This is the route that I'm using for testing this feature: ``` route[TEST] { pv_unset("$(avp(s:pbx_cli)[*])"); $avp(s:pbx_cli) = "1111"; $avp(s:pbx_cli) = "2222"; $avp(s:pbx_cli) = "3333"; $var(num) = "11118765"; if(dp_translate("2", "$var(num)/$avp(s:dest)")) { xlog("$var(num) translated to var $avp(s:dest) \n"); } $var(num) = "0022228765"; if(dp_translate("2", "$var(num)/$avp(s:dest)")) { xlog("$var(num) translated to var $avp(s:dest) \n"); } $var(num) = "00333387"; $(avp(s:cloud_pbx_cli)[*]) = "666"; if(dp_translate("2", "$var(num)/$avp(s:dest)")) { xlog("$var(num) translated to var $avp(s:dest) \n"); } if(dp_translate("1", "$var(num)/$avp(s:dest)")) { xlog("$var(num) translated to var $avp(s:dest) \n"); } } ``` This is the rules at dialplan: ``` sqlite> select * from dialplan; 1|1|1|2|^(00|+)([1-9][0-9]+)$|0|^(00|+)([1-9][0-9]+)$|\2| 2|2|1|2|^$avp(s:caller_cc)$avp(s:caller_ac)([1-9][0-9]+)$|0|^$avp(s:caller_cc)$avp(s:caller_ac)([1-9][0-9]+)$|\1| 3|1|2|0|^(00|+)([1-9][0-9]+)$|0|^(00|+)([1-9][0-9]+)$|\2| 4|2|2|0|^$avp(s:caller_cc)$avp(s:caller_ac)([1-9][0-9]+)$|0|^$avp(s:caller_cc)$avp(s:caller_ac)([1-9][0-9]+)$|\1| 5|1|2|1|^(00|+)([1-9][0-9]+)$|0|^(00|+)([1-9][0-9]+)$|\2| 6|2|2|1|^$avp(s:caller_cc)$avp(s:caller_ac)([1-9][0-9]+)$|0|^$avp(s:caller_cc)$avp(s:caller_ac)([1-9][0-9]+)$|\1| 7|2|2|1|^(00|+)?$(avp(s:pbx_cli)[*])([0-9]{3})$|0|^(00|+)?$(avp(s:pbx_cli)[*])([0-9]{3})$|\2| 8|2|2|1|^(00|+)?$(avp(s:pbx_cli)[*])([1-9]{2})$|0|^(00|+)?$(avp(s:pbx_cli)[*])([1-9]{2})$|$avp(s:cloud_pbx_cli)| ``` As you can see if you use $(avp(s:pbx_cli)[*]) you will get "333 3, 2222, 1111". So the normal behavior for print but not what we want. ``` 4(1425) DEBUG: dialplan [dp_repl.c:77]: dpl_dynamic_pcre(): compiled dynamic pcre expression: ^(00|+)?$(avp(s:pbx_cli)[*])([0-9]{3})$$ (^(00|+)?333 3, 2222, 1111([0-9]{3})$) 2

```

With the patch we are checking every value of the avp, one by one: ``` 4(5674) DEBUG: dialplan [dp_repl.c:249]: dpl_dynamic_pcre_list(): parsed pcre expression: ^(00|+)?$(avp(s:pbx_cli)[*])([0-9]{3})$$ 4(5674) DEBUG: dialplan [dp_repl.c:222]: dpl_dyn_pcre_comp(): compiled dynamic pcre expression: ^(00|+)?3333([0-9]{3})$ 2 4(5674) DEBUG: dialplan [dp_repl.c:222]: dpl_dyn_pcre_comp(): compiled dynamic pcre expression: ^(00|+)?2222([0-9]{3})$ 2 4(5674) DEBUG: dialplan [dp_repl.c:222]: dpl_dyn_pcre_comp(): compiled dynamic pcre expression: ^(00|+)?1111([0-9]{3})$ 2 4(5674) DEBUG: dialplan [dp_repl.c:628]: translate(): match check: [^(00|+)?1111([0-9]{3})$] -1 4(5674) DEBUG: dialplan [dp_repl.c:628]: translate(): match check: [^(00|+)?2222([0-9]{3})$] -1 4(5674) DEBUG: dialplan [dp_repl.c:628]: translate(): match check: [^(00|+)?3333([0-9]{3})$] -1 4(5674) DEBUG: dialplan [dp_repl.c:613]: translate(): regex operator testing over [00333387] ```

--- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/pull/18#issuecomment-72834432

OK, I see now what you were looking for. I will think a bit about, because you remove the option of being able to use $(avp(x)[*]) as it returns for print.

--- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/pull/18#issuecomment-73269648

It still changes the expected behaviour and somehow adds unnecessary complexity as I expect one wants a variant, not the both at the same time.

Perhaps the best is to have another index notation for this case, like: ``` $(avp(x)[+]) ``` or ``` $(avp(x)[**]) ```

The parser function of the index needs to be extended, but that should be easy. Then there are not many pvs using '*' index, thus would not be hard to propagate.

To decide: what to return in case of using the new index in print function (e.g., xlog) -- I would say the first value. CSV is available with '*'.

--- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/pull/18#issuecomment-74041098

ok - it could be defined as PV_IDX_ITR (from iterator).

--- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/pull/18#issuecomment-74056918

It can be merge, testing will be easier afterwards ...

--- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/pull/18#issuecomment-83662801

Merged manually

--- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/pull/18#issuecomment-83687276