All of lore.kernel.org
 help / color / mirror / Atom feed
* status-report expiration
@ 2010-09-06 16:10 Petteri Tikander
  2010-09-07 14:19 ` Denis Kenzior
  0 siblings, 1 reply; 2+ messages in thread
From: Petteri Tikander @ 2010-09-06 16:10 UTC (permalink / raw)
  To: ofono

[-- Attachment #1: Type: text/plain, Size: 1627 bytes --]

Hi Denis,

I'm implementing expiration-logic for status report and have couple of 
questions. 

My expiration function under development goes through every 
addresses and corresponding messages (nodes) from hash-tables, and checks 
the expiration time. But when finding expired message, it cannot remove 
corresponding backup-files, because in this moment backup-file contains the 
whole sms-addressing information (address+numbering plan+number type), but key 
of the hash-table (sms-address) doesn't. So, first I thought simply to add also 
numbering plan+number type in the hash-key. So what do you think, could it be 
a good idea (to include the complete address, but to waste more memory)? OK, I 
can also change the function to find backup-file by not taking into account 
numbering plan+number type at all. For example, I can go through backup-files, 
capture just the sms-address (and message-id), and then search the correct 
node from the hash-table (not necessarily nice solution).

The another question is: is it enough in this moment just to implement 
expiration function (and some unit-tests), as in existing sms-assembly 
solution, when waiting for incoming sms-delivery messages? Seems that in the 
case of sms-assembly, expiration-status for all messages is polled in some 
interval. So, possibly you have already earlier discussed about this issue, 
but how expiration-checking is signaled to sms-util? Will oFono-api take care 
internally of triggering the expiration-status some how? Or, will oFono-api 
receive expiration-checking calls outside oFono via D-bus.

Br, Petteri


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: status-report expiration
  2010-09-06 16:10 status-report expiration Petteri Tikander
@ 2010-09-07 14:19 ` Denis Kenzior
  0 siblings, 0 replies; 2+ messages in thread
From: Denis Kenzior @ 2010-09-07 14:19 UTC (permalink / raw)
  To: ofono

[-- Attachment #1: Type: text/plain, Size: 2493 bytes --]

Hi Petteri,

On 09/06/2010 11:10 AM, Petteri Tikander wrote:
> Hi Denis,
> 
> I'm implementing expiration-logic for status report and have couple of 
> questions. 
> 
> My expiration function under development goes through every 
> addresses and corresponding messages (nodes) from hash-tables, and checks 
> the expiration time. But when finding expired message, it cannot remove 
> corresponding backup-files, because in this moment backup-file contains the 
> whole sms-addressing information (address+numbering plan+number type), but key 
> of the hash-table (sms-address) doesn't. So, first I thought simply to add also 
> numbering plan+number type in the hash-key. So what do you think, could it be 
> a good idea (to include the complete address, but to waste more memory)? OK, I 
> can also change the function to find backup-file by not taking into account 
> numbering plan+number type at all. For example, I can go through backup-files, 
> capture just the sms-address (and message-id), and then search the correct 
> node from the hash-table (not necessarily nice solution).

oFono supports sending SMS messages only to international (prefixed by
+) or 'unknown' (not prefixed by +) numbers.  So you can actually easily
figure out the ton/npi and number based on the address string.  This is
in effect what sms_address_to_hex_string does...

> 
> The another question is: is it enough in this moment just to implement 
> expiration function (and some unit-tests), as in existing sms-assembly 
> solution, when waiting for incoming sms-delivery messages? Seems that in the 
> case of sms-assembly, expiration-status for all messages is polled in some 
> interval. So, possibly you have already earlier discussed about this issue, 
> but how expiration-checking is signaled to sms-util? Will oFono-api take care 
> internally of triggering the expiration-status some how? Or, will oFono-api 
> receive expiration-checking calls outside oFono via D-bus.
> 

For now it is enough just to implement the expiration function and a
unit test.  We still need to hook this up into the core somewhere.  Most
likely it will be a 24 hour timer and we'll also run expiration at
startup if more than 24 hours have elapsed since the last check.

e.g. something like:
LastExpiration = settings.LastExpiration
if (LastExpiration - now > 24 hours)
	run expiration
	LastExpiration = now

set timer until LastExpiration + 24 hours.

Regards,
-Denis

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2010-09-07 14:19 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-09-06 16:10 status-report expiration Petteri Tikander
2010-09-07 14:19 ` Denis Kenzior

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.