[odoo/odoo.git] / addons / project / project.py

# -*- coding: utf-8 -*-
##############################################################################
#
#    OpenERP, Open Source Management Solution
#    Copyright (C) 2004-today OpenERP SA (<http://www.openerp.com>)
#
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU Affero General Public License as
#    published by the Free Software Foundation, either version 3 of the
#    License, or (at your option) any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU Affero General Public License for more details.
#
#    You should have received a copy of the GNU Affero General Public License
#    along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
##############################################################################

from datetime import datetime, date
from lxml import etree
import time

from openerp import SUPERUSER_ID
from openerp import tools
from openerp.addons.resource.faces import task as Task
from openerp.osv import fields, osv
from openerp.tools.translate import _


class project_task_type(osv.osv):
    _name = 'project.task.type'
    _description = 'Task Stage'
    _order = 'sequence'
    _columns = {
        'name': fields.char('Stage Name', required=True, translate=True),
        'description': fields.text('Description'),
        'sequence': fields.integer('Sequence'),
        'case_default': fields.boolean('Default for New Projects',
                        help="If you check this field, this stage will be proposed by default on each new project. It will not assign this stage to existing projects."),
        'project_ids': fields.many2many('project.project', 'project_task_type_rel', 'type_id', 'project_id', 'Projects'),
        'fold': fields.boolean('Folded in Kanban View',
                               help='This stage is folded in the kanban view when'
                               'there are no records in that stage to display.'),
    }

    def _get_default_project_ids(self, cr, uid, ctx={}):
        project_id = self.pool['project.task']._get_default_project_id(cr, uid, context=ctx)
        if project_id:
            return [project_id]
        return None

    _defaults = {
        'sequence': 1,
        'project_ids': _get_default_project_ids,
    }
    _order = 'sequence'


class project(osv.osv):
    _name = "project.project"
    _description = "Project"
    _inherits = {'account.analytic.account': "analytic_account_id",
                 "mail.alias": "alias_id"}
    _inherit = ['mail.thread', 'ir.needaction_mixin']

    def _auto_init(self, cr, context=None):
        """ Installation hook: aliases, project.project """
        # create aliases for all projects and avoid constraint errors
        alias_context = dict(context, alias_model_name='project.task')
        return self.pool.get('mail.alias').migrate_to_alias(cr, self._name, self._table, super(project, self)._auto_init,
            'project.task', self._columns['alias_id'], 'id', alias_prefix='project+', alias_defaults={'project_id':'id'}, context=alias_context)

    def search(self, cr, user, args, offset=0, limit=None, order=None, context=None, count=False):
        if user == 1:
            return super(project, self).search(cr, user, args, offset=offset, limit=limit, order=order, context=context, count=count)
        if context and context.get('user_preference'):
                cr.execute("""SELECT project.id FROM project_project project
                           LEFT JOIN account_analytic_account account ON account.id = project.analytic_account_id
                           LEFT JOIN project_user_rel rel ON rel.project_id = project.id
                           WHERE (account.user_id = %s or rel.uid = %s)"""%(user, user))
                return [(r[0]) for r in cr.fetchall()]
        return super(project, self).search(cr, user, args, offset=offset, limit=limit, order=order,
            context=context, count=count)

    def onchange_partner_id(self, cr, uid, ids, part=False, context=None):
        partner_obj = self.pool.get('res.partner')
        val = {}
        if not part:
            return {'value': val}
        if 'pricelist_id' in self.fields_get(cr, uid, context=context):
            pricelist = partner_obj.read(cr, uid, part, ['property_product_pricelist'], context=context)
            pricelist_id = pricelist.get('property_product_pricelist', False) and pricelist.get('property_product_pricelist')[0] or False
            val['pricelist_id'] = pricelist_id
        return {'value': val}

    def _get_projects_from_tasks(self, cr, uid, task_ids, context=None):
        tasks = self.pool.get('project.task').browse(cr, uid, task_ids, context=context)
        project_ids = [task.project_id.id for task in tasks if task.project_id]
        return self.pool.get('project.project')._get_project_and_parents(cr, uid, project_ids, context)

    def _get_project_and_parents(self, cr, uid, ids, context=None):
        """ return the project ids and all their parent projects """
        res = set(ids)
        while ids:
            cr.execute("""
                SELECT DISTINCT parent.id
                FROM project_project project, project_project parent, account_analytic_account account
                WHERE project.analytic_account_id = account.id
                AND parent.analytic_account_id = account.parent_id
                AND project.id IN %s
                """, (tuple(ids),))
            ids = [t[0] for t in cr.fetchall()]
            res.update(ids)
        return list(res)

    def _get_project_and_children(self, cr, uid, ids, context=None):
        """ retrieve all children projects of project ids;
            return a dictionary mapping each project to its parent project (or None)
        """
        res = dict.fromkeys(ids, None)
        while ids:
            cr.execute("""
                SELECT project.id, parent.id
                FROM project_project project, project_project parent, account_analytic_account account
                WHERE project.analytic_account_id = account.id
                AND parent.analytic_account_id = account.parent_id
                AND parent.id IN %s
                """, (tuple(ids),))
            dic = dict(cr.fetchall())
            res.update(dic)
            ids = dic.keys()
        return res

    def _progress_rate(self, cr, uid, ids, names, arg, context=None):
        child_parent = self._get_project_and_children(cr, uid, ids, context)
        # compute planned_hours, total_hours, effective_hours specific to each project
        cr.execute("""
            SELECT project_id, COALESCE(SUM(planned_hours), 0.0),
                COALESCE(SUM(total_hours), 0.0), COALESCE(SUM(effective_hours), 0.0)
            FROM project_task
            LEFT JOIN project_task_type ON project_task.stage_id = project_task_type.id
            WHERE project_task.project_id IN %s AND project_task_type.fold = False
            GROUP BY project_id
            """, (tuple(child_parent.keys()),))
        # aggregate results into res
        res = dict([(id, {'planned_hours':0.0, 'total_hours':0.0, 'effective_hours':0.0}) for id in ids])
        for id, planned, total, effective in cr.fetchall():
            # add the values specific to id to all parent projects of id in the result
            while id:
                if id in ids:
                    res[id]['planned_hours'] += planned
                    res[id]['total_hours'] += total
                    res[id]['effective_hours'] += effective
                id = child_parent[id]
        # compute progress rates
        for id in ids:
            if res[id]['total_hours']:
                res[id]['progress_rate'] = round(100.0 * res[id]['effective_hours'] / res[id]['total_hours'], 2)
            else:
                res[id]['progress_rate'] = 0.0
        return res

    def unlink(self, cr, uid, ids, context=None):
        alias_ids = []
        mail_alias = self.pool.get('mail.alias')
        analytic_account_to_delete = set()
        for proj in self.browse(cr, uid, ids, context=context):
            if proj.tasks:
                raise osv.except_osv(_('Invalid Action!'),
                                     _('You cannot delete a project containing tasks. You can either delete all the project\'s tasks and then delete the project or simply deactivate the project.'))
            elif proj.alias_id:
                alias_ids.append(proj.alias_id.id)
            if proj.analytic_account_id and not proj.analytic_account_id.line_ids:
                analytic_account_to_delete.add(proj.analytic_account_id.id)
        res = super(project, self).unlink(cr, uid, ids, context=context)
        mail_alias.unlink(cr, uid, alias_ids, context=context)
        self.pool['account.analytic.account'].unlink(cr, uid, list(analytic_account_to_delete), context=context)
        return res

    def _get_attached_docs(self, cr, uid, ids, field_name, arg, context):
        res = {}
        attachment = self.pool.get('ir.attachment')
        task = self.pool.get('project.task')
        for id in ids:
            project_attachments = attachment.search(cr, uid, [('res_model', '=', 'project.project'), ('res_id', '=', id)], context=context, count=True)
            task_ids = task.search(cr, uid, [('project_id', '=', id)], context=context)
            task_attachments = attachment.search(cr, uid, [('res_model', '=', 'project.task'), ('res_id', 'in', task_ids)], context=context, count=True)
            res[id] = (project_attachments or 0) + (task_attachments or 0)
        return res
    def _task_count(self, cr, uid, ids, field_name, arg, context=None):
        res={}
        for tasks in self.browse(cr, uid, ids, context):
            res[tasks.id] = len(tasks.task_ids)
        return res
    def _get_alias_models(self, cr, uid, context=None):
        """ Overriden in project_issue to offer more options """
        return [('project.task', "Tasks")]

    def _get_visibility_selection(self, cr, uid, context=None):
        """ Overriden in portal_project to offer more options """
        return [('public', 'Public project'),
                ('employees', 'Internal project: all employees can access'),
                ('followers', 'Private project: followers Only')]

    def attachment_tree_view(self, cr, uid, ids, context):
        task_ids = self.pool.get('project.task').search(cr, uid, [('project_id', 'in', ids)])
        domain = [
             '|',
             '&', ('res_model', '=', 'project.project'), ('res_id', 'in', ids),
             '&', ('res_model', '=', 'project.task'), ('res_id', 'in', task_ids)]
        res_id = ids and ids[0] or False
        return {
            'name': _('Attachments'),
            'domain': domain,
            'res_model': 'ir.attachment',
            'type': 'ir.actions.act_window',
            'view_id': False,
            'view_mode': 'kanban,tree,form',
            'view_type': 'form',
            'limit': 80,
            'context': "{'default_res_model': '%s','default_res_id': %d}" % (self._name, res_id)
        }

    # Lambda indirection method to avoid passing a copy of the overridable method when declaring the field
    _alias_models = lambda self, *args, **kwargs: self._get_alias_models(*args, **kwargs)
    _visibility_selection = lambda self, *args, **kwargs: self._get_visibility_selection(*args, **kwargs)

    _columns = {
        'active': fields.boolean('Active', help="If the active field is set to False, it will allow you to hide the project without removing it."),
        'sequence': fields.integer('Sequence', help="Gives the sequence order when displaying a list of Projects."),
        'analytic_account_id': fields.many2one('account.analytic.account', 'Contract/Analytic', help="Link this project to an analytic account if you need financial management on projects. It enables you to connect projects with budgets, planning, cost and revenue analysis, timesheets on projects, etc.", ondelete="cascade", required=True),
        'members': fields.many2many('res.users', 'project_user_rel', 'project_id', 'uid', 'Project Members',
            help="Project's members are users who can have an access to the tasks related to this project.", states={'close':[('readonly',True)], 'cancelled':[('readonly',True)]}),
        'tasks': fields.one2many('project.task', 'project_id', "Task Activities"),
        'planned_hours': fields.function(_progress_rate, multi="progress", string='Planned Time', help="Sum of planned hours of all tasks related to this project and its child projects.",
            store = {
                'project.project': (_get_project_and_parents, ['tasks', 'parent_id', 'child_ids'], 10),
                'project.task': (_get_projects_from_tasks, ['planned_hours', 'remaining_hours', 'work_ids', 'stage_id'], 20),
            }),
        'effective_hours': fields.function(_progress_rate, multi="progress", string='Time Spent', help="Sum of spent hours of all tasks related to this project and its child projects.",
            store = {
                'project.project': (_get_project_and_parents, ['tasks', 'parent_id', 'child_ids'], 10),
                'project.task': (_get_projects_from_tasks, ['planned_hours', 'remaining_hours', 'work_ids', 'stage_id'], 20),
            }),
        'total_hours': fields.function(_progress_rate, multi="progress", string='Total Time', help="Sum of total hours of all tasks related to this project and its child projects.",
            store = {
                'project.project': (_get_project_and_parents, ['tasks', 'parent_id', 'child_ids'], 10),
                'project.task': (_get_projects_from_tasks, ['planned_hours', 'remaining_hours', 'work_ids', 'stage_id'], 20),
            }),
        'progress_rate': fields.function(_progress_rate, multi="progress", string='Progress', type='float', group_operator="avg", help="Percent of tasks closed according to the total of tasks todo.",
            store = {
                'project.project': (_get_project_and_parents, ['tasks', 'parent_id', 'child_ids'], 10),
                'project.task': (_get_projects_from_tasks, ['planned_hours', 'remaining_hours', 'work_ids', 'stage_id'], 20),
            }),
        'resource_calendar_id': fields.many2one('resource.calendar', 'Working Time', help="Timetable working hours to adjust the gantt diagram report", states={'close':[('readonly',True)]} ),
        'type_ids': fields.many2many('project.task.type', 'project_task_type_rel', 'project_id', 'type_id', 'Tasks Stages', states={'close':[('readonly',True)], 'cancelled':[('readonly',True)]}),
        'task_count': fields.function(_task_count, type='integer', string="Tasks",),
        'task_ids': fields.one2many('project.task', 'project_id',
                                    domain=[('stage_id.fold', '=', False)]),
        'color': fields.integer('Color Index'),
        'alias_id': fields.many2one('mail.alias', 'Alias', ondelete="restrict", required=True,
                                    help="Internal email associated with this project. Incoming emails are automatically synchronized"
                                         "with Tasks (or optionally Issues if the Issue Tracker module is installed)."),
        'alias_model': fields.selection(_alias_models, "Alias Model", select=True, required=True,
                                        help="The kind of document created when an email is received on this project's email alias"),
        'privacy_visibility': fields.selection(_visibility_selection, 'Privacy / Visibility', required=True,
            help="Holds visibility of the tasks or issues that belong to the current project:\n"
                    "- Public: everybody sees everything; if portal is activated, portal users\n"
                    "   see all tasks or issues; if anonymous portal is activated, visitors\n"
                    "   see all tasks or issues\n"
                    "- Portal (only available if Portal is installed): employees see everything;\n"
                    "   if portal is activated, portal users see the tasks or issues followed by\n"
                    "   them or by someone of their company\n"
                    "- Employees Only: employees see all tasks or issues\n"
                    "- Followers Only: employees see only the followed tasks or issues; if portal\n"
                    "   is activated, portal users see the followed tasks or issues."),
        'state': fields.selection([('template', 'Template'),
                                   ('draft','New'),
                                   ('open','In Progress'),
                                   ('cancelled', 'Cancelled'),
                                   ('pending','Pending'),
                                   ('close','Closed')],
                                  'Status', required=True, copy=False),
        'doc_count': fields.function(
            _get_attached_docs, string="Number of documents attached", type='integer'
        )
     }

    def _get_type_common(self, cr, uid, context):
        ids = self.pool.get('project.task.type').search(cr, uid, [('case_default','=',1)], context=context)
        return ids

    _order = "sequence, id"
    _defaults = {
        'active': True,
        'type': 'contract',
        'state': 'open',
        'sequence': 10,
        'type_ids': _get_type_common,
        'alias_model': 'project.task',
        'privacy_visibility': 'employees',
    }

    # TODO: Why not using a SQL contraints ?
    def _check_dates(self, cr, uid, ids, context=None):
        for leave in self.read(cr, uid, ids, ['date_start', 'date'], context=context):
            if leave['date_start'] and leave['date']:
                if leave['date_start'] > leave['date']:
                    return False
        return True

    _constraints = [
        (_check_dates, 'Error! project start-date must be lower then project end-date.', ['date_start', 'date'])
    ]

    def set_template(self, cr, uid, ids, context=None):
        return self.setActive(cr, uid, ids, value=False, context=context)

    def set_done(self, cr, uid, ids, context=None):
        return self.write(cr, uid, ids, {'state': 'close'}, context=context)

    def set_cancel(self, cr, uid, ids, context=None):
        return self.write(cr, uid, ids, {'state': 'cancelled'}, context=context)

    def set_pending(self, cr, uid, ids, context=None):
        return self.write(cr, uid, ids, {'state': 'pending'}, context=context)

    def set_open(self, cr, uid, ids, context=None):
        return self.write(cr, uid, ids, {'state': 'open'}, context=context)

    def reset_project(self, cr, uid, ids, context=None):
        return self.setActive(cr, uid, ids, value=True, context=context)

    def map_tasks(self, cr, uid, old_project_id, new_project_id, context=None):
        """ copy and map tasks from old to new project """
        if context is None:
            context = {}
        map_task_id = {}
        task_obj = self.pool.get('project.task')
        proj = self.browse(cr, uid, old_project_id, context=context)
        for task in proj.tasks:
            # preserve task name and stage, normally altered during copy
            defaults = {'stage_id': task.stage_id.id,
                        'name': task.name}
            map_task_id[task.id] =  task_obj.copy(cr, uid, task.id, defaults, context=context)
        self.write(cr, uid, [new_project_id], {'tasks':[(6,0, map_task_id.values())]})
        task_obj.duplicate_task(cr, uid, map_task_id, context=context)
        return True

    def copy(self, cr, uid, id, default=None, context=None):
        if default is None:
            default = {}
        context = dict(context or {})
        context['active_test'] = False
        proj = self.browse(cr, uid, id, context=context)
        if not default.get('name'):
            default.update(name=_("%s (copy)") % (proj.name))
        res = super(project, self).copy(cr, uid, id, default, context)
        self.map_tasks(cr, uid, id, res, context=context)
        return res

    def duplicate_template(self, cr, uid, ids, context=None):
        context = dict(context or {})
        data_obj = self.pool.get('ir.model.data')
        result = []
        for proj in self.browse(cr, uid, ids, context=context):
            parent_id = context.get('parent_id', False)
            context.update({'analytic_project_copy': True})
            new_date_start = time.strftime('%Y-%m-%d')
            new_date_end = False
            if proj.date_start and proj.date:
                start_date = date(*time.strptime(proj.date_start,'%Y-%m-%d')[:3])
                end_date = date(*time.strptime(proj.date,'%Y-%m-%d')[:3])
                new_date_end = (datetime(*time.strptime(new_date_start,'%Y-%m-%d')[:3])+(end_date-start_date)).strftime('%Y-%m-%d')
            context.update({'copy':True})
            new_id = self.copy(cr, uid, proj.id, default = {
                                    'name':_("%s (copy)") % (proj.name),
                                    'state':'open',
                                    'date_start':new_date_start,
                                    'date':new_date_end,
                                    'parent_id':parent_id}, context=context)
            result.append(new_id)

            child_ids = self.search(cr, uid, [('parent_id','=', proj.analytic_account_id.id)], context=context)
            parent_id = self.read(cr, uid, new_id, ['analytic_account_id'])['analytic_account_id'][0]
            if child_ids:
                self.duplicate_template(cr, uid, child_ids, context={'parent_id': parent_id})

        if result and len(result):
            res_id = result[0]
            form_view_id = data_obj._get_id(cr, uid, 'project', 'edit_project')
            form_view = data_obj.read(cr, uid, form_view_id, ['res_id'])
            tree_view_id = data_obj._get_id(cr, uid, 'project', 'view_project')
            tree_view = data_obj.read(cr, uid, tree_view_id, ['res_id'])
            search_view_id = data_obj._get_id(cr, uid, 'project', 'view_project_project_filter')
            search_view = data_obj.read(cr, uid, search_view_id, ['res_id'])
            return {
                'name': _('Projects'),
                'view_type': 'form',
                'view_mode': 'form,tree',
                'res_model': 'project.project',
                'view_id': False,
                'res_id': res_id,
                'views': [(form_view['res_id'],'form'),(tree_view['res_id'],'tree')],
                'type': 'ir.actions.act_window',
                'search_view_id': search_view['res_id'],
                'nodestroy': True
            }

    # set active value for a project, its sub projects and its tasks
    def setActive(self, cr, uid, ids, value=True, context=None):
        task_obj = self.pool.get('project.task')
        for proj in self.browse(cr, uid, ids, context=None):
            self.write(cr, uid, [proj.id], {'state': value and 'open' or 'template'}, context)
            cr.execute('select id from project_task where project_id=%s', (proj.id,))
            tasks_id = [x[0] for x in cr.fetchall()]
            if tasks_id:
                task_obj.write(cr, uid, tasks_id, {'active': value}, context=context)
            child_ids = self.search(cr, uid, [('parent_id','=', proj.analytic_account_id.id)])
            if child_ids:
                self.setActive(cr, uid, child_ids, value, context=None)
        return True

    def _schedule_header(self, cr, uid, ids, force_members=True, context=None):
        context = context or {}
        if type(ids) in (long, int,):
            ids = [ids]
        projects = self.browse(cr, uid, ids, context=context)

        for project in projects:
            if (not project.members) and force_members:
                raise osv.except_osv(_('Warning!'),_("You must assign members on the project '%s'!") % (project.name,))

        resource_pool = self.pool.get('resource.resource')

        result = "from openerp.addons.resource.faces import *\n"
        result += "import datetime\n"
        for project in self.browse(cr, uid, ids, context=context):
            u_ids = [i.id for i in project.members]
            if project.user_id and (project.user_id.id not in u_ids):
                u_ids.append(project.user_id.id)
            for task in project.tasks:
                if task.user_id and (task.user_id.id not in u_ids):
                    u_ids.append(task.user_id.id)
            calendar_id = project.resource_calendar_id and project.resource_calendar_id.id or False
            resource_objs = resource_pool.generate_resources(cr, uid, u_ids, calendar_id, context=context)
            for key, vals in resource_objs.items():
                result +='''
class User_%s(Resource):
    efficiency = %s
''' % (key,  vals.get('efficiency', False))

        result += '''
def Project():
        '''
        return result

    def _schedule_project(self, cr, uid, project, context=None):
        resource_pool = self.pool.get('resource.resource')
        calendar_id = project.resource_calendar_id and project.resource_calendar_id.id or False
        working_days = resource_pool.compute_working_calendar(cr, uid, calendar_id, context=context)
        # TODO: check if we need working_..., default values are ok.
        puids = [x.id for x in project.members]
        if project.user_id:
            puids.append(project.user_id.id)
        result = """
  def Project_%d():
    start = \'%s\'
    working_days = %s
    resource = %s
"""       % (
            project.id,
            project.date_start or time.strftime('%Y-%m-%d'), working_days,
            '|'.join(['User_'+str(x) for x in puids]) or 'None'
        )
        vacation = calendar_id and tuple(resource_pool.compute_vacation(cr, uid, calendar_id, context=context)) or False
        if vacation:
            result+= """
    vacation = %s
""" %   ( vacation, )
        return result

    #TODO: DO Resource allocation and compute availability
    def compute_allocation(self, rc, uid, ids, start_date, end_date, context=None):
        if context ==  None:
            context = {}
        allocation = {}
        return allocation

    def schedule_tasks(self, cr, uid, ids, context=None):
        context = context or {}
        if type(ids) in (long, int,):
            ids = [ids]
        projects = self.browse(cr, uid, ids, context=context)
        result = self._schedule_header(cr, uid, ids, False, context=context)
        for project in projects:
            result += self._schedule_project(cr, uid, project, context=context)
            result += self.pool.get('project.task')._generate_task(cr, uid, project.tasks, ident=4, context=context)

        local_dict = {}
        exec result in local_dict
        projects_gantt = Task.BalancedProject(local_dict['Project'])

        for project in projects:
            project_gantt = getattr(projects_gantt, 'Project_%d' % (project.id,))
            for task in project.tasks:
                if task.stage_id and task.stage_id.fold:
                    continue

                p = getattr(project_gantt, 'Task_%d' % (task.id,))

                self.pool.get('project.task').write(cr, uid, [task.id], {
                    'date_start': p.start.strftime('%Y-%m-%d %H:%M:%S'),
                    'date_end': p.end.strftime('%Y-%m-%d %H:%M:%S')
                }, context=context)
                if (not task.user_id) and (p.booked_resource):
                    self.pool.get('project.task').write(cr, uid, [task.id], {
                        'user_id': int(p.booked_resource[0].name[5:]),
                    }, context=context)
        return True

    def create(self, cr, uid, vals, context=None):
        if context is None:
            context = {}
        # Prevent double project creation when 'use_tasks' is checked + alias management
        create_context = dict(context, project_creation_in_progress=True,
                              alias_model_name=vals.get('alias_model', 'project.task'),
                              alias_parent_model_name=self._name)

        if vals.get('type', False) not in ('template', 'contract'):
            vals['type'] = 'contract'

        project_id = super(project, self).create(cr, uid, vals, context=create_context)
        project_rec = self.browse(cr, uid, project_id, context=context)
        ir_values = self.pool.get('ir.values').get_default( cr, uid, 'project.config.settings', 'generate_project_alias' )
        values = { 'alias_parent_thread_id': project_id, 'alias_defaults': {'project_id': project_id}}
        if ir_values:
            values = dict(values, alias_name=vals['name'])
        self.pool.get('mail.alias').write(cr, uid, [project_rec.alias_id.id], values, context=context)
        return project_id

    def write(self, cr, uid, ids, vals, context=None):
        # if alias_model has been changed, update alias_model_id accordingly
        if vals.get('alias_model'):
            model_ids = self.pool.get('ir.model').search(cr, uid, [('model', '=', vals.get('alias_model', 'project.task'))])
            vals.update(alias_model_id=model_ids[0])
        return super(project, self).write(cr, uid, ids, vals, context=context)


class task(osv.osv):
    _name = "project.task"
    _description = "Task"
    _date_name = "date_start"
    _inherit = ['mail.thread', 'ir.needaction_mixin']

    _mail_post_access = 'read'
    _track = {
        'stage_id': {
            # this is only an heuristics; depending on your particular stage configuration it may not match all 'new' stages
            'project.mt_task_new': lambda self, cr, uid, obj, ctx=None: obj.stage_id and obj.stage_id.sequence <= 1,
            'project.mt_task_stage': lambda self, cr, uid, obj, ctx=None: obj.stage_id.sequence > 1,
        },
        'user_id': {
            'project.mt_task_assigned': lambda self, cr, uid, obj, ctx=None: obj.user_id and obj.user_id.id,
        },
        'kanban_state': {
            'project.mt_task_blocked': lambda self, cr, uid, obj, ctx=None: obj.kanban_state == 'blocked',
            'project.mt_task_ready': lambda self, cr, uid, obj, ctx=None: obj.kanban_state == 'done',
        },
    }

    def _get_default_partner(self, cr, uid, context=None):
        project_id = self._get_default_project_id(cr, uid, context)
        if project_id:
            project = self.pool.get('project.project').browse(cr, uid, project_id, context=context)
            if project and project.partner_id:
                return project.partner_id.id
        return False

    def _get_default_project_id(self, cr, uid, context=None):
        """ Gives default section by checking if present in the context """
        return (self._resolve_project_id_from_context(cr, uid, context=context) or False)

    def _get_default_stage_id(self, cr, uid, context=None):
        """ Gives default stage_id """
        project_id = self._get_default_project_id(cr, uid, context=context)
        return self.stage_find(cr, uid, [], project_id, [('fold', '=', False)], context=context)

    def _resolve_project_id_from_context(self, cr, uid, context=None):
        """ Returns ID of project based on the value of 'default_project_id'
            context key, or None if it cannot be resolved to a single
            project.
        """
        if context is None:
            context = {}
        if type(context.get('default_project_id')) in (int, long):
            return context['default_project_id']
        if isinstance(context.get('default_project_id'), basestring):
            project_name = context['default_project_id']
            project_ids = self.pool.get('project.project').name_search(cr, uid, name=project_name, context=context)
            if len(project_ids) == 1:
                return project_ids[0][0]
        return None

    def _read_group_stage_ids(self, cr, uid, ids, domain, read_group_order=None, access_rights_uid=None, context=None):
        stage_obj = self.pool.get('project.task.type')
        order = stage_obj._order
        access_rights_uid = access_rights_uid or uid
        if read_group_order == 'stage_id desc':
            order = '%s desc' % order
        search_domain = []
        project_id = self._resolve_project_id_from_context(cr, uid, context=context)
        if project_id:
            search_domain += ['|', ('project_ids', '=', project_id)]
        search_domain += [('id', 'in', ids)]
        stage_ids = stage_obj._search(cr, uid, search_domain, order=order, access_rights_uid=access_rights_uid, context=context)
        result = stage_obj.name_get(cr, access_rights_uid, stage_ids, context=context)
        # restore order of the search
        result.sort(lambda x,y: cmp(stage_ids.index(x[0]), stage_ids.index(y[0])))

        fold = {}
        for stage in stage_obj.browse(cr, access_rights_uid, stage_ids, context=context):
            fold[stage.id] = stage.fold or False
        return result, fold

    def _read_group_user_id(self, cr, uid, ids, domain, read_group_order=None, access_rights_uid=None, context=None):
        res_users = self.pool.get('res.users')
        project_id = self._resolve_project_id_from_context(cr, uid, context=context)
        access_rights_uid = access_rights_uid or uid
        if project_id:
            ids += self.pool.get('project.project').read(cr, access_rights_uid, project_id, ['members'], context=context)['members']
            order = res_users._order
            # lame way to allow reverting search, should just work in the trivial case
            if read_group_order == 'user_id desc':
                order = '%s desc' % order
            # de-duplicate and apply search order
            ids = res_users._search(cr, uid, [('id','in',ids)], order=order, access_rights_uid=access_rights_uid, context=context)
        result = res_users.name_get(cr, access_rights_uid, ids, context=context)
        # restore order of the search
        result.sort(lambda x,y: cmp(ids.index(x[0]), ids.index(y[0])))
        return result, {}

    _group_by_full = {
        'stage_id': _read_group_stage_ids,
        'user_id': _read_group_user_id,
    }

    def _str_get(self, task, level=0, border='***', context=None):
        return border+' '+(task.user_id and task.user_id.name.upper() or '')+(level and (': L'+str(level)) or '')+(' - %.1fh / %.1fh'%(task.effective_hours or 0.0,task.planned_hours))+' '+border+'\n'+ \
            border[0]+' '+(task.name or '')+'\n'+ \
            (task.description or '')+'\n\n'

    # Compute: effective_hours, total_hours, progress
    def _hours_get(self, cr, uid, ids, field_names, args, context=None):
        res = {}
        cr.execute("SELECT task_id, COALESCE(SUM(hours),0) FROM project_task_work WHERE task_id IN %s GROUP BY task_id",(tuple(ids),))
        hours = dict(cr.fetchall())
        for task in self.browse(cr, uid, ids, context=context):
            res[task.id] = {'effective_hours': hours.get(task.id, 0.0), 'total_hours': (task.remaining_hours or 0.0) + hours.get(task.id, 0.0)}
            res[task.id]['delay_hours'] = res[task.id]['total_hours'] - task.planned_hours
            res[task.id]['progress'] = 0.0
            if (task.remaining_hours + hours.get(task.id, 0.0)):
                res[task.id]['progress'] = round(min(100.0 * hours.get(task.id, 0.0) / res[task.id]['total_hours'], 99.99),2)
            # TDE CHECK: if task.state in ('done','cancelled'):
            if task.stage_id and task.stage_id.fold:
                res[task.id]['progress'] = 100.0
        return res

    def onchange_remaining(self, cr, uid, ids, remaining=0.0, planned=0.0):
        if remaining and not planned:
            return {'value': {'planned_hours': remaining}}
        return {}

    def onchange_planned(self, cr, uid, ids, planned=0.0, effective=0.0):
        return {'value': {'remaining_hours': planned - effective}}

    def onchange_project(self, cr, uid, id, project_id, context=None):
        if project_id:
            project = self.pool.get('project.project').browse(cr, uid, project_id, context=context)
            if project and project.partner_id:
                return {'value': {'partner_id': project.partner_id.id}}
        return {}

    def onchange_user_id(self, cr, uid, ids, user_id, context=None):
        vals = {}
        if user_id:
            vals['date_start'] = fields.datetime.now()
        return {'value': vals}

    def duplicate_task(self, cr, uid, map_ids, context=None):
        mapper = lambda t: map_ids.get(t.id, t.id)
        for task in self.browse(cr, uid, map_ids.values(), context):
            new_child_ids = set(map(mapper, task.child_ids))
            new_parent_ids = set(map(mapper, task.parent_ids))
            if new_child_ids or new_parent_ids:
                task.write({'parent_ids': [(6,0,list(new_parent_ids))],
                            'child_ids':  [(6,0,list(new_child_ids))]})

    def copy_data(self, cr, uid, id, default=None, context=None):
        if default is None:
            default = {}
        if not default.get('name'):
            current = self.browse(cr, uid, id, context=context)       
            default['name'] = _("%s (copy)") % current.name
        return super(task, self).copy_data(cr, uid, id, default, context)
    
    def _is_template(self, cr, uid, ids, field_name, arg, context=None):
        res = {}
        for task in self.browse(cr, uid, ids, context=context):
            res[task.id] = True
            if task.project_id:
                if task.project_id.active == False or task.project_id.state == 'template':
                    res[task.id] = False
        return res

    def _get_task(self, cr, uid, ids, context=None):
        result = {}
        for work in self.pool.get('project.task.work').browse(cr, uid, ids, context=context):
            if work.task_id: result[work.task_id.id] = True
        return result.keys()

    _columns = {
        'active': fields.function(_is_template, store=True, string='Not a Template Task', type='boolean', help="This field is computed automatically and have the same behavior than the boolean 'active' field: if the task is linked to a template or unactivated project, it will be hidden unless specifically asked."),
        'name': fields.char('Task Summary', track_visibility='onchange', size=128, required=True, select=True),
        'description': fields.text('Description'),
        'priority': fields.selection([('0','Low'), ('1','Normal'), ('2','High')], 'Priority', select=True),
        'sequence': fields.integer('Sequence', select=True, help="Gives the sequence order when displaying a list of tasks."),
        'stage_id': fields.many2one('project.task.type', 'Stage', track_visibility='onchange', select=True,
                        domain="[('project_ids', '=', project_id)]", copy=False),
        'categ_ids': fields.many2many('project.category', string='Tags'),
        'kanban_state': fields.selection([('normal', 'In Progress'),('blocked', 'Blocked'),('done', 'Ready for next stage')], 'Kanban State',
                                         track_visibility='onchange',
                                         help="A task's kanban state indicates special situations affecting it:\n"
                                              " * Normal is the default situation\n"
                                              " * Blocked indicates something is preventing the progress of this task\n"
                                              " * Ready for next stage indicates the task is ready to be pulled to the next stage",
                                         required=False, copy=False),
        'create_date': fields.datetime('Create Date', readonly=True, select=True),
        'write_date': fields.datetime('Last Modification Date', readonly=True, select=True), #not displayed in the view but it might be useful with base_action_rule module (and it needs to be defined first for that)
        'date_start': fields.datetime('Starting Date', select=True, copy=False),
        'date_end': fields.datetime('Ending Date', select=True, copy=False),
        'date_deadline': fields.date('Deadline', select=True, copy=False),
        'date_last_stage_update': fields.datetime('Last Stage Update', select=True, copy=False),
        'project_id': fields.many2one('project.project', 'Project', ondelete='set null', select=True, track_visibility='onchange', change_default=True),
        'parent_ids': fields.many2many('project.task', 'project_task_parent_rel', 'task_id', 'parent_id', 'Parent Tasks'),
        'child_ids': fields.many2many('project.task', 'project_task_parent_rel', 'parent_id', 'task_id', 'Delegated Tasks'),
        'notes': fields.text('Notes'),
        'planned_hours': fields.float('Initially Planned Hours', help='Estimated time to do the task, usually set by the project manager when the task is in draft state.'),
        'effective_hours': fields.function(_hours_get, string='Hours Spent', multi='hours', help="Computed using the sum of the task work done.",
            store = {
                'project.task': (lambda self, cr, uid, ids, c={}: ids, ['work_ids', 'remaining_hours', 'planned_hours'], 10),
                'project.task.work': (_get_task, ['hours'], 10),
            }),
        'remaining_hours': fields.float('Remaining Hours', digits=(16,2), help="Total remaining time, can be re-estimated periodically by the assignee of the task."),
        'total_hours': fields.function(_hours_get, string='Total', multi='hours', help="Computed as: Time Spent + Remaining Time.",
            store = {
                'project.task': (lambda self, cr, uid, ids, c={}: ids, ['work_ids', 'remaining_hours', 'planned_hours'], 10),
                'project.task.work': (_get_task, ['hours'], 10),
            }),
        'progress': fields.function(_hours_get, string='Working Time Progress (%)', multi='hours', group_operator="avg", help="If the task has a progress of 99.99% you should close the task if it's finished or reevaluate the time",
            store = {
                'project.task': (lambda self, cr, uid, ids, c={}: ids, ['work_ids', 'remaining_hours', 'planned_hours', 'state', 'stage_id'], 10),
                'project.task.work': (_get_task, ['hours'], 10),
            }),
        'delay_hours': fields.function(_hours_get, string='Delay Hours', multi='hours', help="Computed as difference between planned hours by the project manager and the total hours of the task.",
            store = {
                'project.task': (lambda self, cr, uid, ids, c={}: ids, ['work_ids', 'remaining_hours', 'planned_hours'], 10),
                'project.task.work': (_get_task, ['hours'], 10),
            }),
        'reviewer_id': fields.many2one('res.users', 'Reviewer', select=True, track_visibility='onchange'),
        'user_id': fields.many2one('res.users', 'Assigned to', select=True, track_visibility='onchange'),
        'delegated_user_id': fields.related('child_ids', 'user_id', type='many2one', relation='res.users', string='Delegated To'),
        'partner_id': fields.many2one('res.partner', 'Customer'),
        'work_ids': fields.one2many('project.task.work', 'task_id', 'Work done'),
        'manager_id': fields.related('project_id', 'analytic_account_id', 'user_id', type='many2one', relation='res.users', string='Project Manager'),
        'company_id': fields.many2one('res.company', 'Company'),
        'id': fields.integer('ID', readonly=True),
        'color': fields.integer('Color Index'),
        'user_email': fields.related('user_id', 'email', type='char', string='User Email', readonly=True),
    }
    _defaults = {
        'stage_id': _get_default_stage_id,
        'project_id': _get_default_project_id,
        'date_last_stage_update': fields.datetime.now,
        'kanban_state': 'normal',
        'priority': '1',
        'progress': 0,
        'sequence': 10,
        'active': True,
        'reviewer_id': lambda obj, cr, uid, ctx=None: uid,
        'user_id': lambda obj, cr, uid, ctx=None: uid,
        'company_id': lambda self, cr, uid, ctx=None: self.pool.get('res.company')._company_default_get(cr, uid, 'project.task', context=ctx),
        'partner_id': lambda self, cr, uid, ctx=None: self._get_default_partner(cr, uid, context=ctx),
    }
    _order = "priority, sequence, date_start, name, id"
    
    def _check_recursion(self, cr, uid, ids, context=None):
        for id in ids:
            visited_branch = set()
            visited_node = set()
            res = self._check_cycle(cr, uid, id, visited_branch, visited_node, context=context)
            if not res:
                return False

        return True

    def _check_cycle(self, cr, uid, id, visited_branch, visited_node, context=None):
        if id in visited_branch: #Cycle
            return False

        if id in visited_node: #Already tested don't work one more time for nothing
            return True

        visited_branch.add(id)
        visited_node.add(id)

        #visit child using DFS
        task = self.browse(cr, uid, id, context=context)
        for child in task.child_ids:
            res = self._check_cycle(cr, uid, child.id, visited_branch, visited_node, context=context)
            if not res:
                return False

        visited_branch.remove(id)
        return True

    def _check_dates(self, cr, uid, ids, context=None):
        if context == None:
            context = {}
        obj_task = self.browse(cr, uid, ids[0], context=context)
        start = obj_task.date_start or False
        end = obj_task.date_end or False
        if start and end :
            if start > end:
                return False
        return True

    _constraints = [
        (_check_recursion, 'Error ! You cannot create recursive tasks.', ['parent_ids']),
        (_check_dates, 'Error ! Task end-date must be greater then task start-date', ['date_start','date_end'])
    ]

    # Override view according to the company definition
    def fields_view_get(self, cr, uid, view_id=None, view_type='form', context=None, toolbar=False, submenu=False):
        users_obj = self.pool.get('res.users')
        if context is None: context = {}
        # read uom as admin to avoid access rights issues, e.g. for portal/share users,
        # this should be safe (no context passed to avoid side-effects)
        obj_tm = users_obj.browse(cr, SUPERUSER_ID, uid, context=context).company_id.project_time_mode_id
        tm = obj_tm and obj_tm.name or 'Hours'

        res = super(task, self).fields_view_get(cr, uid, view_id, view_type, context, toolbar, submenu=submenu)

        if tm in ['Hours','Hour']:
            return res

        eview = etree.fromstring(res['arch'])

        def _check_rec(eview):
            if eview.attrib.get('widget','') == 'float_time':
                eview.set('widget','float')
            for child in eview:
                _check_rec(child)
            return True

        _check_rec(eview)

        res['arch'] = etree.tostring(eview)

        for f in res['fields']:
            if 'Hours' in res['fields'][f]['string']:
                res['fields'][f]['string'] = res['fields'][f]['string'].replace('Hours',tm)
        return res

    def get_empty_list_help(self, cr, uid, help, context=None):
        context = dict(context or {})
        context['empty_list_help_id'] = context.get('default_project_id')
        context['empty_list_help_model'] = 'project.project'
        context['empty_list_help_document_name'] = _("tasks")
        return super(task, self).get_empty_list_help(cr, uid, help, context=context)

    # ----------------------------------------
    # Case management
    # ----------------------------------------

    def stage_find(self, cr, uid, cases, section_id, domain=[], order='sequence', context=None):
        """ Override of the base.stage method
            Parameter of the stage search taken from the lead:
            - section_id: if set, stages must belong to this section or
              be a default stage; if not set, stages must be default
              stages
        """
        if isinstance(cases, (int, long)):
            cases = self.browse(cr, uid, cases, context=context)
        # collect all section_ids
        section_ids = []
        if section_id:
            section_ids.append(section_id)
        for task in cases:
            if task.project_id:
                section_ids.append(task.project_id.id)
        search_domain = []
        if section_ids:
            search_domain = [('|')] * (len(section_ids) - 1)
            for section_id in section_ids:
                search_domain.append(('project_ids', '=', section_id))
        search_domain += list(domain)
        # perform search, return the first found
        stage_ids = self.pool.get('project.task.type').search(cr, uid, search_domain, order=order, context=context)
        if stage_ids:
            return stage_ids[0]
        return False

    def _check_child_task(self, cr, uid, ids, context=None):
        if context == None:
            context = {}
        tasks = self.browse(cr, uid, ids, context=context)
        for task in tasks:
            if task.child_ids:
                for child in task.child_ids:
                    if child.stage_id and not child.stage_id.fold:
                        raise osv.except_osv(_("Warning!"), _("Child task still open.\nPlease cancel or complete child task first."))
        return True

    def _delegate_task_attachments(self, cr, uid, task_id, delegated_task_id, context=None):
        attachment = self.pool.get('ir.attachment')
        attachment_ids = attachment.search(cr, uid, [('res_model', '=', self._name), ('res_id', '=', task_id)], context=context)
        new_attachment_ids = []
        for attachment_id in attachment_ids:
            new_attachment_ids.append(attachment.copy(cr, uid, attachment_id, default={'res_id': delegated_task_id}, context=context))
        return new_attachment_ids

    def do_delegate(self, cr, uid, ids, delegate_data=None, context=None):
        """
        Delegate Task to another users.
        """
        if delegate_data is None:
            delegate_data = {}
        assert delegate_data['user_id'], _("Delegated User should be specified")
        delegated_tasks = {}
        for task in self.browse(cr, uid, ids, context=context):
            delegated_task_id = self.copy(cr, uid, task.id, {
                'name': delegate_data['name'],
                'project_id': delegate_data['project_id'] and delegate_data['project_id'][0] or False,
                'stage_id': delegate_data.get('stage_id') and delegate_data.get('stage_id')[0] or False,
                'user_id': delegate_data['user_id'] and delegate_data['user_id'][0] or False,
                'planned_hours': delegate_data['planned_hours'] or 0.0,
                'parent_ids': [(6, 0, [task.id])],
                'description': delegate_data['new_task_description'] or '',
                'child_ids': [],
                'work_ids': []
            }, context=context)
            self._delegate_task_attachments(cr, uid, task.id, delegated_task_id, context=context)
            newname = delegate_data['prefix'] or ''
            task.write({
                'remaining_hours': delegate_data['planned_hours_me'],
                'planned_hours': delegate_data['planned_hours_me'] + (task.effective_hours or 0.0),
                'name': newname,
            }, context=context)
            delegated_tasks[task.id] = delegated_task_id
        return delegated_tasks

    def set_remaining_time(self, cr, uid, ids, remaining_time=1.0, context=None):
        for task in self.browse(cr, uid, ids, context=context):
            if (task.stage_id and task.stage_id.sequence <= 1) or (task.planned_hours == 0.0):
                self.write(cr, uid, [task.id], {'planned_hours': remaining_time}, context=context)
        self.write(cr, uid, ids, {'remaining_hours': remaining_time}, context=context)
        return True

    def set_remaining_time_1(self, cr, uid, ids, context=None):
        return self.set_remaining_time(cr, uid, ids, 1.0, context)

    def set_remaining_time_2(self, cr, uid, ids, context=None):
        return self.set_remaining_time(cr, uid, ids, 2.0, context)

    def set_remaining_time_5(self, cr, uid, ids, context=None):
        return self.set_remaining_time(cr, uid, ids, 5.0, context)

    def set_remaining_time_10(self, cr, uid, ids, context=None):
        return self.set_remaining_time(cr, uid, ids, 10.0, context)

    def _store_history(self, cr, uid, ids, context=None):
        for task in self.browse(cr, uid, ids, context=context):
            self.pool.get('project.task.history').create(cr, uid, {
                'task_id': task.id,
                'remaining_hours': task.remaining_hours,
                'planned_hours': task.planned_hours,
                'kanban_state': task.kanban_state,
                'type_id': task.stage_id.id,
                'user_id': task.user_id.id

            }, context=context)
        return True

    # ------------------------------------------------
    # CRUD overrides
    # ------------------------------------------------

    def create(self, cr, uid, vals, context=None):
        context = dict(context or {})

        # for default stage
        if vals.get('project_id') and not context.get('default_project_id'):
            context['default_project_id'] = vals.get('project_id')
        # user_id change: update date_start
        if vals.get('user_id') and not vals.get('start_date'):
            vals['date_start'] = fields.datetime.now()

        # context: no_log, because subtype already handle this
        create_context = dict(context, mail_create_nolog=True)
        task_id = super(task, self).create(cr, uid, vals, context=create_context)
        self._store_history(cr, uid, [task_id], context=context)
        return task_id

    def write(self, cr, uid, ids, vals, context=None):
        if isinstance(ids, (int, long)):
            ids = [ids]

        # stage change: update date_last_stage_update
        if 'stage_id' in vals:
            vals['date_last_stage_update'] = fields.datetime.now()
        # user_id change: update date_start
        if vals.get('user_id') and 'date_start' not in vals:
            vals['date_start'] = fields.datetime.now()

        # Overridden to reset the kanban_state to normal whenever
        # the stage (stage_id) of the task changes.
        if vals and not 'kanban_state' in vals and 'stage_id' in vals:
            new_stage = vals.get('stage_id')
            vals_reset_kstate = dict(vals, kanban_state='normal')
            for t in self.browse(cr, uid, ids, context=context):
                write_vals = vals_reset_kstate if t.stage_id.id != new_stage else vals
                super(task, self).write(cr, uid, [t.id], write_vals, context=context)
            result = True
        else:
            result = super(task, self).write(cr, uid, ids, vals, context=context)

        if any(item in vals for item in ['stage_id', 'remaining_hours', 'user_id', 'kanban_state']):
            self._store_history(cr, uid, ids, context=context)
        return result

    def unlink(self, cr, uid, ids, context=None):
        if context == None:
            context = {}
        self._check_child_task(cr, uid, ids, context=context)
        res = super(task, self).unlink(cr, uid, ids, context)
        return res

    def _generate_task(self, cr, uid, tasks, ident=4, context=None):
        context = context or {}
        result = ""
        ident = ' '*ident
        for task in tasks:
            if task.stage_id and task.stage_id.fold:
                continue
            result += '''
%sdef Task_%s():
%s  todo = \"%.2fH\"
%s  effort = \"%.2fH\"''' % (ident,task.id, ident,task.remaining_hours, ident,task.total_hours)
            start = []
            for t2 in task.parent_ids:
                start.append("up.Task_%s.end" % (t2.id,))
            if start:
                result += '''
%s  start = max(%s)
''' % (ident,','.join(start))

            if task.user_id:
                result += '''
%s  resource = %s
''' % (ident, 'User_'+str(task.user_id.id))

        result += "\n"
        return result

    # ---------------------------------------------------
    # Mail gateway
    # ---------------------------------------------------

    def _message_get_auto_subscribe_fields(self, cr, uid, updated_fields, auto_follow_fields=None, context=None):
        if auto_follow_fields is None:
            auto_follow_fields = ['user_id', 'reviewer_id']
        return super(task, self)._message_get_auto_subscribe_fields(cr, uid, updated_fields, auto_follow_fields, context=context)

    def message_get_reply_to(self, cr, uid, ids, context=None):
        """ Override to get the reply_to of the parent project. """
        tasks = self.browse(cr, SUPERUSER_ID, ids, context=context)
        project_ids = set([task.project_id.id for task in tasks if task.project_id])
        aliases = self.pool['project.project'].message_get_reply_to(cr, uid, list(project_ids), context=context)
        return dict((task.id, aliases.get(task.project_id and task.project_id.id or 0, False)) for task in tasks)

    def message_new(self, cr, uid, msg, custom_values=None, context=None):
        """ Override to updates the document according to the email. """
        if custom_values is None:
            custom_values = {}
        defaults = {
            'name': msg.get('subject'),
            'planned_hours': 0.0,
        }
        defaults.update(custom_values)
        return super(task, self).message_new(cr, uid, msg, custom_values=defaults, context=context)

    def message_update(self, cr, uid, ids, msg, update_vals=None, context=None):
        """ Override to update the task according to the email. """
        if update_vals is None:
            update_vals = {}
        maps = {
            'cost': 'planned_hours',
        }
        for line in msg['body'].split('\n'):
            line = line.strip()
            res = tools.command_re.match(line)
            if res:
                match = res.group(1).lower()
                field = maps.get(match)
                if field:
                    try:
                        update_vals[field] = float(res.group(2).lower())
                    except (ValueError, TypeError):
                        pass
        return super(task, self).message_update(cr, uid, ids, msg, update_vals=update_vals, context=context)

class project_work(osv.osv):
    _name = "project.task.work"
    _description = "Project Task Work"
    _columns = {
        'name': fields.char('Work summary'),
        'date': fields.datetime('Date', select="1"),
        'task_id': fields.many2one('project.task', 'Task', ondelete='cascade', required=True, select="1"),
        'hours': fields.float('Time Spent'),
        'user_id': fields.many2one('res.users', 'Done by', required=True, select="1"),
        'company_id': fields.related('task_id', 'company_id', type='many2one', relation='res.company', string='Company', store=True, readonly=True)
    }

    _defaults = {
        'user_id': lambda obj, cr, uid, context: uid,
        'date': lambda *a: time.strftime('%Y-%m-%d %H:%M:%S')
    }

    _order = "date desc"
    def create(self, cr, uid, vals, context=None):
        if 'hours' in vals and (not vals['hours']):
            vals['hours'] = 0.00
        if 'task_id' in vals:
            cr.execute('update project_task set remaining_hours=remaining_hours - %s where id=%s', (vals.get('hours',0.0), vals['task_id']))
            self.pool.get('project.task').invalidate_cache(cr, uid, ['remaining_hours'], [vals['task_id']], context=context)
        return super(project_work,self).create(cr, uid, vals, context=context)

    def write(self, cr, uid, ids, vals, context=None):
        if 'hours' in vals and (not vals['hours']):
            vals['hours'] = 0.00
        if 'hours' in vals:
            task_obj = self.pool.get('project.task')
            for work in self.browse(cr, uid, ids, context=context):
                cr.execute('update project_task set remaining_hours=remaining_hours - %s + (%s) where id=%s', (vals.get('hours',0.0), work.hours, work.task_id.id))
                task_obj.invalidate_cache(cr, uid, ['remaining_hours'], [work.task_id.id], context=context)
        return super(project_work,self).write(cr, uid, ids, vals, context)

    def unlink(self, cr, uid, ids, context=None):
        task_obj = self.pool.get('project.task')
        for work in self.browse(cr, uid, ids):
            cr.execute('update project_task set remaining_hours=remaining_hours + %s where id=%s', (work.hours, work.task_id.id))
            task_obj.invalidate_cache(cr, uid, ['remaining_hours'], [work.task_id.id], context=context)
        return super(project_work,self).unlink(cr, uid, ids, context=context)


class account_analytic_account(osv.osv):
    _inherit = 'account.analytic.account'
    _description = 'Analytic Account'
    _columns = {
        'use_tasks': fields.boolean('Tasks',help="If checked, this contract will be available in the project menu and you will be able to manage tasks or track issues"),
        'company_uom_id': fields.related('company_id', 'project_time_mode_id', type='many2one', relation='product.uom'),
    }

    def on_change_template(self, cr, uid, ids, template_id, date_start=False, context=None):
        res = super(account_analytic_account, self).on_change_template(cr, uid, ids, template_id, date_start=date_start, context=context)
        if template_id and 'value' in res:
            template = self.browse(cr, uid, template_id, context=context)
            res['value']['use_tasks'] = template.use_tasks
        return res

    def _trigger_project_creation(self, cr, uid, vals, context=None):
        '''
        This function is used to decide if a project needs to be automatically created or not when an analytic account is created. It returns True if it needs to be so, False otherwise.
        '''
        if context is None: context = {}
        return vals.get('use_tasks') and not 'project_creation_in_progress' in context

    def project_create(self, cr, uid, analytic_account_id, vals, context=None):
        '''
        This function is called at the time of analytic account creation and is used to create a project automatically linked to it if the conditions are meet.
        '''
        project_pool = self.pool.get('project.project')
        project_id = project_pool.search(cr, uid, [('analytic_account_id','=', analytic_account_id)])
        if not project_id and self._trigger_project_creation(cr, uid, vals, context=context):
            project_values = {
                'name': vals.get('name'),
                'analytic_account_id': analytic_account_id,
                'type': vals.get('type','contract'),
            }
            return project_pool.create(cr, uid, project_values, context=context)
        return False

    def create(self, cr, uid, vals, context=None):
        if context is None:
            context = {}
        if vals.get('child_ids', False) and context.get('analytic_project_copy', False):
            vals['child_ids'] = []
        analytic_account_id = super(account_analytic_account, self).create(cr, uid, vals, context=context)
        self.project_create(cr, uid, analytic_account_id, vals, context=context)
        return analytic_account_id

    def write(self, cr, uid, ids, vals, context=None):
        if isinstance(ids, (int, long)):
            ids = [ids]
        vals_for_project = vals.copy()
        for account in self.browse(cr, uid, ids, context=context):
            if not vals.get('name'):
                vals_for_project['name'] = account.name
            if not vals.get('type'):
                vals_for_project['type'] = account.type
            self.project_create(cr, uid, account.id, vals_for_project, context=context)
        return super(account_analytic_account, self).write(cr, uid, ids, vals, context=context)

    def unlink(self, cr, uid, ids, context=None):
        proj_ids = self.pool['project.project'].search(cr, uid, [('analytic_account_id', 'in', ids)])
        has_tasks = self.pool['project.task'].search(cr, uid, [('project_id', 'in', proj_ids)], count=True, context=context)
        if has_tasks:
            raise osv.except_osv(_('Warning!'), _('Please remove existing tasks in the project linked to the accounts you want to delete.'))
        return super(account_analytic_account, self).unlink(cr, uid, ids, context=context)

    def name_search(self, cr, uid, name, args=None, operator='ilike', context=None, limit=100):
        if args is None:
            args = []
        if context is None:
            context={}
        if context.get('current_model') == 'project.project':
            project_ids = self.search(cr, uid, args + [('name', operator, name)], limit=limit, context=context)
            return self.name_get(cr, uid, project_ids, context=context)

        return super(account_analytic_account, self).name_search(cr, uid, name, args=args, operator=operator, context=context, limit=limit)


class project_project(osv.osv):
    _inherit = 'project.project'
    _defaults = {
        'use_tasks': True
    }

class project_task_history(osv.osv):
    """
    Tasks History, used for cumulative flow charts (Lean/Agile)
    """
    _name = 'project.task.history'
    _description = 'History of Tasks'
    _rec_name = 'task_id'
    _log_access = False

    def _get_date(self, cr, uid, ids, name, arg, context=None):
        result = {}
        for history in self.browse(cr, uid, ids, context=context):
            if history.type_id and history.type_id.fold:
                result[history.id] = history.date
                continue
            cr.execute('''select
                    date
                from
                    project_task_history
                where
                    task_id=%s and
                    id>%s
                order by id limit 1''', (history.task_id.id, history.id))
            res = cr.fetchone()
            result[history.id] = res and res[0] or False
        return result

    def _get_related_date(self, cr, uid, ids, context=None):
        result = []
        for history in self.browse(cr, uid, ids, context=context):
            cr.execute('''select
                    id
                from
                    project_task_history
                where
                    task_id=%s and
                    id<%s
                order by id desc limit 1''', (history.task_id.id, history.id))
            res = cr.fetchone()
            if res:
                result.append(res[0])
        return result

    _columns = {
        'task_id': fields.many2one('project.task', 'Task', ondelete='cascade', required=True, select=True),
        'type_id': fields.many2one('project.task.type', 'Stage'),
        'kanban_state': fields.selection([('normal', 'Normal'), ('blocked', 'Blocked'), ('done', 'Ready for next stage')], 'Kanban State', required=False),
        'date': fields.date('Date', select=True),
        'end_date': fields.function(_get_date, string='End Date', type="date", store={
            'project.task.history': (_get_related_date, None, 20)
        }),
        'remaining_hours': fields.float('Remaining Time', digits=(16, 2)),
        'planned_hours': fields.float('Planned Time', digits=(16, 2)),
        'user_id': fields.many2one('res.users', 'Responsible'),
    }
    _defaults = {
        'date': fields.date.context_today,
    }

class project_task_history_cumulative(osv.osv):
    _name = 'project.task.history.cumulative'
    _table = 'project_task_history_cumulative'
    _inherit = 'project.task.history'
    _auto = False

    _columns = {
        'end_date': fields.date('End Date'),
        'project_id': fields.many2one('project.project', 'Project'),
    }

    def init(self, cr):
        tools.drop_view_if_exists(cr, 'project_task_history_cumulative')

        cr.execute(""" CREATE VIEW project_task_history_cumulative AS (
            SELECT
                history.date::varchar||'-'||history.history_id::varchar AS id,
                history.date AS end_date,
                *
            FROM (
                SELECT
                    h.id AS history_id,
                    h.date+generate_series(0, CAST((coalesce(h.end_date, DATE 'tomorrow')::date - h.date) AS integer)-1) AS date,
                    h.task_id, h.type_id, h.user_id, h.kanban_state,
                    greatest(h.remaining_hours, 1) AS remaining_hours, greatest(h.planned_hours, 1) AS planned_hours,
                    t.project_id
                FROM
                    project_task_history AS h
                    JOIN project_task AS t ON (h.task_id = t.id)

            ) AS history
        )
        """)

class project_category(osv.osv):
    """ Category of project's task (or issue) """
    _name = "project.category"
    _description = "Category of project's task, issue, ..."
    _columns = {
        'name': fields.char('Name', required=True, translate=True),
    }
# vim:expandtab:smartindent:tabstop=4:softtabstop=4:shiftwidth=4: