Locate missing/extra braces for parsing error

Tell us about your environment

• ESLint Version: 3.9.1
• Node Version: 7.0.0
• npm Version: 3.10.8

What parser (default, Babel-ESLint, etc.) are you using? default

Please show your full configuration:

{
    "rules": {
        "curly": "error"
    }
}

What did you do? Please include the actual source code causing the issue.

function test(a) {
    if (a) 
        doSomething();
    }
    
    doSomethingElse();
}

test(a);

What did you expect to happen?

  2:5  error  Expected { after 'if' condition  curly
  7:1  error  Parsing error: Unexpected token }

✖ 2 problem (2 error, 0 warnings)

What actually happened? Please include the actual, raw output from ESLint.

  7:1  error  Parsing error: Unexpected token }

✖ 1 problem (1 error, 0 warnings)

Actually, I think it works as intended since ESLint have to build the AST first. But it would be nice if we can get the line number of missing braces. The advantage becomes bigger while working on large files.

Perhaps to make the parser generate a “partial AST” for specific parsing error, or let the rule be able to do some additional check that doesn’t depend on the AST.

[platinumazure]

Again, looking at this from a cost-benefit analysis, I don’t think the benefits are anywhere near the massive costs (rewriting the parser and ESLint architecture plus the maintenance headache from maintaining two kinds of rules).

I’m happy to step back and let the TSC weigh in, but I can virtually guarantee this is a non-starter. ESLint depends on a full AST, period, there is no such thing as linting half a file.

[vitorbal]

Thanks for the suggestion, we really appreciate the interest in helping us improve ESLint. However, it seems like the team has a pretty clear position against this suggestion. I’m going to go ahead and close the issue, but feel free to discuss anything you feel is not clear yet.