libvcc: Fix STRING/STRANDS comparison without STRING folding

[nigoroll]

I do not consider myself enough of an expert on VCC and would appreciate reviews by those who do.
Side note: STRING vs STRINGS vs. STRANDS is confusing

Fixes #4146

[nigoroll]

bugwash @walid-git was asked to review, thank you in advance.

[walid-git]

Even though the changes generate syntactically correct C code and that the STRING/STRANDS comparison error is gone, there is still something wrong here:
With these changes, the following VCL code:

 if ("string" == debug.return_strands("string")) {
                    ...
        }

Will generate this C code:

 if (
        (0 == VRT_CompareStrands(TOSTRAND("string"), vrt_null_strands))
  ) {
  ...

The STRANDS part of debug.return_strands("string") is being translated to vrt_null_strands which is wrong. I need to look closer to find which part of the PR is causing this, but that might be related to the first commit. We should also improve test coverage to not only test the syntactic part but also the semantic.

[nigoroll]

@walid-git thank you very much for catching this significant regression. I have force-pushed an update, the delta is

diff --git a/lib/libvcc/vcc_expr.c b/lib/libvcc/vcc_expr.c
index a348e9b83..cd4b03c2f 100644
--- a/lib/libvcc/vcc_expr.c
+++ b/lib/libvcc/vcc_expr.c
@@ -159,7 +159,11 @@ static void
 vcc_strands_edit(const struct expr *e1, const struct expr *e2)
 {
 
-       if (e2->nstr == 0)
+       assert(e2->fmt == STRANDS || e2->fmt == STRINGS);
+
+       if (e2->fmt == STRANDS)
+               VSB_cat(e1->vsb, VSB_data(e2->vsb));
+       else if (e2->nstr == 0)
                VSB_printf(e1->vsb, "vrt_null_strands");
        else if (e2->nstr == 1)
                VSB_printf(e1->vsb, "TOSTRAND(%s)", VSB_data(e2->vsb));

edit: the first force-push applied the fix to the wrong commit

[walid-git]

Thanks @nigoroll. This looks good to me now. I left a comment for test coverage improvement.