How can error reporting in grammars be improved?

by

If your focus is generating messages for users of your grammar, see Generating Good Parse Errors from a Parser and Grammar::ErrorReporting.

The rest of this answer is about development and debugging of grammars.

May 2020 update The obvious choice now is to use the Grammar Live View feature of the Comma IDE.

First, you can embed arbitrary closures (code) in Raku rules (or tokens or regexes). Just type { your code goes here } in the middle of a rule. So you could just sprinkle { say ... } statements where helpful for debugging. (Note that $/ and its relatives $0, $1 etc. and named sub-captures $<foo> etc. are automatically updated to refer to the current Match object and its sub-captures corresponding to the enclosing rule immediately prior to entering the closure. So you can introspect how the match is going at that point in the regex.)

But there are better options.

Are you using the “batteries included” Rakudo Star distribution? (You should be unless you’ve got good reason not to.) If so, you can add the line use Grammar::Tracer; (as described in the slides at http://www.jnthn.net/papers/2011-yapceu-grammars.pdf) to get a full trace of a parse.

My personal preference (edit: this answer was written years before Comma arrived on the scene; Comma also covers debugging of non-grammar code; and indeed a whole heck of a lot else besides) is to use the Rakudo debugger which provides everything Grammar::Tracer does plus a whole lot more besides (single stepping, breakpoints, etc. of all Raku code including regexes and grammars). The debugger is introduced in this video.

If you have any more questions, I recommend you visit the always friendly IRC channel #raku.

More Related Contents:

  1. How to identify whether a grammar is LL(1), LR(0) or SLR(1)?
  2. The Possibility of a “True” Quiz Generator [closed]
  3. Parse error: syntax error, unexpected T_STRING in app/design/frontend/base/default/template/catalog/product/new.phtml on line 37 [closed]
  4. Writing a parser like Flex/Bison that is usable on 8-bit embedded systems
  5. Why can’t C++ be parsed with a LR(1) parser?
  6. How to parse freeform street/postal address out of text, and into components
  7. What is the difference between an Abstract Syntax Tree and a Concrete Syntax Tree?
  8. Practical difference between parser rules and lexer rules in ANTLR?
  9. lexers vs parsers
  10. Parsing command line arguments in R scripts
  11. Import CSV file with mixed data types
  12. How does the ANTLR lexer disambiguate its rules (or why does my parser produce “mismatched input” errors)?
  13. How would you go about parsing Markdown? [closed]
  14. Handling extra operators in Shunting-yard
  15. How to Traverse an NLTK Tree object?
  16. Difference between constituency parser and dependency parser
  17. Looking for a clear definition of what a “tokenizer”, “parser” and “lexers” are and how they are related to each other and used?
  18. Converting a hexadecimal string to a decimal integer
  19. Reached end of file while parsing compilation error
  20. What is a Context Free Grammar?
  21. Use Scala parser combinator to parse CSV files
  22. Problem with parse a LocalDateTime using java 8
  23. Is it a Lexer’s Job to Parse Numbers and Strings?
  24. How to manually parse a floating point number from a string
  25. Parse URL in shell script
  26. Parse clock time in java 8
  27. How can I read and parse CSV files in C++?
  28. preg_match(); – Unknown modifier ‘+’ [duplicate]
  29. Fastest way to retrieve a in PHP
  30. Python urlparse — extract domain name without subdomain

Leave a Comment