Re: Bug#353911: md5sum --check checks only if _all_ are bad

[Jim Meyering]

Patrick Schoenfeld <address@hidden> wrote:
> ehh.. forgotten to write about this one:
>
> On Fri, Nov 14, 2008 at 09:09:09PM +0100, Jim Meyering wrote:
>>     - demonstrate --check --pedantic failing due to invalid first line
>>         (ensure it still processes following ones -- should it warn
>>         about each? or just the first?)
>
> I don't have a strong opinion about this, but I wonder which behaviour
> would be the best: Process every line without stopping on the first
> invalid line or Process every line until one invalid line is found (like
> it is implemented currently). From your sentence I guess you'd prefer
> the first? Is this a strong requirement in order for the patch to be
> included into coreutils? Otherwise I'd like to hear the submitters
> opinion as well.

It should process every line.
My question was whether to emit a diagnostic for each _malformed_ line,
or just for the first or maybe the first few.