RE: Time comparison in wap_push_ppg.c

Yes, that would be definitely more readable.

Aarno

>From: "Wilfried Goesgens" <[email protected]>
>To: [email protected]
>Subject: Time comparison in wap_push_ppg.c
>Date: Mon, 19 Jan 2004 21:23:31 +0100
>
>
>hy, I've browsed through the source of the push gateway stuff, and found a
>real weird funktion parsing the time, comparing it step by step...
>is there any god reason why not just use strptime to parse the string to a
>tm struct, convert that with mktime to time_t, and just compare the two
>resulting longs?
>
>i don't know if the former function uses an maximum ofset...
>and the actualtime now shouldn't be put as tm, so the mktime wouldn't be
>nescessary, but i didn't want to make such a big change...
>
>I've made a cvs diff to the current cvs and mine changes, and i'd be glad 
>if
>it could be committed to the tree...
>cvs diff wap_push_ppg.c
>Index: wap_push_ppg.c
>===================================================================
>RCS file: /home/cvs/gateway/gw/wap_push_ppg.c,v
>retrieving revision 1.64
>diff -r1.64 wap_push_ppg.c
>2735c2735,2738
><
>---
>>       struct tm aftertime;
>>       time_t after_T;
>>       time_t now_T;
>>       char *error;
>2738c2741,2749
><
>---
>>       
>>error=strptime(octstr_get_cstr(after),"%Y-%m-%dT%H:%M:%SZ",&aftertime);
>>     if (NULL!=error   )
>>               return 1; // the string wasn't parseable
>>       after_T=mktime(&aftertime);
>>       now_T=mktime(&now);
>>       if (after_T<now_T) return 0;
>>       else return 1;
>>
>>       /*
>2751,2752c2762,2763
><
><     return 0;
>---
>>
>>     return 0;*/
>
>
>kind regards, Wilfried Goesgens
>

_________________________________________________________________
Help STOP SPAM with the new MSN 8 and get 2 months FREE*  
http://join.msn.com/?page=features/junkmail