IRC log for #sourcefu, 2018-11-10

http://sourcefu.com

← Previous day | Channels | #sourcefu index | Today | Next day → | Search | Google Search | Plain-Text | plain, newest first | summary

All times shown according to UTC.

Time	S	Nick	Message
14:31		pdurbin	prologic (and others, aditsu maybe): any thoughs on lexers and parsers? Please see https://gitter.im/open-source-society/computer-science?at=5be6d1a42138723405e545e7
14:37		aditsu	I'm probably the wrong person to ask about "best practice", but I think the lexer should go ballistic when it finds an error, unless you want to report multiple errors in one file
22:43		prologic	pdurbin aditsu lexers do not generally report errors
22:43		prologic	they usually have very little context except the token and source position
22:44		prologic	best practices normally dictate the lexer capture not only the token stream but the position in the source including line number and column
22:44		prologic	this is so that the parser can accurately report on parsing errors
22:45		aditsu	prologic: what it the lexer is getting something that can't be a token?
22:45		prologic	the only type of error a lexer can/should report is "invalid token on line x, column y" but beyond that any other kind of error is either impossible or useless
22:45		prologic	^^^
22:45		prologic	you didn't let me finish :)
22:45		aditsu	well, of course, that's what it would report
22:46		prologic	but that's the only error it should report on :)
22:46		aditsu	sure
22:46		prologic	lexers are pretty dumb really
22:46		prologic	_some_ language implemtnations skip lexing altogether

← Previous day | Channels | #sourcefu index | Today | Next day → | Search | Google Search | Plain-Text | plain, newest first | summary

http://sourcefu.com