[lojban] parsing with error detection and recovery

["Chris Capel" <lojban-out@lojban.org>]

I'm looking into implementing a friendly PEG parser. The current PEG
grammar (and morphology) are very unfriendly, in that invalid lojban
text is simply not parsable, as opposed to being parsable with
possible errors listed. But a parser with error detection could be
easily based on the existing PEG grammars by adding additional rules
(with lower precedence than any rules for valid Lojban) that are
specially marked and are associated with descriptive error messages.
Adding these rules would also add substantial error recovery/tolerance
to parsers.

For instance, the morphology rules in the BPFK Peg Morphology[1] will
only parse consonants that don't appear in invalid consonant clusters.
If a consonant cluster is invalid, it will stop parsing. But by adding
error rules for consonants that don't check the validity (that only
get matched if the ones that do check don't match) or that check for
specific kinds of invalid pairs, the output of the parser could be
more likely to finish, and could tell the user why the cluster is
invalid.

Composing a good set of these rules would definitely be quite an art,
but seems like a good approach.

So, my question is this: is there an easy way to prove the equivalence
of PEG parser A with the parts of parser B that apply only to valid
input? My first hunch is that as long as B is derived from A by only
adding rules where the added rule is an error condition if ever
matched in an input, and by only modifying existing rules either by
renaming them (and all references to them) or by adding options to the
end that point toward error rules, then parser B will return a parse
tree with no matches on error rules if and only if parser A would be
able to parse the input at all. But I'm not completely sure that's the
case.

Chris Capel
--
"What is it like to be a bat? What is it like to bat a bee? What is it
like to be a bee being batted? What is it like to be a batted bee?"
-- The Mind's I (Hofstadter, Dennet)


To unsubscribe from this list, send mail to lojban-list-request@lojban.org
with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if
you're really stuck, send mail to secretary@lojban.org for help.