Guy Harris
changed
bug 12958
| What |
Removed |
Added |
| Status |
CONFIRMED
|
INCOMPLETE
|
Comment # 11
on bug 12958
from Guy Harris
(In reply to Pascal Quantin from comment #7)
> IMHO any solution selected will need to be in the JSON export specific code:
> we should not pollute the dissectors code with whatever constraint coming
> from JSON format.
Absolutely.
Dissectors should produce protocol trees containing *all* the fields in the
packet; this means that they can produce multiple instances of a given field.
I see nothing in the ECMA specification:
http://www.ecma-international.org/publications/files/ECMA-ST/ECMA-404.pdf
that requires that an "object" have, for any given name value, only one
name/value pari with that name value, so I see nothing in the ECMA
specification that would mean that the JSON in iec104.json file violates the
specification.
I.e., http://jsonlint.com/ is reporting "errors" that do not appear to be
violations of the JSON specification.
However, if something *other* than the ECMA specification imposes such a
restriction, then there isn't a simple mapping from a protocol tree to JSON,
so, *if* that restriction is important, whatever code we have that produces
JSON would need to do *something* to work around that restriction.
So what is jsonlint.com's rationale for considering a duplicate key to be
something worth complaining about?
You are receiving this mail because:
- You are watching all bug changes.