482810 - Sane file I/O in SpiderMonkey

Status: RESOLVED WONTFIX

(Core :: JavaScript Engine, defect)

Description

[Jason Orendorff [:jorendorff]]

The scope of this bug is:

- hg rm jsfile.{h,cpp}
- add a new ByteArray constructor for mutable byte-oriented binary buffers
- add a new File class for reading and writing files in binary mode
- add a few functions for converting ByteArrays to strings and vice versa, in just a few encodings (UTF-8 and Latin-1 works for me)

...stealing design work from Python again.  Of course Mozilla can't use this (already got our own, for one thing; security concerns, for another).  It's for embeddings like Treehydra.

I can comfortably say I'll never get around to this, but the bug should exist.  If anyone wants to put in some coding effort, I can quickly supply a design.

Comment 1

[Brendan Eich [:brendan]]

I will dance a jig when this bug gets fixed. mrbkap may know of volunteer candidates.

/be

Comment 2

[Wesley W. Garland]

The ServerJS pseudo-wg is seeking comments for a cross-engine file system API.

Anybody interested in resolving this bug is encouraged to contribute. If they can come up with a good implementation, there is a high probability it will be implemented for SpiderMonkey and usable for resolving this bug.

http://groups.google.ca/group/serverjs/browse_thread/thread/acf6a054ab5e0b6/69be0f7ec38452af?hl=en&ie=UTF-8&oe=utf-8&q=file+api+brouhaha&pli=1

Comment 3

[Blake Kaplan (:mrbkap) (inactive)]

(In reply to comment #0)
> - add a new ByteArray constructor for mutable byte-oriented binary buffers

This would be really useful in a few other places too (e.g. canvas's GetImageData)... Can we also specify a "no getters" policy (or, at the very least, that getters inherited from ByteArray.prototype are not called) for these?

Comment 4

[Raghav Sagar]

if this bug is still open, i am interested in working on it.

Comment 5

[Nochum Sossonko [:Natch]]

jsfile was removed in bug 545275. It was not replaced with the mechanisms mentioned in this bug, it was just ripped out (see http://groups.google.com/group/mozilla.dev.tech.js-engine/browse_thread/thread/e65bcf72800dcc16).

That said, I don't know if there is a desire to pursue this or not, I'll leave that to one of the js people. Thanks for volunteering!

Comment 6

[Saurabh Anand [:sawrubh]]

@jorendorff and others, Can you guys clarify if this bug is still going to be pursued (See comment 5). I would like to help fix this, once the decision to pursue it has been confirmed.

Thanks.

Comment 7

[David Teller [:Yoric] - still alive but not very active]

By now, we have OS.File in Firefox. Perhaps the simplest path is to make use of it?

Comment 8

[Jason Orendorff [:jorendorff]]

I don't think we really need this anymore.

Comment 9

[David Teller [:Yoric] - still alive but not very active]

(In reply to Jason Orendorff [:jorendorff] from comment #8)
> I don't think we really need this anymore.

Just to be sure I understand: because we do not need file I/O in SpiderMonkey or because there is a solution?