Attached is a modified version of that class that lets you define any
path you want for the servlet and lets you specify which queue to use
like this: Deferred.defer(task, queueName);
(I needed this for my own purposes)
Do with it as you wish :-)
The only other major change I would make is to
I store a record that has a natural Long primary key (a facebook user
id). I currently create a Key with that id and insert this record.
I just noticed this comment on the Entity(Key) constructor: Creating
an entity for the purpose of insertion (as opposed to update) with a
key that has its id
I, for one, am sick and tired of Sun's domineering, suffocating
stranglehold on what is and isn't Java. GAE is a breath of fresh air.
JavaME and JavaEE also impose a variety of limitations on Java.
What's the difference between those and GAE? The difference is that
Sun got a committee of big
There doesn't appear to be any API for detaching objects using the JPA
interface. If you need to serialize entity objects (say, to put them
in memcache), you're stuck with JDO (or one of the alternative
interfaces like Siena, SimpleDS, or the low-level API).
(if there is some undocumented way of
Why isn't there a system property that can be used to detect if a
program is running on appengine?
This is really a glaring oversight. I would expect something like
System.getProperty(google.appengine) to be development or
production (or of course null).
Before someone says check the servlet
On Thu, Dec 3, 2009 at 10:46 AM, Jeffrey Goetsch jeffg@gmail.com wrote:
Could you create a Servlet that is loaded on launch, and set this kind of
property? I think this is a workaround until they can do it officially.
I can't - remember this is in a library, so I would have to force all
Ok, glad to know it's not just me. Unfortunately I'm getting large
numbers of these exceptions, pretty much bringing my app to its knees.
The related code and data haven't changed:
Caused by: com.google.appengine.api.datastore.DatastoreTimeoutException: Unknown
at
H, some issues surrounding this:
* As you've probably figured out, the datastore natively only
supports two types of id, Long and String
* Only Long will autogenerate
* There is no guarantee that autogenerated values will be monotonic
or even valued in the low-order 4 bytes.
So, if you are
Hello?
I've now bugged this:
http://code.google.com/p/googleappengine/issues/detail?id=2519
Jeff
On Mon, Dec 14, 2009 at 4:24 PM, Jeff Schnitzer j...@infohazard.org wrote:
Can anyone comment on this in an official capacity?
I can write some code to test actual behavior, but I'd really rather
and not 500 characters?
Thanks,
Jeff
On Wed, Dec 16, 2009 at 1:34 PM, Ikai L (Google) ika...@google.com wrote:
Jeff,
The limit should be 500 bytes.
On Mon, Dec 14, 2009 at 4:24 PM, Jeff Schnitzer j...@infohazard.org wrote:
Can anyone comment on this in an official capacity?
I can write some
It's not really intuitive behavior. It's basically summed up in this bug:
http://code.google.com/p/googleappengine/issues/detail?id=2530
Jeff
On Sat, Jan 2, 2010 at 6:20 PM, seleronm seler...@gmail.com wrote:
Hi,
Please refer to the following links
It looks like this issue has been forgotten about:
http://groups.google.com/group/google-appengine/browse_thread/thread/ba2f36a160049e4d/c52daefd59156399
After some experimentation, I've concluded that it's impossible to
send HTML email with inline images using GAE. This is a really,
*really*
You have a mistaken understanding of what a Key is. Both values
displayed on the Datastore Viewer
(aghtb2JjYS1zdHIOCxIHQ2hlY2tpbhjiQww and YourEntityKind: id=12345)
are merely alternative visual representations. The actual value
stored in BigTable is a binary representation of this data. It may
better, what the mechanics are here...
Viktor
On Jan 9, 12:11 pm, Jeff Schnitzer j...@infohazard.org wrote:
You have a mistaken understanding of what a Key is. Both values
displayed on the Datastore Viewer
(aghtb2JjYS1zdHIOCxIHQ2hlY2tpbhjiQww and YourEntityKind: id=12345)
are merely
I've been thinking about this issue a little. It's not quite as
straightforward as just keeping an instance warm. Even if you have an
app that gets multiple hits per second, there will still be cold
starts:
* When a new instance comes online to serve more demand.
* When you redeploy a version
This is a tragedy of the commons. It may in fact work for you right
now, and will continue to work... until it doesn't. It will stop
working when enough people figure it out and millions of zombie
applications push the working applications out of memory. Or Google
pulls the plug on this
http://code.google.com/p/objectify-appengine/
I probably should have called this project Goldilocks, because it's
a little bit how I feel. Despite being a longtime Hibernate user
(since the 1.0 days), the JDO/JPA abstraction just doesn't make me
happy on appengine - it's too big, too
On Thu, Jan 14, 2010 at 12:59 PM, Stephan Hartmann hartm...@metamesh.de wrote:
Jeff,
in one point i disagree.
In a high available einvironment you would have a cluster of load balanced
application servers and you would deploy new versions of your app in turn,
one at a time.
So if one
Should my application be cycling even if it receives steady traffic?
I average about 1 hit per second, yet my app seems to bootstrap
several times every hour at an erratic rate.
Appid is mobca-st.
Look in the logs for ConfigurationBootstrap at level Info. This is
the Resteasy bootstrap.
The
JBoss Resteasy works fine without any special customization for GAE.
I use it extensively.
http://www.jboss.org/resteasy
Jeff
On Sat, Jan 16, 2010 at 1:09 PM, ChrisDane gregersen@gmail.com wrote:
Hi there,
Has anyone seen any examples on using only JAX-RS on GAE.
Implementing the
I mentioned Resteasy in your other thread, but you might also want to
consider Hessian. I recently patched the Caucho impl and now both
client server work on GAE (we use Hessian for server-server RPC,
and JAX-RS to the phones). If you're creating internal protocols,
Hessian is way easier than
anything if I figured out to implement my own
javax.ws.rs.Application?
Thanks again
Regards
ChrisDane
On Jan 16, 10:23 pm, Jeff Schnitzer j...@infohazard.org wrote:
JBoss Resteasy works fine without any special customization for GAE.
I use it extensively.
http://www.jboss.org
Or you can just get rid of that trainwreck called Maven.
Jeff
On Mon, Jan 18, 2010 at 12:10 PM, Philippe Marschall
philippe.marsch...@gmail.com wrote:
On Jan 16, 10:23 pm, Jeff Schnitzer j...@infohazard.org wrote:
JBoss Resteasy works fine without any special customization for GAE.
I use
If you call PreparedQuery.asIterator(), you can actually iterate
yourself as far as you want (within the deadline, of course). In
practice I've found you can rarely iterate through more than a couple
thousand entities.
The offset value is limited to 1,000. Basically the datastore is
iterating
Entities persisted with Objectify can be serialized in their entirety
(including keys) through GWT-RPC, so I'd be surprised if you had any
trouble using them with Restlet. I have a number of times rendered
entity objects directly to JSON with JAX-RS, although there's no
reason you couldn't do
Sorry if this is a silly question but...
How do you make an asynchronous URLFetch call in Java? Is this a
question of making calls to fetch() without calling getContent() on
the response?
Jeff
I see no limit of simultaneous [urlfetch] API calls documented on
the quotas page (
On Thu, Jan 21, 2010 at 7:48 AM, Duong BaTien duong.bat...@gmail.com wrote:
While exploring list-property and merge-join from this talk
http://code.google.com/events/io/2009/sessions/BuildingScalableComplexApps.html
i concur with your value proposition.
Neat, I had missed that talk. Good
SimpleDS' api is closer than the others to what I want, but here's
what I don't like:
* Not enough use of generics. Key, Query, and PreparedQuery should
all be generified classes.
* Key should not be used as the id for a class. The class itself
identifies the Kind, so having a Key id carries
On Tue, Jan 26, 2010 at 5:50 AM, a.maza andr.m...@gmail.com wrote:
sorry for this maybe thumb question, but I just want to make sure:
both frameworks are making the needs for open session in view (as
almost required when using jdo/jpa in web applications) obsolete,
or?
Neither Twig nor
On Mon, Jan 25, 2010 at 11:52 PM, John Patterson jdpatter...@gmail.com wrote:
This is why you configure what type of relationship is used using: @Embed,
@Entity(PARENT), @Entity(CHILD) or @Entity(INDEPENDENT)
So you have the flexibility to choose configuration _without_ rewriting your
code.
, especially in the social graph set intersections
and union.
Duong BaTien
DBGROUPS and BudhNet
On Tue, 2010-01-26 at 14:19 -0800, Jeff Schnitzer wrote:
On Mon, Jan 25, 2010 at 11:52 PM, John Patterson jdpatter...@gmail.com
wrote:
This is why you configure what type of relationship
On Wed, Jan 27, 2010 at 2:44 AM, datanucleus andy_jeffer...@yahoo.com wrote:
Exposing Key to a user is not (in my opinion) a requirement for
*providing* a JDO impl on GAE/J. Other datastores have their own
internal id and we don't expose theirs (e.g db4o).
...
And, of course, all of that
On Wed, Jan 27, 2010 at 3:28 PM, datanucleus andy_jeffer...@yahoo.com wrote:
I only represent JDO, as a spec, not how GAE/J have implemented it.
The docs for that are provided in the JDO spec and on the DataNucleus
website. I fail to see anything controversial in how they are
represented in
The easiest way to publish a web service (interpreted as an
http-based rpc protocol) is with Hessian. The 4.0.3 version of
Caucho's hessian jar should work fine as both client and server on
Appengine.
Other solutions (JAX-RS, SOAP) are available but they are all a lot
more complicated. With
and this will
be null
+ if (getClassLoader() == null)
+ return;
+
try {
Enumeration iter;
On Fri, Jan 29, 2010 at 12:26 PM, Jeff Schnitzer j...@infohazard.org wrote:
The fix went into version 4.0.3, which is actually available for
download but the link is missing from
FWIW, it's actually quite easy to check out Resin trunk (which holds
the hessian source code), make the patch, build with ant, and get a
hessian.jar.
Hessian works quite beautifully on appengine. I wish I had an
asynchronous ObjectiveC hessian client so I could ditch my tediously
duplicated
On Thu, Feb 4, 2010 at 11:52 AM, Keith Platfoot kplatf...@google.com wrote:
GWT/App Engine projects will no longer require our SDK library on the
classpath. This means Maven users will be able to pull in JAR files from
their M2 repository as they're accustomed to and the plugin won't mind a
If you add Objectify
(http://code.google.com/p/objectify-appengine/wiki/ObjectifyWithGWT)
to your project, you will be able to use the datastore basic types
(Email, Link, GeoPt, etc) in GWT without any extra effort.
This works even if you don't use the Objectify persistence system.
Jeff
On Mon,
Today we released version 2.0 of Objectify-Appengine. Objectify is a
mid-level persistence API for the Appengine datastore - much simpler
than JDO, much more sophisticated than the Low-Level API, much easier
to use than either!
http://code.google.com/p/objectify-appengine/
If you are unfamiliar
A bit of a blast from the past on this thread, but...
I've always had a trickle of serialization problems in production
using the Deferred servlet. Finally I had one that was consistent and
repeatable. I forced it to base64 encode always (not just dev mode)
and now it works consistently.
FYI.
apps servers.
On Feb 18, 6:25 pm, Ivan Pardo trux...@gmail.com wrote:
At least now I can be certain that it's not a problem with my code. I
can only hope that Google can fix this problem ASAP, as my application
is absolutely useless without a working geocoder.
On Feb 18, 5:44 pm, Jeff
There is this:
http://appscale.cs.ucsb.edu/
http://code.google.com/p/appscale/
I have no idea how mature or stable it is.
Jeff
On Thu, Feb 18, 2010 at 2:58 PM, Greg Marine gregmar...@iccnet.org wrote:
Hello All!
I am evaluating GAE/Java for some of my projects and came across a
scenario I
updates.
On Fri, Feb 19, 2010 at 10:51 AM, Jeff Schnitzer j...@infohazard.org
wrote:
It helps to have some links... please star these issues!
http://code.google.com/p/googleappengine/issues/detail?id=2806
http://code.google.com/p/gmaps-api-issues/issues/detail?id=2186
This problem
I hate to be the dork that argues with the referees, but I'm not fond
of this advice...
* long-id keys can be user-defined just as easily as string-name
keys, so you can enforce uniqueness constraints either way if you have
some sort of natural key.
* Readable urls are a high level construct,
On Wed, Jan 27, 2010 at 10:22 AM, datanucleus andy_jeffer...@yahoo.com wrote:
A List follows the contract of java.util.List, hence obviously
ordering should be preserved. Whether it is JDO or JPA, the whole
point of transparent persistence - people shouldn't need to know
anything about the
On Mon, Feb 22, 2010 at 1:19 PM, Max Ross (Google)
maxr+appeng...@google.com wrote:
user-defined long-id keys are not quite as easily used. You either need to
commit to not letting the datastore generate ids for that kind or you need
to reserve a batch of ids using the
and/or siggestion.
Duong BaTien
DBGROUPS and BudhNet
On Mon, 2010-02-22 at 13:52 -0800, Jeff Schnitzer wrote:
On Mon, Feb 22, 2010 at 1:19 PM, Max Ross (Google)
maxr+appeng...@google.com wrote:
user-defined long-id keys are not quite as easily used. You either need to
commit to not letting
On Mon, Feb 22, 2010 at 2:29 PM, Steve Pritchard steve...@gmail.com wrote:
This reply is really to the Google folks.
Guess:
64Bit Servers running some sort of VMWare.
32bit Linux used as the host operating system.
(4) My guesses could be way off and this whole strategy out to lunch.
I'm
string pk is a necessary part of the
solution.
Regards,
Max
On Mon, Feb 22, 2010 at 1:52 PM, Jeff Schnitzer j...@infohazard.org wrote:
On Mon, Feb 22, 2010 at 1:19 PM, Max Ross (Google)
maxr+appeng...@google.com wrote:
user-defined long-id keys are not quite as easily used. You either
... it just silently overwrites the existing one. In
either case, storing the email as a key or as a field, you would need to do
a query (or get) in a transaction to be sure you were not going to overwrite
an existing record with the same email.
On 23 Feb 2010, at 13:37, Jeff Schnitzer wrote
I think this is actually an interesting question, and brings up a
discussion worth having:
Is datastore performance reasonable?
I don't want to make this a discussion of reliability, which is a
separate issue. It just seems to me that the datastore is actually
kinda pokey, taking seconds to
making sense.
On Wed, Feb 24, 2010 at 5:52 PM, Jeff Schnitzer j...@infohazard.org wrote:
On Wed, Feb 24, 2010 at 1:06 PM, Ikai L (Google) ika...@google.com
wrote:
My point wasn't necessarily that it wasn't possible. makePersistentAll
does
use a batch write, and there are definitely sites
Did anyone ever resolve this in a consistent way?
I recently upgraded from OSX 10.5 to 10.6, and now my Eclipse (Cocoa
64-bit, as I was using before) hangs on most startups. I see the
Updating MyProje... - 1.3.1 in the bottom right corner and the
entire window is locked up. Only way out is to
that the update of war/WEB-INF/lib
occurs on classpath change instead of classpath init. That should alleviate
this sporadic issue. The fix will be available in a forthcoming version of
GPE.
On Mon, Mar 1, 2010 at 2:34 AM, Jeff Schnitzer j...@infohazard.org wrote:
Did anyone ever resolve
The short answer is no, this is not possible. It's not an appengine
issue, IP networks just don't work that way.
Jeff
On Wed, Mar 3, 2010 at 8:47 AM, mscwd01 mscw...@gmail.com wrote:
Hey,
Is there a method in which I can receive a request and then forward
the same request to an external URL
On Thu, Mar 4, 2010 at 1:54 PM, Robert Lancer robert.lan...@gmail.com wrote:
You should join me in commenting on every pro google article on
techcrunch (they're all pro google), im basically accusing them of
being google suck ups for totally missing the app engine outage. this
is prob the
On Thu, Mar 4, 2010 at 4:34 PM, Locke locke2...@gmail.com wrote:
Really? Are you absolutely certain that it is technically impossible
to spin up an app *before* handing it over to users? Are you confident
that the laws of physics prohibit spinning up new instance of apps in
the background
On Thu, Mar 4, 2010 at 6:31 PM, Robert Lancer robert.lan...@gmail.com wrote:
Jeff please watch this http://www.youtube.com/watch?v=aXJklICrFJI
That reinforces my opinion that the GAE team should spend less time
focusing on cold starts and worry more about datastore latency, which
affects *every*
On Fri, Mar 5, 2010 at 12:34 AM, Nacho Coloma icol...@gmail.com wrote:
Having an id field instead of a key makes queries easier. With
Objectify, you can create a query without needing to know that the id
is part of a key field:
query.filter(id , 5000).sort(-id)
This would only work with
If you use the Google data classes (Key, GeoPt, Email, Link, User,
etc) in your entities you will have trouble serializing them through
GWT-RPC. Even though you're using JDO, you can still use Objectify's
GWT integration to solve the problem:
On Sun, Mar 7, 2010 at 11:36 AM, tempy fay...@gmail.com wrote:
Interesting... now I need to go and make sure all my logic still works
if singletons stick around between requests. GAE is full of
surprises! Thanks for the tip =)
It may be true that GAE is full of surprises, but this shouldn't
On Sun, Mar 7, 2010 at 2:30 PM, tempy fay...@gmail.com wrote:
Well I didn't think that GAE would start up a new JVM per request, but
I assumed the GC would clean up any data created between requests, as
there seems to be nothing left holding references to said data. One
of these days I will
On Tue, Mar 9, 2010 at 7:07 PM, Yasuo Higa higaya...@gmail.com wrote:
Slim3 locks an entity group when getting, putting and deleting entities
so that nobody updates the entity group. So it never happens
that the first two entity group transactions commit and the third fails.
What do you do
Create a UserSkill entity with a parent key of User. Do a keysOnly
query for UserSkill objects that match your criteria, then get the
parent keys out of the UserSkill key, then do a batch get on the
parent keys. The Objectify code:
class UserSkill {
@Id Long id;
@Parent KeyUser user;
This is a good segue into something I've been contemplating: Does
anyone actually use the Relation Index Entity pattern? How does it
perform?
I don't mean how do queries perform - obviously those would be fast.
But how expensive in both latency and price is it to write an entity
with
On Tue, Mar 9, 2010 at 8:47 PM, John Patterson jdpatter...@gmail.com wrote:
For the sake of comparison, see below for the Twig equivalent
class UserSkill {
�...@parent User user;
String skill
int ability;
}
Note that no key is required and user is referenced directly
On Wed, Mar 10, 2010 at 6:14 AM, John Patterson jdpatter...@gmail.com wrote:
No that is incorrect. You actually have very fine control over what is
loaded and when using activation [1]
Ah, this Activation annotation does help out. But it still leaves
edge cases (read on).
I also don't think
On Wed, Mar 10, 2010 at 7:39 PM, John Patterson jdpatter...@gmail.com wrote:
On 11 Mar 2010, at 03:40, Jeff Schnitzer wrote:
That is an empty claim with no example or evidence. Every comparison we
have see so far is cleaner and more readable in Twig.
Nonsense. The only example
I'll try to ignore the blatent trolls. But you want things easier to
do in Objectify than Twig, so here's a few:
-
How, in Twig, do you rename a field?
Objectify lets you import data from a variety of historical formats;
as you load and save data it will naturally be transformed in format:
On Thu, Mar 11, 2010 at 8:56 PM, John Patterson jdpatter...@gmail.com wrote:
But for typesafe changes large or small Twig supports data migration in a
much safer, more flexible way than Objectify. Read on for details.
You are increasing my suspicion that you've never actually performed
schema
Scott: Nacho is the author of SimpleDS.
Schema migration is something that Hibernate and RDBMSes actually do
rather poorly. The typical process is to prepare a series of scripts
(ALTER TABLE and then any relevant data transmogrification), shut down
the application, run the scripts, then bring
On Fri, Mar 12, 2010 at 8:35 PM, John Patterson jdpatter...@gmail.com wrote:
On 12 Mar 2010, at 16:28, Jeff Schnitzer wrote:
Look at these graphs:
http://code.google.com/status/appengine/detail/datastore/2010/03/12#ae-trust-detail-datastore-get-latency
http://code.google.com/status
jdpatter...@gmail.com wrote:
On 13 Mar 2010, at 11:00, Jeff Schnitzer wrote:
since you can (and IMNSHO probably should) always disable automatic
activation and refresh the graph manually.
Although I dislike premature optimisations such as this note that you can
configure Activation
One thing you get used to on appengine is that any bulk data work
requires the task queue. You can use a little bit of framework and
make all of these transforms (including deleting data) a question of
just writing a simple task class and firing it off. You'll want a
copy of the Deferred
On Sun, Mar 14, 2010 at 9:16 PM, Max thebb...@gmail.com wrote:
If I would like to find all users that have java level 5 and c++
level 2, then how to write a query in your suggested data model?
Every time this kind of question comes up I usually think to myself
how would an RDBMS do this? and
I'm puzzled by the behavior of custom indexes. I have a simple test
case below, a simple equality filter on one property combined with a
descending sort on another property. If I set the properties with
setUnindexedProperty(), the query fails to find the result. If I set
the properties with
include the property in any custom
indexes interpretation of setUnindexedProperty. Can you post the link
to the continuation post? I'm curious what Googlers have to say about
it.
Tristan
On Mar 18, 11:13 am, Jeff Schnitzer j...@infohazard.org wrote:
This doesn't make sense to me.
Every scrap
On Mon, Mar 22, 2010 at 7:21 AM, Guillermo Schwarz
guillermo.schw...@gmail.com wrote:
In order to load the app, a kind of core file must be read from disk
in order to represent the app state in RAM. That process takes too
long (used to be up to 3 secs, now it is up to 12 seconds AFAIK)
Today we released Objectify v2.1, the latest version of our opensource
replacement for JDO/JPA on the Google App Engine datastore.
This version includes a major new feature, Partial Indexes. If you
aren't sure what partial indexes are, the Wikipedia page
It seems quite unlikely that GAE appservers are going to have public
IP addresses. Even if you can get the IP address of the host (and I
suspect this is impossible), it's going to be a private address.
The best you will probably be able to do is get the NATed apparent
address of the URLFetch
On Tue, Mar 30, 2010 at 8:03 AM, Guillermo Schwarz
guillermo.schw...@gmail.com wrote:
SQL can be run on top of a file system (fseek, read, write) or on top of a
persistent hashmap (datastore).
If you create a SQL interface on top of any of those, then it is a
relational database, not a fake
to access the data. This tool works very well.
Thanks,
Sandeep.
On Tue, Mar 30, 2010 at 4:51 AM, Jeff Schnitzer j...@infohazard.org wrote:
Hey, so one of the things I complained about in Andreas' great article
is the difficulty of doing analytics on appengine. GAE is great for
operational systems
On Wed, Mar 31, 2010 at 6:36 AM, John Patterson jdpatter...@gmail.com wrote:
Unlike some of the other frameworks there is no registration process for
your data model classes - each class is analyzed the first time it is used
per server instance and the meta-data is cached for the duration of
On Wed, Mar 31, 2010 at 6:41 PM, jd jdpatter...@gmail.com wrote:
On Apr 1, 3:14 am, Jeff Schnitzer j...@infohazard.org wrote:
What does Twig do when someone issues a type-less query?
datastore.find().addFilter(color, EQUAL, green). returnResultsNow()
The expression above is actually
On Thu, Apr 1, 2010 at 12:45 AM, John Patterson jdpatter...@gmail.com wrote:
All queries that are possible with the low-level datastore are possible with
Twig because there are low-level to type-safe bridge methods. So you can
simply do a typeless low-level ancestor query and then get Twig to
On Thu, Apr 1, 2010 at 1:57 AM, John Patterson jdpatter...@gmail.com wrote:
You are making a classic premature optimization mistake.
[...]
The only String value that can uniquely identify a class with no
registration is its fully qualified type name. I see no problem using
that kind name as
On Thu, Apr 1, 2010 at 2:19 AM, John Patterson jdpatter...@gmail.com wrote:
From a quick browse of the Objectify source code it looks to me like the
default is to use Class.getSimpleName() for the kind name which will also
break on renaming and doesn't even have the advantage of working
This is a genuine, heartfelt plea: The Builder pattern
(DatastoreServiceConfig, FetchOptions) makes code extra annoying when
layering an API on top of the low-level API.
Let's say you are writing some code by hand that creates a
FetchOptions with a limit and an offset:
FetchOptions opts =
:07 AM, Jeff Schnitzer j...@infohazard.org wrote:
This is a genuine, heartfelt plea: The Builder pattern
(DatastoreServiceConfig, FetchOptions) makes code extra annoying when
layering an API on top of the low-level API.
Let's say you are writing some code by hand that creates a
FetchOptions
, Jeff Schnitzer j...@infohazard.org wrote:
Yes, that would solve the problem. Although I have to wonder, how is
this any better than having a public constructor?
Jeff
On Thu, Apr 1, 2010 at 10:08 AM, Max Ross (Google)
maxr+appeng...@google.com wrote:
Hi Jeff,
Note
Appengine devs have said on a number of occasions that we should not
assume thread-safety of any class not documented as being thread-safe.
I don't see anything in the javadocs indicating that DatastoreService
is thread-safe, therefore keeping an appwide instance of DS is
probably risky - even if
The most important number in my mind is the number of line items in an
order. If you're seeing 20s+ queries, it must be a pretty large
number.
Objectify (and Twig) support queryable collections of embedded
objects, so you can put all the line items in a single order object.
There are some
You can serialize object graphs, but keep in mind that you won't be
able to index/query this data and it will be opaque to the datastore
viewer. It will also be opaque to GAE/Python tools.
As John mentioned, you can get embedded object collections without
Java serialization if you use Objectify
You should never put something in the memcache that you don't mind
losing when the memcache service flushes your data. Players are
probably going to be very unhappy if their scores and current
positions suddenly disappear mid-game.
GAE is just not an appropriate platform for a realtime game.
, Jeff Schnitzer j...@infohazard.org wrote:
You should never put something in the memcache that you don't mind
losing when the memcache service flushes your data. Players are
probably going to be very unhappy if their scores and current
positions suddenly disappear mid-game.
GAE is just
The Caucho folks have been slow to update the hessian download page,
but the actual binaries are still being generated and put up for
download. You can craft the download link by hand.
This version works with GAE:
http://caucho.com/download/hessian-4.0.6.jar
Jeff
On Sun, Apr 18, 2010 at 1:36
3-4s is pretty close to as good as it gets to start a JVM, load your
app, and begin serving pages. I've seen 2s load times but 3-4s is
pretty typical.
Consider yourself lucky! The poor souls running Spring apps usually
wait 15s+ for cold starts...
Jeff
On Mon, Apr 19, 2010 at 4:58 PM, Blake
Looks like you found a bug. I should be able to have a unit test and
a fix checked in tonight. Sorry about that!
Jeff
On Wed, Apr 21, 2010 at 1:43 PM, Larry White ljw1...@gmail.com wrote:
Hi,
I'm trying to pass in a collection of Strings to be used in a filter using
Objectify (which I
Fixed. Although you might prefer to issue a batch get() operation
instead of the query. There might be a performance difference - I'm
not sure.
Jeff
On Wed, Apr 21, 2010 at 6:46 PM, Jeff Schnitzer j...@infohazard.org wrote:
Looks like you found a bug. I should be able to have a unit test
?) but don't know which one.
Can anyone tell me what jar
com/google/appengine/api/datastore/ReadPolicy$Consistency is in?
On Wed, Apr 21, 2010 at 11:10 PM, Jeff Schnitzer j...@infohazard.org
wrote:
Fixed. Although you might prefer to issue a batch get() operation
instead of the query. There might
1 - 100 of 265 matches
Mail list logo
