent to resource

      portal_type    - only take rows in stock table which portal_type
                       is in portal_type parameter
      """
      # XXX For now, consider that from_date and to_date are required
      if (from_date is None) or (to_date is None):
        raise NotImplementedError, \
              "getAvailableTime does not managed yet None values"
      # Calculate portal_type
      if portal_type == []:
        portal_type = self.getPortalCalendarPeriodTypeList()

      simulation_state = self.getPortalCurrentInventoryStateList() + \
                         self.getPortalTransitInventoryStateList() + \
                         self.getPortalReservedInventoryStateList()

      sql_result = self.Person_zGetAvailableTime(
                          from_date=from_date,
                          to_date=to_date,
                          portal_type=portal_type,
                          node=node,
                          resource=resource,
                          simulation_state=simulation_state,
                          src__=src__, **kw)
      if not src__:
        result = 0
        if len(sql_result) == 1:
          result = sql_result[0].total_quantity
      else:
        result = sql_result
      return result

    security.declareProtected(Permissions.AccessContentsInformation, 
                              'getAvailableTimeSequence')
    def getAvailableTimeSequence(self, from_date, to_date,  
                                 portal_type=[], node=[],
                                 resource=[], 
                                 src__=0,
                                 **kw):
      """
      Calculate available time for a node in multiple period of time.
      Each row is the available time for a specific period

      node           - only take rows in stock table which node_uid is
                       equivalent to node

      portal_type    - only take rows in stock table which portal_type
                       is in portal_type parameter

      resource       - only take rows in stock table which resource_uid is
                       equivalent to resource

      from_date (>=) - return period which start >= from_date

      to_date   (<)  - return period which start < to_date

      second, minute,
      hour, day,
      month, year   - duration of each time period (cumulative)
      """
      # Calculate portal_type
      if portal_type == []:
        portal_type = self.getPortalCalendarPeriodTypeList()

      sequence = Sequence(from_date, to_date, **kw)
      for sequence_item in sequence:
        setattr(sequence_item, 'total_quantity',
                self.getAvailableTime(
                          from_date=sequence_item.from_date,
                          to_date=sequence_item.to_date,
                          portal_type=portal_type,
                          node=node,
                          resource=resource,
                          src__=src__))
      return sequence

from Products.ERP5Type.DateUtils import addToDate

class SequenceItem:
  """
  SequenceItem define a time period.
  period.
  """
  def __init__(self, from_date, to_date):
    self.from_date = from_date
    self.to_date = to_date

class Sequence:
  """
  Sequence is a iterable object, which calculate a range of time
  period.
  """
  def __init__(self, from_date, to_date, 
               second=None, minute=None, hour=None,
               day=None, month=None, year=None):
    """
    Calculate a list of time period.
    Time period is a 2-tuple of 2 DateTime, which represent the from date
    and to date of the period.

    The start date of a period is calculated with the rule
        start_date of the previous + period duration

    from_date (>=) - return period which start >= from_date

    to_date   (<)  - return period which start < to_date

    second, minute,
    hour, day,
    month, year   - duration of each time period (cumulative)
                    at least one of those parameters must be specified.
    """
    if not (second or minute or hour or day or month or year):
      raise ValueError('Period duration must be specified')

    self.item_list = []
    # Calculate all time period
    current_from_date = from_date
    while current_from_date < to_date:
      current_to_date = addToDate(current_from_date, 
                                  second=second,
                                  minute=minute,
                                  hour=hour,
                                  day=day,
                                  month=month,
                                  year=year)
      self.item_list.append(SequenceItem(current_from_date, 
                                         current_to_date))
      current_from_date = current_to_date

  def __len__(self):
    return len(self.item_list)

  def __getitem__(self, key):
    return self.item_list[key]

  def __contains__(self, value):
    return (value in self.item_list)

  def __iter__(self):
    for x in self.item_list:
      yield x

InitializeClass(Sequence)
allow_class(Sequence)
InitializeClass(SequenceItem)
allow_class(SequenceItem)

InitializeClass(SimulationTool)