It is define. Oh, wait, my bad: I pass 2>&/dev/null to the command to surpress 
the one message on start-up, and if in this version all error messages go to 
stderr, which they seem to do, it's explained. All's fine and dandy in the 
universe again.



Mar 4, 2023, 22:16 by [email protected]:

> Even if eformat_j_ is undefined or fails, you should get the old terse 
> messages.
>
> Henry Rich
>
> On 3/4/2023 3:12 PM, 'Viktor Grigorov' via General wrote:
>
>> I updated to the official 941 and I'm getting no error messages. I'd had a 
>> eformat.ijs missing error, so I rm -r 'd the /usr/share/j/9sth/addons dir, 
>> and cp -r 'd the one in the zip thereto. No more messages, but also no error 
>> messages. My jsource clone, built yesterday, spits out errors messages for 
>> the same inputs, e.g. length errors for 1 2 * 1 2 3.
>>
>> How can diagnose it? Should I even, that is, should I stick with the source?
>>
>> Also, I noticed -O2 being passed to gcc for linux---how come not -O3?
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to