- From: <bugzilla@jessica.w3.org>
- Date: Wed, 09 Apr 2014 21:38:38 +0000
- To: public-webapps-bugzilla@w3.org
https://www.w3.org/Bugs/Public/show_bug.cgi?id=25081 --- Comment #8 from Arun <arun@mozilla.com> --- (In reply to Arun from comment #7) > (In reply to Anne from comment #6) > > Yeah that would be good. > > > > I think if we want to define FormData and keep the concept of closed blobs > > that can be collected, we should have some kind of "keepalive number" on > > blob. By default it would be 0. Whenever an attempt is made to serialize a > > FormData object it would increase the "keepalive number" of the blobs it > > consist of by 1. Meaning that even if the blobs get closed they would still > > be kept alive because they will be read from. After reading the FormData > > object would be responsible for decreasing the number by 1. It needs to be a > > number as multiple FormData objects might reference the same blob. > > > > Does that make sense? > > > OK, but a few questions: > > 1. Is this really a specification detail, or should it be an implementation > detail? If a specification detail, is there a limit on the keepalive number > for Blobs? > > 2. Other than FormData, which API uses the keepalive property on Blobs? > Should this be internal to FormData? That is, FormData is responsible for > keeping records of Blobs it needs after they are neutered? Bug 25302 is to address Comment 6. I've Cc'd you to that bug, since I have questions. -- You are receiving this mail because: You are the QA Contact for the bug.
Received on Wednesday, 9 April 2014 21:38:40 UTC