Odoo Help

Welcome!

This community is for beginners and experts willing to share their Odoo knowledge. It's not a forum to discuss ideas, but a knowledge base of questions and their answers.

0

How to get users timezone for python?

By
Laura Jarvenpaa
on 9/12/13, 7:34 AM 6,499 views

I have created a module which overwrites some functions on leaves management. This function includes getting employee working schedule from contract and calculating leave length based on the contract (In my company there can be leaves that are less than one day) the times provided by working schedule are just in format: {'time_from': 8.00, 'time_to':18:00} so they don't include any kind of timezone information however dates are transformed into utc time when they are inserted to db and when you compare those times with working schedule times you need to correct timezone to user's timezone.

My code is like this and as you see there is three places where 8 is added on dates to fix this timezone problem which is fine now when our openerp is used in utc+8 but in future we plan to enable using our system outside of our LAN and then there may be people located in different timezones using it so this fix won't help in future:

def _get_number_of_days(self, date_from, date_to, cr, uid, ids, employee_id):

    DATETIME_FORMAT = "%Y-%m-%d %H:%M:%S"
    date_from = datetime.datetime.strptime(date_from, DATETIME_FORMAT)
    date_to = datetime.datetime.strptime(date_to, DATETIME_FORMAT)
    timedelta = date_to - date_from
    working_shift = 86400
    diff_day = float();
    if timedelta.days > 1:
        diff_day += timedelta.days - 1

    if(employee_id):
        contract_pool = self.pool.get('hr.contract')
        active_contract_ids= contract_pool.search(cr, uid, [
        '&',
        ('employee_id', '=', employee_id),
        '|',
        '&',
        ('date_start', '<=', date_to),
        '|',
        ('date_end', '>=', date_to),
        ('date_end', '=', False),
        '&',
        '|',
        ('trial_date_start', '=', False),
        ('trial_date_start', '<=', date_to),
        '|',
        ('trial_date_end', '=', False),
        ('trial_date_end', '>=', date_to),
        ])
        if active_contract_ids:
            contract = contract_pool.browse(cr, uid, active_contract_ids[0])

            cr.execute("select hour_from, hour_to from resource_calendar_attendance where calendar_id = %s and dayofweek = %s order by hour_from;", (contract.working_hours.id, str(date_from.isoweekday())))
            start_timetable = cr.dictfetchall()
            if start_timetable:

                start_working_shift = self.pool.get('resource.calendar').working_hours_on_day(cr, uid, contract.working_hours, date_from)

                if start_working_shift:
                    # add 8 hours to fix timezone                                                                                                        
                    start_time = date_from.hour + 8 + float(date_from.minute)/60
                    if date_to.date() == date_from.date():
                        # add 8 hours to fix timezone                                                                                                    
                        end_time = date_to.hour + 8 + float(date_to.minute)/60
                        if start_time < start_timetable[0]['hour_from'] and end_time > start_timetable[len(start_timetable) - 1]['hour_to']:
                            # leave length is one day because it starts before starting hour and ends after ending hour                                  
                            return 1           
                        else:
                            hours = self._calculate_hours_for_working_day(start_timetable, start_time, end_time)
                            return hours / start_working_shift
                    if start_time < start_timetable[0]['hour_from']:
                            # leave length is one day because it starts before starting hour and ends after ending hour                               
                            diff_day += 1
                    else:
                        end_time = float(start_timetable[len(start_timetable) - 1]['hour_to'])
                        hours = self._calculate_hours_for_working_day(start_timetable, start_time, end_time)
                        diff_day += hours / start_working_shift


            cr.execute("select hour_from, hour_to from resource_calendar_attendance where calendar_id = %s and dayofweek = %s order by hour_from;", (contract.working_hours.id, str(date_to.isoweekday())))
            end_timetable = cr.dictfetchall()
            if end_timetable:
                end_working_shift = self.pool.get('resource.calendar').working_hours_on_day(cr, uid, contract.working_hours, date_to)
                if end_working_shift:
                    start_time = float(end_timetable[0]['hour_from'])
                    # add 8 hours to fix timezone                                                                                                        
                    end_time = date_to.hour + 8 + float(date_to.minute)/60
                    if end_time > start_timetable[len(start_timetable) - 1]['hour_to']:
                        # leave length is one day because it starts before starting hour and ends after ending hour                                      
                        diff_day += 1
                    else:
                        hours = self._calculate_hours_for_working_day(end_timetable, start_time, end_time)
                        diff_day += hours / end_working_shift

        else:
            # tell user that employee don't have contract so leaves can't be  calculated                                                                 
            raise osv.except_osv(_('Error'), _('Employee doesn\'t have contract. Leave request can\'t be done.'))
    else:
        diff_day = timedelta.days + float(timedelta.seconds) / working_shift
    return diff_day

Tried to fix this by adding parameter context for onchange function of the date selection and using pytz to transform these times to right timezone:

active_tz = pytz.timezone(context.get("tz","UTC")

this did not work (found this solution from another module but well as their module didn't work properly either it's not too big surprise that this did not work).

3
Jeroen Mollers
On 9/12/13, 10:00 AM

To get the user's timezone you could do something like this in your code:

Version 7.0 code:

import pytz
from openerp import SUPERUSER_ID

# get user's timezone
user_pool = self.pool.get('res.users')
user = user_pool.browse(cr, SUPERUSER_ID, uid)
tz = pytz.timezone(user.partner_id.tz) or pytz.utc

# get localized dates
date_from = pytz.utc.localize(datetime.datetime.strptime(date_from, DATETIME_FORMAT)).astimezone(tz)
date_to   = pytz.utc.localize(datetime.datetime.strptime(date_to, DATETIME_FORMAT)).astimezone(tz)

Users that use version 6.1 should change the code line:

tz = pytz.timezone(user.partner_id.tz) or pytz.utc

into code line:

tz = pytz.timezone(user.context_tz) or pytz.utc

Thanks this was helpful but at least in version 7 I had to do following modification on your code: user_pool = self.pool.get('res.partner') user = user_pool.browse(cr, SUPERUSER_ID, uid) tz = pytz.timezone(user.tz) or pytz.utc because timezone info is now in res.partner not in res.users and timezone info provided by that table had to be converted into timezone as it is only string. If you add these modifications to your answer I can mark it as right.

Laura Jarvenpaa
on 9/13/13, 12:16 AM

You are correct. My code was for version 6.1. In your suggested modification you retrieve a partner record based on a user's ID. I suggest to modify my original code from "tz = pytz.timezone(user.context_tz) or pytz.utc" to "tz = pytz.timezone(user.partner_id.tz) or pytz.utc". I have modified my original answer and also included the version 6.1 one in case someone is interested in that one.

Jeroen Mollers
on 9/13/13, 7:03 AM

Ok, good to know.

Laura Jarvenpaa
on 9/16/13, 1:06 AM

when we print "date_from" its display like '2014-08-14 05:25:55+05:30' i am using time-zone 'Asia/Kolkata'

Ashish Singh
on 9/20/14, 8:19 AM

Your Answer

Please try to give a substantial answer. If you wanted to comment on the question or answer, just use the commenting tool. Please remember that you can always revise your answers - no need to answer the same question twice. Also, please don't forget to vote - it really helps to select the best questions and answers!

About This Community

This community is for professionals and enthusiasts of our products and services. Read Guidelines

Question tools

0 follower(s)

Stats

Asked: 9/12/13, 7:34 AM
Seen: 6499 times
Last updated: 3/16/15, 8:10 AM