Search code examples
parsingpeggrako

Why does the Grako parsing process fail if my grammar contains an expression that consists of many or-concatenated subexpressions?

I am using Grako. In my EBNF grammar, I have an expression that consists of a lot of subexpressions that are concatenated using the OR-operator, like so:

expression = subexpressionA | subexpressionB | ...  | subexpressionZ;

The parsing process always fails if the input string contains one of the latter subexpressions, say subexpressionZ. When I rewrite the grammar like this

expression = subexpressionZ | subexpressionB |  ...  | subexpressionA;

the parsing process finishes successfully if the input string contains subexpressionZ but will now fail if it contains subexpressionA.

Has anyone ever had a similar problem? Is that a bug in Grako (I am using 3.6.3.) or am I doing something wrong?

Thanks a lot for any ideas!

Solution

  • I solved my problem - a long time ago :) - by splitting up the expressions in a number of sub-expressions like so:

    expression1 = subexpressionA | subexpressionB | subexpressionC;
expression2 = subexpressionD | subexpressionE | ...  | subexpressionZ;
expression = expression1 | expression2;

    For some reason, this works...