Variables should be accessed by adjacent level

[asatklichov]

Currently, variables can be accessed by any order, like below, and without any error. See the workspace shared.
Right behavior is expecting error message - 'Unknown identifier 'CITY''

Also, the "repeated declaration" error is reported (see screen shot), but variable is uniquely defined.

CUST_RECORD.CITY = 'Nisa';
INQCUST_COMMAREA.INQCUST_DOB_MM = 'Ani';

 DCL 1 CUST_RECORD,  /*repeated definition*/
         2 CUSTOMER_ADDR,
           3 HOUSE_NUM       CHAR(5),
           3 STREET_NAME     CHAR(15),
           3 CITY            CHAR(15),
         2 FULL_NAME         CHAR(30),
         2 ACCOUNT           PIC '9999999V99',
         2 FIELD             CHAR(15);

[msujew]

@asatklichov Do you also have this nested.pli file in your workspace?

 %DCL CSTPLI CHARACTER;
 %DCL EDA  CHARACTER;

 %CSTPLI = 'myInc.pli';
 %EDA = 'nest2'; 
 
 %INSCAN CSTPLI;
 %INSCAN EDA;


 %INSCAN CSTPLI;

It uses %INSCAN CSTPLI; twice. If you remove one of those, the errors will disappear. This is working as intended.

[asatklichov]

I tried it, but the error is still there. Actually error is in unrelated file 'preprocessor.pli'

…

On Mon, Jul 28, 2025 at 3:51 PM Mark Sujew ***@***.***> wrote: *msujew* left a comment (zowe/zowe-pli-language-support#303) <#303 (comment)> @asatklichov <https://github.com/asatklichov> Do you also have this nested.pli file in your workspace? %DCL CSTPLI CHARACTER; %DCL EDA CHARACTER; %CSTPLI = 'myInc.pli'; %EDA = 'nest2'; %INSCAN CSTPLI; %INSCAN EDA; %INSCAN CSTPLI; It uses %INSCAN CSTPLI; twice. If you remove one of those, the errors will disappear. This is working as intended. — Reply to this email directly, view it on GitHub <#303 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/APCOB6SMQPLX45Y7MU6HPR33KYTENAVCNFSM6AAAAACCQ6ATYKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMRXGM2DANJSGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- *Azat Satklycov* Senior R&D Software Engineer | code4z | test4z ***@***.*** ***@***.***>

------------------------------------------------------------------- Broadcom CA CZ, s.r.o. ICO: 25694073 V Parku 2314/8 | Prague 4 Chodov, 148 00 | Czech Republic

[asatklichov]

Yes repeated declaration goes, but variable access issue is there On Mon, Jul 28, 2025 at 3:56 PM Azat Satklycov ***@***.***> wrote:

…

I tried it, but the error is still there. Actually error is in unrelated file 'preprocessor.pli' On Mon, Jul 28, 2025 at 3:51 PM Mark Sujew ***@***.***> wrote: > *msujew* left a comment (zowe/zowe-pli-language-support#303) > <#303 (comment)> > > @asatklichov <https://github.com/asatklichov> Do you also have this > nested.pli file in your workspace? > > %DCL CSTPLI CHARACTER; > %DCL EDA CHARACTER; > > %CSTPLI = 'myInc.pli'; > %EDA = 'nest2'; > > %INSCAN CSTPLI; > %INSCAN EDA; > > > %INSCAN CSTPLI; > > It uses %INSCAN CSTPLI; twice. If you remove one of those, the errors > will disappear. This is working as intended. > > — > Reply to this email directly, view it on GitHub > <#303 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/APCOB6SMQPLX45Y7MU6HPR33KYTENAVCNFSM6AAAAACCQ6ATYKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMRXGM2DANJSGQ> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> > -- *Azat Satklycov* Senior R&D Software Engineer | code4z | test4z ***@***.*** ***@***.***> ------------------------------------------------------------------- Broadcom CA CZ, s.r.o. ICO: 25694073 V Parku 2314/8 | Prague 4 Chodov, 148 00 | Czech Republic

-- *Azat Satklycov* Senior R&D Software Engineer | code4z | test4z ***@***.*** ***@***.***>

------------------------------------------------------------------- Broadcom CA CZ, s.r.o. ICO: 25694073 V Parku 2314/8 | Prague 4 Chodov, 148 00 | Czech Republic

[msujew]

but variable access issue is there

Which issue? The only issue visible in the images is the duplicate declaration (due to the duplicate INSCAN).

[asatklichov]

but variable access issue is there

Which issue? The only issue visible in the images is the duplicate declaration (due to the duplicate INSCAN).

Here, you can access variable which is not adjacent

CUST_RECORD.CITY = 'Nisa';
INQCUST_COMMAREA.INQCUST_DOB_MM = 'Ani';

[msujew]

@asatklichov But that is correct. In PLI, as long as there's no ambiguity, you can "skip" (i.e. partially qualify) variables to access them. This is working as it should 👍

[asatklichov]

Here is the outcome on different workspace, I expect this is the correct behavior. So we have two different behavior, above one is a bug.