Re: [SR-Users] rfc: accounting records time details

2013-09-05 Thread Daniel-Constantin Mierla
For the records, this has been implemented in master branch - a matter of time_mode parameter for acc, time value can be stored in various formats. Cheers, Daniel On 5/10/13 9:50 AM, Daniel-Constantin Mierla wrote: Hello, any more comments on time storage for acc records? Obviously, it has

Re: [SR-Users] rfc: accounting records time details

2013-05-10 Thread Daniel-Constantin Mierla
Hello, any more comments on time storage for acc records? Obviously, it has to be done via something that is configurable and extensible. If no more comments, I will go ahead and develop soon a configurable framework for seconds and seconds, microseconds, then others can come and add more.

Re: [SR-Users] rfc: accounting records time details

2013-05-01 Thread Daniel-Constantin Mierla
Hello, On 4/29/13 11:47 AM, Thilo Bangert wrote: More suggestions? Pro or cons opinions? I'd just save one timestamp, ie TAI64 or as java does miliseconds since epoch, in a single, new field. saving the timestamp in two fields seems messy. it can be one field only, it is a matter of module

Re: [SR-Users] rfc: accounting records time details

2013-05-01 Thread Daniel-Constantin Mierla
Hello, On 4/29/13 1:42 PM, Alex Hermann wrote: On Monday 29 April 2013 11:05:36 Daniel-Constantin Mierla wrote: Here are some suggestions presented so far. 1) store seconds.miliseconds as double - there is a patch (which Please do not use floating point respresentations for values that will

Re: [SR-Users] rfc: accounting records time details

2013-05-01 Thread Daniel-Constantin Mierla
Hello, On 4/29/13 2:15 PM, Andreas Granig wrote: Hi, On 04/29/2013 11:47 AM, Thilo Bangert wrote: I'd just save one timestamp, ie TAI64 or as java does miliseconds since epoch, in a single, new field. saving the timestamp in two fields seems messy. Agreed. the option to get current

Re: [SR-Users] rfc: accounting records time details

2013-04-29 Thread Thilo Bangert
More suggestions? Pro or cons opinions? I'd just save one timestamp, ie TAI64 or as java does miliseconds since epoch, in a single, new field. saving the timestamp in two fields seems messy. miliseconds since epoch is probably preferable, since it can be converted to human readable dates

Re: [SR-Users] rfc: accounting records time details

2013-04-29 Thread Alex Hermann
On Monday 29 April 2013 11:05:36 Daniel-Constantin Mierla wrote: Here are some suggestions presented so far. 1) store seconds.miliseconds as double - there is a patch (which Please do not use floating point respresentations for values that will be used in accounting. Floating point is

Re: [SR-Users] rfc: accounting records time details

2013-04-29 Thread Andreas Granig
Hi, On 04/29/2013 01:42 PM, Alex Hermann wrote: On Monday 29 April 2013 11:05:36 Daniel-Constantin Mierla wrote: 1) store seconds.miliseconds as double - there is a patch (which Please do not use floating point respresentations for values that will be used in accounting. Floating point is

Re: [SR-Users] rfc: accounting records time details

2013-04-29 Thread Andreas Granig
Hi, On 04/29/2013 11:47 AM, Thilo Bangert wrote: I'd just save one timestamp, ie TAI64 or as java does miliseconds since epoch, in a single, new field. saving the timestamp in two fields seems messy. Agreed. miliseconds since epoch is probably preferable, since it can be converted to human