I believe, that between inserts/updates and selects you should do commit On Friday, September 1, 2017 at 10:15:00 PM UTC+3, Dave S wrote: > > > > On Friday, September 1, 2017 at 10:03:48 AM UTC-7, Ben Lawrence wrote: >> >> Also, if i am inserting within a for loop, should I place the db.commit() >> within the for loop too? >> >> On Friday, September 1, 2017 at 9:59:11 AM UTC-7, Ben Lawrence wrote: >>> >>> Good point, but I put the commits after the function returns as there >>> are a number of inserts in a variety of tables. So is it preferred to do a >>> commit after every insert instead of grouping them together for one big >>> commit? >>> TIA Ben >>> >> > > Part of the answer has to do with rolling back if there is an error. If > you need ALL changes undone if any one change fails, then the commit comes > last. If you need the first changes to survive even if the later changes > fail, then sprinkle commits in between. Usually, the first choice is what > people make. > > /dps > > >>> On Friday, September 1, 2017 at 12:35:01 AM UTC-7, Val K wrote: >>>> >>>> from the book: >>>> Remember to call db.commit() at the end of every task if it involves >>>> inserts/updates to the database. web2py commits by default at the end of a >>>> successful action but the scheduler tasks are not actions. >>>> >>>> On Friday, September 1, 2017 at 8:30:00 AM UTC+3, Ben Lawrence wrote: >>>>> >>>>> Thanks guys for your time. No I do not know what a _before_insert hook >>>>> is, so I doubt that there is one here. >>>>> >>>>> here is >>>>> >>>>> ################################################################################ >>>>> >>>>> def insert_company(valid_name, valid_email): >>>>> """ Inserts a company record if there is a new domain_tag >>>>> """ >>>>> c_id = 0 # is the reference to the company record >>>>> # what is the company domain ? >>>>> email_tup = valid_email.split(u'@') >>>>> if len(email_tup) > 1: # it is an email address >>>>> c_domain = email_tup[1] >>>>> elif len(email_tup) == 1: >>>>> c_domain = valid_email >>>>> else: # give up >>>>> c_domain = valid_name.upper() >>>>> # use this to find the company, >>>>> c_rows = db(db.company.domain_tag == c_domain).select() >>>>> if len(c_rows)==0: # add the company >>>>> c_id = db.company.update_or_insert( >>>>> name = get_company_name(c_domain), >>>>> e_mail = valid_email, >>>>> domain_tag = c_domain ) >>>>> else: # length MUST be one >>>>> c_id = c_rows.last().id >>>>> return c_id >>>>> >>>>> ################################################################################ >>>>> >>>>> You know what, >>>>> I am running all this on ubuntu 16 in an LXD container with software >>>>> raid on the metal. I am really starting to wonder if DAL and postgres and >>>>> lxd and software raid are somehow getting ahead of themselves in timing? >>>>> Perhaps there is no feedback or not enough feedback of "operation >>>>> completed" somewhere? >>>>> >>>>> >>>>> On Wednesday, August 30, 2017 at 1:59:04 AM UTC-7, Val K wrote: >>>>>> >>>>>> As I see, it's a scheduler task - show all code of insert_company >>>>>> Is there any _before_insert hook? >>>>>> >>>>>> On Wednesday, August 30, 2017 at 1:22:57 AM UTC+3, Ben Lawrence wrote: >>>>>>> >>>>>>> 2.15.3-stable+timestamp.2017.08.07.07.32.04 >>>>>>> (Running on nginx/1.10.3, Python 2.7.12) >>>>>>> >>>>>>> >>>>>>> Hi >>>>>>> I hesitate to ask this because it will betray my stupidity >>>>>>> >>>>>>> I have this: >>>>>>> >>>>>>> *c_rows = db(db.company.domain_tag == c_domain).select()* >>>>>>> if *len(c_rows)==0*: # add the company >>>>>>> c_id = db.company.insert( >>>>>>> name = get_company_name(c_domain), >>>>>>> e_mail = valid_email, >>>>>>> *domain_tag = c_domain )* >>>>>>> else: # length MUST be one >>>>>>> c_id = c_rows.last().id >>>>>>> >>>>>>> and this the error: >>>>>>> >>>>>>> File "applications/remail/models/scheduler.py", line 120, in >>>>>>> insert_company >>>>>>> domain_tag = c_domain ) >>>>>>> File "/home/www-data/web2py/gluon/packages/dal/pydal/objects.py", >>>>>>> line 734, in insert >>>>>>> ret = self._db._adapter.insert(self, row.op_values()) >>>>>>> File >>>>>>> "/home/www-data/web2py/gluon/packages/dal/pydal/adapters/base.py", line >>>>>>> 486, in insert >>>>>>> raise e >>>>>>> IntegrityError: duplicate key value violates unique constraint >>>>>>> "company_domain_tag_key" >>>>>>> DETAIL: Key (domain_tag)=(platronics.com) already exists. >>>>>>> >>>>>>> >>>>>>> Wha?! if domain_tag is already there would len(c_rows) > 0 and so >>>>>>> would not try to insert the record? >>>>>>> >>>>>>> >>>>>>>
-- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
