Fw: Memory change in 5.0.15

[Bill Akers]

Richard Hane wrote:
> Forwarded to fp list
> 
> 
> ----- Forwarded Message ----
> From: Richard Hane <yoresoft at sbcglobal.net>
> To: Nancy Palmquist <nlp at vss3.com>
> Cc: Ray Hoover <rhoover at fptech.com>
> Sent: Thursday, May 15, 2008 9:39:59 AM
> Subject: Re: Memory change in 5.0.15
> 
> 
> [ Sorry for the top post ]
>  
> Nancy, Thanks for the idea but that did not work.  
> Ray, no the dif file is not bad.
>  
> Here is what I found.  If the Excel dif file has more than 40K rows 

The magic number here is somewhere between 43K and 44K rows and is 
apparently an Excel limitation. I had a similar problem during a sales 
tax audit and had to insist that the state would only get an ascii file, 
not Excel spreadsheets like they wanted. I just wasn't interested in 
chopping the data up into multiple spreadsheets or figuring out what was 
causing the problem. It would apparently only have to do with the 
computer memory if it could not find enough ram+page memory to actually 
allow the sheet to be constructed.

>it does not work. It goes through the file but gets no data (Won't even display it with a show statement). This has happened on 3 of the dif files I import.  All others are under 30K rows and work fine.
>  
> Now I have made a work around.  Using selection criteria in the Excel querry (odbc from an SQL file) I am able to limit the records select to keep all rows on all dif files under 35K.  Tested and all worked correctly.
>  
> BTW.. I checked the available memory while running one of these processes and I still had 166 mb free.
>  
> I would like to find out what is causing this but I am functional (insert joke here) at this point.
>  
> Thank you,
> Rick Hane
> Controller
> Deluxe Stitcher Company Inc
> www.deluxestitcher.com
>  
>  
> ----- Original Message ----
> From: Nancy Palmquist <nlp at vss3.com>
> To: Richard Hane <yoresoft at sbcglobal.net>
> Cc: filepro-list at lists.celestial.com
> Sent: Wednesday, May 14, 2008 2:14:38 PM
> Subject: Re: Memory change in 5.0.15
> 
> Richard Hane wrote:
>> Ray,
>>   
>> Here is the batch file that is used.
>>   
>> set PFDSK=V
>> set PFPROG=V:
>> set PFDATA=V:
>> set PFDIR=
>> set PFCMARK=20
>> set PFLMHOST=server2007
>> cd fp
>> dprodir bom -ka
>> rreport acttime -f getbom -s one
>> cd\
>> cls
>> Note: dirve V: is mapped to \\server2007\F\fpodbc 
>> <file://\\server2007\F\fpodbc>
>> The batch file resides in the fpodbc and the bom.dif resides in \fp folder
>>   
>> There is one other dif file that has more rows but I hav not checked it 
>> yet.  I will do so tonight and advise.
>>   
>> The process runs one 1 record in a file called acttime (which is where I 
>> run all my import processes).
>>   
>> Here is the process...
>> 1 (label) = getit  Then: import dif fpo = bom
>> 2. If: not fpo  Then: end
>> 3. Then: a(20,*);q(9,.4);d(10,mdyy/);b(20,*)
>> 4. Then: q=fpo(3); b=fpo(2); a=fpo(1); d=fpo(4)
>> 5. Then: np(5,.0)=np+"1"
>> 6. Then: show ("7","10")  "processing part number"<a
>> 7. Then: show ("9","10")  "no. of parts"<np
>> 8. Then: lookup miv = bom r=free -n
>> 9. If: not miv  Then: errorbox "Part"<a<"Not Created";end
>> 10. Then: miv(1)=a;miv(2)=b;miv(3)=q;miv(4)=d
>> 11. Then write miv
>> 12. Then: goto getit
>> 13: Then: end
>>   
>> When the process does not run correctly it creates 47k blank records.
>>   
>> Thanks
>> Rick
>>   
>>   
>>
> 
> Rick,
> 
> Try:
> 
> PFIMPBUF= 65536
> 
> to increase the maximum record length to 65K for an imported record.
> 
> Nancy
> 
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Filepro-list mailing list
> Filepro-list at lists.celestial.com
> http://mailman.celestial.com/mailman/listinfo/filepro-list