--- Dominique Devienne [EMAIL PROTECTED] wrote:
On 5/17/07, Matt Benson [EMAIL PROTECTED]
wrote:
Example name for the task we've talked about
before
but have not written for inclusion, the
counter-argument being that the correct thing to
do
here would be write a custom task
--- Kevin Jackson [EMAIL PROTECTED] wrote:
Hi Steve,
Not sure about this. Why take a string? Its not
being resolved relative
to the project base, and is doing work that ant
should do.
This is why I wanted someone to check over my
resource stuff as I
wasn't sure it was correct.
This probably isn't your whole problem, but right off
the bat you generally need to make your inner types
static.
HTH,
Matt
--- Kevin Jackson [EMAIL PROTECTED] wrote:
Hi all,
I've just spent a hugely frustrating 2 hours or so
trying to get a
simple nested element to work within a task.
--- Steve Loughran [EMAIL PROTECTED] wrote:
I'm just prototypting resource support for
scriptdef, so it can load
scripts off the classpath, etc...all the work goes
on in ScriptHelper
One question. Should I resource enable a task by
adding support for
ResourceCollections inst
I am staggering under the weight of this (in a good
way). It looks cool. :) This is exactly the type of
thing I had hoped would just occur to us after we got
the resources stuff in place. Great work, Steve. :)
-Matt
--- [EMAIL PROTECTED] wrote:
Author: stevel
Date: Fri Apr 27 07:37:56
--- Steve Loughran [EMAIL PROTECTED] wrote:
Matt Benson wrote:
I am staggering under the weight of this (in a
good
way). It looks cool. :) This is exactly the
type of
thing I had hoped would just occur to us after we
got
the resources stuff in place. Great work, Steve
It occurs to me, Darin, that nobody would complain if
you wanted to open a new issue and immediately close
as fixed just for documentation's sake.
-Matt
--- Darin Swanson [EMAIL PROTECTED] wrote:
Thanks Kevin.
I agree that it is a regression between 1.6.5 and
1.7.0
As well there is that
--- Steve Loughran [EMAIL PROTECTED] wrote:
The fact that people are running automation
experiments on bugzilla
makes me think that now is a good time to move to
Jira. The way to do
this is we file a bugrep with infrastructure
pointing to a vote that
made the decision. I think they
--- Steve Loughran [EMAIL PROTECTED] wrote:
Matt Benson wrote:
FTR, since I had sent a note to the list
announcing my
intent to use IDEA on my MBP, I eval'd IDEA and it
informed me of random problems that didn't really
exist. I then attempted to use NetBeans but its
apparent lack
This is getting on my nerves, personally. I found
(and closed) one he had submitted for httpd as well.
I already contacted sakhatech, and was kind enough to
do so in private. Since we have heard nothing back
and Manjunath was still making trouble as of last
night, we should think on what
FTR, since I had sent a note to the list announcing my
intent to use IDEA on my MBP, I eval'd IDEA and it
informed me of random problems that didn't really
exist. I then attempted to use NetBeans but its
apparent lack of a workspace concept (feel free to
show me where I'm wrong here, anyone) was
--- Steve Loughran [EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] wrote:
AFAIK:
- Stefan : DE (Düsseldorf)
- Jan: DE (Düsseldorf)
- Antoine: US (NY)
- DD : FR (Paris)
- Steve : GB
jesse Tilly: US
Tilly?
Matt Benson: US?
Guilty--TN specifically i.e. in driving range
It's weird, I've sent requests to [EMAIL PROTECTED]
and Joe Schaefer told me he disabled ParisHilton but
she just keeps on coming!
self-pity
Why must I be responsible and withhold my snide
remarks about this subject in a public forum???
/self-pity
--- Jeffrey E Care [EMAIL PROTECTED] wrote:
Thanks, Jan! I haven't tested on 1.2--I think this
will work, and I will test if someone else doesn't
beat me to it.
Thanks,
Matt
--- [EMAIL PROTECTED] wrote:
Author: mbenson
Date: Thu Apr 12 10:20:19 2007
New Revision: 528047
URL:
http://svn.apache.org/viewvc?view=revrev=528047
Log:
Thanks, Jeff!
-Matt
--- Apache Wiki [EMAIL PROTECTED] wrote:
Dear Wiki user,
You have subscribed to a wiki page or wiki category
on Ant Wiki for change notification.
The following page has been changed by JeffCare:
http://wiki.apache.org/ant/AntTutorials
The comment on the change
--- Kevin Jackson [EMAIL PROTECTED] wrote:
Hi Matt,
I'm just getting my new Macbook Pro set up and I
wanted to give IDEA a try since we get our big
free
OSS license and all. Can anybody give me any
pointers
on how best to get set up for Ant development in
IDEA?
I've done Ant
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Fri, 23 Mar 2007, Matt Benson
[EMAIL PROTECTED] wrote:
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
I guess we'd need something dynamic, i.e. there
has to be a way to
register a ressource prefix with RU so that I can
have string
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Wed, 21 Mar 2007, Matt Benson
[EMAIL PROTECTED] wrote:
I think what is needed is a well-known syntax
specification for
specifying a string representation of
[resource-type][String to pass
to constructor] e.g. file?/foo/bar.baz,
url
Hi all--
I'm just getting my new Macbook Pro set up and I
wanted to give IDEA a try since we get our big free
OSS license and all. Can anybody give me any pointers
on how best to get set up for Ant development in IDEA?
I've done Ant dev from the command line for years
now...
TIA,
Matt
--- Steve Loughran [EMAIL PROTECTED] wrote:
Stefan Bodewig wrote:
On Wed, 21 Mar 2007, Matt Benson
[EMAIL PROTECTED] wrote:
I think what is needed is a well-known syntax
specification for
specifying a string representation of
[resource-type][String to pass
to constructor] e.g
whew, I'll try this: :)
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Wed, 21 Mar 2007, Matt Benson
[EMAIL PROTECTED] wrote:
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
We would want to have a FileRessource remain a
FileRessource even
when mapping them - same for the other
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
Sorry for dropping the ball.
Not a problem. To add another metaphor to the
conversation, my plate's been so full lately I seem to
be dropping various balls... ;)
On Tue, 6 Mar 2007, Matt Benson
[EMAIL PROTECTED] wrote:
--- Stefan Bodewig
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Fri, 2 Mar 2007, Matt Benson
[EMAIL PROTECTED] wrote:
I noticed that in order to pick the archivefileset
parameters on
e.g. tar (I'm assuming zip and family work this
way as well) the
archivefileset has to be a direct child
I noticed that in order to pick the archivefileset
parameters on e.g. tar (I'm assuming zip and family
work this way as well) the archivefileset has to be a
direct child of the archive task. This means that
specifying e.g.
resources id=archiveResources
fileset dir=foo excludes=bar /
I appear to be missing emails... :(
Yahoo! Music Unlimited
Access over 1 million songs.
http://music.yahoo.com/unlimited
-
To unsubscribe,
All right... maybe I _am_ ready to prototype this
thing (see the userlist if you're not up to speed). I
finally came up with a name I like, so maybe that's a
sign.
So what's the deal with starting a sandbox Antlib? Do
I need 3 committers when I write the proposal, or
later if/when it's ready
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Wed, 7 Feb 2007, Matt Benson
[EMAIL PROTECTED] wrote:
All right... maybe I _am_ ready to prototype this
thing (see the userlist if you're not up to
speed).
Will need to do my reading later, but ...
So what's the deal with starting
Sorry! ;)
--- [EMAIL PROTECTED] wrote:
Author: jhm
Date: Wed Jan 24 01:10:41 2007
New Revision: 499323
URL:
http://svn.apache.org/viewvc?view=revrev=499323
Log:
JDK 1.2 compatibility
Modified:
ant/core/trunk/src/main/org/apache/tools/ant/taskdefs/Concat.java
Modified:
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Fri, 12 Jan 2007, [EMAIL PROTECTED] wrote:
add FAQ about deleting directory contents while
preserving top-level
directory.
you also need to log in to people.apache.org and do
an svn up in
/www/ant.apache.org for your changes to have any
--- Kevin Jackson [EMAIL PROTECTED] wrote:
Hi Matt,
I have checked in code to fix this. Those
experiencing the problem might want to test
against a
build from SVN HEAD to verify the problem is fixed
for
them. It wouldn't hurt to remind the Maven folks
to
set the Project instance
That looks like a bug in the Maven task. The only way
for an NPE to be triggered here would seem to indicate
the task passed null as the project argument to
getDirectoryScanner(). Looking at Ant 1.6.5's code it
appears that the same exception would be thrown,
however, so I am confused how this
Sorry... to go farther down the stacktrace, I should
correct myself: it appears that the Maven guys don't
set a project on their fileset when they create it, or
something. So it's a bug for them, but possibly a
regression for Ant after all, if you want to look at
it that way.
-Matt
--- Matt
--- Peter Reilly [EMAIL PROTECTED] wrote:
It looks like it is some strange stuff with
path through the code that just avoids an NPE in
ant.1.6.5.
[SNIP]
I am not sure how to fix this in ant.
I've got it. It is indeed a case of the Maven code
not having set Project references on their
--- Peter Reilly [EMAIL PROTECTED] wrote:
On 1/3/07, Steve Loughran [EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] wrote:
+ h4a name=pathshortcutPath
Shortcut/a/h4
+ p
+In Ant 1.6 a shortcut for converting paths
to OS specific strings
+in properties has been added. One
--- Martijn Kruithof [EMAIL PROTECTED] wrote:
Have we already created an ANT_17_BRANCH?
Not that I know of. We did pretty well restraining
ourselves from committing to HEAD just before the
release. :) Should we wait on branching until nearer
to 1.7.1, or until we know we want to put
is a pain.
Regards,
Antoine
Original-Nachricht
Datum: Fri, 22 Dec 2006 22:05:12 +0100
Von: Martijn Kruithof [EMAIL PROTECTED]
An: Ant Developers List dev@ant.apache.org
Betreff: Re: Branch
Matt Benson schreef:
--- Martijn Kruithof [EMAIL PROTECTED]
wrote
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Wed, 20 Dec 2006, Steve Loughran
[EMAIL PROTECTED] wrote:
Does this also means we can go to 1.0 with
antunit?
Do we want to?
I haven't seen anybody complain about missing
features in beta2, is
AntUnit ready for a 1.0 release?
I was
--- Peter Reilly [EMAIL PROTECTED] wrote:
My comment was on the define feature in macrodef
it has been in the build for a while (~a year). It
was placed
there as an experiment, but I do not think that
people have
used it (as there is no comments, and it does not
fullfill
its usecase - it
--- Kev Jackson [EMAIL PROTECTED] wrote:
with first last, we should also consider
renaming to
head tail as suggested by (IIRC) Jan or
Antoine.
I'd prefer a head tail naming style too, it
matches functional
list programming style, but your the main dev on
this stuff it's up
--- Peter Reilly [EMAIL PROTECTED] wrote:
first and last seem fine to me (:-)
In which case, no name change and last would be a
convenience addition post 1.7.0. I guess that works.
-Matt
Peter
On 12/12/06, Matt Benson [EMAIL PROTECTED]
wrote:
--- Kev Jackson [EMAIL PROTECTED
--- Peter Reilly [EMAIL PROTECTED] wrote:
[SNIP]
Hopefully, we will figure out a solution for scratch
properties for Ant 1.8.0.
I still think you were on the right track with those
property stacks... if we can just figure out what it
was about the ThreadLocal scope branching that was
leaking
--- Peter Reilly [EMAIL PROTECTED] wrote:
On 12/12/06, Matt Benson [EMAIL PROTECTED]
wrote:
--- Peter Reilly [EMAIL PROTECTED]
wrote:
[SNIP]
Hopefully, we will figure out a solution for
scratch
properties for Ant 1.8.0.
I still think you were on the right track with
those
--- Peter Reilly [EMAIL PROTECTED] wrote:
Should the cache attribute of
BaseResourceCollectionContainer not
be defaulted to false. I fear that there may be
other similar cases -
id a collection, print a collection, change the
collection, use the collection.
Maybe it should default to
Damn it... do we think this is a blocker for 1.7.0?
Or should we leave this known hole for 1.7.1? :( I
won't be able to get back to this until Monday... and
isn't Antoine building RC2 tomorrow? :( Is it wrong
to build an RC that we already know won't be the
final?
-Matt
[SNIP]
---
--- Steve Loughran [EMAIL PROTECTED] wrote:
[SNIP]
#1 automatic proxy setup on Java 1.5
[SNIP]
Proposal: pull the auto proxy unless you say
-noproxy; add a -autoproxy
option to turn it on if you want it. Fix the
documentation coverage to
match. Users who live behind silly firewalls (me)
Currently,
globmapper id=g to=x* from=y* /
mapper refid=g /
doesn't work, because mapper expects its refid
always to point to a mapper. Should the above
example work? If so, it's a very small/quick fix.
Further, should(n't) e.g. globmapper support a refid
as well?
-Matt
Hi Prashant,
I'm still watching the video, and I agree that
VirtualAnt looks pretty cool, as did ANTLRStudio. :)
It strikes me that you must have done a lot of
explicit setup for e.g. the javac task support. I'm
sure it would be a fine line between staying generic
enough to handle whatever a
--- Peter Reilly [EMAIL PROTECTED] wrote:
On 10/14/06, Matt Benson [EMAIL PROTECTED]
wrote:
--- Peter Reilly [EMAIL PROTECTED]
wrote:
the uber-antlib could be called ant-contrib!,
I think that this task could belong to
ant-contrib,
along with some of Rainer's other classloader
--- Peter Reilly [EMAIL PROTECTED] wrote:
Opps, good catch.
Peter
That other PR reminded me of it, so I went to look in
WHATSNEW so I'd know which bug was the duplicate.
Went to check it out and it was something else
entirely, so I found the right number in SVN comments.
:)
-Matt
You mean like
copy
filterchain
expandproperties /
/filterchain
/copy
?
-Matt
--- Shekhar Yadav [EMAIL PROTECTED] wrote:
Hello,
I needed to generate files ( configuration files,
template files )
during our release process. Ant helped me in making
the process similar,
but I
--- Dominique Devienne [EMAIL PROTECTED] wrote:
[SNIP]
AFAIK, *any* task output is prefixed by the task's
name. A line
separator is output in my book ;-) I fail to see why
it would be good
to make an exception for echo/ somehow.
FWIW, I agree.
-Matt
Go ahead and fix the doc bug if you
editing a task? Do you mean writing your own? It
depends on what you want to do. Redirectors were
created to (1) make it possible to declare the I/O
handling of your task separately from the task itself
(they are ref/reusable); and (2) specifically to allow
mapped output destinations in the case
--- Antoine Levy-Lambert [EMAIL PROTECTED] wrote:
Original-Nachricht
Datum: Tue, 24 Oct 2006 09:59:05 +0100
Von: Steve Loughran [EMAIL PROTECTED]
[SNIP]
wow, you have been a busy little bee :)
Happens sometimes. Had an argument with my wife.
Compensated by workaholism.
Steve: Did this ever work for you?
I'm such a dolt, I'm like.. hmm... FileTokenizer,
what's that? Doesn't look familiar... and I'm the one
who extracted it to a public class less than a month
ago. :|
But anyway...
--- Steve Loughran [EMAIL PROTECTED] wrote:
I cant get my local copy to
--- Antoine Levy-Lambert [EMAIL PROTECTED] wrote:
Hi,
Sorry to send this long email.
Running the antunit tests under Win2K, cygwin, JDK
1.5, I get these
error failures against SVN head.
Weird. I am running on WinXP, cygwin, JDK 1.5 and I
get no errors. I can't believe Win2K vs. WinXP
--- Xavier Hanin [EMAIL PROTECTED] wrote:
On 10/20/06, Steve Loughran [EMAIL PROTECTED]
wrote:
Xavier Hanin wrote:
On 10/20/06, Steve Loughran [EMAIL PROTECTED]
wrote:
Stephane Bailliez wrote:
Antoine Levy-Lambert wrote:
Hi,
I am going to be the champion and the
--- Peter Reilly [EMAIL PROTECTED] wrote:
On 10/13/06, Matt Benson [EMAIL PROTECTED]
wrote:
As promised, I reworked Peter's task to use
ResourceCollection stuff better, made a stupid
error
when writing my tests, and forgot to come back to
it
until now. The thing works great. Pasting
As promised, I reworked Peter's task to use
ResourceCollection stuff better, made a stupid error
when writing my tests, and forgot to come back to it
until now. The thing works great. Pasting inline...
are we still thinking it's possible to put this in 1.7
? Or this thing is self-contained
--- [EMAIL PROTECTED] wrote:
Author: mbenson
Date: Thu Oct 12 08:45:53 2006
New Revision: 463287
URL:
http://svn.apache.org/viewvc?view=revrev=463287
Log:
fix failures
Sorry for the failures all; I thought I ran the tests
but I must have forgotten to copy over the new classes
first. :|
--- Peter Reilly [EMAIL PROTECTED] wrote:
I have run into a bc problem with resources.
If ant-contrib is compiled with ant1.7, and used
with
ant1.6 all the tasks that use Path
cannot be used:
C:\Documents and
Settings\reilly\learning\a\for\build.xml:3: Could
not
create type
--- Peter Reilly [EMAIL PROTECTED] wrote:
On 10/11/06, Matt Benson [EMAIL PROTECTED]
wrote:
--- Peter Reilly [EMAIL PROTECTED]
wrote:
I have run into a bc problem with resources.
If ant-contrib is compiled with ant1.7, and used
with
ant1.6 all the tasks that use Path
cannot
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Tue, 10 Oct 2006, Rick Genter
[EMAIL PROTECTED] wrote:
I think ASCII NUL is a valid character in a Unix
filename, at least
for some flavors of Unix.
While looking for something else I came accros
Mono's implementation
of System.IO.Path
When I run the antunit tests against core HEAD, I get
failures on the matcher stuff due to the
org.apache.tools.ant.util.regexp.Jdk14RegexpRegexp
class not being found. For some reason this happens
even when I set ant.regexp.matcherimpl to e.g.
org.apache.tools.ant.util.regexp.JakartaOroMatcher .
--- Matt Benson [EMAIL PROTECTED] wrote:
When I run the antunit tests against core HEAD, I
get
failures on the matcher stuff due to the
org.apache.tools.ant.util.regexp.Jdk14RegexpRegexp
class not being found. For some reason this happens
even when I set ant.regexp.matcherimpl to e.g
--- [EMAIL PROTECTED] wrote:
Author: mbenson
Date: Wed Oct 11 15:17:47 2006
New Revision: 463002
URL:
http://svn.apache.org/viewvc?view=revrev=463002
Log:
Fix failure by making each test execute in its own
Project just like we said.
To clarify, this is not addressing whatever's
--- Matt Benson [EMAIL PROTECTED] wrote:
--- Matt Benson [EMAIL PROTECTED] wrote:
When I run the antunit tests against core HEAD, I
get
failures on the matcher stuff due to the
org.apache.tools.ant.util.regexp.Jdk14RegexpRegexp
class not being found. For some reason this
happens
Jan: Yes, but files is like fileset without a
basedir, so absolute paths are required in that
context.
-Matt
--- [EMAIL PROTECTED] wrote:
Havent tested that, but would
fileset dir=somedir includesfile=fs.inc/
with fs.inc containing
/
work? On my first view that would result in
I couldn't figure out how to create a file with a NUL
in the name on a FreeBSD system. I had forgotten that
you couldn't use the alt-numpad trick on DOS-based
systems with 0. So a NUL would work on DOS, if it
would be hard to read. Can anyone verify a system on
which NUL can appear in a
Could you distill this down to a smaller example and
submit a report in Bugzilla?
Thanks,
Matt
--- Vlastimil Babka [EMAIL PROTECTED] wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
I've run into the following problem while compiling
commons-logging-1.1
with 1.7.0_beta2:
-
Ah... nice try, but pattern characters such as /, \,
*, and ? will not work either since we are discussing
includes/excludes.
Thanks,
Matt
--- Rick Genter [EMAIL PROTECTED] wrote:
/ (forward slash) is not a valid filename
character on either Unix or
Windows.
--
Rick Genter
Principal
The current state of Ant's JUnit integration seems to
crash itself when running its own test. I haven't
looked into why, but it's adorable in an ironic sort
of way...
-Matt
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection
--- Peter Reilly [EMAIL PROTECTED] wrote:
On 10/5/06, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
I propose to change the logic slightly in that
the nameless
project will set the ant.file.null property
always and not
just when there is no current project name.
What to do if you have
+1
-Matt
--- Antoine Levy-Lambert [EMAIL PROTECTED] wrote:
Hi,
I propose to build a third beta on Sunday, October
15th
the tag in Subversion would be : ANT_170_B3
the version returned by ant -version would be : ant
1.7.0Beta3
Regards,
Antoine
--- Stefan Bodewig [EMAIL PROTECTED] wrote:
On Tue, 3 Oct 2006, Matt Benson
[EMAIL PROTECTED] wrote:
The AntUnit documentation says:
Each test target is run in a fresh Ant project;
i.e.
each test target has a fresh set of properties and
references.
I think it should better
--- Dominique Devienne [EMAIL PROTECTED] wrote:
What to do if you have multiple included
projects
without names?
Exactly as it does at the moment, silently
overwrite
ant.file.null
I think it makes more sense that nameless projects
just don't get such consideration. A nameless
--- Dominique Devienne [EMAIL PROTECTED] wrote:
The user's omission of the project name
implies consent to this.
Not quite. [...]
I still don't see how this conflicts with my
earlier statement.
Where I disagree is that you assume the imported
build writer is the
same as the
Several of us seem to have been finding time for Ant
lately. :) But I personally feel like we have so
many ideas flying around that a branch seems
indicated. I am not calling for a feature freeze yet,
but I am pretty sure I am not the only one who has
stuff to work on that is probably too
Hadn't seen any mail, so...
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
JUnit 4 and TestNG have a Before/AfterClass concept
for expensive setup/tearDown applicable to most or all
of a test class. AntUnit needs something similar;
e.g. before/after[File|Build|Resource|All]... I think
it's useful to retain the before/after nomenclature
for familiarity bred from JUnit 4
Hmm... sounds reasonable. ;) -Matt
--- Peter Reilly [EMAIL PROTECTED] wrote:
before/afterTests?
Peter
On 10/3/06, Dominique Devienne [EMAIL PROTECTED]
wrote:
On 10/3/06, Matt Benson [EMAIL PROTECTED]
wrote:
JUnit 4 and TestNG have a Before/AfterClass
concept
for expensive
(minus the memory leaks) for Ant
1.7.1-1.8 , to alleviate the discomfort of having to
choose new properties for every test (see the apply
testcase in core).
Thoughts?
-Matt
--- Matt Benson [EMAIL PROTECTED] wrote:
JUnit 4 and TestNG have a Before/AfterClass concept
for expensive setup
--- Antoine Levy-Lambert [EMAIL PROTECTED] wrote:
Matt Benson wrote:
Hmm... nobody knows about (1) and nobody minds
about (2)? Going once...
-Matt
- Original Message
From: Matt Benson [EMAIL PROTECTED]
To: Ant Developers List dev@ant.apache.org
Sent: Friday, September
:) I am accustomed to other people not seeing things the way I do. The
absence of the parentheses doesn't bother me, and satisfies my, again,
pointless, desire to eliminate as many non-essential characters as possible. I
have no problem putting these back if it makes everyone else happier
Hmm... nobody knows about (1) and nobody minds about (2)? Going once...
-Matt
- Original Message
From: Matt Benson [EMAIL PROTECTED]
To: Ant Developers List dev@ant.apache.org
Sent: Friday, September 29, 2006 2:00:27 PM
Subject: nodeps jar
I'm curious why the launcher classes
FYI all, this AM I did think of another point which heavily favors
application-level resolution: we have already set a precedent for this with
our handling of the antlib: pseudo-protocol.
-Matt
- Original Message
From: Matt Benson [EMAIL PROTECTED]
To: Ant Developers List dev
AM
Subject: Re: Resource.getURL() - example
On 9/29/06, Matt Benson [EMAIL PROTECTED] wrote:
FYI all, this AM I did think of another point which heavily favors
application-level resolution: we have already set a precedent for this with
our handling of the antlib: pseudo-protocol.
Spring also
+1
-Matt
- Original Message
From: Peter Reilly [EMAIL PROTECTED]
To: Ant Developers List dev@ant.apache.org
Sent: Friday, September 29, 2006 10:23:22 AM
Subject: VOTE: deferring resolving of references
Hi, I started a thead last week that did not end in a conclusion.
should we defer
Peter said:
In ant 1.8, all conditions will be resolved by using void add(Condition
x),
so tasks that want to have nested conditions just need to extend Task and
implement add(Condition x).
Wait... how? With componentdef?
-Matt
Peter
Subject: Re: svn commit: r451000 - in /ant/core/trunk/src:
main/org/apache/tools/ant/taskdefs/
main/org/apache/tools/ant/taskdefs/condition/ tests/junit/org/apache/tools/ant/
On 9/29/06, Matt Benson [EMAIL PROTECTED] wrote:
Peter said:
In ant 1.8, all conditions will be resolved by using
I'm curious why the launcher classes are in this jar. Do they really need to
be?
Additionally, any arguments against moving (while retaining API BC) some of
the always-available optional tasks e.g. echoproperties to oata.taskdefs?
-Matt
Doh! Thanks!
-Matt
--- [EMAIL PROTECTED] wrote:
Author: kevj
Date: Wed Sep 27 20:44:19 2006
New Revision: 450679
URL:
http://svn.apache.org/viewvc?view=revrev=450679
Log:
- fix broken build - change to
ConcatResourceInputStream wasn't completed here in
Concat
Modified:
--- [EMAIL PROTECTED] wrote:
[SNIP] (RE removing parens from ternary ops)
What is the advantage apart from decreasing
readability?
Sorry, just my irrational resentment of unnecessary
bytes... we can put them back if you like and I'll try
to control myself.
-Matt
Jan
--- Scott Stirling [EMAIL PROTECTED] wrote:
On 9/28/06, Stephen McConnell [EMAIL PROTECTED]
wrote:
[...]
This difference in scope an example that
demonstrates an area where existing
protocols and software are insufficient.
OK, if you say so. In my experience the statement
existing
It strikes me that we don't have a utility to
declaratively build custom conditions from others;
macrodef would seem the obvious choice for, e.g.:
macrodef name=isEmptyFile
attribute name=file /
sequential
and
available file=@{file} type=file /
length file=@{file} length=0 /
When writing antunit tests, one cannot use the
log-inspection assertions or the new au:logcontent
resource outside the context of an AntUnit test. This
is understandable/unavoidable, but it does mean that
there are some tests that cannot be developed with
'ant -f ...'. Obviously you could write
=org.me.SimpleLang/
/jar
Do not know if it is possible.
Peter
On 9/26/06, Alexey Solofnenko [EMAIL PROTECTED]
wrote:
+1 Good idea, Matt.
- Alexey.
On 9/26/06, Matt Benson [EMAIL PROTECTED]
wrote:
It strikes me that we don't have a utility to
declaratively build custom
Hi Kev,
this all sounds like closures or other form of first
class functions to me. I know I've seen a lot of
discussion at IIRC JavaLobby etc. on whether Java
needs closures... but in any event new language
features don't help us in Ant development because we
stick to BC. I was thinking of
It was a design decision for Ant's logging system to
work in such a way that only the logger instances
themselves know their log level. This was probably
intended to specifically _prevent_ tasks having to
know too much about such things; also it means that
different attached loggers can log at
--- Dominique Devienne [EMAIL PROTECTED] wrote:
A URLResource could then be created using:
new URLResource( name, uri,
handlerClassResourceName );
The upside of this approach is the elimination of
the entire subject of
system classloader mutation, and a framework for
binding custom
--- Stephen McConnell [EMAIL PROTECTED] wrote:
-Original Message-
From: Dominique Devienne
[mailto:[EMAIL PROTECTED]
Sent: Monday, 25 September 2006 10:32 PM
To: Ant Developers List
Subject: Re: Resource.getURL()
The question is why would you need to get back
a
301 - 400 of 1052 matches
Mail list logo
