990830 - [openc 1.3]The phone number matching rule for is not fit for Venezuela

Status: RESOLVED FIXED

(Firefox OS Graveyard :: Runtime, defect)

Description

[zhangdapeng]

User Agent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; TCO_20140402084134)

Steps to reproduce:

here saves a number 04149113619 named TEST
call 9113619 ,it matched TEST
call 49113619，it didn't match， 
call 4149113619 it matched
call 84149113619 it didn't match




Expected results:

for Venezuela,they acquire a match rule as 10 numbers from end
that is if 10 numbers from the end is the same ,it will match

Comment 1

[Jason Smith [:jsmith]]

> here saves a number 04149113619 named TEST
> call 9113619 ,it matched TEST

This is working as expected.

> call 49113619，it didn't match， 

This is a bug.

> call 4149113619 it matched

This is working as expected.

> call 84149113619 it didn't match

This is working as expected.

Comment 2

[Jason Smith [:jsmith]]

Can we see if we can confirm if we can reproduce the bug portion called out in comment 1 on 1.3 in house?

Comment 3

[zhangdapeng]

Attachment: matchrule.bmp

see attachment matchrule.bmp
as different countries has different match rules
now for Peru,the equipment recognize 8 numbers in the contact list, but must be 7 numbers.
and for VE,it should be 10 numbers

as i see in gecko/dom/phonenumberutils/PhoneNumberMetaData.jsm
every country has a Regular expressions as match rule
and in gecko/b2g/app/b2g.js
there is code like this
// Shortnumber matching needed for e.g. Brazil:
// 01187654321 can be found with 87654321
pref("dom.phonenumber.substringmatching.BR", 8);
pref("dom.phonenumber.substringmatching.CO", 10);
pref("dom.phonenumber.substringmatching.VE", 7);


so please tell us 
if we want to change the match number to 7 numbers foe PE
how shuold i do
thanks

Comment 4

[Juan Perez-Bedmar [:juanpbf]]

Hi, 

Just to clarify and avoid confusion, this matching shall be done between the incoming calls and the numbers stored in the contact list.

Steven, can you help us here?

Comment 5

[Juan Perez-Bedmar [:juanpbf]]

ni to Steven per above comment

Comment 6

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #2)
> Can we see if we can confirm if we can reproduce the bug portion called out
> in comment 1 on 1.3 in house?

Clearing - this is a country-specific issue, not something to reproduce in house.

Comment 7

[zhangdapeng]

see in gecko/dom/phonenumberutils/PhoneNumberMetaData.jsm
the match rule for PE is

"51": '["PE","19(?:1[124]|77|90)00","0",,,"($NP$FG)","\\d{6,9}","[14-9]\\d{7,8}",[["(1)(\\d{7})","$1 $2","1",,],["([4-8]\\d)(\\d{6})","$1 $2","[4-7]|8[2-4]",,],["(\\d{3})(\\d{5})","$1 $2","80",,],["(9\\d{2})(\\d{3})(\\d{3})","$1 $2 $3","9","$FG",]]]',

for VE is
"58": '["VE","00","0",,,"$NP$FG","\\d{7,10}","[24589]\\d{9}",[["(\\d{3})(\\d{7})","$1-$2",,,]]]',

and now it is recognize 8 numbers in the contact for PE
if we want 7 numbers for PE and 10 numbers for VE
how can we modify this two expressions

Comment 8

[Juan Perez-Bedmar [:juanpbf]]

Hi Vance, could you please help my colleague Dapeng with his doubts? 
To fix this is critical and extremely urgent for us, it's a certification blocker and we need to deliver a new Official build ASAP to customer for IOT

Comment 9

[Vance Chen [:vchen][skype:vance.lucida][[e-mail]]]

Hi Francisco  -

Could you kindly provide some suggestions regarding how to modify and generate the PhoneNumberMetaData.jsm to meet Venezuela's requirement?

Thanks for your help

Comment 10

[Jason Smith [:jsmith]]

Cert blocker - blocking+

Comment 11

[Gregor Wagner [:gwagner]]

Lets focus on Venezuela here and file a another bug if we have issues with Peru please.

How did you test this? Do you have a SIM from Venezuela in the device or are you testing with a device connected to a network in Venezuela? Note that 04149113619 doesn't include the country code for VE and we have no way to identify this number as a local number in VE.

Our current implementation does a 7 digit matching if you have a SIM from VE. If the last 7 digits match, we can find the contact via the contacts API.

Not all numbers in Venezuela have 10 digits right? So how can we require a 10 digit matching?
Afaik, local numbers have 7 digits like 5551212.

Comment 12

[zhangdapeng]

i have file another bug for Peru
Bug 991989 - [zte openc 1.3]The phone number matching rule for is not fit for peru 

please Pay attention to  and fix it together

Comment 13

[zhangdapeng]

as i see in gecko/dom/phonenumberutils/mcc_iso3166_table.jsm
it distinguish countries from mcc
so i think if the number does not have country code 
it could be identified as many calls and SMS number does not have a country code
so i think this may not a problem

Comment 14

[Juan Perez-Bedmar [:juanpbf]]

(In reply to Gregor Wagner [:gwagner] from comment #11)
> Lets focus on Venezuela here and file a another bug if we have issues with
> Peru please.
> 
> How did you test this? Do you have a SIM from Venezuela in the device or are
> you testing with a device connected to a network in Venezuela? Note that
> 04149113619 doesn't include the country code for VE and we have no way to
> identify this number as a local number in VE.
>
> Our current implementation does a 7 digit matching if you have a SIM from
> VE. If the last 7 digits match, we can find the contact via the contacts API.
> 
> Not all numbers in Venezuela have 10 digits right? So how can we require a
> 10 digit matching?
> Afaik, local numbers have 7 digits like 5551212.

I'm not sure how this test is performed, I will contact the test house in Venezuela.
What is clear is that this is a requirement from the operator, critical for them.

Comment 15

[Jason Smith [:jsmith]]

(In reply to Vance Chen [:vchen][[e-mail]] from comment #9)
> Hi Francisco  -
> 
> Could you kindly provide some suggestions regarding how to modify and
> generate the PhoneNumberMetaData.jsm to meet Venezuela's requirement?
> 
> Thanks for your help

I think Gregor is the person you are looking for here - he's already fixing the other bug that's similar to this, so he probably knows how to do this.

Comment 16

[Gregor Wagner [:gwagner]]

(In reply to zhangdapeng from comment #13)
> as i see in gecko/dom/phonenumberutils/mcc_iso3166_table.jsm
> it distinguish countries from mcc
> so i think if the number does not have country code 
> it could be identified as many calls and SMS number does not have a country
> code
> so i think this may not a problem

Yes, this should already work.

If you still see problems please enable the debug output in dom/phonenumberutils/ PhoneNumberService.js and PhoneNumberUtils.jsm by changing 'const DEBUG = false;' to true.

Comment 17

[Jason Smith [:jsmith]]

Gregor - Since you are already looking into bug 991989, would you be interested in taking this bug as well?

Comment 18

[zhangdapeng]

 (In reply to zhangdapeng from comment #13)
> as i see in gecko/dom/phonenumberutils/mcc_iso3166_table.jsm
> it distinguish countries from mcc
> so i think if the number does not have country code 
> it could be identified as many calls and SMS number does not have a country
> code
> so i think this may not a problem

Yes, this should already work.

If you still see problems please enable the debug output in dom/phonenumberutils/ PhoneNumberService.js and PhoneNumberUtils.jsm by changing 'const DEBUG = false;' to true.


sorry i didn't say it clear
i mean the number didn't have a country code may not the reason for this match prbolem!
as you say in bug 991989
for VE
change 
pref("dom.phonenumber.substringmatching.VE", 10);
and  fake the country name 
i will have a try

Comment 19

[Gregor Wagner [:gwagner]]

(In reply to zhangdapeng from comment #18)
>  (In reply to zhangdapeng from comment #13)
> > as i see in gecko/dom/phonenumberutils/mcc_iso3166_table.jsm
> > it distinguish countries from mcc
> > so i think if the number does not have country code 
> > it could be identified as many calls and SMS number does not have a country
> > code
> > so i think this may not a problem
> 
> Yes, this should already work.
> 
> If you still see problems please enable the debug output in
> dom/phonenumberutils/ PhoneNumberService.js and PhoneNumberUtils.jsm by
> changing 'const DEBUG = false;' to true.
> 
> 
> sorry i didn't say it clear
> i mean the number didn't have a country code may not the reason for this
> match prbolem!
> as you say in bug 991989
> for VE
> change 
> pref("dom.phonenumber.substringmatching.VE", 10);
> and  fake the country name 
> i will have a try

Any update here?

Comment 20

[zhangdapeng]

hello when i change  pref("dom.phonenumber.substringmatching.VE", 10);
it works right from our college says

Comment 21

[Vance Chen [:vchen][skype:vance.lucida][[e-mail]]]

(In reply to zhangdapeng from comment #20)
> hello when i change  pref("dom.phonenumber.substringmatching.VE", 10);
> it works right from our college says

So now everything is OK in 990830? We can close this one?

Comment 22

[Vance Chen [:vchen][skype:vance.lucida][[e-mail]]]

Confirmed with partner, this one has already been fixed, we can close it now

Comment 23

[KTucker [:KTucker][Inactive 3/4/2016]]

Test case has been added here in Moztrap: https://moztrap.mozilla.org/manage/case/13758/