Welcome back!
On Sun, Apr 26, 2020 at 1:27 PM Kalle Korhonen
wrote:
> Hi all,
>
> just wanted to say hello as I'm back working on Tapestry & Tynamo, at least
> for a fleeting moment! It all started when I was brought to steward an
> unruly, JSP-based(!) codebase that used Spring ACL as its secur
Tapestry is not quite smart enough to know that .min.css means a minimized
CSS file.
You should have access to the non-minimized CSS file, so change it from
"context:bootstrap/bootstrap.min.css" to "context:bootstrap/bootstrap.css"
(or equivalent).
The whole concept of Tapestry doing the minimiza
I'm working on it now. It's going to take a bit to rebuild from the
source, I haven't touched this code in a while and Gradle has changed a bit.
On Tue, Apr 14, 2015 at 8:27 AM, Howard Lewis Ship wrote:
> I'll find some time later today to work on this. It got lost in t
I'll find some time later today to work on this. It got lost in the
shuffle when I was moving hosting providers.
On Tue, Apr 14, 2015 at 7:50 AM, John Owen Atala
wrote:
> Hi,
>
> tapx maven site is not working.
>
> I try to download git source, but i cannot build it (i think it fails
> because
>>> snapshots should work fine. And furthermore - how would you like this
>>>> changed? If it's just adding a Page as a threadlocal, perhaps you can
>>>> just
>>>> write a patch for it.
>>>>
>>>> Kalle
>>>>
>>&
Possibly we need something more advanced; our own reference type that can
react to memory pressure by discarding pages that haven't been used in
configurable amount of time.
Or perhaps we could just assume that any page that has been used once need
to be used in the future and get rid of the SoftR
No, we've been having problems adapting to some changes in Apache
infrastructure.
You can build locally.
beta-28 is available as a preview release as well, and it's about 24 hours
old.
https://repository.apache.org/content/groups/staging
On Thu, Feb 12, 2015 at 6:39 AM, Sven Homburg wrote:
>
For modules, which can't be fingerprinted with a checksum in the URL,
Tapestry applies a short expiration time in development, and a much longer
one in production. I suspect you are hitting that. A hard refresh,
clearing the local cache, should force the browser to request the updated
resource.
S
Further, for both modules and normal assets, there's etag support ... so
most requsts for a module get a 304 and the browser can use its local cache.
On Tue, Jan 27, 2015 at 10:55 AM, Howard Lewis Ship
wrote:
> Modules can't have a far-future expires header, because we can't p
Modules can't have a far-future expires header, because we can't put a
fingerprint (asset checksum) into a module URI.
All other assets have a checksum in the URI and get the far future expires
header.
The handling of this is different between development mode and production
mode. Short or no exp
This release is now available for download, or from Maven central.
Notable fixes and improvements since the previous beta:
- Fixed problems with tracking validation errors on fields inside Ajax
updates
- Improvements to the exception report page, and the ability to write an
exception
The mask is configurable, that is, it can be disabled. That being said,
perhaps it should dismiss after a fixed period of time (a couple of
seconds) or even if any initializations throw an exception.
On Wed, Jan 14, 2015 at 12:48 AM, nhhockeyplayer nashua <
nhhockeypla...@hotmail.com> wrote:
> Th
Tapestry loads all the JavaScript in the background; libraries are loaded
sequentially, modules will load in parallel. Your is executing
immediately, before any of this background loading has started.
On Fri, Jan 9, 2015 at 9:53 AM, nhhockeyplayer nashua <
nhhockeypla...@hotmail.com> wrote:
> Ok
You know, I sometimes post that link to "How to ask questions ..." but
here's another observation: play more face-to-face board games if you want
better support.
Playing a F2F game (Settlers of Catan and its thousands of descendants)
teaches you non-emotional empathy: the ability to view the worl
This is usually because of how a Loop inside the Form (and around your
snippet) is configured. You are probably forcing the currentObject itself
to be serialized, and if that has references to other serializable objects,
it can get large. Generally, you should just be serializing an object's id.
Se
The main motivation behind URLEncoder (side note: really wish I had used
the term "Transcoder" for these things) was because Jetty, Tomcat, and
my-weak-understanding-of the servlet spec didn't agree on how and when
URL-encoded data in the path is decoded.
On Thu, Dec 25, 2014 at 10:49 AM, Thiago
ably.
> >>>
> >>> On Thu, Dec 18, 2014 at 6:49 PM, Ilya Obshadko <
> ilya.obsha...@gmail.com>
> >>> wrote:
> >>>
> >>> Thanks Howard, that's making a lot of sense.
> >>>>
> >>>> However my
Actually, Clojure interop with Java is very good, so a Clojure function
could be passed a Java object:
(defn frobnicate-the-request
[^Request request]
(.setAttribute request "xyzzyx" (compute-the-magic-name)))
The ^Request part is a type hint, it allows the Clojure compiler to
generate proper
Well, right off the bat, any NPE like this (with no additional message to
explain it) is a failure.
Was there anything special about the page being rendered? It seems like
the MIME type of the page is null, that's what might cause this NPE.
On Tue, Dec 9, 2014 at 11:01 AM, George Christman
wrot
y5/exception-frame.css
> >- …z9cac79b1/tapestry5/tapestry-console.css
> >- …/assets/meta/z123f049/tapestry5/tree.css
> >- …uto/assets/meta/zdf3b597a/css/layout.css
> >- …to/assets/meta/z1a175d43/css/desktop.css
> >- …ets/meta/core/z4225cc8a/Autocomplete
Tapestry puts all of its main JavaScript at the bottom of the page. It
includes a tiny amount of JavaScript at the top of the page, to introduce
an element that prevents unwanted user input until the page is fully
loaded. On newer browsers, it will animate in a translucent overlay and a
cycling "lo
By contrast; the Clojure/Conj and Strange Loop conferences are getting
videos up on the same day as the talk! Strange Loop even had a
stenographer doing live subtitles.
On Sun, Dec 7, 2014 at 8:09 AM, Thilo Tanner
wrote:
> Hi,
>
> Unfortunately not. They recorded audio, but not video and as far
It's a very deliberate security design implement by all browsers otherwise
a invisible hidden form field could read any file on your filesystem
On Friday, December 5, 2014, Thiago H de Paula Figueiredo <
thiag...@gmail.com> wrote:
> On Fri, 05 Dec 2014 16:41:55 -0200, martijn.list
> wrote:
>
>
Sometimes I'll use curl to get the module, to see if it has recompiled ...
and then, yes, it's usually the browser caching when I really want it not
to.
In addition, modules are streamed with Cache-Control header that gives them
a short time to live (I believe a minute by default). Otherwise, eac
That's odd, but I haven't played with the 5.4 quickstart yet. First thing
to check is that the HTML text fields have the data- arguments that trigger
validation on the client. If those are missing, it's a server-side problem
(perhaps with the annotations). If they are present but not acted upon,
th
The change in JavaScript is to support several goals:
- A good transition from Prototype.js to jQuery (or perhaps something else
in the future)
- Reduce the total amount of JavaScript sent to the client
- Allow more of the JavaScript to transfer in parallel
- Make it easier to override Tapestry's b
I've been thinking of a blog post about everything I would do differently
in Tapestry if I had perfect foreknowledge. Trust me, it would be
different, and simpler in many ways ... and based more heavily on
immutability (& controlled mutability) ala Clojure.
Parallel rendering would be nice though
Coda: The application version IS still used in exception reports (both the
HTML version and, if memory servers, the new text file written to the file
system).
On Thu, Oct 23, 2014 at 3:33 PM, Paul Stanton wrote:
> great!
>
>
> On 24/10/2014 8:37 AM, Howard Lewis Ship wrote:
>
>
That is correct; tapestry.application-version is no longer needed. Each
individual asset gets its own content fingerprint. This has huge
implications for upgrades of you application, as often, all the unchanged
assets will maintain the same fingerprint, and already be present in the
end-user's bro
Just a survey ... does ANYBODY use the FormInjector component directly?
It's going to be a lot of work to upgrade it for 5.4 and I'm very tempted
to just remove it outright.
--
Howard M. Lewis Ship
Creator of Apache Tapestry
The source for Tapestry training, mentoring and support. Contact me to
This needs to be mentioned in the release notes, if not done so already.
On Wed, Oct 1, 2014 at 8:22 AM, Chris Mylonas wrote:
> Ah OK fair call.
>
> It's another change of behaviour from upgrading from 5.3 to 5.4 I noticed
> - 5.3 came with log4j-slf4j stuff included.
> 5.4 comes with plug-your-
I'm working my way through the issue lists, trying to hit low hanging
fruit, and get rid of duplicates and other things.
I'm resolving many issues as "won't fix", not because they are invalid, but
because the problems described are small, or the fix is highly disruptive,
or because the committer t
This is not currently possible with Tapestry; in fact some people use just
the opposite approach, using a Zone update to keep the session live.
It is difficult to tell, on the server side, what the "intent" of a request
is, and Tapestry has had no concept of "user generated action" as the
intent.
I've added a new symbol to control this behavior, so that you can have it
for development, but be safe in production.
On Wed, Jul 30, 2014 at 2:19 PM, Lance Java
wrote:
> Note, this could also break non GAE environments since it calls mkdirs() on
> @Symbol(WebResourcesSymbols.CACHE_DIR). For ex
Check the recent commits, I think this got taken care of last week.
On Mon, Jul 14, 2014 at 1:14 PM, George Christman
wrote:
> Hi guys, I'm wondering if there is a chance we could upgrade the
> typeahead.js to the latest version? It seems as if we are a little more
> than a year out of date.
>
If you getter method is "is", not "get", prefixed then the type MUST be
boolean, not Boolean. That's well defined in the Java Beans specification.
On Thu, May 29, 2014 at 12:41 AM, Lance Java
wrote:
> Eg: https://gist.github.com/uklance/0d461f9618b3a131ca3b
> On 28 May 2014 12:43, "Lance Java
Is there any possibility that there are two different Alerts components on
the page?
On Thu, Apr 24, 2014 at 2:20 PM, Jon Williams
wrote:
> Still broken in the beta5.
> Pretty sure this is a bug, i'll make a JIRA issue.
>
--
Howard M. Lewis Ship
Creator of Apache Tapestry
The source for Tap
That's standard HTML browser behavior; when you hit enter in a text field,
is searches forward for a submit and clicks it. You can perhaps address
this by putting an event handler on the text field itself. I'd have to do
experimentation/research to find the correct event.
On Thu, Apr 17, 2014 a
http://catb.org/esr/faqs/smart-questions.html
The problem(s) you are having is that you are treating Tapestry component
templates and assets as if they are simple files. They are not. The URLs
for accessing a page, and accessing an asset (such as the d3.js script and
your JSON file) are different
Look in the documentation about moving the regexp out of the binding and
into the page's message catalog.
On Tue, Apr 1, 2014 at 3:00 PM, Richard Frovarp wrote:
> This looks to be a bug, but I can't find any good documentation, so I
> thought I'd bring it here first.
>
> I through a really quic
Note that this will work differently, and more simply, in 5.4, where it's
just a matter of putting them into the correct sub-folder of
META-INF/assets.
On Fri, Mar 28, 2014 at 2:09 PM, Thiago H de Paula Figueiredo <
thiag...@gmail.com> wrote:
> On Fri, 28 Mar 2014 04:55:17 -0300, Nathan Quirynen
For normal assets, the URL directly includes a content hash value. This
covers stylesheets, images, fonts, and JavaScript libraries. In addition,
assets provide both far-future expires headers and ETags, to assist with
caching in the client or intermediate servers.
Modules are a bit different be
Just a follow on ... following a successful vote, 5.4-beta-3 is now
available for download (binary, source, and API documentation) from
http://www.apache.org/dist/tapestry/ ... and is now available from the
central Maven repository as well.
--
Howard M. Lewis Ship
Creator of Apache Tapestry
The
Tapestry is starting to stabilize; this preview release includes quite a
few bug fixes, including those from Thiago's "Month of Tapestry".
The beta-3 preview also includes upgrades to dependencies:
- Bootstrap 3.1.0
- CoffeeScript 1.7.1
- Less4J 1.2.4
Unfortunately, the tapestry-cdi module was pu
I'd suggest monkey-patching the client-side code to match your design needs.
On Wed, Feb 12, 2014 at 1:45 PM, nhhockeyplayer nashua <
nhhockeypla...@hotmail.com> wrote:
> Thanks for the fast feedback.
>
> Can anyone suggest a reimplementation for the original code block?
>
> @Inject
> pr
Validation decoration is now done on the client-side, and is mostly done by
toggling CSS classes on particular elements (containers of the field).
On Tue, Feb 11, 2014 at 10:07 AM, nhhockeyplayer nashua <
nhhockeypla...@hotmail.com> wrote:
> Folks,
>
> I have some code using
>
> ValidationDecora
I agree; this is an ugly hack to satisfy the arbitrary CSP requirement.
On Thu, Feb 6, 2014 at 5:20 AM, Lance Java wrote:
> In theory, the data URL approach sounds perfect.
>
> But in reality we'd be swapping this:
>alert('hello');
>
> For this:
>
>
> As you mentioned, it's likely that a
It's easy to say "it should be cached" until you realize that the
application could be running in a cluster. At that point, you need to have
that cached data available across all servers in the cluster ... that means
you need to store it in the HttpSession, which is exactly the opposite
approach th
Also, you can contribute overrides of modules to the ModuleManager service;
this allows you to provide any old resource, including a patched version of
a core module. This is less error-prone that having two different files on
the classpath with the same path and name ... you aren't beholden to a
ser with the perception the app were slow.
>
> cheers,
> Kris
>
>
> On Fri, Jan 17, 2014 at 8:46 PM, Howard Lewis Ship
> wrote:
>
> > That said, in a HTTP 2.0 world, there isn't such an advantage to
> > aggregation, especially with Tapestry's asset URL
That said, in a HTTP 2.0 world, there isn't such an advantage to
aggregation, especially with Tapestry's asset URLs (e.g., the encoding of
the content hash into the URL).
On Fri, Jan 17, 2014 at 11:45 AM, Howard Lewis Ship wrote:
> Google Closure is best in a situation wh
Google Closure is best in a situation where *all* the JavaScript that will
be part of the page is known statically; it goes beyond other minimizers in
that, if you follow certain rules, it can perform real optimizations, such
as dead-code analysis and function inlining. It can work with a couple o
What are you looking for? Advanced minimization, unfortunately, doesn't
make sense when you are minimizing multiple files (there can always be a
mix of stacks, libraries, and modules).
On Mon, Jan 13, 2014 at 12:29 PM, Kristian Marinkovic <
kristian.marinko...@gmail.com> wrote:
> Hi all,
>
> ho
The outer function is a "hygenic wrapper" created by the CoffeeScript
compiler. Occasionally, it places a few things that should be local to the
individual file inside the function and before the define(). I haven't
checked yet whether the Google Closure compiler is smart enough to unwrap
these, wh
Looks great!
On Mon, Jan 6, 2014 at 4:44 AM, Lance Java wrote:
> Haha... I did see some similarities. Parts of tapestry-stitch were inspired
> by Jumpstart You scratch my back, I'll scratch yours!
> On 6 Jan 2014 10:00, "Geoff Callender" <
> geoff.callender.jumpst...@gmail.com>
> wrote:
>
>
Which version of Tapestry?
On Wed, Dec 18, 2013 at 8:15 AM, George Christman
wrote:
> Hello, I've created a modal dialog box mixin that is attached to my
> submit actions. On form submission, is there a way to detect whether
> or not clientside validation errors have occurred in js so I can
> pr
The key to performance with Grid is to look to the GridDataModel and
provide a smarter caching version that limits the amount of data taken from
the database.
That being said, Grid is an scaffolding component ... much like
BeanEditForm. It is meant to be useful but if a grid view of some form is
e
Following a beta release vote, I've generated the first beta preview
release of Apache Tapestry 5.4.
The bundles version of Bootstrap has been updated to 3.0.3, and jQuery to
1.10.2.
Tapestry now bundles moment.js, and adds new components LocalDate and
TimeInterval. LocalDate displays a date/time
The short form is: trigger an events.zone.update event on the zone's
client-side element to update it immediately, using available content (a
string, or DOM elements).
Or, trigger events.zone.refresh event when you know a URL and additional
parameters; Tapestry will run the Ajax request and updat
Getting very close to a beta ...
I've done some re-work on assets and modules; its a bit simpler with
shorter URLs now.
When a JavaScriptStack includes a module, any time the module is required
(via JavaScriptSupport), the stack itself is imported. This
parallels existing logics for JavaScript li
On Wed, Nov 27, 2013 at 8:07 AM, Thiago H de Paula Figueiredo <
thiag...@gmail.com> wrote:
> On Wed, 27 Nov 2013 13:53:09 -0200, Norman Franke
> wrote:
>
> I hope it works out.
>>
>
> Thanks!
>
>
> Anything you can add to help porting from prototype-based Tapestry JS
>> code to T5.4 would be we
gt; Emmanuel
>
> PS: Are shaving machines so expensive at where you live? You need to do
> something about that beard. The last time I saw you live, you were looking
> very much like the deceased Iraqi dictator Saddam Hoessein.
>
>
> On Tue, Nov 26, 2013 at 6:14 PM, Howard Lewis
g-td5724691.html#a5724745
>
>
> On Mon, Nov 25, 2013 at 8:46 PM, Howard Lewis Ship
> wrote:
>
> > Getting very close to a beta ...
> >
> > This release fixes a latent bug in the jQuery event support that would
> > manifest as forms with input validations getti
Getting very close to a beta ...
This release fixes a latent bug in the jQuery event support that would
manifest as forms with input validations getting submitted via a LinkSubmit,
and perhaps other ways.
I've hopefully fixed a bug that affects Windows/Eclipse users, where the
application will no
A couple of years have gone by, and our favorite Tapestry mailing list
troll is back. Please ignore him. Do not respond to his taunts and he will
return to his normal state of total irrelevance. It is not possible to
"engage" a troll in discussion.
--
Howard M. Lewis Ship
Creator of Apache Tape
I get nervous about creating new binding factories that then try to parse
the (rather complex) property binding syntax, that's all. We've kind of
used up all the punctuation and delimiters.
On Sun, Nov 17, 2013 at 9:30 AM, Matthias wrote:
> @Howard
> Thanks for your reply. I know that it was po
What's you environment, including exact version of Tapestry?
On Thu, Nov 14, 2013 at 1:26 PM, Eugen wrote:
> Good day,
> when I try to require the bootstrap/popover Module I become an error
> in Firefox console:
>
> [22:05:14.435] "RequireJS error: scripterror: Script error for:
> bootstrap/pop
You do it in Java code. public String getLinkClass() { return isActive ?
"active" : "notActive"; } and just reference ${linkClass} in the template.
On Sun, Nov 17, 2013 at 6:30 AM, Matthias wrote:
> Hi, if theres something in tapestry for inline conditional expressions? I
> think of something
Add tapestry-webresources as a dependency and all shall work.
On Tue, Nov 12, 2013 at 9:27 AM, Matthias E. wrote:
> Hi, im trying to include some *.less css with 5.4-alpha-25. My first try
> was to use the @Import and it looked like: @Import(stylesheet =
> "context:css/bootstrap-datetimepicker.l
class listening for zone
> changes or initialize something once the library is loaded.
>
>
>
You should have the markup that needs that include an @Import(module="xxx")
or an explicit call to JavaScriptSupport.require().
> On Tue, Nov 12, 2013 at 12:01 AM, Howard Lewis Sh
This updates a number of bundled dependencies:
- Bootstrap 3.0.2
- RequireJS 2.1.9
- Less4J 1.2.0
- Closure Compiler v20131014
It also fixes a bug that prevent JavaScript stacks from being minimized.
There's also been a little bit of juggling to make Underscore a library
within the core JavaScript
My mistake; I mistook weird end-of-line handling for minification. I'll
look into this.
On Mon, Nov 11, 2013 at 10:38 AM, Howard Lewis Ship wrote:
> Works for me! I turned off JS aggregation and minification in a client app
> and it works exactly as expected.
>
> You probab
at 9:08 AM, Howard Lewis Ship wrote:
> That sounds like an oversight, and I'm a bit surprised!
>
>
> On Sun, Nov 10, 2013 at 6:55 AM, Bård Magnus Kvalheim
> wrote:
>
>> Hi guys.
>>
>> Just noised that stack js in not minified in 5.4.
>> Tried
That sounds like an oversight, and I'm a bit surprised!
On Sun, Nov 10, 2013 at 6:55 AM, Bård Magnus Kvalheim wrote:
> Hi guys.
>
> Just noised that stack js in not minified in 5.4.
> Tried on 5.4.22 and 5.4.24 and seems to be missing on both.
>
> I'm in production mode and have also set
> combi
tages.MixinsSolver.buildMixinReferenceReplacement(MixinsSolver.java:127)
> at
>
> com.github.sommeri.less4j.core.compiler.stages.ReferencesSolver.solveMixinReferences(ReferencesSolver.java:128)
> at
>
> com.github.sommeri.less4j.core.compiler.stages.ReferencesSolver.unsafeDoSolveReferences(ReferencesSolver.java:81)
>
>
>
A few highlights in this release:
- Tapestry can now aggregate modules into JavaScript stacks; in fact, the
only use for a contributing modules to a stack is for aggregation(*).
- Bundle Bootstrap 3.0.1
- Exception trace now presented as a more readable table
- RequireJS is now part of the core Ja
That's an interesting find; I've never seen that before. Does it happen
consistently?
That being said, it's highly unlikely we can reproduce it without some
insight into what modules you are using and (a guess from what I'm seeing
in the stack trace) whats in your @Startup module methods.
On Thu
The WRO4J project includes plugins for Maven that allow you to do the same
conversions at runtime; the CoffeeScript compiler will even use the node
version of CoffeeScript for much faster compilation.
In an ideal world, you could use runtime compilation in development and
pre-compilation for produ
Does this only occur during testing? I've found that we sometimes see this
during testing when Selenium starts loading a new page while there are
outstanding AMD module requests; the requests fail as if 404 just before
the new page content loads.
On Thu, Sep 26, 2013 at 5:51 AM, Peter Hvass wro
You can also test your theory by disabling GZIP compression.
On Thu, Sep 26, 2013 at 9:59 AM, Howard Lewis Ship wrote:
> Does this only occur during testing? I've found that we sometimes see
> this during testing when Selenium starts loading a new page while there are
> outstand
t; > at
> >
> org.apache.tapestry5.ioc.internal.ModuleImpl$9.transform(ModuleImpl.java:494)
> > at
> >
> org.apache.tapestry5.plastic.PlasticManager.createProxy(PlasticManager.java:235)
> > at
> >
> org.apache.tapestry5.ioc
c elements, is the main difference.
On Tue, Sep 24, 2013 at 2:10 PM, Bård Magnus Kvalheim wrote:
> Thanks, that sounds good and would go a long way.
>
> Will make me worry less currently moving from stacks to modules - and it
> does seem cleaner..
>
> On Tue, Sep 24, 2013 a
Actually, I hope to extend the stack concept to allow modules to be
included in the stack.
RequireJS's optimize does a bit more; you specify some module names and it
figures out the transitive dependencies and builds a single script. That
would be great, but I'm not sure I can pull it off.
On Tu
That's a problem; any idea if the ASM code inside Plastic (for Tapestry
5.3+) works on JDK 8?
Javassist is 100% gone in 5.4.
I've seen some troubling discussions about bytecode changes in JDK 8 that
are going to cause a lot of havoc.
On Thu, Sep 19, 2013 at 11:12 AM, Kalle Korhonen wrote:
> J
This is part of using Boostrap; which is mobile-first and therefore likes
everything stacked vertically and full width. Add "form-horizontal" to the
[BeanEdit]Form and the layout will look nicer. You have to wrap the
form-control elements in .col-md-N (md for medium device size) to keep them
from
Some module, hopefully not an of the t5/core modules, has a dependency on
"bootstrap" ... but the Boostrap JS has been broken up into a number of
small modules, such as "bootstrap/modal". The modules don't export
anything, but can be declared as module dependencies to ensure they are
loaded.
On
I've actually created a series of preview releases since the last notice.
There's been a lot of changes to work with Bootstrap 3 correctly.
There's now an option to run the pop-up JavaScript console in invisible
mode, which is useful for testing. In visible mode, the console can
interfere with te
You can store the file together; you just need to update your build tool to
package those files along with the compiled classes. This is not the
default for Maven or Gradle, but can be easily added. You may also need to
do some tweeks on how you are running the application inside your IDE, to
ensu
On Wed, Sep 11, 2013 at 8:44 AM, Lenny Primak wrote:
> I believe if you keep prototype, there is a compatibility module that
> keeps that namespace. I am not 100% sure about that because I never tested
> that scenario.
>
> On Sep 11, 2013, at 11:34 AM, Norman Franke wrote:
>
> > What about the mi
On Tue, Sep 10, 2013 at 12:08 PM, Lenny Primak wrote:
> Hi guys,
>
> I've been recently trying to convert some of our client apps to tapestry
> 5.4
> I am using alpha-17 as of this report.
>
> All of these depend on the FlowLogix library so one of the first tasks is
> to migrate the library.
> Her
The checked exception behavior was explicit and on-purpose; it is in-line
with what the EJB containers do: a checked exception represents a failure
from which some recovery is possible, as opposed to a runtime exception
which leaves the system in an unknown state (therefore, rolling back the
transa
ling:
> - click outside the datepicker and have it disappear
> - type in gibberish into the date field and still be able to pull up
> datepicker
> - not have page re-flow when date picker is popped up
>
>
> On Sep 5, 2013, at 7:17 PM, Howard Lewis Ship wrote:
>
> > Works f
You can leave the asset in the old location, and Tapestry will find it,
with a warning.
Or you can move it to the new location; /assets/LIB-NAME/path.js ... the
LIB-NAME is determined from the library containing the component.
The error message only reports the 5.3 location; it should be updat
Works for me :-)
What browser?
On Thu, Sep 5, 2013 at 1:39 PM, Lenny Primak wrote:
> FYI:
>
> There was a whole bunch of bugs that are filed against datepicker that are
> once again broken in T5.4 (alpha-17 at least)
>
> Here are the JIRAs:
>
> https://issues.apache.org/jira/browse/TAP5-805 (c
This should probably go into the main code as well; possibly with a
compatibility code to make it easy to turn off for users that don't need
IE8 support (i.e., so as to streamline their apps in modern browsers).
On Wed, Sep 4, 2013 at 8:00 AM, Peter Hvass wrote:
> Hello,
>
>
> Just a heads up i
StreamableResource is a kind of wrapper around Resource that includes
additional information needed to generate asset URLs and a bunch of other
related concerns.
There's a StreamableResourceSource service.
Oops, that's unfortunate; you'll need the internal ResourceChangeTracker
service (which imp
Yes, seems to have a repackaged copy of part of the Rhino parser.
On Tue, Sep 3, 2013 at 12:12 PM, Lenny Primak wrote:
> Thanks, looks like the clojure compiler integrates Rhino correctly, and it
> won't interfere
> with the official rhino.
>
> On Sep 3, 2013, at 3:06 P
Tapestry 5.4 switched from YUICompressor to Google Closure, and things have
been much better. We use Closure for just simple minimizations (removal of
whitespace, local variable renames, etc.).
We do use the YUICompressor CSS minimizer.
On Tue, Sep 3, 2013 at 11:27 AM, Lenny Primak wrote:
> Hi
Looks like the code that checks for case mismatch is broken when the path
separator is different. It must be hard coded to look for "/", which is a
problem on Windows.
On Fri, Aug 30, 2013 at 6:55 AM, Ville Virtanen <
ville.virta...@orientimport.fi> wrote:
> Dear list,
>
>
>
> just tried to upg
This is something being actively pursued in the Typeahead.js forums.
http://jsfiddle.net/needathinkle/MA7Ep/
On Thu, Aug 29, 2013 at 9:00 AM, Peter Hvass wrote:
> Hello,
>
>
> We've been playing around with the new autocomplete from alpha 15 today
> and discovered
> some gnarly style issues.
>
1 - 100 of 3334 matches
Mail list logo
