Re: misleading error message from variable modifier

[don fong]

Chet Ramey wrote:
> What are the most important features that you consider to lack unit tests?

are there any tests that cover the variable modifiers, either unit tests or
functional tests?

is any of subst.c covered by unit tests?

how do unit tests get run on this project?  i don't see a unit test target
in
the Makefile.  the "make tests" target mentioned in the INSTALL file,
seems to do only functional tests.




On Sun, Mar 4, 2018 at 1:59 PM, Chet Ramey <chet.ramey@case.edu> wrote:

> On 3/2/18 4:45 PM, don fong wrote:
> > Chet, thanks for the explanation about CHANGES.  i am not familiar with
> the
> > bash release process.
> >
> > as for this:
> >
> >> I didn't think the tests were necessary.
> >
> > what standard are you using to judge whether tests are necessary?  does
> the
> > bash project have any coverage metrics?
>
> It's admittedly subjective, but the bash test suite is very large and
> concentrates on testing code that has undergone more changes than this,
> and is more heavily used.
>
> In this case, I didn't think a fix that added two lines of code to a
> relatively stable function needed a test to ensure that it doesn't change.
>
> > as far as i can tell, the vast majority of the C code here seems to lack
> > unit tests as well.
>
> What are the most important features that you consider to lack unit tests?
>
> --
> ``The lyf so short, the craft so long to lerne.'' - Chaucer
>                  ``Ars longa, vita brevis'' - Hippocrates
> Chet Ramey, UTech, CWRU    chet@case.edu    http://tiswww.cwru.edu/~chet/
>