Interesting on replace.. Ive always had an AGI script to cleanse data.. Never ran macros more than a couple deep.. Always would rather come back before I get in too deep or use AEL which is more suited for lots of loopty loop hopping around...
I didn’t realize replaced works like you describe.. Does it work better if you put the search and replace in quotes so they are treated as one rather than as individual strings as it appears to be doing?, or does it take the quotes as literals? On 2/10/12 5:14 PM, "Lonnie Abelbeck" <[email protected]> wrote: >Hi Chris, > >Yes, GoSub() is favored over Macro() in Asterisk 1.8, but Macro() still >works fine. The problem is Macro()'s can't be nested over 5 or so deep >without maxing out the stack. GoSub() has no such limitations. > >As you say Asterisk 1.8 has a new REPLACE function, but I think that >would be just as clumsy as using URIENCODE / URIDECODE and less >effective. Speaking of REPLACE, it is useful for deleting unwanted >characters, but not much more than that since it doesn't work like the >UNIX 'tr', ie, foo=xcatx ${REPLACE(foo,cat,CAT)} returns xCCCx and not >xCATx as you might expect. > >Lonnie > > >On Feb 10, 2012, at 1:42 PM, Chris Abnett wrote: > >> I never did understand why asterisk went from the pipe to the comma.. >> Just for that reason.... >> >> When I ran into this I had to cleanse my caller ID names before letting >> scripts get hold of them... Esp since my caller ID names are often >>keyed >> into hotel PMS systems by people.. They use whatever delimiter they feel >> like to separate names... >> >> You can try the asterisk function replace.. It is supposed to find a >> character and replace it.. don¹t know if it is accruate, I have not >>tried >> it.. >> >> I also thought that gosub is the preferred method instead of macro for >>all >> versions of asterisk going forward... >> -Christopher >> >> On 2/10/12 2:29 PM, "Lonnie Abelbeck" <[email protected]> wrote: >> >>> Update, >>> >>> I'm thinking this may be useful info, enclosed is an update. As David >>> suggests this appears to be a bug using Macro (and parsing), so let's >>>try >>> GoSub()... >>> -- >>> [macro-test] >>> exten => s,1,NOOP(ARG1=${ARG1}) >>> exten => s,n,NOOP(ARG2=${ARG2}) >>> exten => s,n,NOOP(ARG3=${ARG3}) >>> >>> [sub-test] >>> exten => start,1,NOOP(ARG1=${ARG1}) >>> same => n,NOOP(ARG2=${ARG2}) >>> same => n,NOOP(ARG3=${ARG3}) >>> same => n,Return() >>> -- >>> >>> Test1 code: pass ${CALLERID(name)} by itself >>> -- >>> exten => 9,1,Set(CALLERID(name)=Abelbeck, Lonnie) >>> exten => 9,n,Macro(test,${CALLERID(name)}) >>> exten => 9,n,GoSub(sub-test,start,1(${CALLERID(name)})) >>> == Output == >>> -- Executing [9@default:1] Set("SIP/10.10.50.1-0000000e", >>> "CALLERID(name)=Abelbeck, Lonnie") in new stack >>> -- Executing [9@default:2] Macro("SIP/10.10.50.1-0000000e", >>> "test,Abelbeck, Lonnie") in new stack >>> -- Executing [s@macro-test:1] NoOp("SIP/10.10.50.1-0000000e", >>> "ARG1=Abelbeck") in new stack >>> -- Executing [s@macro-test:2] NoOp("SIP/10.10.50.1-0000000e", "ARG2= >>> Lonnie") in new stack >>> -- Executing [s@macro-test:3] NoOp("SIP/10.10.50.1-0000000e", "ARG3=") >>>in >>> new stack >>> >>> -- Executing [9@default:3] Gosub("SIP/10.10.50.1-0000000e", >>> "sub-test,start,1(Abelbeck, Lonnie)") in new stack >>> -- Executing [start@sub-test:1] NoOp("SIP/10.10.50.1-0000000e", >>> "ARG1=Abelbeck") in new stack >>> -- Executing [start@sub-test:2] NoOp("SIP/10.10.50.1-0000000e", "ARG2= >>> Lonnie") in new stack >>> -- Executing [start@sub-test:3] NoOp("SIP/10.10.50.1-0000000e", >>>"ARG3=") >>> in new stack >>> -- Executing [start@sub-test:4] Return("SIP/10.10.50.1-0000000e", "") >>>in >>> new stack >>> == >>> Note both Macro and GoSub fail in the same way, the comma in the data >>>is >>> parsed. >>> >>> >>> Test2 code: add quotes around "${CALLERID(name)}" when passing >>> -- >>> exten => 9,1,Set(CALLERID(name)=Abelbeck, Lonnie) >>> exten => 9,n,Macro(test,"${CALLERID(name)}") >>> exten => 9,n,GoSub(sub-test,start,1("${CALLERID(name)}")) >>> == Output == >>> -- Executing [9@default:1] Set("SIP/10.10.50.1-0000000b", >>> "CALLERID(name)=Abelbeck, Lonnie") in new stack >>> -- Executing [9@default:2] Macro("SIP/10.10.50.1-0000000b", >>> "test,"Abelbeck, Lonnie"") in new stack >>> -- Executing [s@macro-test:1] NoOp("SIP/10.10.50.1-0000000b", >>> "ARG1="Abelbeck") in new stack >>> -- Executing [s@macro-test:2] NoOp("SIP/10.10.50.1-0000000b", "ARG2= >>> Lonnie"") in new stack >>> -- Executing [s@macro-test:3] NoOp("SIP/10.10.50.1-0000000b", "ARG3=") >>>in >>> new stack >>> >>> -- Executing [9@default:3] Gosub("SIP/10.10.50.1-0000000b", >>> "sub-test,start,1("Abelbeck, Lonnie")") in new stack >>> -- Executing [start@sub-test:1] NoOp("SIP/10.10.50.1-0000000b", >>> "ARG1="Abelbeck, Lonnie"") in new stack >>> -- Executing [start@sub-test:2] NoOp("SIP/10.10.50.1-0000000b", >>>"ARG2=") >>> in new stack >>> -- Executing [start@sub-test:3] NoOp("SIP/10.10.50.1-0000000b", >>>"ARG3=") >>> in new stack >>> -- Executing [start@sub-test:4] Return("SIP/10.10.50.1-0000000b", "") >>>in >>> new stack >>> == >>> Note Macro fails in the same way as Test1 with quoting but GoSub now >>> "works" (the comma is no longer parsed) but the downside is the "'s >>>chars >>> are now added to ARG1. >>> >>> >>> Test3 code: pass ${CALLERID(name)} with URIENCODE() then URIDECODE() >>> -- >>> [macro-test] >>> exten => s,1,NOOP(ARG1=${URIDECODE(${ARG1})}) >>> exten => s,n,NOOP(ARG2=${ARG2}) >>> exten => s,n,NOOP(ARG3=${ARG3}) >>> >>> [sub-test] >>> exten => start,1,NOOP(ARG1=${URIDECODE(${ARG1})}) >>> same => n,NOOP(ARG2=${ARG2}) >>> same => n,NOOP(ARG3=${ARG3}) >>> same => n,Return() >>> >>> exten => 9,1,Set(CALLERID(name)=Abelbeck, Lonnie) >>> exten => 9,n,Macro(test,${URIENCODE(${CALLERID(name)})}) >>> exten => 9,n,GoSub(sub-test,start,1(${URIENCODE(${CALLERID(name)})})) >>> == Output == >>> -- Executing [9@default:1] Set("SIP/10.10.50.1-00000014", >>> "CALLERID(name)=Abelbeck, Lonnie") in new stack >>> -- Executing [9@default:2] Macro("SIP/10.10.50.1-00000014", >>> "test,Abelbeck%2C%20Lonnie") in new stack >>> -- Executing [s@macro-test:1] NoOp("SIP/10.10.50.1-00000014", >>> "ARG1=Abelbeck, Lonnie") in new stack >>> -- Executing [s@macro-test:2] NoOp("SIP/10.10.50.1-00000014", "ARG2=") >>>in >>> new stack >>> -- Executing [s@macro-test:3] NoOp("SIP/10.10.50.1-00000014", "ARG3=") >>>in >>> new stack >>> >>> -- Executing [9@default:3] Gosub("SIP/10.10.50.1-00000014", >>> "sub-test,start,1(Abelbeck%2C%20Lonnie)") in new stack >>> -- Executing [start@sub-test:1] NoOp("SIP/10.10.50.1-00000014", >>> "ARG1=Abelbeck, Lonnie") in new stack >>> -- Executing [start@sub-test:2] NoOp("SIP/10.10.50.1-00000014", >>>"ARG2=") >>> in new stack >>> -- Executing [start@sub-test:3] NoOp("SIP/10.10.50.1-00000014", >>>"ARG3=") >>> in new stack >>> -- Executing [start@sub-test:4] Return("SIP/10.10.50.1-00000014", "") >>>in >>> new stack >>> == >>> Note the code is messy, but it now works with both Macro and GoSub >>> without any added chars to the args. >>> >>> For this special case, moving the ${CALLERID(name)} call into the Macro >>> rather than passing as an argument is the simplest. >>> >>> The moral to this story is in Asterisk 1.4 the pipe (|) character was >>> uncommon in the data and not much of a problem, while in Asterisk 1.8 >>>the >>> comma (,) is quite common in data leading to unexpected results. >>> >>> Lonnie > >-------------------------------------------------------------------------- >---- >Virtualization & Cloud Management Using Capacity Planning >Cloud computing makes use of virtualization - but cloud computing >also focuses on allowing computing to be delivered as a service. >http://www.accelacomm.com/jaw/sfnl/114/51521223/ >_______________________________________________ >Astlinux-users mailing list >[email protected] >https://lists.sourceforge.net/lists/listinfo/astlinux-users > >Donations to support AstLinux are graciously accepted via PayPal to >[email protected]. ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Astlinux-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/astlinux-users Donations to support AstLinux are graciously accepted via PayPal to [email protected].
