[lmi] Dependent files for 'make check' [Was: Running unit tests in less time]

[Greg Chicares]

On 2017-05-06 22:22, Vadim Zeitlin wrote:
> On Fri, 5 May 2017 22:31:09 +0000 Greg Chicares <address@hidden> wrote:
>

[... unit tests depend on finding all these files
  sample.ill
  single_cell_document.xsd
[both of which are in .git/../ ]
  configurable_settings.xml
and a set of product files in /opt/lmi/data/ ]

> GC> ...couldn't all those problems could be solved by installing lmi into
> GC> '/opt/lmi/data'?
> 
>  They probably could, but especially under Linux it would be really nice if
> the program could be tested without requiring root permissions, which are
> normally necessary for installing it under /opt.

We could write copies (or symlinks) elsewhere. We already use
  /etc/opt/lmi
for other purposes. These files don't really belong there, though.
What would be the most natural place? $HOME should be available, but
doesn't seem appropriate. How about /var/lmi ?

> But even leaving the
> permission issues aside, it's very unusual to require installing before
> being able to run the tests. I.e. the traditional workflow is "make && make
> check && make install".

Surely this sort of situation must have arisen with other projects;
how do they handle it?

The source directory .git/../ already contains two of the files
mentioned above as well as
  cgi.touchstone
which is used by 'cgi_tests'; is it considered normal for 'make check'
to copy such files from the source directory?

The most natural way to generate product files is to build and run
'product_files'. Is there a conventional way of handling that? It
seems far better to run that program than to maintain repository
copies of the files it creates.

Then there's this file, created as a here-doc by 'install_msw.sh':
  cat >/opt/lmi/data/configurable_settings.xml <<EOF
Is there a convention for files like that?