Re: Lexer and syntax highlighter powered by LPeg

[Peter Odding <xolox@...>]

mitchell wrote:
> I am having trouble understanding the following. (I don't think it's
> documented very clearly)
You mean my code isn’t documented clearly enough? I’ll add some more comments, 
though my problem is always deciding which parts need them, since LPeg already 
has clearly defined operators and spelling everything out in human language 
quickly gets redundant.

> > P "'" * ((1 - S "'\r\n\f\\") + (P '\\' * 1)) ^ 0 * "'" 
> What is the P '\\' * 1 for? Does it match a \ and then any single character?
Correct. A plain number matches/consumes it’s own number of characters. I should 
probably have matched the full escape but I didn’t because I don’t think there 
are any error cases to handle. This would be the correct pattern:

local escape = S ""\r\n\f\\"
local sq_string = P "'" * ((1-escape) + (P'\\' * escape))^0 * "'"

also just noticed that the above could be made faster (I think):

local sq_string = P "'" * ((1-escape)^0 + (P'\\' * escape))^0 * "'"

Have a nice day,

 - Peter Odding