Using the -u switch for diff in regress.sh

Started by Frank Ridderbuschalmost 28 years ago3 messages
#1Frank Ridderbusch
ridderbusch.pad@sni.de

Hi,

has somebody tried to replace the -c switch with the -u switch
(available with GNU diffutils) in the script regess.sh?

Well, I did it and I find the resulting output much easier on the eye
to compare the current and the expected results.

This change might be worth a little hint in the regress/README file.

--
Frank

#2Bruce Momjian
maillist@candle.pha.pa.us
In reply to: Frank Ridderbusch (#1)
Re: [HACKERS] Using the -u switch for diff in regress.sh

Hi,

has somebody tried to replace the -c switch with the -u switch
(available with GNU diffutils) in the script regess.sh?

Well, I did it and I find the resulting output much easier on the eye
to compare the current and the expected results.

This change might be worth a little hint in the regress/README file.

regress/checkresults has those new flags.

-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)
#3Zeugswetter Andreas SARZ
Andreas.Zeugswetter@telecom.at
In reply to: Bruce Momjian (#2)
Re: [HACKERS] Using the -u switch for diff in regress.sh

has somebody tried to replace the -c switch with the -u switch
(available with GNU diffutils) in the script regess.sh?

Well, I did it and I find the resulting output much easier on the eye
to compare the current and the expected results.

This change might be worth a little hint in the regress/README file.

regress/checkresults has those new flags.

Unfortunately, not all diff's have the -u option (AIX 4.1 :-( so I have to
get a gnu diff)
but it looks like diff -wC3 will produce similar results. Actually
since the diff is so large I prefer diff -wC1 which only gives one line of
context
usually the sql statement.

Andreas