This is needed to be able to round trip enums.
If there is no codegen for an enum that is read from the datastore we
now return a types.Enum which can be serialized back to the same
sequence.
The generated code for typed structs now uses a Go struct which
implements Value directly. The fields in this struct uses the "user"
type. (The union value still uses types.Value though.)
When a typed struct is created by the decoder, it asks for a struct
builder which returns a channel that the values of the fields of the
struct are sent to.
The output file should always be emitted relative to -out-dir. I was
incorrectly appending the path to the input file onto that provided in
-out-dir to construct the path to the output file.
We'd wound up in a spot where serialization code used 'TypeRefKind' to
mean one of two very different things...either an actual value that
describes some Noms type, or a reference to a type definition that
lives somewhere else. To get rid of this ambiguity, we introduce
'UnresolvedKind' to take over the latter meaning. Now, TypeRefKind
means _only_ a value that describes a type. If you want to point off
to a type definition elsewhere in the type package, or in another
type package, use UnresolvedKind.
grammar.peg didn't get updated along with grammar.peg.go in arv's last patch,
so that needed to be fixed. Also, pkg.Parsed had its own field named Types, which
shadowed the one it got by embedded types.Package. This only came into play when
generating code for packages pulled out of a dataset. Since arv had to manually
patch up all generated code in his last patch, he never hit this issue and I
missed it in review.
Now, go generate passes once more. Yay