u 1/19/2006 23:14
To: Ant Developers List
Cc:
Subject: Re: project dependencies
Thanks Phil.
But can I specify their dependency ?
--- Phil Weighill Smith
<[EMAIL PROTECTED]> wrote:
> Ted, this
Ted, this is something that should be addressed to the user list, not
the developer list. However, short answer: no. But remember that you can
call the ant build script anything you like. You could have a main
"build.xml" that invokes targets in or imports (as appropriate) the two
other build scrip
easier. Otherwise a simple typo would be hidden and it will be a painful
manual process to see why a build script does not do what it is
supposed to
do.
- Alexey.
On 10/25/05, Phil Weighill-Smith <[EMAIL PROTECTED]> wrote:
>
&
e Nachricht-
>Von: Phil Weighill-Smith [mailto:[EMAIL PROTECTED]
>Gesendet: Dienstag, 25. Oktober 2005 09:45
>An: Ant Developers List
>Betreff: RE: fails if inner fileset directory does not exist.
>
>My point is more about filesets
ould always set this flag on all provided filesets.
- Alexey.
On 10/24/05, Phil Weighill-Smith <[EMAIL PROTECTED]> wrote:
>
> Which has always given me grief. It has always seemed sensible to me
that
> a fileset would simply ev
Which has always given me grief. It has always seemed sensible to me that a
fileset would simply evaluate to no files if the base directory for the fileset
doesn't exist (and perhaps generate a verbose log mentioning that fact) in the
same way it does when the pattern(s) doesn't match anything.
I would suggest having a separate (external) "typedef" (or whatever the
appropriate Ant concept would be):
...
Referencing this value by ID would allow programmatic and build script
level references to be made to such a repository without the need to
know the exact de
On the understanding this is a "work in progress", I'd like to point out
that a project can now contain pretty much anything (if not everything)
that a target can (essentially creating an unnamed target that all other
targets in the file depend on implicitly) making the "project" element
definition
Jim,
What you've said sounds wrong to me; and are in the
"default namespace" (a declared namespace that has no prefix), whatever
that has been defined to be.
This could be the same as n1 if the URI for n1 is also associated with
the default namespace, but that seems unlikely.
Phil :n.
On Mon,
Would be handy. I'm assuming that the "name" could create a property of
that name, set to the process ID for example. You can then just "kill"
that process using this ID...
Phil :n.
On Fri, 2005-08-12 at 15:38 +0100, Steve Loughran wrote:
> I'm just reviewing what major changes I am needing to do
You may well have already included this sort of thing, but I'd cover
(not really esoteric):
* Immutability and overriding of properties (an often mis-
understood aspect)
* Conditional targets
* Usage of id/refid
* Multiple source tree support (e.g. for production an
Why not simply put two calls to javac in your build script and split the
source tree in two in the same way that you have in your new task,
passing one tree to the first call and the other to the second?
Clearly you need to ensure that the first call compiles "pre-requisite"
code for the second ca
On Mon, 2005-07-04 at 12:56 +0100, Steve Loughran wrote:
> Xpath is profound once you apply to
> object trees
Interestingly, and orthogonal to the initial discussion, JXPath has been
able to apply XPaths to object trees for quite some time. (Clearly XJ is
something else again and definitely inter
I was thinking about "local" properties, having read the discussions
relating to these in the last few months.
Perhaps we could introduce the concept of "property scopes" instead.
The project would provide a standard "global" property scope by default.
It would also provide the ability to push an
The key here is likely to be Sockets.
On Thu, 2005-06-23 at 10:05 +0200, Peter Nabbefeld wrote:
> Hello,
>
> I've asked this question on the users list, but it seems nobody knows there:
> Is it possible to create two java processes (with fork="true") and pipe
> the output of one process into the
Steve Loughran wrote:
> Phil Weighill-Smith wrote:
> > As to your varying new content, simply handle this with parameters to
> > your XSLT or something like that (perhaps even having different
> > conditional targets that handle different EJB configuration setups in
> >
started working with ant now.
> What I understood is of ant do the necessary processing and
> modifications.
>
> But if the value of this particular tag
> varies from EJB to EJB then
>
> srini.
>
> -----Original Message-
> From: Phil Weighill-Smith [mailto:
EJB then
>
> srini.
>
> -----Original Message-
> From: Phil Weighill-Smith [mailto:[EMAIL PROTECTED]
> Sent: Thursday, June 09, 2005 5:07 PM
> To: Ant Developers List
> Subject: Re: Can you help me!!!
>
>
> Try using to extract the XML file from the EAR t
.xml inside the
> .EAR and
> Element should be modified per EJB basis.
>
> What do you suggest?
>
> Regards,
> Srini.
>
> -Original Message-
> From: Phil Weighill-Smith [mailto:[EMAIL PROTECTED]
> Sent: Sunday, May 29, 2005 8:42 PM
>
Sun 29/05/2005 16:06
To: Ant Developers List
Cc:
Subject: Re: A possible solution for conditional execution of tasks?
Phil Weighill-Smith wrote:
>My opinion regarding the disadvantages of this approach:
>
>
My opinion regarding the disadvantages of this approach:
* Antcall has to create a whole new Project in memory in order to work
and is therefore an inefficient task
* If something invoked via Antcall depends on a target that is also
depended on by something depending on the target in
On Thu, 2005-05-26 at 10:06 +0100, Steve Loughran wrote:
> > Hmm. Currently the task is only writing the first
> > child as the root element of the output XML document,
> > and including the XML declaration. AFAIK a
> > non-well-formed fragment should forego the
> > declaration. If so, we can do
att Benson <[EMAIL PROTECTED]> wrote:
> --- Phil Weighill Smith
> <[EMAIL PROTECTED]> wrote:
> > More along the lines of will it work with
More along the lines of will it work with something like:
On Wed, 2005-05-25 at 05:04 -0700, Matt Benson wrote:
> --- Phil Weighill Smith
> <[EMAIL PROTECTED]> wrote:
> > Does it support document fragments?
> >
> Do yo
Does it support document fragments?
Phil :n.
On Wed, 2005-05-25 at 11:38 +0100, Steve Loughran wrote:
> Matt Benson wrote:
> > --- Alexey Solofnenko <[EMAIL PROTECTED]> wrote:
> >
> >>Why not to add this functionality into normal
> >>?
> >
> >
> > I suppose if you wanted to use entities,
> >
Rather than having a specific element, could this not
be achieved via a naming convention such as "-target-name" (the leading
"-" actually makes the target uninvocable from the command line anyway
and therefore pretty good and private from the CLI point-of-view)?
Or even a new attribute on target
On Fri, 2005-05-13 at 11:38 +0100, Jose Alberto Fernandez wrote:
> I would advocate to allow the importer the specify the aliasing name
> it
> wants to use for the imported things, so one has something like:
>
> foo.xml
>
> ...
>
>
> bar.xml
>
>
> ...
>
+1 to the 'as=""'.
Would we s
On Fri, 2005-05-13 at 10:25 +0100, Steve Loughran wrote:
> -when you override a target, you dont get access to its dependents.
> Workaround: many pseudo-targets that only model dependencies.
In fact we have "real" targets that don't have dependencies and it is
these that we override. These are al
On Fri, 2005-05-13 at 09:00 +0200, Stefan Bodewig wrote:
> > But say the importER explicitly depends on bar.foo . Isn't this
> > still going to pollute the log in the opposite way my implementation
> > would? :) i.e.
> >
> > [foo]:
> >
> > [bar.foo]:
>
> Yes. But this is less likely than havi
I missed the beginning of this thread but just want to say that I personally
think that import is the best feature in Ant today (apart from Ant's being in
the first place, that is)!
Phil :n)
-Original Message-
From: Matt Benson [mailto:[EMAIL PROTECTED]
Sent:
Try GMail... it works pretty well and is a cool web-based e-mail system.
Phil :n)
On Tue, 2005-05-10 at 08:59 -0500, Dominique Devienne wrote:
> And here comes the newly Company-wide instituted automatic
> disclaimer inserted at the outbound email gateway...
>
> I'm really sorry about this. And
On this topic, what do people think of the ability to define a filter
set from a property set, where each filter has the property's name and
the property's value?
Phil :n.
On Wed, 2005-04-20 at 16:48 -0500, Dominique Devienne wrote:
> > From: Matt Benson [mailto:[EMAIL PROTECTED]
> >
> > Any tho
Use of "core" as a package/directory name is mildly off in a
UNIX environment as the directory might be confused with a core dump!
;n)
Phil :n)
-Original Message-
From: Dominique Devienne [mailto:[EMAIL PROTECTED]
Sent: Wed 20/04/2005 15:28
To: Ant Developers List
Cc:
Subje
rs List
Cc:
Subject:Re: Ant symbolic debugger?
At least some JBuilder 2005 (SE+?) have ANT debugger that allows also
debug custom tasks.
- Alexey.
Phil Weighill-Smith wrote:
>I wonder if there's any mileage in developing a symbolic debugger for Ant
>scripts? While the debug out
I wonder if there's any mileage in developing a symbolic debugger for Ant
scripts? While the debug output is often adequate for simple(ish) scripts I've
had a number of cases where I really could do with a symbolic debugger (like
just now!).
Phil :n.
---
Steve,
I've not written any code to help here, but thought the following issue
could also be considered:
When executing a test class that has no test methods what-so-ever (e.g.
someone comments out all tests for some reason) the runner currently
barfs instead of ignoring the test class.
Phil :n.
I believe that using an undeclared prefix doesn't make the XML not well-formed
it makes the XML "invalid" (i.e. a validating parser will barf) [1]. I think
"well formed" simply means all the right syntax (element and attribute
declarations are correct, and are all correctly closed/nested).
[1]
Won't argue with your comment in general, but Xxx Yyy wrote:
>>
>>
>>
which seems to be a case where overwrite is appropriate.
Phil :n)
On Tue, 2005-03-08 at 16:44, Peter Reilly wrote:
> Phil Weighill-Smith wrote:
>
> >An alternative, and far more
PY fails.
>
> Temporary work-around is to override the granularity attribute on COPY.
> But the root cause of the problem (bad test in FileUtils) needs to be
> fixed.
>
> http://issues.apache.org/bugzilla/show_bug.cgi?id=33906
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
|FIXED
>
>
>
>
> --- Additional Comments From [EMAIL PROTECTED] 2005-03-02 01:15
> ---
> I used the javac "fork" option with "memoryMaximumSize=800m" and it
> worked fine.
> Now I feel silly.
>
> Anyone know how to pass to javac -Xmx800m without forking?
>
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
:
Subject: Re: Launching Ant again
Phil Weighill-Smith <[EMAIL PROTECTED]> writes:
> You could make your build generate a "build-targets.xml" from the
> project.xml then have a standard build.xml that:
>
> 1.
'build.xml' with the same command line
>
> Is case '3' really feasible. If yes, how can I do ?
>
> I'm almost sure case '4' is feasible. But how can I get the current process
> command line to invoke a new Ant ?
>
> Is there another way to do the complete process or only one step ?
>
> Thank you in advance for your help
> Regards
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
Lists and also
> > happens to be performant with ArrayLists, to do the
> > particular traversal
> > that I'm doing.
> >
> > Kev
> >
> > -
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
> >
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
o add a
special "use environment's class path" option to Ant's invocation to
allow for backwards compatibility).
Phil :n.
On Tue, 2005-01-25 at 09:37, Kev Jackson wrote:
> Phil Weighill-Smith wrote:
>
> >On Tue, 2005-01-25 at 09:22, Erik Hatcher wrote:
> >
> >
On Tue, 2005-01-25 at 09:22, Erik Hatcher wrote:
> You may consider creating a custom ant command script which unsets
> CLASSPATH before invoking the executable class.
>
Exactly what we do!
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
gt;
> > Peter
>
> Following the VBness of "let", how about "dim"? ;)
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
If we require an object parameter to be non-null our methods test for this and
throw an IllegalArgumentException (note that we don't throw a
NullPointerException). Perhaps the same could apply in Ant?
-Original Message-
From: Kev Jackson [mailto:[EMAIL PROTECTED]
ependencies)
> >
> > * utilize
> > something like XSLT to generate Ant scripts from simple XML files
> > that define the inter-dependencies within such a multi-part build
> >
> > * use a repository accessor (along the lines of the "get libraries"
I'm not trying to rain on anyone's parade, but have you considered...
It seems to me that there are as many completely different ways of using Ant as
there are projects in this world. If you were to supply a "standard" build
script, ready for customization, what input file structure would it ex
sedir"./" and basedir="/./" both doesnot work.
>
> So, Please let me know other solutions, if it is possible
>
> Thanks
> sudhakar
>
> -----
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
good if a little stange initially.
>
> 4) this is good as well, if one is useing elements with macrodefs a lot.
> It may
> be a bit confusing.
>
> I would say go for it!
>
> Peter
>
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
pC.java
>
> Peter
> <cid:[EMAIL PROTECTED]>
>
> >
> > Kev
> >
> > -
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
> >
> >
>
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
_
> Libero ADSL: navighi gratis a 1.2 Mega, senza canone e costi di attivazione.
> Abbonati subito su http://www.libero.it
>
>
>
> -----
> To unsubscribe, e-mail: [
; It does not work.If I add E: (the volume where all the file paths starts)does
> not work as well
> Help
>
> -- Initial Header ---
>
> From : "Phil Weighill-Smith" [EMAIL PROTECTED]
> To : "Ant Developers List" [EMAIL PROTECTED]
> Cc :
gt;
> > >
> > >
> > >
> > >
> > >> > includesfile="${file.set.toJar}" >
> > >
> > >
> > >> > valu
value="${jar.classpath.dir}"/>
>
>
>
>
>
> -- Initial Header ---
>
> From : "Phil Weighill-Smith" [EMAIL PROTECTED]
> To : "Ant Developers List" [E
__
> Libero ADSL: navighi gratis a 1.2 Mega, senza canone e costi di attivazione.
> Abbonati subito su http://www.libero.it
>
>
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
&
attribute on the
"junit" task invocation (again, it would seem sensible if this task used
JAVA_HOME to resolve the executable by default).
What do people think about the defaults for javac executable and junit
jvm?
Has anyone tried to do this sort of thing (different Ant and javac/junit
Javas)? If so, any hints or tips welcome!
Phil :n.
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
would allow any sub-part of a target (including the default
target) to be made conditional without resorting to the use of properties.
Have I missed an existing Ant capability, or would this be a worthy enhancement?
Phil Weighill-Smith
-
child mapper elements. So the answer
> is you would nest this way for a composite:
>
>
>
>
>
>
>
>
> and this way for a chained mapper:
>
>
>
>
>
>
>
>
>
> HTH,
> Matt
>
> --- Phi
mposite" and "chained" as appropriate types.
Have I missed something or is this a bug with the Mapper.MapperType
enumeration?
Phil :n.
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
Having got to this point in the conversation, it seems to me that the
documentation and the implementation do not match. So, is this a bug or should
the documentation be updated?
Phil :n.
-Original Message-
From: Alexey N. Solofnenko [mailto:[EMAIL PROTECTED]
Perhaps the regexp mapper would do this?
Phil :n(
-Original Message-
From: Dominique Devienne [mailto:[EMAIL PROTECTED]
Sent: Fri 13/08/2004 19:07
To: Ant Developers List
Cc:
Subject: RE: Bug in the "glob" file mapper?
x27;s todir +
/internal/ant.jarenv/apache-ant-1.6.2/lib.
I.e. if you don't use a "*" in the to attribute of the mapper, the "*"
match is suffixed onto the "to". This seems wrong to me. Is this a bug?
Phil :n.
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Vola
t have you tested on?
> >
>
> -----
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
r example, velocimacros.
>
> I am also thinking on Conditionals providing a subinterface that allows
> simpler glueing into the evaluation engine. So we would be able to say
> something like:
>
> if="${cond: $available.att("file",
> ${myFile}).att("
do that in
> Ant's core.
>
> Stefan
>
> -----
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
>
> -----
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
.theaimsgroup.com/?l=ant-user&m=105523577505707&w=2
> > >
> > > Peter
> > >
> > > On Thursday 09 October 2003 16:48, you wrote:
> > > > Hi Peter,
> > > >
> > > > I haven't looked at in details, but I'm wondering if it
> > >
> > > could
> > >
> > > > be adapted to implement arbitrary interfaces instead of just a task?
> > > >
> > > > With such a facility, a COBOL filename case changing mapper could be
> > > > implemented on the fly (and easily posted to the ML), and have to code
> > >
> > > in
> > >
> > > > Java a bunch of tiny classes, and suffer from the overhead associated
> > >
> > > for
> > >
> > > > most users (compiling and packaging and declaring it the build file).
> > > >
> > > > Scripted Mappers, FilterReaders, Selectors, property manipulators,
> > >
> > > etc...
> > >
> > > > would simply things tremendously, and give unlimited power to Ant the
> > >
> > > world
> > >
> > > > ;-) --DD
>
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
E: n = n.toLowerCase(); break;
> case EXTENSION: e = e.toLowerCase(); break;
> case BOTH: n = n.toLowerCase(); e = e.toLowerCase(); break;
> }
> }
> break;
> }
>
> File src = new File(srcDir, files[i]);
> File dst = new File(srcDir, n+"."+e);
>
> boolean renamed = src.renameTo(dst);
>
> if(!renamed) {
> log("source "+src+" destination "+dst+" not renamed!");
> }
> } // (int i = 0; i < files.length; i++) {
> }
> }
>
>
>
>
>
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
> .
>
>
>
>
> IncrediMail - Email has finally evolved - Click Here
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
ot; -> "
> + value, Project.MSG_DEBUG);
> userProperties.put(name, value);
>
>
>
>
> Jan
>
> -----
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
--
Phil Weighill-Smith <[EMAIL PROTECTED]>
Volantis Systems
n Thu, 2003-09-25 at 18:00, Steve Loughran wrote:
> Phil Weighill-Smith wrote:
> > This attribute would be rather handy for tools such as IntelliJ IDEA
> > which allows the graphical display of available targets for invocation
> > (within its Ant integration). It would be grea
execution. For it to do this, the "public/private" attribute would be
needed.
Phil Weighill-Smith
On Thu, 2003-09-25 at 17:02, [EMAIL PROTECTED] wrote:
> I don´t see the need for such an attribute. And if introduced it should work
> not only from commandline. It should work too
or
> > properties file.
> >
> > I can supply my example Ant/test files if you'd like. In
> > addition, I can supply a JUnit test case if you think it's
> > worth having as a new optional task.
> >
> > -----
> > Email provided by
74 matches
Mail list logo
