DOS ain't dead

bencollver 04.06.2024, 05:21	uncsv 0.9.1 (Announce) Post reply
	uncsv 0.91. Requires an 8086 CPU or better. uncsv is a filter command converting the lines of a CSV file into a delimited file... it will also replace the carriage return and new-line characters found in quoted values, allowing you to use awk to play with your CSVs. https://archive.org/details/uncsv-0.91-for-dos gopher://tilde.pink/1/~bencollver/files/dos/util/uncsv/

bencollver 07.06.2024, 03:55 @ bencollver	uncsv 0.9.1 Post reply
	I posted uncsv091b.zip, which increases the csv.exe line length limit from 4096 to 16382 bytes. Note that uncsv.exe does not have a line length limit.

Rugxulo

Usono,
07.06.2024, 04:20

@ bencollver

uncsv 0.9.1

Post reply

> allowing you to use awk to play with your CSVs.

I only briefly looked at your NAWK builds, and there you seemed to stick to 20230909, right? Even the man file seemed to imply it was an older version (1988 book reference only), but the 2.0 book (from last year) added Unicode and CSV. I guess you knew that already??

* https://www.amazon.com/Programming-Language-Addison-Wesley-Professional-Computing/dp/0138269726

(I privately built it with DJGPP a few months ago but haven't heavily tested it.)

bencollver

07.06.2024, 16:05

@ Rugxulo

uncsv 0.9.1

Post reply

> > allowing you to use awk to play with your CSVs.
>
> I only briefly looked at your NAWK builds, and there you seemed to stick to
> 20230909, right?

Correct. This version seemed to work better for me.

> (1988 book reference only), but the 2.0 book (from last year) added Unicode
> and CSV. I guess you knew that already??

Yes, i built the latest NAWK with Unicode and CSV support. It took some doing in Watcom because NAWK's Unicode support assumes 32-bit int but the Watcom DOS target has 16-bit int. My result would run hello world, but i did not test it heavily.

My DJGPP build crashed a lot. SYMIFY always showed the crash happening somewhere in the C library. I had trouble debugging it because i could not reproduce the crash in GDB. In my experience, this usually means the code assumes that malloc() zeroes out the allocated memory.

One nice thing about uncsv is that it will work with any AWK. FreeDOS has the UTF8TOCP utility to decode Unicode. Another fun utility is json2tsv, which allows you to use AWK to play with JSON data.

uncsv 0.9.1 (Announce)

uncsv 0.9.1

uncsv 0.9.1

uncsv 0.9.1