egghelp/eggheads community

Run a proc delayed

2010-09-21T05:59:50-04:00

And why not call your procedure again. I mean

Code:

proc somename {args} {....::http::register https 443 ::tls::socketif {![catch {set http_handle [http::geturl $the_url -binary 1 -timeout $::check(timeout)]} error]} {   set http_status [http::status $http_handle]   if {$http_status == "ok"} {      if {[http::ncode $http_handle] == 200} {         set http_data [::http::data $http_handle]         ::http::cleanup $http_handle         set File [::mysql::escape $http_data]         ::mysql::encoding $db_handle binary         set sql [::mysql::sel $db_handle "INSERT INTO ...      }   } elseif {$http_status == "timeout"} {      sendmsg $channel "\002\00304\[ERROR\]\002\003 \002Connection timed out - $the_url ... Retrying ..."     somename $args; #you have already defined all that you need   } elseif {[http::ncode $http_handle] == 404} { ......}

if you use you method you must re-define error like -timeout again and again

, and nml375 is right after block eggdrop for some time but -timeout option all so use after in http.tcl

Statistics: Posted by game_over — Tue Sep 21, 2010 5:59 am

Run a proc delayed

2010-09-19T15:13:56-04:00

Elfriede,
What you need to do, is to split your proc up into several pieces, thus allowing you to call the second http-transaction separately from the first one using a timer.

Statistics: Posted by nml375 — Sun Sep 19, 2010 3:13 pm

Run a proc delayed

2010-09-19T09:52:07-04:00

I really appreciate ur answer and im quite sure i do "understand", but im unsure how to use that on my proc. Actually it looks like:

Code:

::http::register https 443 ::tls::socketif {![catch {set http_handle [http::geturl $the_url -binary 1 -timeout $::check(timeout)]} error]} {set http_status [http::status $http_handle]if {$http_status == "ok"} {if {[http::ncode $http_handle] == 200} {set http_data [::http::data $http_handle]::http::cleanup $http_handleset File [::mysql::escape $http_data]::mysql::encoding $db_handle binaryset sql [::mysql::sel $db_handle "INSERT INTO ...}} elseif {$http_status == "timeout"} {sendmsg $channel "\002\00304\[ERROR\]\002\003 \002Connection timed out - $the_url ... Retrying ..."set http_data [::http::data $http_handle]::http::cleanup $http_handleset File [::mysql::escape [getPage $the_url]]::mysql::encoding $db_handle binaryset sql [::mysql::sel $db_handle "INSERT INTO...} elseif {[http::ncode $http_handle] == 404} {

The first part, as shown here, uses a timeout of 3 seconds. In case of timeout, ur example code should retry xx seconds delayed, but i dunno the best way to use it. May u please help

Statistics: Posted by Elfriede — Sun Sep 19, 2010 9:52 am

Run a proc delayed

2010-09-18T12:14:48-04:00

I would not recommend using the after command in blocking mode, as eggdrop is single-threaded, and blocking behavior prevents the eggdrop from taking any actions at all during the block.

I'd rather recommend using the utimer command to schedule a piece of code to be executed at a later time, or if possible, use the -command option with your http-request to use an event-driven mode:

Code:

::http::register https 443 ::tls::socket ::http::geturl "http://www.example.com/" -binary 1 -command handle_responseproc handle_response {token} {  if {[::http::status $token] == "ok"} {    set File [::mysql::escape [::http::data $token]]    utimer 20 [list handle_final $File]  }  ::http::cleanup}proc handle_final {File} {  #this is called 20 seconds after the http-transaction is completed...  putlog "The MySQL-escaped filename was $File"}

Statistics: Posted by nml375 — Sat Sep 18, 2010 12:14 pm

Run a proc delayed

2010-09-18T10:55:19-04:00

Code:

} elseif {$http_status == "timeout"} {    sendmsg # "\002\00304\[ERROR\]\002\003 \002Connection timed out - $the_url ... Retrying ..."    set http_data [::http::data $http_handle]    ::http::cleanup $http_handle    set File [::mysql::escape [getPage $the_url]]    ..... proc getPage { the_url } { putlog "wait"after 3000 ;# Simulates deep thoughtputlog "now"   ::http::register https 443 ::tls::socket    set token [::http::geturl $the_url -binary 1 -timeout 10000]    set data [::http::data $token]    ::http::cleanup $token    return $data }

you could give this a go

Statistics: Posted by doggo — Sat Sep 18, 2010 10:55 am

Run a proc delayed

2010-09-17T03:05:59-04:00

Hi : Its me again ^^

Here's my code:

Code:

} elseif {$http_status == "timeout"} {sendmsg # "\002\00304\[ERROR\]\002\003 \002Connection timed out - $the_url ... Retrying ..."set http_data [::http::data $http_handle]::http::cleanup $http_handleset File [::mysql::escape [getPage $the_url]].....proc getPage { the_url } {::http::register https 443 ::tls::socketset token [::http::geturl $the_url -binary 1 -timeout 10000]set data [::http::data $token]::http::cleanup $tokenreturn $data}

As u can see i have a timeout of 10s in the getPage proc. What i'm looking for is that this part:
set File [::mysql::escape [getPage $the_url]]
would be called for eg lets say 20 seconds later, instead of needing/using such a high timeout in my getPage proc. Everything ive tried till yet ended up in errors :/

Statistics: Posted by Elfriede — Fri Sep 17, 2010 3:05 am