Convert ~ and ! prefix tokens to esprima #276
Conversation
| type.isAssign) { | ||
|
|
||
| type.isAssign || | ||
| (type.prefix && (token.value === "~" || token.value === "!")) |
There was a problem hiding this comment.
Do we need to filter further if type.prefix is true? Is there ever a situation where a prefix isn't a punctuation? Or are there other punctuator prefixes that this should also cover?
There was a problem hiding this comment.
I looked through the acorn code and I'm pretty sure that currently if type.prefix is true, the value is either ! or ~. I only added the value comparison, so that in future if acorn adds another prefix (which might not be a Punctator?) it is not immediately converted to Punctator.
Do you think it is safe to assume that prefix is always a Punctator?
There was a problem hiding this comment.
I think you can safely say if the first character of the token is not an identifier start, then it's a punctuator.
There was a problem hiding this comment.
I just checked again: typeof, void and delete are also prefixes. So a differentiation is necessary.
|
It seems that my change also makes the testsuite hang on the XMLHttpRequest test. I can't figure out why my change is breaking stuff. Actually it looks like the assert is hanging when comparing the results. |
|
I finally fixed the tests. They were hanging because i forgot to update the library results and assert can't handle diffs that big. |
|
This looks great, thanks! I'll do a release later today. |
The two prefixes are now converted to Punctator like esprima is returning them.
fixes #274