RE: Problem with based data structures -- RPG400-L

Huh?  What do you mean by "you _CAN'T_ sort/search on subfields"?

I sort on DS array subfields in several apps, though I've gone away from
that technique due to overhead.  Can you elaborate on why you say this
_CAN'T_ be done??!?  I'll not disagree that bsearch and qsort are effective
alternatives, but in most cases, I've been able to accomplish my goals quite
well with sorta and %lookup()...

Eric DeLong
Sally Beauty Company
MIS-Project Manager (BSG)
940-297-2863 or ext. 1863



-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx]On Behalf Of Wilt, Charles
Sent: Thursday, February 09, 2006 6:59 AM
To: RPG programming on the AS400 / iSeries
Subject: RE: Problem with based data structures


Tim,

Why other to do the copy from the fixed version of the array to the
variable size one?

Instead of trying to use %lookup and SORTA, which don't work well with
DS arrays ( you _CAN'T_ sort/search on subfields)*, use the bsearch and
qsort C runtime functions.

There are plenty of examples on the web and in the archives of using the
bsearch and qsort from within RPGLE.

*note: I've seen some interesting hoops here on the people have jumped
through trying to work around this, some work, some don't really.
However, using qsort and bsearch are the best solutions.  They are easy
and 100% effective.  

HTH,

Charles Wilt
--
iSeries Systems Administrator / Developer
Mitsubishi Electric Automotive America
ph: 513-573-4343
fax: 513-398-1121
  

> -----Original Message-----
> From: rpg400-l-bounces@xxxxxxxxxxxx 
> [mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Tim Kredlo
> Sent: Wednesday, February 08, 2006 9:19 PM
> To: RPG400-L@xxxxxxxxxxxx
> Subject: Problem with based data structures
> 
> Help!
> 
>  
> 
> I am having difficulty accomplishing the following:
> 
> I need to create and fill a data structure that will hold an 
> array of up to
> approximately 2500 elements.
> 
> The exact number will not be known until run time.
> 
> Each element of the array will be like (likeds) a 94 character data
> structure.
> 
> I need to be able to resize the 2500 element data structure 
> to remove any
> unfilled elements.
> 
> I need to be able to sort the data structure by any of the sub-fields.
> 
>  
> 
> This is what I have done so far:
> 
>  
> 
> Declared a data structure 'template' for each element of both 
> arrays (Dlr
> and Dlr1):
> 
> D DlrDs           Ds                  Qualified          
> 
> D  Idx                                Like(Int10)        
> 
> D  Code                               Like(CsRecd.CsDele)
> 
> D  Type                               Like(A_1)          
> 
> D  Grp#                               Like(CsRecd.CsGrp#)
> 
> D  Dlr#                               Like(CsRecd.CsDlr#)
> 
> D  Name                               Like(CsRecd.CsName)
> 
> D  Stat                               Like(DaRecd.DaStat)
> 
> D  City                               Like(DaRecd.DaCity)
> 
> A_1 is declared as 1A
> 
> CsRecd.CsDele is declared as 1A
> 
> CsRecd.CsGrp# is declared as 6P 0
> 
> CsRecd.CsDlr# is declared as 6P 0
> 
> CsRecd.CsName is declared as 40A
> 
> DaRecd.DaStat is declared as 10A
> 
> DaRecd.DaCity is declared as 20A
> 
>  
> 
> Declared 2 pointers:
> 
> D DlrBase         S               *   Inz 
> 
> D DlrBas1         S               *   Inz
> 
>  
> 
> Declared two data structures based on the above pointers:
> 
> D                 Ds                  Based(DlrBase)      
> 
> D  Dlr                          94A   Varying             
> 
> D                                     Dim(2500)           
> 
> D                                     Ascend              
> 
> D  DlrIdx                             Like(DlrDs.Idx)     
> 
> D                                     Overlay(Dlr : 1)    
> 
> D  DlrCode                            Like(DlrDs.Code)    
> 
> D                                     Overlay(Dlr : *Next)
> 
> D  DlrType                            Like(DlrDs.Type)    
> 
> D                                     Overlay(Dlr : *Next)
> 
> D  DlrGrp#                            Like(DlrDs.Grp#)    
> 
> D                                     Overlay(Dlr : *Next)
> 
> D  DlrDlr#                            Like(DlrDs.Dlr#)    
> 
> D                                     Overlay(Dlr : *Next)
> 
> D  DlrName                            Like(DlrDs.Name)    
> 
> D                                     Overlay(Dlr : *Next)
> 
> D  DlrCity                            Like(DlrDs.City)    
> 
> D                                     Overlay(Dlr : *Next)
> 
> D  DlrStat                            Like(DlrDs.Stat)    
> 
> D                                     Overlay(Dlr : *Next)
> 
>  
> 
> D                 Ds                  Based(DlrBas1)       
> 
> D  Dlr1                         94A   Dim(2500)            
> 
> D                                     Ascend               
> 
> D  Dlr1Idx                            Like(DlrDs.Idx)      
> 
> D                                     Overlay(Dlr1 : 1)    
> 
> D  Dlr1Code                           Like(DlrDs.Code)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
> D  Dlr1Type                           Like(DlrDs.Type)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
> D  Dlr1Grp#                           Like(DlrDs.Grp#)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
> D  Dlr1Dlr#                           Like(DlrDs.Dlr#)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
> D  Dlr1Name                           Like(DlrDs.Name)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
> D  Dlr1City                           Like(DlrDs.City)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
> D  Dlr1Stat                           Like(DlrDs.Stat)     
> 
> D                                     Overlay(Dlr1 : *Next)
> 
>  
> 
> I had hoped that I could have declared these using 'likeds', 
> but I don't
> think I could have made them 'varying' or been able to sort by the
> sub-fields. Please tell me I am wrong about that, because it 
> sure would make
> maintenance easier.
> 
> What I would really like to be able to do is "SortArray
> (QlfdDs((QlfdDs.Field1:Ascend)(QlfdDs.Field2:Descend)))".
> 
> (In my dreams)
> 
>  
> 
> I then created 2 user spaces, initialized with blanks('x'40') 
> and retrieved
> their pointers (DlrBase and DlrBas1).
> 
> I filled elements of Dlr1, the non-varying array, from files 
> till done and
> kept the number of elements filled (Idx = 1820 today).
> 
>  
> 
> The following statement was then executed to fill Dlr and 
> resize it to only
> contain the filled elements:
> 
> %SubArr( Dlr : 1 : Idx ) = %SubArr ( Dlr1 : 1 : Idx );
> 
>  
> 
> Immediately after that statement was executed, debug evals showed the
> following:
> 
> > EVAL dlr1(1)                                                
>            
> 
>   DLR1(1) =                                                   
>            
> 
>             
> ....5...10...15...20...25...30...35...40...45...50...55...60 
> 
>        1   '   -AA000000000050MISCELLANEOUS                   
>           '
> 
>       61   '                                  '               
>            
> 
> > EVAL dlr(1)                                                 
>            
> 
>   DLR(1) =                                                    
>            
> 
>             
> ....5...10...15...20...25...30...35...40...45...50...55...60 
> 
>        1   '   -AA000000000050MISCELLANEOUS                   
>           '
> 
>       61   '                                    ' 
> 
> I don't know if the e-mail formatting will show this 
> correctly, but the end
> "'" for DLR(1) is in position 95 and the end "'" for DLR(1) 
> is in position
> 97.
> 
>  
> 
> I then did the following debug evals:                       
> 
> > EVAL dlr1name(1)                                        
> 
>   DLR1NAME(1) = 'MISCELLANEOUS                           '
> 
> > EVAL dlrname(1)                                         
> 
>   DLRNAME(1) = '50MISCELLANEOUS                         '
> 
>  
> 
> Notice that the two data structures (Dlr and Dlr1) are 
> declared identically,
> but the DlrName field now appears to start 2 characters prior 
> to Dlr1name
> field.
> 
>  
> 
> I thought that perhaps this was an 'align' issue, so stuck 
> 'align' on the
> declaration for both Dlr and Dlr1 and recompiled. (I don't 
> know if there is
> more that I would need to do.)There were no warnings in the 
> compile listing
> about mis-aligned fields. I re-ran and received the same results.
> 
>  
> 
> I am stuck as to where to look next.
> 
>  
> 
> All code and debug info was 'cut and pasted' into this email. 
> So if you see
> what appears to be a typo, it is probably wrong in my code 
> and needs to be
> pointed out.
> 
> Thanks in advance for any help. 
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
> -- 
> This is the RPG programming on the AS400 / iSeries (RPG400-L) 
> mailing list
> To post a message email: RPG400-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
> or email: RPG400-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/rpg400-l.
> 
>