On 2014/12/01 20:15:33, Dmitry Lomov (chromium) wrote:
On 2014/12/01 20:11:59, marja wrote:
> Parsing can be done on a background thread, so we cannot access
anything on
the
> heap (Handle<String>s) anyway :(
>
> So, like you say, we could store only offsets here and then do the raw
string
> construction run-time... what do you mean by "every single time a tagged
> template is invoked"; would we need to do it in cases where the user
doesn't
> access t.raw too?
But the user will.
I think the right solution is for scanner to generate raw strings in
addition
to
cooked strings.
(My understanding is that you cannot recover a raw string from a cooked
string,
right?)
Right. But the scanner has has access to the underlying string. I'll look
into
getting the data straight from the scanner instead.
https://codereview.chromium.org/768203002/
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
