what's the difference between
warDir, webappDirectory, workDir,workingDirectory, codeServerWorkDir ,
devModeWorkDir, tempDir?
i mean it's hard to follow all those around.
On Mon, Oct 19, 2020 at 2:58 PM Elhanan Maayan wrote:
> hey thomas, the thing is, i wanna do it right without a
hey thomas, the thing is, i wanna do it right without any hacks, but if i
demine the pom as gwt-app, the output is just gwt compiled code with no
static html or css resources, if i use war in the pom type, the output is
just the static resources. i want the entire artifact to be considered as
both
look at that...
>
> Elhanan Maayan schrieb am Mo., 19. Okt. 2020,
> 10:00:
>
>> no i meant, that when i do regular compile i still see *.devmode.* files
>> on the target dir, and when i try to use just a normal server, it says it
>> can't see the codeserver
anspiles everything first
> before it starts with gwt:devmode? ... Actually it doesn't need to, I
> agree... maybe @tbroyer could explain?
>
> Elhanan Maayan schrieb am So., 18. Okt. 2020,
> 23:58:
>
>> thanks, btw i don't understand why the maven plugin insists on compil
thanks, btw i don't understand why the maven plugin insists on compiling it
for devmode
On Mon, Oct 19, 2020 at 12:44 AM lofid...@gmail.com
wrote:
> Yes, the main thing you have a fast turn around time for *compile - test
> cycle* for UI development. Because if you are working on UI you don‘t
>
essentially i could use all the real static files in the client side
On Mon, Oct 19, 2020 at 12:30 AM lofid...@gmail.com
wrote:
> Yes and you can control it using Maven assembly as I showed above
>
> https://github.com/gwtboot/domino-rest-enum-date/blob/master/domino-rest-enum-date-client/src/as
yes, but in addition to the html file your also have the css files as
well..
On Sun, Oct 18, 2020 at 10:54 PM lofid...@gmail.com
wrote:
> Yes correct, HTML file is your host file.
>
> I always put a simple HTML file to be able to launch / debug gwt webapp on
> the client... see:
> https://github
i missing something basic here with all the code server/dev mode/super dev
mode/launch dir/workdir/startup url/s
at the end of the day you have to have an html file to use, when you launch
for debug, but the gwt plugins are in the client module which doesn't have
the html file on one hand, but they
but the webui-server contains all the static resources like the html page
On Sun, Oct 18, 2020 at 8:24 PM lofid...@gmail.com
wrote:
> --webui
> webui-client ---> this is the GWT code
> webui-server ---> this is server-side code like Servlet, Spring Boot,
> JavaEE servers
>
> This is the
i did just that only it's more like
main_pom
--mod1
--mod2
--webui
webui-client
webui-server
--builder-mod that collects all jars from other modes into a directory for
final deploy and installation
what i wanna do is also be able to build only the webui project, and
possibility to use the
i'm assuming those 2 need to be in there own aggregator pom (i.e my main
parent pom will a web ui pom, which will have those submodules) , because i
want to be able to build just that webproject alone without the need to
build the whole project again (this is because that in order to develop for
co
yes, i understand that gwt-maven-plugin is "opinionated" (i.e forces me to
do it) and i'm in the process of doing so, the thing is there's no "Server
part" it's meant to operate as a page under cockpit ui
On Sun, Oct 18, 2020 at 6:00 PM lofid...@gmail.com
wrote:
> Hmm it is not good to have "ser
i don't know, right now i'm having trouble to even compile everything, i
mean, i can either compile all the static files (html WEB-INF) into a war,
OR i can compile the gwt sources, depending on what i place in pom.xml
(gwt-app or war) i was trying to have it all in one module
On Sun, Oct 18, 202
cockpit guys reached out, they told me i have inline css which
cockpit disables those, once i enabled CSP in cockpit it was ok. , they
also recommended i should try elemento because it's more like patternfly
On Tue, Jul 14, 2020 at 7:42 PM Rogelio Flores
wrote:
> Looks like you have css on your
i just want to use cockpit's authentication, (meaning when they log into
cockpit, they already have ot use the os's user/pass) so i don't have
create my own credentials system
my current main problem, is that debugging as i was told cockpit pages
aren't developed on windows.
in addition the same
Actually i don't want to integrate much woth cockpit just want to call out
some rest api to another server. Why would it be hard?
On Thu, Jul 9, 2020, 00:45 Rogelio Flores wrote:
> Of course GWT can be used, but IMO is probably not needed and would be
> easier not to use it, unless you want to p
*yea, i think there's a bug somewhere about it. *
On Wed, Jan 4, 2012 at 7:26 PM, Thomas Broyer wrote:
> I suspect it's *Eclipse* which messes things up actually.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To view this discu
You mean everything must be public?
On Jan 4, 2012 12:23 PM, "Thomas Broyer" wrote:
> The getStartDate() getter is not public in your class, but mapped on the
> proxy.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To view this d
What about iT ?
On Jan 4, 2012 12:19 PM, "Thomas Broyer" wrote:
> How about the getStartDate()?
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/google-web-
ryPKProxy directly?
>
> RF will ignore proxies it can't reach, (ie there's no method
> returning/using them anywhere) but that shouldn't be the case unless they
> are un-used. Can you post the place where the JobHistoryPKProxy is used ?
>
>
> On Tue, Jan 3, 2012 at 8:
the zero stuff is intended to be package level (hell i only placed because
rf),. but even with public ctor it doesn't work.
On Wed, Jan 4, 2012 at 12:42 AM, Thomas Broyer wrote:
> Your zero-args ctor and getter are not public, is this intended? RF only
> uses public members and actors (that last
yep i had that as well.. but still the same exception.
@ProxyFor(value = JobHistoryPK.class)
public interface JobHistoryPKProxy extends ValueProxy {
java.util.Date getStartDate();
}
i've been trying to dig around in the deobfuscator code and
the domainToClientType map doesn't even show the Valu
check that, actually i a slightly different error only difference is
instead of JobHistory, i get JobHistoryPk can't be sent to the client.
On Sun, Jan 1, 2012 at 7:56 AM, Elhanan Maayan wrote:
> ok, they really REALLY need to fix that thing, i changed the directory of
> .apt_g
ok, they really REALLY need to fix that thing, i changed the directory of
.apt_generated to .apt_gen and got an exception on the serve side (some
stupid bug in eclipse link which doesn't make it compatible with jpa), i
dunno if this was the original problem, but after fixing it still get the
origin
and btw here's what i thought of...
it's still in the rough, but basically it's a combination between
ValueListBox and SuggestionBox
(hence ValueSuggestionBox :) ), i've tried in my small sourceforge project
and seems to be working (at least with requestFactory EntityProxies), so
now i can use an e
yes, it's called "too-much-values-to-be-displayed-in-one-list"...
On Mon, Dec 26, 2011 at 8:36 AM, -sowdri- wrote:
> Is there anything that 's preventing you from creating
> a ValueListBox)??
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolkit
yes, and this is why i posted originally, i already seen these 2 and i was
wondering if i'm down between choosing something that is mature but 50%
slower then something of version 0.3 and in incubation mode..
On Sat, Dec 24, 2011 at 11:23 AM, Eric Charles
wrote:
> Hi,
> I never used apache beanva
i mean a SEPARATE validation framework, if i use eclipseLINK for example
On Thu, Dec 22, 2011 at 10:31 PM, Patrick Julien wrote:
> I use OpenJPA and I know it works, it's able to validate them.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Web Toolk
ok, when does one find such information, i mean, clearly these kind of
things, even SkipInterfaceValidation do not appear on the developer guide,
is there a secret cabal of gwt-isters, gathering somewhere in the ether?
On Tue, Dec 13, 2011 at 11:39 AM, Thomas Broyer wrote:
> A DeobfuscatorBuilde
do you have any examples on DeobfuscatorBuilder and ServiceLayerDecorator ?
On Tue, Dec 13, 2011 at 12:58 AM, Thomas Broyer wrote:
> If you're really sure about what you're doing, you can annotate the
> property with @SkipInterfaceValidation; and if you need to generate the
> DeobfuscatorBuilde
ok let's assume i'm using the server side validation ,is there any easy to
get the control which has the wrong value (i'm assuming the use of path is
needed) from the editorError?
On Wed, Nov 30, 2011 at 12:38 PM, Thomas Broyer wrote:
> RequestFactoryEditorDriver#flush simply copies the data fro
you mean keep a reference to ctx2 and call persist on it after the flush ,
becouse persist requires the proxy, i think i finally got the bit about
queing, and i'm going for the edit-persist sequence, and flush-fire (which
won't make much sense readability wise), but then again, neither is most of
r
but that's the thing i don't get, you refer to the point AFTER i persisted
the entity, but what i'm talking about is how do i GET the new changed
entity proxy so i could send it the persist method. i mean just using the
fire won't magically persist the thing. unlike the find method in the
entity lo
i know i CAN use guice, but i CAN't use it, becouse our CTO is against
integrating anything else aside from GWT (yea i know using GWT alone is an
anti-pattern, but he doesn't want any added risks, hell he won't let us even
use Activity places)
so i was wondering if there is any underlying mechanism
oh and we can't use GXT.
On Wed, Jul 20, 2011 at 5:15 PM, Elhanan Maayan wrote:
> before i check out, does this framework implements MVP on it's own? becouse
> i don't want to duplicate the work of allready in GWT (for example MVP4G ,
> which seems great but doesn't
before i check out, does this framework implements MVP on it's own? becouse
i don't want to duplicate the work of allready in GWT (for example MVP4G ,
which seems great but doesn't use places and activities and thus bind me
more to it's code) , the same goes for the editing framework, i would like
are there ANY open api's in gwt?
On Tue, Jul 19, 2011 at 5:01 PM, Thomas Broyer wrote:
>
>
> On Tuesday, July 19, 2011 12:18:58 PM UTC+2, Elhanan wrote:
>>
>> what do you mean by open api?
>>
>
> A "web API" (similar to Twitter's API, GData API, etc.) that's clearly and
> entirely specified so y
what do you mean by open api?
btw on a similar note, what's exactly Titanium Appcelerator , seems like a
every CTO's wet web dream.
On Tue, Jul 19, 2011 at 1:13 PM, Thomas Broyer wrote:
> You're right. RequestFactory is a protocol whose wire-format is based on
> JSON, just like XML-RPC or SOAP
if there's one thing haven't figured out completely it's the layouts, i
mean, i wanna use the DockLayoutPanel in the ui designer for example with
only 3 panels inside it (one containing a tree, and other textbox and the
last a table) , and yet i can't see the panels i've inserted.
what's more if i
i think google should clarify it's own philosophy in a special article in
the developer guide, because as of right now , there way too many "black
holes" on the subject of transports and frameworks that use them
GWT-RPC
JSON
RequestFactory
AutoBeans
XML.
i know that autobeans and rtf both uses js
woo hoo! mabye i'll get to convince our CTO to use RF, he wants to avoid
that becouse he says if the client will be in sliverlight, that won't work.
On Fri, Jul 15, 2011 at 7:27 PM, David Chandler wrote:
> We jumped the gun a bit, but 2.4 is pretty close to becoming "latest" :-)
>
> /dmc
>
>
> On
considering it's talking about requestfactory-server.jar hmm, pretty much
yea:) (is 2.4 THAT close?)
On Fri, Jul 15, 2011 at 11:38 AM, Thomas Broyer wrote:
> Hmm, shouldn't it have been for "trunk"? I believe @ExtraTypes and
> polymorphism support are not in 2.3, only in the upcoming 2.4.
>
>
does this mean i cannot use validation if i'm not using requestFactory?
http://code.google.com/webtoolkit/doc/latest/DevGuideRequestFactory.html#validation
On Tue, Jul 12, 2011 at 9:57 PM, Nick Chalko wrote:
> GWT 2.4 beta include support for built in Validation take a look at
> http://code.goo
i've been reading about security issues, although it make not seem related
embedding "foreign" javascript in gwt is considered a security risk, you
should look into that also.
On Wed, Jul 13, 2011 at 3:39 AM, mike b wrote:
> The BIRT dev environment has a way of exporting javascript to embed in
i don't think sticking session applies here in case of clustering xsrf
protection as i understand it is meant to ensure each REQUEST came from the
actual client, and not adjacent website which activates your request via
javascript code injection (which cannot really get your cookie because of
SOP)
is there another way to reflective recursive search in a form?
it's just occured to me, that won't solve my problem with buttons and things
that are not editors.
it would the developer would have would have somehow "annotated" the element
of the widget with something and i would have to create a
can i use it with my normal classes i'm looking for a way to have a
singleton class and obtain it from a shared location, said class will
contain, eventsbus,ActivityManager,ActivityMapper etc..
i also rely in many cases on static variable which are androgynous clases, i
failed to to see that press
actually i allready took a look at the dev guide, that's where i got the
idea from
i wish i had some example on how to use the RPC class in this way, i
understand i still gonna need to create a servlet to extend a remote servlet
only this time it would without any interfaces or async interfaces.
hi..
assume that i have a Dynamic Web Project configured with google sdk, which
references another project for domain classes, which itself references a 3rd
project for persistence utility classes.
i'm trying to use tomcat as debug servlet , but the only way i'm able to
reslove ClassNotFoundExce
actually we've decided to abandon RequestFactory framework for the reason
that it cannot be ported elsewhere.
that is if we decided to have other clients such as silverlight/flex, those
won't have request factory capabilities . so it kinda binds us to gwt.
On Mon, Jul 4, 2011 at 12:56 PM, Robert W
and does jetty support that? btw what DOES jetty support in terms of jee? we
are planning for websphere 7..
On Mon, Jul 4, 2011 at 8:00 AM, Juan Pablo Gardella <
gardellajuanpa...@gmail.com> wrote:
> Use a Servlet 3.0. You can configurate via annotations.
>
>
> 2011/7/2 iker
>
>> It is possible
if i may add my 2 cents, i think that Places-Activities-Framework(PAF?
anyone? , each time somebody adds another component, and it's beginning to
be absurd, so why not go for PAF :) ). should NOT be considred as an MVP
framework, nor should be considered as history handling one.
for example our CTO
our CTO wants to generate jpa stuff, can you bring some articles on how to
do stuff with xtext?
On Fri, Jul 1, 2011 at 1:55 AM, dd wrote:
> ### AutoBeans can be used on client *and* server side:
>
> static interface MyAutoBeanFactory extends AutoBeanFactory {
>
> AutoBean login();
>
> }
i believe to treat rf api , is to first think about how would do it with
normal java dao, and persistence api, then begin to translate them to rf
api's becouse that's basically what it tries to do.
by persistence api, i mainly thinkking about DAO's classes, where each DAO
usually revolves around a
i wonder if there are open source projects for that. actually it would a
good tooling addition.
On Tue, Jun 28, 2011 at 11:36 AM, Thomas Broyer wrote:
>
>
> On Tuesday, June 28, 2011 7:29:30 AM UTC+2, Elhanan wrote:
>>
>> how can you automate writing proxies?
>>
>
> We wrote a code generator (in
in regards to to re-using pojo's that may sound like a good idea in the
begining (hell, i like it myself) however bare in mind that your'e
restricting yourself to the jre emulation limitations on the client side,
meaning you won't be able to use any cross-cutting domain services (like
persistence,a
how can you automate writing proxies?
in regards to deserliziing why not use the solution i wrote before, get a
string, turn it to autobean and then use the visitor pattern with
ReflectionHelper to copy data autobean to the bean?
On Fri, Jun 24, 2011 at 4:34 PM, Thomas Broyer wrote:
>
>
> On Fr
can this embedded jetty be used with PersistentUnit servlet injections ?
On Tue, Jun 28, 2011 at 2:05 AM, Thomas Broyer wrote:
> The embedded Jetty should only be taken for what it's for: a simple servlet
> container. It's only meant to run servlets, you shouldn't even care whether
> it's Jetty
i allready did, and it works, i just wanted to make sure.
employee does implement IEmployee,i'm using the visitor becouse i wanted to
avoid all the boiler code for set=get stuff, this way i have a generic
class. my original intent was to use jpa's meta model api's and then
dynamically call autob
ok, so you could at least place this snippet of info at the TOP of the
devguide?
On Mon, Jun 27, 2011 at 5:29 PM, David Chandler wrote:
> Hi Eric,
>
> We don't publish the GWT docs for offline browsing, but instead of manual
> copying, you can use
>
> wget -r -p -np -k
> http://code.google.com/we
as far as i know there are NO downloadable docs, there IS however an open
issue about that, with a gagilion votes, i guess google is too busy being
sued by oracle...
On Sun, Jun 26, 2011 at 12:44 AM, Eric Atkinson <
eric.atkin...@strategicinternationalsystems.com> wrote:
> I have been able to dow
yea, but that means i can ever use driver.edit without using request.edit
first, and driver.edit already does request.edit
On Fri, Jun 24, 2011 at 8:41 PM, isern wrote:
> Just an idea, I'd try this:
>
> proxy = request.edit(proxy);
>
> driver.edit(proxy);
>
> If I got RF right, the edited refere
thanks, btw, is there any problem that a service will contain it's own
locator?
what i've done is uinify the Service and Entity Locator into one class
(where you have a generic Entity locator which implements the finder
methods and then a service locator which extends it with the new Instance
met
doubt that, as java JEE is mature and advances in a steady rate, GWT is
still "young" and exploded with new patterns and technologies.
i mean have you seen any current documentation worth it's salt aside from
the unprintable developer guide?
essential gwt is the only book on gwt 2 , but doesn't inc
> 2011/6/22 Nicolas Antoniazzi
>
>> Here, we use Shiro since it is really easy to integrate.
>> Moreover, Shiro is not tight to Spring... And that's a good thing for us
>> since we use Gin on client and Guice on Server (to not multiply API).
>>
>> 2011/6/22 E
we are exploring apache shiro with gwt integration, the reason is first of
all, you don't spring beans for it, and 2nd is that also provides for
session management which spring security does not.
even folks at spring forums say shiro has much broader scope then security
..
On Wed, Jun 22, 2011 a
first off you probably mean the strong type definitions provided by GWT,
not java right?
second, that's not what i'm suggesting, i don't want to use events for
enable disable buttons, but i do want to use to create logical
messages, that a work process has ended/started, or basically whatever
ca
i'm unfamiliar with SOP and gwt policy files,
i havn't tried much of rpc aside form the fact that each service should only
object of input and object of output both extending RequesInput and
RequestOutput interfaces (generic wise) to avoid break method signature
incase of changes .
i then moved o
actually that's EXACTLY what i did, only on in the place itself as thomas
thinks i did, there is only ONE place class which contains as single token
string.
the reasons for this is override gwt's type system for the places where each
place class is by default a new place, which goes against oo pr
actually i think there should be no code activity mapper as the place itself
should contain the logic on what activity it should be refferred, for each
new place activity combo i have to go though lots of boiler plate code which
is really redudent..
On Mon, Jun 20, 2011 at 3:43 PM, Thomas Broyer
70 matches
Mail list logo