I am writing a lexer for a custom language and want to match a single line of comment starting with # like in python but here comment always starts from the beginning of line. Example:
#this is a comment
and #this is not a comment
I have tried this: "^#" (~["\n"])* ("\n")? but not working. How can I achieve this?
Unfortunately JavaCC does not support the ^ for start of line, nor the $ for end of line.
You can use lexical states. Use the default state * for the start of a line. For example:
// If at the start of a line, skip a # and all other charaters on the line.
// Stay in the same state.
<*> SKIP : { <COMMENT : "#" (~["\n"])* ("\n") > : * }
// Skip any newlines and return to (or stay in) the start of line state.
<*, MIDLINE> SKIP : { <NEWLINE : "\n" > : * }
// Skip any other white space.
<*, MIDLINE> SKIP : { <WHITESPACE : " " | "\t" > : MIDLINE }
// If not at the start of line, a # is a token.
<MIDLINE> TOKEN : { <HASH : "#" > : MIDLINE }
// Numbers can be at the start of a line or the middle of a line.
// After a number, we are no longer at the start of line.
<*, MIDLINE> TOKEN : { <NUMBER : ["0"-"9"]+ > : MIDLINE }
and so on.
