Hi,
Did you tried to inspect the PPFailure in the Moose? There is a tab with a tree structure giving you a pretty good overview, what is going wrong... Or as an alternative, one can call: myParser enableDebug parse:myInput
Otherwise, PetitParser really needs some nice error reporting system. I would have integrated one, but I am not aware of any suitable solution :(
Cheers, Jan
On Fri, Jun 12, 2015 at 8:07 AM Francisco Garau francisco.garau@gmail.com wrote:
Hi Camile
On 11 Jun 2015, at 10:47, Camille camille.teruel@gmail.com wrote:
On 11 Jun 2015, at 09:30, Jan Kurš kurs@iam.unibe.ch wrote:
That sounds really cool and useful extension.
Thank you Jan!
Regarding the furthest failure, the core of the problem is the distinction between an error and a failure. Error reports on a problem in the input, while failure is information for choice parser combinator. In general, the furthest failure is a better approximation of an error than the last failure, so we use it.
I am not sure what exactly is the problem in case of PrattParser. I guess the last failure gives better results for a user?
Yes, with furthest failure I get errors from tokenization instead of my errors. For exemple with the calculator grammar I gave in my first mail when I parse ‘1+’ the furthestFailure is ‘$- expected at 2’ whereas I return ‘expression expected at 2’ because there's a whole expression missing. Same thing with ‘(1+2’ that returns ‘digit expected at 4’ instead of ‘$) expected at 4’.
But furthest failure gives wrong messages in other cases to. Consider this sequence parser:
keyword := #letter asParser plus , $: asParser. keyword parse: ‘foo'
This returns 'letter expected at 3’, but no matter how many letters I add to the end I’ll still get ‘letter expected’. I want to get what is really missing: '$: expected at 3’.
Any of those error messages wouldn't be too bad if the failing production rule were also mentioned. Something like "keyword rule failed with xx expected at 3"
But yeah, that would require annotating the production rules which can pollute the clarity of the grammar definition...
Maybe returning the “latest furthest failure” instead of the “first furthest failure” could solves the problem here (i.e. replacing > with >= in PPContext>>#noteFailure:)?
One has to consider a pratt parser included in the normal parser, e. g. Expressions parsed by pratt in a Java Grammar. Depending where an error occurs, different strategy for choosing the proper failure is necessary :-/
Indeed, my hack (redefining #parseWithContext:) works only when the Pratt parser is the top parser, but a soon as I compose it I’m screwed because only parseOn: is sent to the Pratt parser. That’s why I wonder if letting the parser decide what to return wouldn’t solve the problem: by default the furthest failure but special parsers can still decide.
Regarding tokenization, there is a message token, that returns PPTokenParser, which transforms a parsed input into the PPToken object. Perhaps this might be helpful?
The Pratt tokens are special: a token points back to the parser that generated it (its “kind”). PPTokenKind subclasses PPFlatteningParser and generates instances of PPPrattToken. A PPTokenKind stores the precedence, the action to be executed when a token of this kind is met at the start of an expression (for terminals and prefixes) and the action to be executed when a token is met in the middle of an expression (for postfixes and infixes).
Cheers, Camille
Cheers Jan
On Wed, Jun 10, 2015, 20:52 Richard Sargent < richard.sargent@gemtalksystems.com> wrote:
camille teruel wrote
On 10 Jun 2015, at 19:11, Chris Cunningham <
cunningham.cb@
> wrote:
Inteteresting....
On Wed, Jun 10, 2015 at 9:36 AM, Camille <
camille.teruel@
<mailto:
camille.teruel@
>> wrote:
Hello Pharoers and Moosers,
I did a Pratt parser extension for PetitParser.
<snip>
@PP Devs: I had trouble with the PPContext furthestFailure that is taken into account instead of the failures I return, so I had to redefine #parseWithContext: to return the failures I want. The results given by furthestFailure were not very meaningful in my case (the same is true
for
PPExpressionParser btw). But I guess it was introduced because it gives good results in other cases. So would it be possible to change this behavior to let the parser
decide
if it returns the furthestFailure or the original failure?
The intent behind the furthestFailure is that it give the failure that gets the furthest into the source stream. It is most useful when there are embedded choice operators in the parser - the original/pre furthest behaviour would return the last failure, which depending on the
incoming
stream and the order of the choice options could be significantly not useful.
I ran into this when working with the sql parser, which started off
with
the outer choice of (by memory): ^ selectStatement / insertStatement / updateStatement / deleteStatement If I was trying to part a select statement that had an error at the
very
end of the statement, the parser would return an error talking about
how
the incoming stream failed in deleteStatement. Not useful.
I would be saddened if this further failure was not available.
Yes in that case returning the furthest failure gives better results. However, this don’t give meaningful messages in all cases. For exemple with the calculator I gave in my previous message, if I
parse
‘1+’ I want to get ‘expression expected at: 2’ but instead it returns
‘$-
expected at 2'. I’m not proposing to remove this feature but to let parsers decide to
use
it or not. Something like (changes in bold):
PPParser>>parseWithContext: context | result | context initializeFor: self. result := self parseOn: context.
"Return the furthest failure, it gives better results than thelast
failure" (result isPetitFailure and: [ self wantsFurthestFailure and: [
context
furthestFailure notNil ] ]) ifTrue: [ ^ context furthestFailure ]. ^ result
This screams at me. Why not just delegate to the context and use a context that returns the preferred failure? e.g. end with: ^context preferredResultFor: result.
PPParser>>wantsFurthestFailure ^ true
Like this, one can return the failures he wants.
PPPrattParser>>wantsFurthestFailure ^ false
Camille
-cbc
-- View this message in context: http://forum.world.st/Pratt-Parsers-for-PetitParser-tp4831456p4831486.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.
On 12 Jun 2015, at 09:36, Jan Kurš kurs@iam.unibe.ch wrote:
Hi,
Did you tried to inspect the PPFailure in the Moose? There is a tab with a tree structure giving you a pretty good overview, what is going wrong... Or as an alternative, one can call: myParser enableDebug parse:myInput
I didn’t know about these features, there are cool :) Thanks!
Otherwise, PetitParser really needs some nice error reporting system. I would have integrated one, but I am not aware of any suitable solution :(
I would be happy to discuss any idea you’ll have on that subject :)
Cheers, Jan
On Fri, Jun 12, 2015 at 8:07 AM Francisco Garau <francisco.garau@gmail.com mailto:francisco.garau@gmail.com> wrote:
Hi Camile
On 11 Jun 2015, at 10:47, Camille <camille.teruel@gmail.com mailto:camille.teruel@gmail.com> wrote:
On 11 Jun 2015, at 09:30, Jan Kurš <kurs@iam.unibe.ch mailto:kurs@iam.unibe.ch> wrote:
That sounds really cool and useful extension.
Thank you Jan!
Regarding the furthest failure, the core of the problem is the distinction between an error and a failure. Error reports on a problem in the input, while failure is information for choice parser combinator. In general, the furthest failure is a better approximation of an error than the last failure, so we use it.
I am not sure what exactly is the problem in case of PrattParser. I guess the last failure gives better results for a user?
Yes, with furthest failure I get errors from tokenization instead of my errors. For exemple with the calculator grammar I gave in my first mail when I parse ‘1+’ the furthestFailure is ‘$- expected at 2’ whereas I return ‘expression expected at 2’ because there's a whole expression missing. Same thing with ‘(1+2’ that returns ‘digit expected at 4’ instead of ‘$) expected at 4’.
But furthest failure gives wrong messages in other cases to. Consider this sequence parser:
keyword := #letter asParser plus , $: asParser. keyword parse: ‘foo'
This returns 'letter expected at 3’, but no matter how many letters I add to the end I’ll still get ‘letter expected’. I want to get what is really missing: '$: expected at 3’.
Any of those error messages wouldn't be too bad if the failing production rule were also mentioned. Something like "keyword rule failed with xx expected at 3"
But yeah, that would require annotating the production rules which can pollute the clarity of the grammar definition...
Maybe returning the “latest furthest failure” instead of the “first furthest failure” could solves the problem here (i.e. replacing > with >= in PPContext>>#noteFailure:)?
One has to consider a pratt parser included in the normal parser, e. g. Expressions parsed by pratt in a Java Grammar. Depending where an error occurs, different strategy for choosing the proper failure is necessary :-/
Indeed, my hack (redefining #parseWithContext:) works only when the Pratt parser is the top parser, but a soon as I compose it I’m screwed because only parseOn: is sent to the Pratt parser. That’s why I wonder if letting the parser decide what to return wouldn’t solve the problem: by default the furthest failure but special parsers can still decide.
Regarding tokenization, there is a message token, that returns PPTokenParser, which transforms a parsed input into the PPToken object. Perhaps this might be helpful?
The Pratt tokens are special: a token points back to the parser that generated it (its “kind”). PPTokenKind subclasses PPFlatteningParser and generates instances of PPPrattToken. A PPTokenKind stores the precedence, the action to be executed when a token of this kind is met at the start of an expression (for terminals and prefixes) and the action to be executed when a token is met in the middle of an expression (for postfixes and infixes).
Cheers, Camille
Cheers Jan
On Wed, Jun 10, 2015, 20:52 Richard Sargent <richard.sargent@gemtalksystems.com mailto:richard.sargent@gemtalksystems.com> wrote: camille teruel wrote
On 10 Jun 2015, at 19:11, Chris Cunningham <
cunningham.cb@
> wrote:
Inteteresting....
On Wed, Jun 10, 2015 at 9:36 AM, Camille <
camille.teruel@
<mailto:
camille.teruel@
>> wrote:
Hello Pharoers and Moosers,
I did a Pratt parser extension for PetitParser.
<snip>
@PP Devs: I had trouble with the PPContext furthestFailure that is taken into account instead of the failures I return, so I had to redefine #parseWithContext: to return the failures I want. The results given by furthestFailure were not very meaningful in my case (the same is true for PPExpressionParser btw). But I guess it was introduced because it gives good results in other cases. So would it be possible to change this behavior to let the parser decide if it returns the furthestFailure or the original failure?
The intent behind the furthestFailure is that it give the failure that gets the furthest into the source stream. It is most useful when there are embedded choice operators in the parser - the original/pre furthest behaviour would return the last failure, which depending on the incoming stream and the order of the choice options could be significantly not useful.
I ran into this when working with the sql parser, which started off with the outer choice of (by memory): ^ selectStatement / insertStatement / updateStatement / deleteStatement If I was trying to part a select statement that had an error at the very end of the statement, the parser would return an error talking about how the incoming stream failed in deleteStatement. Not useful.
I would be saddened if this further failure was not available.
Yes in that case returning the furthest failure gives better results. However, this don’t give meaningful messages in all cases. For exemple with the calculator I gave in my previous message, if I parse ‘1+’ I want to get ‘expression expected at: 2’ but instead it returns ‘$- expected at 2'. I’m not proposing to remove this feature but to let parsers decide to use it or not. Something like (changes in bold):
PPParser>>parseWithContext: context | result | context initializeFor: self. result := self parseOn: context.
"Return the furthest failure, it gives better results than the lastfailure" (result isPetitFailure and: [ self wantsFurthestFailure and: [ context furthestFailure notNil ] ]) ifTrue: [ ^ context furthestFailure ]. ^ result
This screams at me. Why not just delegate to the context and use a context that returns the preferred failure? e.g. end with: ^context preferredResultFor: result.
PPParser>>wantsFurthestFailure ^ true
Like this, one can return the failures he wants.
PPPrattParser>>wantsFurthestFailure ^ false
Camille
-cbc
-- View this message in context: http://forum.world.st/Pratt-Parsers-for-PetitParser-tp4831456p4831486.html http://forum.world.st/Pratt-Parsers-for-PetitParser-tp4831456p4831486.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com http://nabble.com/.
-
Camille -
Jan Kurš