In wireshark you'll be looking to use a normal HTTP rule as the JSON
will just be passed as entity body in the request. Doing a filter for
HTTP port 80 should be sufficient.
Cheers
Tim
--~--~-~--~~~---~--~~
You received this message because you are subscribed
Okay... I've issolated the problem. Lots of GC items being sent. This is
causing a stack overflow. I've got a test for it and I'll fix the problem
today.
On Mon, Feb 16, 2009 at 1:33 AM, Tim Perrett he...@timperrett.com wrote:
In wireshark you'll be looking to use a normal HTTP rule as the
On Sun, Feb 15, 2009 at 2:42 PM, askiadevelo...@gmail.com
askiadevelo...@gmail.com wrote:
Dave,
I am having the same issue using eclipse. Can you please elaborate
your paragraph below.
What do you mean lift is built against Jetty 6?
Jetty 7 changed some APIs. Lift's RunWebApp is failing
What's the impact overall? Is this a result of upgrade to JQuery
1.3.1?
Thanks, Tim
On Feb 16, 4:51 pm, David Pollak feeder.of.the.be...@gmail.com
wrote:
Folks,
It turns out that there are issues with jQuery's methods that insert/replace
HTML and using XHTML on Firefox. Sigh.
We should be
On Mon, Feb 16, 2009 at 9:04 AM, Tim Perrett he...@timperrett.com wrote:
What's the impact overall?
The impact is: use innerHTML directly rather than going through jQuery's
HTML re-writing.
Is this a result of upgrade to JQuery
1.3.1?
No. This is a latent Firefox bug and has to do with
Marius,
Please feel encouraged to clean these up for 1.0.
Thanks,
David
On Sun, Feb 15, 2009 at 4:34 AM, Marius marius.dan...@gmail.com wrote:
Hi,
Building latest code I'm getting
[WARNING] /media/WORK/marius/repository/github/liftweb/lift/src/main/
Lifted,
Does anyone have experience with using the hibernate reverse-engineering
stuff under maven? Everything i've stumbled on is ant-driven.
Best wishes,
--greg
--
L.G. Meredith
Managing Partner
Biosimilarity LLC
806 55th St NE
Seattle, WA 98105
+1 206.650.3740
I'm looking for a reason to build a distributed computation and storage
system on top of Yahoo!'s ZooKeeper. Kind of like Google's big table, but
with the ability to perform optimistically locked transactions on the data.
On Mon, Feb 16, 2009 at 1:15 PM, Chris Richardson
Lee: thanks for the pointer to what looks like an excellent report.
Scala might be great for cloud computing and it would be interesting
to extend Lift in that direction.
There is a ton of closed and open source work going on now on scalable
storage and it would be interesting to see whether
How about adding a class-attribute called gc to everything that's
supposedly collectible?
Then doing a scrape using jQuery(.gc) would be just as feasible.
Would it solve the problem or can it be extended to solve it fully?
On Mon, Feb 16, 2009 at 10:37 PM, David Pollak
what would the raw XHTML look like?
Also, we can't rely on jQuery to scrape because we have to support other
libraries (e.g., YUI) without jQuery.
On Mon, Feb 16, 2009 at 1:45 PM, Viktor Klang viktor.kl...@gmail.comwrote:
How about adding a class-attribute called gc to everything that's
Alan,
We did the vast majority of the breaking API changes between 0.9 and 0.10.
0.11 is almost identical (except in some corner cases that you will not
likely be coding against) to 0.10
I strongly advise upgrading to 0.11 as it's also 1.0 RC, so we're testing
the heck out of it prior to the 1.0
On Mon, Feb 16, 2009 at 1:53 PM, Alan M alan.morten...@gmail.com wrote:
I'm still working away on my web service project (mentioned many
months ago when I first lobbied to get Scala used) and we rely heavily
on Put requests to create new resources. My first attempt involving
Put and I've
This might be a bit of a pain to deal with, but something like:
tag class=lift-gc lift-gc-[uuid] /
Of course, you'd want to append to the 'class' attribute, not replace it
entirely.
--j
On Mon, Feb 16, 2009 at 1:49 PM, David Pollak feeder.of.the.be...@gmail.com
wrote:
what would the raw
I wonder if 1,000 missing classes will cause the CSS engines to barf.
Anyone want to mock up a page or two and feed them into IE6, etc.?
On Mon, Feb 16, 2009 at 2:15 PM, Jorge Ortiz jorge.or...@gmail.com wrote:
This might be a bit of a pain to deal with, but something like:
tag
Would it not be because of:
ServletFileUpload.isMultipartContent(request) // Req.scala line 79
If you then reference the commons lib, the source of that method looks
like:
public static final boolean isMultipartContent(
HttpServletRequest request) {
if
What kind of deployment scripts did you have in mind?
Also, I run lift in EC2 with no problems and its actually really
really great :-)
Cheers, Tim
* Deployment: perhaps lift should release deployment scripts, or even
EC2 VM instances?
Food for thought.
Sorry Alan,
I assumed too much - my fault not yours.
Generally, handling web services is done via dispatching. for
instance:
// Boot.scala
LiftRules.dispatch.prepend {
case r @ Req(some :: path :: Nil, , PutRequest) = () =
myHandlerFunction(r)
}
def myHandlerFunction(req: Req):
I can understand why he wants to use PUT - its more resource
orientated.
I think we should stop talking about file uploads in the traditional
sense, as the current lift design is doing the right thing for
normal form based uploads.. However, lets talk about passing large
entity bodies with web
Yeah but when I do the PUT, the req.body is an Empty box (no byte
array). Also note, I'm using the statelessDispatch not dispatch.. but
other than that, what you wrote code-wise is almost exactly one of the
ways I tried. (the other way is to look at uploadedFiles instead of
body)
Alan
On Feb
Oh.. so I can stuff data in the body, even if the message is a PUT
message? Not sure how to do that with curl without sending a bogus
file along for the ride.. but if that's what you are saying then I
guess that'd be acceptable. With the exception of some GIF/PNG/JPEG
images now and again, the
Alan,
I'm modifying Lift. It turns out that it treated everything
like application/x-www-form-urlencoded if there was not another handler.
I'm changing the code so that if the submission is not marked with
Content-Type application/x-www-form-urlencoded, you will have access to the
post body.
On Mon, Feb 16, 2009 at 3:34 PM, Alan M alan.morten...@gmail.com wrote:
I'm using maven for builds.. on .11-SNAPSHOT.. will this go there or
will I have to switch over to building my own?
It'll get built on Hudson and if you do a mvn -U, you'll get it.
See
Cool. That will make WebDAV and CalDAV support easier :)
On 17/02/2009, at 10:38 AM, David Pollak wrote:
On Mon, Feb 16, 2009 at 3:34 PM, Alan M alan.morten...@gmail.com
wrote:
I'm using maven for builds.. on .11-SNAPSHOT.. will this go there or
will I have to switch over to building my
It appears that DPP is solving your issue now :-)
However, for the record, I do exactly what you describe in terms of
PUT / POST - this is a fairly normal ROA (but with objective-c
clients, not javascript). FYI... if you have xml messages, you can
access the xml automagically in your
On Mon, Feb 16, 2009 at 4:34 PM, Alan M alan.morten...@gmail.com wrote:
OK just to wrap up for me, because I didn't really follow all the
implications of the change being made.. If I send a Put to a lift web
service I should not be able to see the file content in the body of
the Request?
On 17/02/2009, at 3:18 PM, Josh Suereth wrote:
Marc, Also try running the scalac command directly. Here's what
Maven is trying to run on your machine:
The result is:
Mac:liftweb marc$ sh cmd
cmd: line 1: 10785 Segmentation fault /System/Library/Frameworks/
re: deployment scripts, cloudtools looks interesting. I think it
would be impressive if you could install lift, generate a hello world
app, and deploy it to a new amazon cluster in three steps. Perhaps
lift could bundle a production cloudtools configuration or two in the
mvn archetypes?
Lee
I agree -- a web framework that works well with scalable storage is
the most interesting challenge.
I haven't thought much about dynamic scaling, and it's more cutting
edge then most would need. That said, if there's a scaling tool that
lift could connect to, it would be neat if lift
29 matches
Mail list logo
