Re: SPARK Examiner -- visibility of Ada.Strings.Unbounded (and probably lots of other packages)

[roderick.chapman@googlemail.com]

Yeah - to process Bar, the Examiner needs to have previously seen
(or at least know where to find) the specification of Foo.

There's no "default" way to do this, since the Examiner is
designed to deal with the program-library structure and naming
scheme of _any_ Ada compiler - we can't just assume that people
name their compilation units and files like GNAT prefers to, for
example.

So...you either need to specify both files on the command line - this
works for small example but rapidly becomes tedious!  e.g.

 spark bar.ads foo.ads

The better solution is to add Bar and Foo to the Examiner's
"Index file" - this establishes a mapping from compilation
unit to filenames that the Examiner can use to automatically
find the specifications of other units.

The index file is set by the -index option so if it
says -index=spark, then you need to add the following
two lines to the spark.idx file

 foo spec is in foo.ads
 bar spec is in bar.ads

OR...you can use the "sparkmake" tool to generate the index file
for you automatically from a group of files.

More details in the book chapter 9 and section 4 of
the Examiner User Manual (which you should have - Examiner_UM.pdf).

I strongly recommend that you read through at least the first
4 chapters of the Examiner User Manual - full of useful info
I promise! :-)

All the best,
 Rod