Re: Does QSHELL sed have a limitation on the file size

Hello Alan,

Am 30.11.2021 um 22:56 schrieb Alan Shore via MIDRANGE-L <midrange-l@xxxxxxxxxxxxxxxxxx>:

In this instance, the process dropped almost half of the records - but original the file was larger than normal

"Drop" as in "They were in the input file but not in the output file" — or "they were unaltered in the output file"?

The only advice I have is to run the input file in question through the sed commands one by one and see where it fails.

Aside from that, efficiency might improve slightly when you delete the CR at the beginning of the pipe (place tr to the begin), and use only one call of sed with multiple -e parameters for each regex to be applied in order.

And - to make matters worse - when we rerun this process - using the same original file, the number of records that are created is different each time
Hence my question - Does QSHELL sed have a limitation on the file size

All sed implementations process input line by line, as does tr — this is what pipes most often are about. As long as your lines aren't hundreds of Megabytes "long", I doubt you have a size problem on the PASE side of affairs.

:wq! PoC