rack-devel archive mirror (unofficial) https://groups.google.com/group/rack-devel
 help / color / mirror / code / Atom feed
From: Eric Wong <normalperson@yhbt.net>
To: rack-devel@googlegroups.com
Subject: Re: Not cleaning up tempfiles for multipart?
Date: Thu, 18 Mar 2010 02:54:09 -0700	[thread overview]
Message-ID: <20100318095409.GB15049@dcvr.yhbt.net> (raw)
In-Reply-To: <f04d2211003170841o5f97169cy165e0fedb3ebe552@mail.gmail.com>

Charles Oliver Nutter <headius@headius.com> wrote:
> On Sun, Mar 7, 2010 at 7:12 PM, Eric Wong <normalperson@yhbt.net> wrote:
> > I guess this deals with wonky multipart uploads that browsers still
> > generate these days[1].  Ugh, yeah, it's nasty...
> > The comment below in _call is very important.
> >
> >
> > 1.  Ensure all tempfiles created by Rack go into an array in env,
> >    probably env["rack.tempfiles"]:
> >
> >    tempfile = Tempfile.new("foo")
> >    (env["rack.tempfiles"] ||= []) << tempfile
> 
> I'm with you so far...
> 
> > 2.  Have a middleware wrap everything, including the response body:
> 
> I'm a bit new to the middleware thing. Does this mean everyone would
> have to configure middleware on their setups to get this
> tempfile-closing behavior? And your comment below...does that mean
> there are situations where this wouldn't work?
> 
> Closing and removing tempfiles when they're no longer needed should be
> the default behavior, not something you have to configure. It's a bug
> to not close and remove them (or at least, a bug to keep creating new
> ones and expecting GC to clean everything up eventually).

I would expect major frameworks to stick this into the default
middleware stack as a convenience, but some users want a more bare-bones
Rack stack can still opt out.  I see it as needless overhead as the vast
majority of HTTP requests do not create tempfiles.

I don't make decisions for Rack, though.

<snip>

> >      # the Rack server should call this (when we're the body)
> >      def close
> >        tempfiles = env["rack.tempfiles"]
> >        if tempfiles
> >          tempfiles.each { |tmp| tmp.close! rescue nil }
> >        end
> >      end
> 
> By "the Rack server should call this" do you mean that if the server
> doesn't call this, tempfiles Rack creates will not be cleaned up until
> GC runs?

Yes, a Rack-compliant server should call body.close if it responds to
body.close at the end of the response cycle for every individual
request.

I put this in the wrapped body because body.each {} could be relying on
the tempfiles to generate the response.  body.close is the absolute last
action for any Rack application.

> Shouldn't Rack itself be guaranteeing it doesn't leave garbage files around?
> 
> >      # wrap the normal application call, saving env
> >      def _call(env)
> >        self.env = env
> >
> >        # XXX VERY IMPORTANT:
> >        # you need to ensure env stays the same throughout the request,
> >        # some middlewares overwrite/replace it instead of merge!-ing into it
> >        status, headers, body = app.call(env)
> >        self.body = body
> >        [ status, headers, self ]
> >      end
> >    end
> 
> Same question as above...can badly-written middleware now cause
> tempfiles to linger?

Yes, it might be safer to do this before app.call above:

   env["rack.tempfiles"] ||= []

That way if env gets replaced down the stack, e.g. via:

   app.call(env.merge("foo.hello" => "world"))
   # env.merge! would be correct above

any use of env["rack.tempfiles"] will still point to the same array.
Well, almost...

Then again some bad middleware could do:
   env["rack.tempfiles"] += [ tmp_a, tmp_b ]

Instead of what they _should_ do:
   env["rack.tempfiles"].concat([ tmp_a, tmp_b ])

So yes, discourage future middleware authors from replacing
the rack.tempfiles array.

-- 
Eric Wong

  reply	other threads:[~2010-03-18  9:54 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-05 14:40 Charles Oliver Nutter
2010-03-05 14:47 ` Charles Oliver Nutter
2010-03-05 14:48   ` Charles Oliver Nutter
2010-03-06 10:20   ` Hongli Lai
2010-03-07 14:25     ` Charles Oliver Nutter
2010-03-06  7:55 ` Eric Wong
2010-03-06 10:25   ` Hongli Lai
2010-03-07 14:34     ` Charles Oliver Nutter
2010-03-08  0:22       ` Eric Wong
2010-03-08  1:12         ` Eric Wong
2010-03-17 15:41           ` Charles Oliver Nutter
2010-03-18  9:54             ` Eric Wong [this message]
2014-01-06 22:45               ` Wojtek Kruszewski
2014-02-11 21:05                 ` Eric Wong
2014-03-27 21:40                   ` Lenny Marks
2010-03-07 23:53     ` Eric Wong
2010-03-08 11:26       ` Hongli Lai
2010-03-08 11:30         ` Hongli Lai
2010-03-08 14:33           ` Randy Fischer
2010-03-08 14:43           ` Charles Oliver Nutter
2010-03-08 14:49             ` James Tucker
2010-03-17  2:37           ` Eric Wong
2010-03-08 13:22         ` Charles Oliver Nutter
2010-03-08 14:42         ` James Tucker
2010-03-08 17:24           ` Hongli Lai
2010-03-09  7:43             ` Charles Oliver Nutter
2010-03-08 10:05     ` James Tucker
2010-03-07 14:27   ` Charles Oliver Nutter
2010-03-08  0:18     ` Eric Wong
2010-03-08 10:07       ` James Tucker

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-list from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://groups.google.com/group/rack-devel

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100318095409.GB15049@dcvr.yhbt.net \
    --to=rack-devel@googlegroups.com \
    --subject='Re: Not cleaning up tempfiles for multipart?' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

rack-devel archive mirror (unofficial) https://groups.google.com/group/rack-devel

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://public-inbox.org/rack-devel

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V1 rack-devel rack-devel/ https://public-inbox.org/rack-devel \
		rack-devel@googlegroups.com
	public-inbox-index rack-devel

Example config snippet for mirrors.
Newsgroups are available over NNTP:
	nntp://news.public-inbox.org/inbox.comp.lang.ruby.rack.devel
	nntp://7fh6tueqddpjyxjmgtdiueylzoqt6pt7hec3pukyptlmohoowvhde4yd.onion/inbox.comp.lang.ruby.rack.devel
	nntp://news.gmane.io/gmane.comp.lang.ruby.rack.devel
 note: .onion URLs require Tor: https://www.torproject.org/

code repositories for project(s) associated with this inbox:

	https://80x24.org/mirrors/rack.git

AGPL code for this site: git clone https://public-inbox.org/public-inbox.git