Re: Parse one fiile, send the records to two different files

Near the end, you have push @data, \@fields;, then earlier, you do: print $ERR_FH join (',', @$_), $/ for @data;, which from what I can tell, is repeatedly printing to the error file all elements of @data, which is being compounded (which is probably why you're seeing 8mil entries).

Should that be:

print $ERR_FH join(',', $_) for @fields;
[download]

Comment on Re: Parse one fiile, send the records to two different files Select or Download Code

Replies are listed 'Best First'.
Re^2: Parse one fiile, send the records to two different files by BigRedEO (Acolyte) on May 27, 2016 at 16:46 UTC
Thank you both - I was simply copying my last PRINT statement at the end of the script when I was adding the IF/ELSE statement. But now I've run into a new problem - I ran the script after making that simple change and it stopped at record 140, which is the first "error" record in the file (line 4 of my sample code in my original post) and is giving me this message, then stopping - Can't use string ("650187016") as an ARRAY ref while "strict refs" in use at AlterData.pl line 39, <$FH> line 140.	[reply]
Re^3: Parse one fiile, send the records to two different files by stevieb (Canon) on May 27, 2016 at 16:53 UTC
Are you still using `@$_` in your join statement? ie. `print $ERR_FH join(',', @$_) for @fields; __^^^__` [download] If so, remove the `@` so that you use the contents of `$_` as-is, without attempting to dereference it as an array: `print $ERR_FH join(',', $_) for @fields; __^^^__` [download]	[reply] [d/l] [select]
Re^4: Parse one fiile, send the records to two different files by BigRedEO (Acolyte) on May 27, 2016 at 17:24 UTC
Nope! STILL didn't work! Now it jumbled it all the fields from the "error" records together into one just line with spaces and no commas. My errorFiles.csv has just one line with every field from the error records. `if (!length $fields[28]) { print $ERR_FH join (',', $_) for @fields; } else` [download] ...is what I'm using. So what else could be missing?	[reply] [d/l]
Re^5: Parse one fiile, send the records to two different files by BrowserUk (Patriarch) on May 27, 2016 at 18:32 UTC
Re^5: Parse one fiile, send the records to two different files by stevieb (Canon) on May 27, 2016 at 18:33 UTC
Re^6: Parse one fiile, send the records to two different files by BigRedEO (Acolyte) on May 27, 2016 at 18:51 UTC
Some notes below your chosen depth have not been shown here
Re^5: Parse one fiile, send the records to two different files by stevieb (Canon) on May 27, 2016 at 18:25 UTC
Re^6: Parse one fiile, send the records to two different files by BigRedEO (Acolyte) on May 27, 2016 at 18:32 UTC
Re^4: Parse one fiile, send the records to two different files by BigRedEO (Acolyte) on May 27, 2016 at 17:16 UTC
Well, the script now ran through and again put only the correct files in my "clean" file, but for the error file, I now have 1138 lines, where I should only have 272 lines. When I did a "head" on that error file, I find that it's listing each field as a separate line and no comma delimiters i.e. - 650187016 2 1 checked out under cash 650200678 1 1 HIT CASH TWICE 650096506 1 whereas I want the lines put back in their original form, all commas included, i.e. 650187016,2,1,checked out under cash ,,,,,,,,,,,,,,,,,,,,,,,,,,,,	[reply]
Re^4: Parse one fiile, send the records to two different files by BigRedEO (Acolyte) on May 27, 2016 at 17:19 UTC
Wait - never mind! Think I caught what else I left in that statement that caused the problem. Am running it again now to verify.	[reply]