AW: task submition question
I have been developing this Task that I am thinking of submitting to the ANT community. Always good for the community to get feedback and code :-) But spending to Ant is not the only possibility - see the list of external task libraries at http://ant.apache.org/external.html It would be an optional task for SCM. Which? I have a question though...my Task uses an external library. That is, it needs a librabry for a Java API from the SCM tool that I am developing the task for. The problem is, that Java API is not open source, but it is available to customers who have purchased the SCM tool. Not unusual. But two things: - license We have to be careful whether it is allowed to write an open source application (the task) against that library. For example: if the lib is under GPL then the library has to be licensed under GPL as well - which is not ok for the ASF. In that case we use reflection for using the lib´s functionality. - technically There are many optional tasks (see ANT_HOME/lib/ant-*.jar) which rely on external libs. While creating the distro (or compiling your own) the external jars have to be there. Otherwise Ants own buildfile deselects the relying classes. When not using these jars - nothing happens: only Ant´s internal classpath is longer. When using an optional jar without the underlying lib you´ll get an error message by Ant Problem: failed to create task or type ... Cause: the class ... was not found. ... Should I be emaling that jar file too? or how does it work? Decide yourself if you want to contribute to Ant or maintain it yourself - Ant: - open a bug on bugzilla and submit the code - we couldnt deliver that external lib, so dont upload that - provide a patch to docs/manual/install.html#librarydependencies describing your dependency - Yourself: - create a webpage containing the infos and downloads - personally I would only provide a link to the 3rd party lib, seems to be more secure (legally) - create a path for xdocs/external.xml and send it via eMail to [EMAIL PROTECTED] I am actually working for the company that distributes this SCM tool. I was assigned to integrate it with the ANT community. So if there anything we need to do, I would need to know. Your company could deliver the task with the SCM tool ... (like Tomcat its jsp compiler) I wasn't really sure where to send my questions... That´s ok. but I have been reading this mailing list for a bit and I figured there are a lot of smart people here who can help me. Thanks for the flowers :-) Jan
AW: cvs commit: ant/docs/manual/CoreTypes resources.html
ok - 1st version is 8days old, so no problems with BC :-)
Jan
-Ursprüngliche Nachricht-
Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Log:
Renamed file resource's base attribute to basedir,
for consistency / familiarity.
Revision ChangesPath
1.2 +15 -15
ant/src/main/org/apache/tools/ant/types/resources/FileResource.java
Index: FileResource.java
===
-public void setBase(File b) {
+public void setBaseDir(File b) {
AW: Online Bookstore
You seem to be using a slightly different form for the Powerll's link, but the right page comes up, so I suppose it's all right. The affiliate ID is correct. -Ted. Page is online at http://ant.apache.org/resources.html#books If Amazon and others do not sponsor us, we can just put a single link to www.addall.com http://www.addall.com for people to find the best price. - Alexey. I think we shouldnt do that. To many shops could start Shop X is listed, why I´m not? So only shops are there where the ASF could earn some $$. Jan
AW: Ant Logging isVerbose()?
In log4j, commons-logging, etc. a common pattern is
if (isDebugEnabled()) {
// some expensive string building to put message together
log.debug(expensiveMessage);
}
I don't see such functionality in Ant
Because Ant doesn't have a way to determine isDebugEnabled().
XmlLogger, for example, logs everything and ignores the command line
switches. So the only thing which would know it is the listeners
themselves.
Since the listener API doesn't expose the verbosity - and
changing the
interface is no good idea either - I don't see how we could do it.
I've always wondered how much overhead the verbose/debug log
info takes
up. It would make sense to determine the cost before addressing the
issue. And, as you say, the only solution is changing the
interface (or
cheating, using reflection on the side).
Or introducing a new ...
public interface BuildLogger2 extends BuildLogger {
public boolean isVerboseEnabled();
...
}
Jan
AW: Online Bookstore
No, if you add the isbn flag, the links are generated for all. If some publishers only support certain books you have to have a matrix (publisher-book) just for generating simple links. That´s too much work IMO. Who could maintain this matrix? Adding manually would be easier then. On a more enterprise you could have that matrix and a servlet which provides the vital information of that book and all supporting resellers if you call it with isbn. But that would mean much more work ... Final question: are these links ok? Then I will commit that (hopefully) today. Jan -Ursprüngliche Nachricht- Von: Ted Husted [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 26. Mai 2005 01:57 An: Ant Developers List Betreff: Re: Online Bookstore Looks good, Jan. So, aside from the retailers, some of the publishers, like O'Reilly and Manning, have affiliates programs too. Would the approach you're using let you add publisher links for just certain books. For example, the Manning link would apply to Java Development with Ant, but not the others. Not a vital feature, just wondering. The links you have are great! -Ted. On 25 May 2005 15:26:19 -, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Ok, I uploaded the example to http://www.apache.org/~jhm/bookstore/ cheers Jan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- HTH, Ted. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Online Bookstore
mmh - doesnt come through. I´ll upload that later today Jan -Ursprüngliche Nachricht- Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 25. Mai 2005 14:42 An: dev@ant.apache.org Cc: [EMAIL PROTECTED] Betreff: AW: Online Bookstore I tried a little bit ... What do you think? Ted, are that links ok? Jan docs/resources.html - generated (test) page xdocs/resources.xml (testversion, only two books) - added isbn attribute to subsection subsection name=Ant: The Definitive Guide, 2nd edition isbn=0596006098 xdocs/stylesheets/site.vsl - add´s marked with ##MAT-bookstore - new macro for inserting the link to external partners - invocation of that macro from #subsection, only if there is an isbn-attribute #macro ( subsection $subsection) ... #if ($subsection.getAttributeValue(isbn)) #set ($isbn = $subsection.getAttributeValue(isbn)) #isbn ($isbn) #end #end #macro ( isbn $isbn) pbAvailable from:/bbr a href=... /$!isbn/apachesoftwar-20/ target=_blankAmazon.com/a | a href=... amp;sourceid=41462544amp;bfpid=$!isbn target=_blankBarnes amp; Noble/a | a href=... ?affiliateId=Apacheamp;isbn=$!isbn target=_blankBook Sense/a | a href=... ?isbn=$!isbnamp;partner_id=29693 target=_blankPowells.com/a br/p #end xdocs/stylesheets/templates.vm - add´s marked with ##MAT-bookstore - if that info (isbn) is present, why not show that in the table #macro ( table $table) ... #if ($table.getParent().getAttributeValue(isbn)) #set ($isbn = $table.getParent().getAttributeValue(isbn)) trthISBN:/thtd$!isbn/td/tr #end /table #end -Ursprüngliche Nachricht- Von: Stefan Bodewig [ mailto:[EMAIL PROTECTED] mailto:[EMAIL PROTECTED] ] Gesendet am: Dienstag, 24. Mai 2005 14:46 An: dev@ant.apache.org Betreff: Re: Online Bookstore On Tue, 24 May 2005, Jan Materne [EMAIL PROTECTED] wrote: The idea is - store the ISBN in an extra tag in the xdoc thISBN:/th th id=isbn ... /th - use the VM makro to generate the link passage Ahh, thanks. +1 Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [Fwd: Some Thoughts on Ant 1.3 and 2.0]
(*) Extension tasks
Make extending Ant as simple as possible, i.e. drop a jar file in a
specific directory (ANT_HOME/ext globally or ~/.ant/ext on a user by
user basis).
This jar file must include taskdefs to define the tasks, Schema
files to describe the tasks and documentation in the XML format we
agreed on.
The Schema file is there to support GUI tools for example. They could
prompt for required attributes, let you select enumerated attributes
from a list, require integer attributes to really contain numbers ...
First is done with AntLibs and autoloading via xml-namespaces.
But I like the 3rd paragraph - but could be be done via java source?
- annotations and XDoclet/apt generates such a file
- something like
public interface ConfigurationCheck {
/** Returns all names of required attributes. */
public static String[] getRequiredAttributes();
/** Returns all names of required nested elements. */
public static String[] getRequiredElements();
/** This instance is valid, means all required values are set
correctly. */
public boolean validate();
}
How the IDEs solve that at the moment?
Jan
AW: Online Bookstore
Especially for Ant we have a page [1] with books about Ant. That page is generated via Anakia and it should be possible just to change the template (maybe minor changes on the xdoc) to get links to resellers and info site. What´s your opinion? Jan [1] http://ant.apache.org/resources.html#books -Ursprüngliche Nachricht- Von: Ted Husted [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 16. Mai 2005 01:54 An: dev@ant.apache.org Betreff: Online Bookstore The Apache Software Foundation is a not-for-profit corporation funded solely through donations from the community. To help offset the costs of ASF infrastructure, we could make it possible for ASF supporters to buy books about ASF project software through commission-program links. The commissions for such links is usually 5% or more. Given the vast number of books ASF supporters purchase, the commissions from these sales could become a respectable source of revenue. Setting up an online bookstore to benefit the ASF has been a longstanding itch of mine, which I've finally gotten a chance to scratch. A prototype bookstore is now available as a Confluence wiki. You can check it out at: * http://opensource.atlassian.com/confluence/oss/display/BOOKS/Home
AW: Online Bookstore
We keep the data for the resource page [1] in xdocs [2]. The html page is generated via Anakia stylesheet [3]. Ted provided an example [4]: at the buttom is a box Available from where are some links to several book resellers. The idea is - store the ISBN in an extra tag in the xdoc thISBN:/th th id=isbn ... /th - use the VM makro to generate the link passage Jan Example: resources.xml ---88888888- ... subsection name=Ant: The Definitive Guide, 2nd edition pPublished April 2005, and covers Ant release 1.6.1./p pThis is a complete rewrite of the first edition; this book is now 290 pages and so covers Ant in more depth than its predecessor./p pIt also mixes reference information (tables) with text explanation on how to use the tasks. Contents includes JUnit, CVS, execution, basic deployment, Web application development and XDoclet. There is also coverage of XDoclet, and a chapter on how to extend Ant in Java./p table class=externals tr thAuthor:/th tdSteve Holzner/td /tr tr thURL:/th td a href=http://www.oreilly.com/catalog/anttdg2/;http://www.oreilly.com/catalo g/anttdg2//a /td /tr tr thISBN:/th th0596006098/th /tr /table /subsection ... ---88888888- templates: TBD resources.html ---88888888- ... tr th colspan=1 rowspan=1 valign=top align=leftISBN:/th td colspan=1 rowspan=1 valign=top align=left0596006098/td /tr tr th colspan=1 rowspan=1 valign=top align=leftAvailable from:/th td colspan=1 rowspan=1 valign=top align=leftul lia href=http://www.amazon.com/exec/obidos/tg/detail/-/0596006098/apachesoftwar -20/Amazon.com/a/li lia href=http://service.bfast.com/bfast/click?bfmid=2181amp;bfmtype=bookamp;s ourceid=41462544amp;bfpid=0596006098Barnes amp; Noble/a/li lia href=http://www.booksense.com/product/info.jsp?affiliateId=Apacheamp;isbn= 0596006098Book Sense/a/li lia href=http://www.powells.com/cgi-bin/biblio?inkey=61-0596006098-0amp;partne r_id=29693Powells.com/a/li /ul/td /tr ... ---88888888- [1] http://ant.apache.org/resources.html#books [2] http://cvs.apache.org/viewcvs.cgi/*checkout*/ant/xdocs/resources.xml?content -type=text%2Fplain [3] http://cvs.apache.org/viewcvs.cgi/ant/xdocs/stylesheets/templates.vm?view=ma rkup [4] http://opensource.atlassian.com/confluence/oss/pages/viewpage.action?pageId= 231 -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 24. Mai 2005 09:24 An: dev@ant.apache.org Betreff: Re: Online Bookstore On Tue, 24 May 2005, Jan Materne [EMAIL PROTECTED] wrote: Especially for Ant we have a page [1] with books about Ant. That page is generated via Anakia and it should be possible just to change the template (maybe minor changes on the xdoc) to get links to resellers and info site. What´s your opinion? I'm not sure I understand what you want to do. What kind of links (where-to) do you want to add to which part of which page? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant/src/testcases/org/apache/tools/ant/types/opti onal/depend ClassFileSetTest.java
src/resources/org/apache/tools/ant/types/resources/selectors Do you mean, we have to rewrite our file-selectors? Jan
AW: ResourceCollections WAS Re: AW: cvs commit: ant/src/testcases /org/apache/tools/ant/types/opti onal/depend ClassFileSetTest.java
I think about that by myself :) You said that a Resource could could give Input/OutputStream. So it would be possible just to add another Cache implementiation based on that. I thought of implementing the ResourceSelector interface additionally ... Jan -Ursprüngliche Nachricht- Von: Matt Benson [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 24. Mai 2005 17:43 An: Ant Developers List Betreff: ResourceCollections WAS Re: AW: cvs commit: ant/src/testcases/org/apache/tools/ant/types/opti onal/depend ClassFileSetTest.java --- [EMAIL PROTECTED] wrote:h src/resources/org/apache/tools/ant/types/resources/selectors Do you mean, we have to rewrite our file-selectors? Any fileselector continues to work with fileset/dirset as well as files. Most fileselectors should still be relevant as resourceselectors, and in those cases it might be nice to delegate file selection to the corresponding resourceselector, just to decrease the number of places where an implementation of a check exists. I didn't attempt to create a resource version of modified because a) I don't understand it :) and b) I wasn't sure how to handle the properties file approach that exists w/ the file version. It is probably translatable, but I hoped maybe you could do it :) or help me to. -Matt Jan __ Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. http://mobile.yahoo.com/learn/mail - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [VOTE] Ant 1.6.5 release
+1 Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 23. Mai 2005 09:24 An: dev@ant.apache.org Betreff: Re: [VOTE] Ant 1.6.5 release On Mon, 23 May 2005, Antoine Levy-Lambert [EMAIL PROTECTED] wrote: I propose to release Ant 1.6.5 on Thursday, June 2nd. +1 Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [VOTE] Shut down the 1.6 branch after 1.6.5
+1 Jan -Ursprüngliche Nachricht- Von: Antoine Levy-Lambert [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 23. Mai 2005 09:50 An: Ant Developers List Betreff: Re: [VOTE] Shut down the 1.6 branch after 1.6.5 This leads me to the subject of this vote. Let's get rid of the branch, stabelize CVS HEAD and release 1.7.0-beta in a reasonable time-frame. Cheers Stefan +1 Antoine PS: I also intend to start a vote that branches shouldn't live as long as the 1.5 and 1.6 branches did but we do new releases from HEAD more quickly. This will wait until this vote has been decided. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: echoxml
http://www.oopsconsultancy.com/software/xmltask/index.html :-)
Ok, you mean a core task. I dont think that we have something else.
property echoproperties
xmlproperty ???
ATM that´s only a kind of templating the output. Works - but hasnt much
to do with xml. Some ideas ... (but from this early morning, so excuse me)
- using an xml object model
- typedef that
- instantiate an object
- write that out via jaxb
Jan
-Ursprüngliche Nachricht-
Von: Steve Loughran [mailto:[EMAIL PROTECTED]
Gesendet am: Montag, 23. Mai 2005 18:15
An: Ant Developers List
Betreff: echoxml
We need an echo task that takes well formed XML and spits it out. Is
this really the best we have? escaped cdata in an echo task?
echo file=${target.pom}![CDATA[project
modelVersion4.0.0/modelVersion
groupId${m2.groupID}/groupId
artifactId${artifact.name}/artifactId
packagingjar/packaging
version${Version}/version
/project
]]/echo
-steve
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant/docs/manual/OptionalTasks ftp.html
Little test on IE6 and Firefox 1.0.4 html body And now test html: pre lt;html style=test lt;body lt;/html /pre /body /html That works for me. You have to mask the character. After that the closing sign is ignored, because no opening sign was there. Jan -Ursprüngliche Nachricht- Von: Alexey N. Solofnenko [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 18. Mai 2005 06:05 An: Ant Developers List Betreff: Re: cvs commit: ant/docs/manual/OptionalTasks ftp.html Actually, gt; is also not required. - Alexey. Steve Cohen wrote: To be honest, I never thought about it. The previous version of the page used them and I just assumed they were required, and followed the pattern with my new examples. I didn't even assume, actually, I just followed the pattern unthinkingly. But you're quite right. The quot; are not necessary. The lt; and gt;, however, are. The source file is an html page. We aren't seriously suggesting formatting these emails, are we? To me, that makes no sense at all. This is a cvs-generated diff. Modifying it would be incorrect, making the diff unusable as a patch, which is, I guess, why these emails include them. I will, however, remove the unnecessary quot; marks. -- -- -- / Alexey N. Solofnenko home: http://trelony.cjb.net/ / - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: BugZilla Antidote
I just see that Antidote is just a valid Component in BugZilla. 9 Bug entries, all marked as fixed, invalid or wont (and wont any more after dropping the whole subproject ;) Should we delete that component? You can only delete a component in Bugzilla if you re-assign all bugs (closed or not) to a different component. Which would that be? Ok, let it there :-) Jan
AW: cvs commit: ant/xdocs external.xml
Modified:docs external.html resources.html xdocsexternal.xml Log: Roxes is now Orangevolt 1.87 +89 -13ant/docs/resources.html Index: resources.html === a name=Books/a Books /h3 -a name=Programmieren für Ant/a -Programmieren für Ant +a name=Programmieren f�r Ant/a +Programmieren f�r Ant Some characters broken? Jan
AW: cvs commit: ant/xdocs external.xml
I found only that one in the cvs mail. Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 18. Mai 2005 12:02 An: dev@ant.apache.org Betreff: Re: cvs commit: ant/xdocs external.xml On Wed, 18 May 2005, Jan Materne [EMAIL PROTECTED] wrote: Some characters broken? Looks as if Steve's editor had saved the file as UTF-8 instead of ISO-8859-1. Should be fixed now. Has there been more than this one character? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs Co pyPa th.java defaults.properties
When dealing with paths and filesets I only handle paths and simply add the fileset to a dummy path ... And from the user point of view it´s more clear... Jan -Ursprüngliche Nachricht- Von: Steve Loughran [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 18. Mai 2005 11:00 An: Ant Developers List Betreff: Re: AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs CopyPa th.java defaults.properties [EMAIL PROTECTED] wrote: Mmh, would it make more sense to add nested paths to copy? See 20635 [1]. I looked at copy, and felt it was complex enough as it was. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs CopyPa th.java defaults.properties
Added: src/main/org/apache/tools/ant/taskdefs CopyPath.java
Log:
utterly without documents or tests, a utility task to
extract a path to a dir, by way of a mapper.
Index: defaults.properties
===
+copypath=org.apache.tools.ant.taskdefs.CopyPath
Index: CopyPath.java
===
package org.apache.tools.ant.taskdefs;
/**
* Copy the contents of a path to a destination, using the mapper of
choice
* @since Ant 1.7
*
* @ant.task category=filesystem
*/
public class CopyPath extends Task {
/**
* Set the path to be used when running the Java class.
*
* @param s an Ant Path object containing the path.
*/
public void setPath(Path s) {
createPath().append(s);
}
/**
* This is a very minimal derivative of the nomal copy logic.
*
* @throws BuildException if something goes wrong with
the build.
*/
public void execute() throws BuildException {
validateAttributes();
String[] sourceFiles = path.list();
if (sourceFiles.length == 0) {
log(Path is empty, Project.MSG_VERBOSE);
return;
}
for (int sources = 0; sources sourceFiles.length;
sources++) {
String sourceFileName = sourceFiles[sources];
File sourceFile=new File(sourceFileName);
String[] toFiles = (String[])
mapper.mapFileName(sourceFileName);
for (int i = 0; i toFiles.length; i++) {
String destFileName = toFiles[i];
File destFile=new File(destDir,destFileName);
if (sourceFile.equals(destFile)) {
log(Skipping self-copy of + sourceFileName,
Project.MSG_VERBOSE);
continue;
}
try {
log(Copying + sourceFile + to + destFile,
Project.MSG_VERBOSE);
fileUtils.copyFile(sourceFile, destFile, null,
null, false,
preserveLastModified, null,
null, getProject());
} catch (IOException ioe) {
String msg = Failed to copy +
sourceFile +
to +
destFile
+ due to + ioe.getMessage();
if (destFile.exists() !destFile.delete()) {
msg += and I couldn't delete the
corrupt + destFile;
}
throw new BuildException(msg, ioe,
getLocation());
}
}
}
}
}
Mmh, would it make more sense to add nested paths to copy? See 20635
[1].
Jan
[1] http://issues.apache.org/bugzilla/show_bug.cgi?id=20635
AW: Commercial products on the related products page?
I had hopes that we'd use the wiki for this sort of stuff in the future, allowing it to be self-serve. The downside is that it's not included in the current distributions, but we could probably do some sort of get to pull wiki content into distributions as a snapshot. We had a similarly discussion earlier [1] :-) But using the Wiki (only with registration!) sounds good - but we should have a special eye on that when including into the distro. Jan [1] http://marc.theaimsgroup.com/?l=ant-devm=105952175522328w=2
BugZilla Antidote
I just see that Antidote is just a valid Component in BugZilla. 9 Bug entries, all marked as fixed, invalid or wont (and wont any more after dropping the whole subproject ;) Should we delete that component? Jan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Commercial products on the related products page?
Should be ok. License information is for each entry present. Maybe we should split externalrelated into free/commercial sections ... if there are more entries. Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Freitag, 13. Mai 2005 14:12 An: dev@ant.apache.org Betreff: Commercial products on the related products page? Hi, Slava Imeshev of Viewtier Systems sent me a private email asking whether it was OK to include links to commercial products on our related projects page. The original email - minus the actual patch and Slava's phone number - is reproduced below (with Slava's permission, of course). We already point to commercial IDEs for example, so I don't see any reason why we shouldn't do so for a buildserver, but maybe you feel different about it? Cheers Stefan From: Slava Imeshev [EMAIL PROTECTED] Subject: Patch to /ant/xdocs/projects.xml To: Stefan Bodewig [EMAIL PROTECTED] Date: Thu, 12 May 2005 23:49:35 -0700 Organization: Viewtier Systems, Inc. Hello Stefan, I was wondering if you accept commercial additions to the Ant Related Projects page at /ant/xdocs/projects.xml. If this is the case, I'd like to suggest a patch (enclosed) that adds Parabuild to /ant/xdocs/projects.xml. Parabuild is a multiplatform build automation server. Ant users are our main target audience. The URL of the product home is http://www.viewtier.com/products/parabuild/index.htm Please don't hesitate to contact me if you have any questions. Regards, Slava Imeshev President Viewtier Systems, Inc. 800 West El Camino Real, Suite 180 Mountain View, CA, 94040 - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Commercial products on the related products page?
What - the most important snippet you had forgotten??? :-O Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Freitag, 13. Mai 2005 14:20 An: dev@ant.apache.org Betreff: Re: Commercial products on the related products page? A snippet I forgot to forward 8-) Using this chance I'd like to say you guys did really good job with Ant - I think it is one of the best Java products out there. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [patch] FTP.java - adding support for new features in com mons-net 1.4.0 and performance improvement
But everybody will have a different opinion what makes up this core. copy? You bet. chmod? For those RPM builders probably yes. war? _I_ don't think so. Why chmod - works only on *nix ;-) Jan
AW: [VOTE] Ant 1.6.4 release
+1 Jan - would be a birthday present for my girl friend :-) -Ursprüngliche Nachricht- Von: Antoine Levy-Lambert [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 10. Mai 2005 08:49 An: Ant Developers List Betreff: [VOTE] Ant 1.6.4 release Do we want to release ant 1.6.4 on Thursday, May 19th (this would at least suit Eclipse) [ ] Yes [ ] No Cheers, Antoine - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant build.xml
Betreff: cvs commit: ant build.xml
bodewig 2005/05/10 06:57:48
Modified:.build.xml
Log:
Add SHA1 checksums, prepare stuff for
http://www.apache.org/dist/java-repository/
+copy todir=${java-repository.dir}
+ fileset dir=${dist.name}/lib
+include name=ant*.jar/
+ /fileset
+ mapper type=glob from=*.jar to=*-${version}.jar/
+/copy
Wouldnt it be better to have the version in front of the name to get all
files together?
ant-1.6.3.jar
ant-1.6.4.jar
ant-jsch-1.6.3.jar
ant-jsch-1.6.4.jar
ant-optional-1.6.3.jar
ant-optional-1.6.4.jar
1.6.3-ant.jar
1.6.3-ant-jsch.jar
1.6.3-ant-optional.jar
1.6.4-ant.jar
1.6.4-ant-jsch.jar
1.6.4-ant-optional.jar
better would another directory ...
1.6.3\ant.jar, ...
1.6.4\ant.jar, ...
Just my 2ct
Jan
AW: cvs commit: ant build.xml
The structure my patch creates is the structure the existing repo expects. Oki-doki. Facts overtrump. Jan
AW: Environment and PropertySet
I knew there were a discussion about that ... http://marc.theaimsgroup.com/?l=ant-userm=106278198725781w=2 Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 10. Mai 2005 16:36 An: dev@ant.apache.org Betreff: Re: Environment and PropertySet On Mon, 9 May 2005, Matt Benson [EMAIL PROTECTED] wrote: Variable can accept a nested PropertySet OR have its key value attributes set as usual I wonder about the feeling of the dev community on such a change... ? Works for me, although we better use a different name than propertyset in the user-visible part. variableset? I've never been good at names, but using property in the context of environment variables feels wrong. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [Fwd: [ANNOUNCEMENT] Commons-Net 1.4.0 Released]
I admit I know next to nothing about antlib, and in fact it was nothing until I googled it just now. If the code for such an ftp task were made to reside in such a structure (rather than in commons-net itself) there would be no circularity. However, unless there is a general move to decouple ant, which I would support on general principles, I think these changes belong on the main stem of ant going forward. There has been more than one request/bug report concerning these features, and these requests should be granted now that there is finally a means of doing so. For the time being, Neeme Praks' contributions or something like them fill the gap until the next release. These are not subsidiary features of commons-net and they should not be subsidiary features of the ant ftp task. There is no general decouple-strategy yet, but - as sometimes said - with SVN and AntLibs we have the infrastructure for more fine grained development. E.g. we could make the ftp to it´s own AntLib and grant access to that AntLib to the commons-net-team ... Jan
AW: [patch] FTP.java - adding support for new features in com mons-net 1.4.0 and performance improvement
Neeme Praks wrote: Ok, commons-net 1.4.0 has been released now. How can we proceed? Since there is apparently an Ant 1.6.4 version coming out on May 19th and since Neeme Praks has already submitted a patch to accomodate it, why not try to get this into that release? As the author of the commons-net code that Mr. Praks is relying on, and an Ant committer, I would be happy to take a look at his code and sponsor it for the 1.6.4 release. I realize that there may be other deadlines here. What do other Ant committers think? Steve Cohen If the tests pass - why not? :-) Jan
AW: PropertySet.iterator()
Title: AW: PropertySet.iterator() I had a project where a code generator inserts one interface statement - and that what missing. So I created just an empty interface to make the compiler happy. Couldnt we backport the Iterable interface? I played a little with these... - build.xml - src\oata\Iterable.java - src\oata\MyClass.java Jan -Ursprüngliche Nachricht- Von: Dominique Devienne [mailto:[EMAIL PROTECTED]] Gesendet am: Donnerstag, 21. April 2005 00:05 An: Ant Developers List Betreff: RE: PropertySet.iterator() From: Matt Benson [mailto:[EMAIL PROTECTED]] I will take your opinion as gospel, since you are the father of PropertySet, and since you agree with me ;) Note that I can't think of any use case for iterating over the property names selected by a propertyset, but I'm sure someone somewhere sometime will find one ;-) --DD - 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]
AW: PropertySet.iterator()
Ok - inlined ...
Jan
build.xml
---88888888-
project default=compile
available property=onJDK15 classname=java.lang.Iterable/
echoonJDK15: ${onJDK15}/echo
target name=jdk15.1 if=onJDK15
echoUsing Java 5 stuff/echo
replace dir=src token=/*JDK5-backport*/ value=extends
java.lang.Iterable/
/target
target name=jdk15.2 unless=onJDK15
echoHiding Java 5 stuff/echo
replace dir=src token=extends java.lang.Iterable
value=/*JDK5-backport*//
/target
target name=compile depends=jdk15.1,jdk15.2
delete dir=classes/
mkdir dir=classes/
javac srcdir=src destdir=classes/
/target
/project
---88888888-
src\oata\Iterable.java
---88888888-
package oata;
import java.util.Iterator;
public interface Iterable /*JDK5-backport*/ {
public Iterator iterator();
}
---88888888-
src\oata\MyClass.java
---88888888-
package oata;
import java.util.Iterator;
public class MyClass implements Iterable {
public Iterator iterator() {
return null;
}
}
---88888888-
-Ursprüngliche Nachricht-
Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Gesendet am: Donnerstag, 21. April 2005 07:58
An: [EMAIL PROTECTED]
Betreff: AW: PropertySet.iterator()
I had a project where a code generator inserts one interface statement - and
that
what missing. So I created just an empty interface to make the compiler
happy. Couldnt
we backport the Iterable interface?
I played a little with these...
- build.xml
- src\oata\Iterable.java
- src\oata\MyClass.java
Jan
-Ursprüngliche Nachricht-
Von: Dominique Devienne [mailto:[EMAIL PROTECTED]
Gesendet am: Donnerstag, 21. April 2005 00:05
An: Ant Developers List
Betreff: RE: PropertySet.iterator()
From: Matt Benson [mailto:[EMAIL PROTECTED]
I will take your opinion as gospel, since you are the
father of PropertySet, and since you agree with me ;)
Note that I can't think of any use case for iterating
over the property names selected by a propertyset, but
I'm sure someone somewhere sometime will find one ;-) --DD
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
AW: AW: PropertySet.iterator()
Compile under 1.4 and you only have a comment. The oata.Iterable defines the same behaviour so external tools could use - reflection for 1.4 compiled code - direct access for 1.5 compiled code Jan -Ursprüngliche Nachricht- Von: Peter Reilly [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 21. April 2005 12:31 An: Ant Developers List Betreff: Re: AW: PropertySet.iterator() [EMAIL PROTECTED] wrote: I had a project where a code generator inserts one interface statement - and that what missing. So I created just an empty interface to make the compiler happy. Couldnt we backport the Iterable interface? I played a little with these... - build.xml - src\oata\Iterable.java - src\oata\MyClass.java Nice, but not useful in ant core (kernel) until we move to using java 1.5+ as we would not be able to use java 1.5+ to create an ant distribution that can be used for java 1.4 and previous. Peter Jan -Ursprüngliche Nachricht- Von: Dominique Devienne [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 21. April 2005 00:05 An: Ant Developers List Betreff: RE: PropertySet.iterator() From: Matt Benson [mailto:[EMAIL PROTECTED] I will take your opinion as gospel, since you are the father of PropertySet, and since you agree with me ;) Note that I can't think of any use case for iterating over the property names selected by a propertyset, but I'm sure someone somewhere sometime will find one ;-) --DD - 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] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Antlib package names
We shouldnt plan the name only for the svn-package. That´s a general question for future extensions. Common prefix org.apache.ant seems to be approved. Old stuff keeps under org.apache.tools.ant for BWC. Jan -Ursprüngliche Nachricht- Von: Peter Reilly [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 20. April 2005 11:09 An: Ant Developers List Betreff: Re: Antlib package names Dominique Devienne wrote: From: Stefan Bodewig [mailto:[EMAIL PROTECTED] for the svn tasks I've used org.apache.tools.ant.taskdefs.svn, for the newer antunit I've used org.apache.ant.antlibs.antunit. I'd like to drop the tools for antlibs and I think antlibs should be somewhere in the package names, so I'd turn the former into org.apache.ant.antlibs.svn. Any other ideas, preferences? I'm all for dropping tools. But I don't like the sounds of ant.antlibs. I just don't like the repetition. I personally prefer either ant.libs or even just org.apache.antlibs, although the later being a new top-level name is probably not appropriate. No strong feeling really, but reading ant.antlibs is painful to me somehow ;-) --DD Perhaps it could be just org.apache.ant.svn - i.e. drop antlibs. Peter - 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]
AW: BuildException wrapping in macrodef
No - loaded from an AntLib a macro should behave as a usual task. So it doesnt matter in what language the task was written :-) Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 19. April 2005 17:38 An: [EMAIL PROTECTED] Betreff: BuildException wrapping in macrodef Hi, using project xmlns:au=antlib:org.apache.ant.antlib.antunit au:assertEquals expected=bar actual=baz/ /project I get /tmp/au.xml:2: The following error occurred while executing this line: jar:file:/home/bodewig/ASF/jakarta/ant/sandbox/antlibs/antunit /trunk/build/lib/ant-antunit.jar!/org/apache/ant/antlib/antuni t/antlib.xml:40: Expected 'bar' but was 'baz' what I really want is /tmp/au.xml:2: Expected 'bar' but was 'baz' The reason for this is that assertEquals is a macrodef and MacroInstance unconditionally wraps any BuildExecption into a new one adding the macro definition's location. For tasks provided in antlibs, the user probably doesn't care whether it is a macrodef or a plain task. Would anybody object if I added an attribute to macrodef that you could use to turn off the wrapping (default would certainly be with wrapping)? Another side effect of this is that an ExitStatusException thrown by fail nested into macrodef gets turned into a plain BuildException again. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Antlib package names
The 'tools' came from old Jakarta times I think and should be in for bwc reasons. But writing complete new libraries should fit to our project structure (top level project org.apache.ant and components antunit). So I would agree using the new package name. Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 19. April 2005 17:40 An: [EMAIL PROTECTED] Betreff: Antlib package names Hi, for the svn tasks I've used org.apache.tools.ant.taskdefs.svn, for the newer antunit I've used org.apache.ant.antlibs.antunit. I'd like to drop the tools for antlibs and I think antlibs should be somewhere in the package names, so I'd turn the former into org.apache.ant.antlibs.svn. Any other ideas, preferences? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: We have SVN
I would also prefer that cvs/svn/wiki messages not go to the dev-list, because - you could easier write your mail rules - not all people want to get these mails - not all must be hold on archives (it´s archived in the repo) Jan -Ursprüngliche Nachricht- Von: Dominique Devienne [mailto:[EMAIL PROTECTED] Gesendet am: Freitag, 15. April 2005 21:39 An: Ant Developers List Betreff: RE: We have SVN From: Stefan Bodewig [mailto:[EMAIL PROTECTED] Can we keep old [EMAIL PROTECTED] email address for SVN updates? The commit messages just spam dev build list. Actually I thought we'd be getting the CVS commit mails there ([EMAIL PROTECTED]) as well. Does the majority of us want the commit mails go to a different list than [EMAIL PROTECTED] This is a simple configuration issue at least Conor and myself can handle (I think you have to be a PMV chairman to change the mailer configuration). Wiki changes also go to ant-cvs (so there's a precedent for non-CVS stuff to go there). I would also prefer SVN commit emails to go there, as I have a rule to handle ant-cvs, but if not I can always add more rules to process [EMAIL PROTECTED] with svn commit: prefix in the header. But I kind of like to have ant-cvs separate, and when we reply to a message from ant-cvs, the message to go to [EMAIL PROTECTED] It separates the auto-generated messages from the discussions about them (or the changes they report). ant-cvs messages go to a different email folder than [EMAIL PROTECTED] for me. --DD - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [VOTE] Alexey Solofnenko
+1 Jan -Ursprüngliche Nachricht- Von: Erik Hatcher [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 7. April 2005 10:58 An: Ant Developers List Betreff: Re: [VOTE] Alexey Solofnenko +1 On Apr 6, 2005, at 1:28 AM, Conor MacNeill wrote: I would like to propose Alexey Solofnenko as an Ant committer. Alexey is a long time contributor to Ant and is also active on our user list helping people with good suggestions and ideas. here's my +1 Conor - 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]
AW: [EMAIL PROTECTED]: Project test-ant (in module ant) failed
yep - since a long long time ... have heard on the gump list that 600+ project should be infected ... (if i remember right) Jan -Ursprüngliche Nachricht- Von: Steve Loughran [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 30. März 2005 11:44 An: Ant Developers List Betreff: Re: [EMAIL PROTECTED]: Project test-ant (in module ant) failed Gump Integration Build wrote: To whom it may engage... This is an automated request, but not an unsolicited one. For more information please visit http://gump.apache.org/nagged.html, and/or contact the folk at [EMAIL PROTECTED] Project test-ant has an issue affecting its community integration. This issue affects 1 projects. The current state of this project is 'Failed', with reason 'Build Failed'. For reference only, the following projects are affected by this: - test-ant : Java based build tool Xalan? [junit] Caused by: javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found [junit] at javax.xml.transform.TransformerFactory.newInstance(Unknown Source) [junit] at org.apache.tools.ant.taskdefs.optional.junit.Xalan2Executor.i nit(Xalan2Executor.java:41) - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Validation of instance variables
Nice idea, but I think that´s a little overkill for just checking against
null.
Just checking the javadoc [1]: inherited fields are not catched by that
method. Without
having checked that, I could think that the AbstractCvsTask (I think there
is one...)
could contain some must-not-null-fields. Same could be for the Continuus
tasks.
But if we introduce that, I wouldnt name that validate because it just
checks against
null. So something like
assertNotNull(Class c)
assertNotNull(Class c, Collection optionalFields)
would be more meaningful.
That check doesnt seem to need any project relevant data - just the class
itself which is
provided as argument. We could provide that as a util class
oata.util.ClassUtils (Singleton
as FileUtils et al).
Jan
[1]
http://java.sun.com/j2se/1.4.2/docs/api/java/lang/Class.html#getDeclaredFiel
ds()
-Ursprüngliche Nachricht-
Von: Kev Jackson [mailto:[EMAIL PROTECTED]
Gesendet am: Freitag, 25. März 2005 10:01
An: Ant Developers List
Betreff: Validation of instance variables
We usually have to check to ensure that various values aren't null
before calling execute() in the tasks.
I've been thinking about this, and it seems that at the moment every
task handles it differently. Some check the values as part of
theexecute method, some have a seperate private method that
gets called
before execute. I'd like to unify this checking, just out of
cleanliness, so how about something like...
public void validate(Class c) throws BuildException {
Field[] fields = c.getDeclaredFields();
for (int i=0; ifields.length; i++) {
if (fields[i] == null) {
throw new BuildException(+fields[i].getName()+
attribute required);
}
}
}
would work fine, if all attributes were required, but some are only
optional, so
public void validate(Class c, Collection optional) throws
BuildException {
Field[] fields = c.getDeclaredFields();
for (int i=0; ifields.length; i++) {
if (fields[i] == null
!optional.contains(fields[i].getName())) {
throw new BuildException(+fields[i].getName()+
attribute required);
}
}
}
may be more appropriate. Then you could have this as a method in the
Task and simply call
super.validate(this, new ArrayList());
if all fields are required, or
super.validate(this, new ArrayList(optionalProp1, optionalProp2));
if all fields except optionalProp1 and optionalProp2 are required
This would move the vaildation up into the Task (where I
think not-null
checking makes more sense) and it makes it simple for tasks
to override
if more complex (this one can only be null if this one is
set) kind of
checks are required.
I've attached the patched Task and a (trivial) example.
Thoughts? Comments?
Kev
AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs SignJa r.java
+/** + * error string for unit test verification: [EMAIL PROTECTED] + */ public static final String ERROR_TODIR_AND_SIGNEDJAR = 'destdir' and 'signedjar' cannot both be set; IDEA is telling me that @value isnt allowed here. though I've used in other projects. Is there something obvious I am doing wrong? Never seen that - should JavaDoc include the value that field? Mmh ... let´s have a look at the javadoc manual ... http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/[EMAIL PROTECTED] Ah - yes. Comes with 1.4 - maybe you´re using 1.2 for checking our compile time minimum :-) Jan
AW: [VOTE] Retire Antidote
+1 Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 21. März 2005 17:02 An: [EMAIL PROTECTED] Betreff: [VOTE] Retire Antidote This again is a vote that needs two-third of all active committers, oh my. Since Antidote's development has been stalled for a long time now - and there doesn't seem to be a big need for an Ant GUI given the great IDE support we have - I hereby propose to retire the Antidote subproject. Actions to be taken: * announce that development has been stopped, both on Ant's website and the user list. * remove the Antidote pages from the website. * zip up the ant-antidote CVS module and offer it via archive.apache.org to interested parties. * ask infrastructure to remove the ant-antidote CVS module. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [VOTE] Retire Antidote
IDE integration is so good these days, our work is done. No - but we have to work on another area :-) Jan
AW: [VOTE] Ant 1.6.3 release candidate - new date proposed : Marc h 31st
Since several committers seemed to want a bit more time for fixes, let's vote for Ant 1.6.3 rc1 on Thursday, March 31st. (and cancel the previous vote of course) [ ] Yes [ ] No Let me begin with my +1 +1 Jan
AW: Doc spelling?
AFAIK we have no strictly rule about that. I would use customize - but I also would understand the other :-) Mmmh ... there was a thread about that ... http://marc.theaimsgroup.com/?l=ant-devm=107796038614327w=2 Jan -Ursprüngliche Nachricht- Von: Kev Jackson [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 16. März 2005 04:22 An: Ant Developers List Betreff: Doc spelling? US or British spelling? Customize/Customise? Kev - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Doc spelling?
Leave it as it is, I'd say. Open to all? Main point is that you have to know what the author meant? Jan
AW: Future of Antidote?
Nowadays most of the IDEs (all?) have an editor for Ant buildfiles. Antidote was touched 17 months ago (as I can see), for Jack Woehr´s I18N-patch. But I dont think that its much used and supported. So it would be a corpse in the repository. Same would be the Ant2-stuff. We have decided to work on 1.x, most of 2.x-thoughts are realized on 1.x-basis. Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 16. März 2005 11:58 An: [EMAIL PROTECTED] Betreff: Future of Antidote? Is there one? I'm leaning towards retiring antidote instead of moving it over to subversion. What are your thoughts? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [VOTE] Ant 1.6.3 release candidate
+1 (means yes) Jan -Ursprüngliche Nachricht- Von: Kev Jackson [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 15. März 2005 06:53 An: Ant Developers List Betreff: Re: [VOTE] Ant 1.6.3 release candidate Do we want to make a ant 1.6.3 RC1 on Wednesday, March 23d ? [x ] Yes (if we've addressed all the bugs/issues that we can) [ ] No - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: AW: [VOTE] Ant 1.6.3 release candidate
Only committers could do a binding vote :-) [1] other opinions are welcome, too. And because Antoine has started the vote, I should think about my vote [2]. Jan [1] http://ant.apache.org/bylaws.html#Decision%20Making [2] http://ant.apache.org/contributors.html -Ursprüngliche Nachricht- Von: Kev Jackson [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 15. März 2005 08:34 An: Ant Developers List Betreff: Re: AW: [VOTE] Ant 1.6.3 release candidate [EMAIL PROTECTED] wrote: +1 (means yes) I thought only committers could +1 something on a vote? Ok, +1 if I can/makes any difference Kev - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [VOTE] Ant 1.6.3 release candidate
What about direcly after easter weekend? We could place an easter egg ant -easteregg would redirect to http://java.sun.com/developer/Quizzes/misc/ant.html :-) Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 15. März 2005 13:24 An: [EMAIL PROTECTED] Betreff: Re: [VOTE] Ant 1.6.3 release candidate On Tue, 15 Mar 2005, Antoine Levy-Lambert [EMAIL PROTECTED] wrote: Do we want to make a ant 1.6.3 RC1 on Wednesday, March 23d ? I somehow managed to miss that you are *not* talking about tomorrow. Next week would work for me, I don't think I'll be allowed to code on Easter weekend. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant/src/main/org/apache/tools/ant AntClassLoader. java
Removed superfluous cast
Revision ChangesPath
1.91 +1 -1
ant/src/main/org/apache/tools/ant/AntClassLoader.java
Index: AntClassLoader.java
if (cons.length 0 cons[0] != null) {
final String[] strs = new String[NUMBER_OF_STRINGS];
try {
-cons[0].newInstance((Object[])strs);
+cons[0].newInstance(strs);
// Expecting an exception to be thrown by this
call:
// IllegalArgumentException: wrong number of
Arguments
} catch (Exception e) {
Mmh - but why it´s introduces just before?
Revision 1.91 Removed superfluous cast by jkf
Revision 1.90 deal with javac1.5 whinings by stevel
Log messages
http://cvs.apache.org/viewcvs.cgi/ant/src/main/org/apache/tools/ant/AntClass
Loader.java?rev=1.91view=log
only cast added in 1.90
http://cvs.apache.org/viewcvs.cgi/ant/src/main/org/apache/tools/ant/AntClass
Loader.java?r1=1.89r2=1.90diff_format=h
Jan
AW: how can we find which jre home is in use by ant script?
echo${java.home}/echo
You can print out each system property (see java.lang.System.getProperties)
Jan
-Ursprüngliche Nachricht-
Von: IndianAtTech [mailto:[EMAIL PROTECTED]
Gesendet am: Montag, 14. März 2005 13:16
An: Ant Developers List
Cc: user@ant.apache.org
Betreff: how can we find which jre home is in use by ant script?
Hi All,
how can we find which jre home is in use by ant script?
Thanks
Sudhakar
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
AW: [patch] mainly name-hiding
No doubt - I'm not sure that Jikes should dictate the code. Personally, I primarily use terneraries for situations such as parameter passing, where I need to pass one of two values. In general, I much prefer readable code with good variable names over smaller code with abbrv vrbl nms. You can configure Eclipse for finding name hiding, too. But I also think that WE should name our variables not just a tool, because WE have to read the code - not only the compiler. So it´s the same with if ternaries. If it´s just a choice between two static values its ok (short, readable). But you should not create cascades of them. :-) what about return (callee != null ? calle : super).handleInput(buffer, offset, length); ? I think we could change the names for getting rid of name hiding, but then choose readable names. Jan
AW: Ant SVN repo URLs
Do you envision any use for sandbox outside of the Ant libraries context? testing new tasks -- future in /lib testing complex modifications/enhancements of core -- future in /core Jan
AW: JIRA or Bugzilla?
Asking infrastructure for their preferences would be good - maybe they want to kill bugzilla as cvs :-) And IMO we could wait for starting the new project and bugtracking until they´ve solved the problems ... if it needs only a few weeks. But we should have a the same infrastructure for the whole ant project - same scm system for ant core and ant libs (svn) - same bugtracking for ant core and ant libs (whatever) Personally I´d prefer Bugzilla, because I am familiar with that. Jira would be complete new to me. Nearly the same as svn. But that´s a solvable problem. Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 10. März 2005 11:24 An: [EMAIL PROTECTED] Betreff: JIRA or Bugzilla? Hi, lurking on the infrastructure list shows that we do have stability problems with JIRA and Tomcat on issues.apache.org, so right now I'd prefer to stick with Bugzilla. Even without that I'd suggest we ask the infrastructure team what they'd prefer us to use before we make any decision. Comments? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: When to move Ant to SVN?
I would stay on cvs until 1.6 branch is closed, means 1.7 is before the door. - develop on cvs 1.6.3 - develop on cvs 1.7 until release - migrate the 1.7 release to svn - ship 1.7 from svn Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 10. März 2005 11:26 An: [EMAIL PROTECTED] Betreff: When to move Ant to SVN? The question really is when, not if. Shall we do it together with the creation of the antlibs repo or should we wait a bit longer? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Ant documentation
Not unless we move to Java1.5 attributes :) This may be a while No - I wouldnt change from 1.2 to 1.5 as minimum for Ant ;-) Jan
AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs Execut e.java
+/**
+ * This is the operation to get our environment.
+ * It is a notorious troublespot pre-Java1.5, and should be
approached
+ * with extreme cuation.
--^^
Typo!
Jan
+ * @return
+ */
private static String[] getProcEnvCommand() {
AW: [VOTE] Start a subproject for Ant libraries
+1 Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 8. März 2005 10:02 An: [EMAIL PROTECTED] Betreff: [VOTE] Start a subproject for Ant libraries Hi all, as threatened, here is the proposal. According to our bylaws it needs +1s froms two thirds of all active committers to pass. By my count that is 12 out of 18 (all PMC members listed on the contributors page plus all committers listed there plus Jose Alberto who hasn't added himself to the page yet). Even if I subtract the two PMC members who've stated they want to go to emeritus state (Costin and Magesh) two thirds would still mean more than 11. The vote runs for one week. The actual propsal (copy-paste of the Wiki page, slightly reformatted): Proposal to Create a Ant-Libraries Sub-Project in Apache Ant (0) rationale Ant itself has accumulated lots and lots of tasks over time. So many, that Ant developers have become reluctant to adding new task. Furthermore any new task in Ant would be tied to Ant's release schedule which is too slow for a thriving, fresh piece of code. The proposal allows Ant tasks and types to be developed under the Ant umbrella by Ant developers but have much shorter release cycles than Ant itself. In addition it would new committers who would have commit access to a single Ant library instead of the whole of Ant. (1) scope of the subproject The subproject shall create and maintain libraries of Ant tasks and types. Each library will be managed in the same manner as the Ant project itself, the PMC is ultimately responsible for it. Common Java libraries that only happen to provide Ant tasks as well are out of scope of the subproject. Providing the tasks or types has to be the primary goal of the library. To further this goal, the subproject shall also host a workplace for Ant committers. (1.5) interaction with other subprojects (1.5.1) the sandbox The subproject will host a SVN repository available to all Ant committers as a workplace for new Ant libraries. Before a library can have a public release it has to get promoted to the proper Ant libraries subproject. This also means it has to match the requirements of an Ant library as defined in section (4) under Guidelines below. The status of any library developed in the sandbox shall be reviewed after six months and the library gets either promoted or removed - or it has to be re-evaluated after another six months. (2) identify the initial source from which the subproject is to be populated Some Ant committers have developed tasks or libraries inside of the Ant CVS module under the proposal/sandbox directory. Committers are free to move them over to the new sandbox subproject or remove them completely. Libraries expected to move to the sandbox subproject initially are * the .NET tasks under proposal/sandbox/dotnet * the Subversion support tasks under proposal/sandbox/svn (3) identify the initial Apache resources to be created (3.1) mailing list(s) None. At least at the beginning we don't expect too much traffic and the existing mailing lists of the Ant projects will be used. (3.2) SVN repositories Create http://svn.apache.org/repos/asf/ant/ Expected are sub-directories antlibs/ | - proper/ || |- library1 ||| ||--- trunk ||--- tags ||--- branches |- library2 | | | --- trunk | --- tags | --- branches | - sandbox/ | - library1 || |--- trunk |--- tags |--- branches - library2 | --- trunk --- tags --- branches And potentially collections of all-trunks using svn:external as shown by the current Jakarta Commons structure. (3.3) Bugzilla New components under product Ant for each new library. (4) identify the initial set of committers All current Ant PMC members plus the active Ant committers who are not PMC members yet. Guidelines -- Note: * is, has, will, shall, must - required. * may, should, are encouraged - optional but recommended. (1) The primary unit of reuse and release is the Ant library. (2) The library is not a framework or a general library but a collection of Ant tasks and types. (3) Each library must have a clearly defined purpose, scope, and API. (4) Each library is treated as a product in its own right. (4.1) Each library has its own status file, release schedule, version number, QA tests, documentation, bug category, and individual JAR. (4.2) Each library must clearly specify any
AW: [VOTE] Start a subproject for Ant libraries
The infrastrucutre team asked us to migrate to svn until end of 2005 (or earlier) [1]. So the future for Ant is svn. Svn tasks from sandbox should be part of the new antlib subproject (just in process of vote for creation, this thread). So it should be fine with you :-) Jan [1] http://marc.theaimsgroup.com/?l=ant-devm=110803361106920w=2 -Ursprüngliche Nachricht- Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 8. März 2005 14:34 An: Ant Developers List Betreff: Re: [VOTE] Start a subproject for Ant libraries +1 I'd rather have only one version control system though, (and a version control system that is supported by a non sandbox part of ant). This is not meant as a no against svn, but more as ant should in that case also be migrated to svn lets say this calender year, and the svn tasks should be moved to a normal ant library before that. Martijn - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Ant SVN repo URLs (WAS: RE: [VOTE] Start a subproject for Ant libraries)
For Ant its not our decision. SVN is the repository the ASF uses in the near future. Jan -Ursprngliche Nachricht- Von: Alexey N. Solofnenko [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 8. Mrz 2005 18:04 An: Ant Developers List Betreff: Re: Ant SVN repo URLs (WAS: RE: [VOTE] Start a subproject for Ant libraries) Have you tried darcs? It is very nice and it allows non-commiters to have their own repositories. - Alexey. Dominique Devienne wrote: From: Stefan Bodewig [mailto:[EMAIL PROTECTED] (3.2) SVN repositories Create http://svn.apache.org/repos/asf/ant/ If we are going to use SVN for this, I think we should migrate Ant to SVN as well and consider where to put it in your proposed structure. Actually I consider this a first step for the migration since more committers will get their feet wet with SVN. IMHO Ant itself fits in there as http://svn.apache.org/repos/asf/ant/ant/ How about http://svn.apache.org/repos/asf/ant/core/ http://svn.apache.org/repos/asf/ant/libs/ instead? --DD - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- / Alexey N. Solofnenko home: http://trelony.cjb.net/ / - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: manual with css
I found another style:
/* first row */
table tr:first-child td {
background-color: cc;
color:00;
}
which uses a pseudo-class selector [1] of CSS 2.1. The table header in
subant
has this special colors, but the old manual (javac) has only a
tabletrtdb so
it´s difficult to select the cells. Works with Firefox 1.0, IE 6.0 ignores
that.
By searching I found two interesting resources: CSS hacks browser
detection [2] with
nice infos and links and browser support - Selectors [3] which lists the
browsesr and
their support of selectors (other features in separate tables linked there).
Jan
[1] http://www.w3.org/TR/CSS21/selector.html#first-child
[2]
http://www.webcredible.co.uk/user-friendly-resources/css/hacks-browser-detec
tion.shtml
[3]
http://www.westciv.com/style_master/academy/browser_support/selectors.html
-Ursprüngliche Nachricht-
Von: Stefan Bodewig [mailto:[EMAIL PROTECTED]
Gesendet am: Montag, 7. März 2005 13:53
An: [EMAIL PROTECTED]
Betreff: Re: manual with css
On Mon, 07 Mar 2005, Peter Reilly [EMAIL PROTECTED] wrote:
I have been looking at using a little css to make the hand-made ant
manual pages look a bit like the generated ant manual pages.
+1
I would like to add a reference to the css file for
the non-generated pages. It probablly should be
a different stylesheet file that the antmanual.css as
that one should also be used by the generated pages.
Wouldn't it be cleaner to
(1) add that to antmanual.css
(2) make the xdocs proposal reference antmanual.css
(3) make the xdocs proposal not create any bgcolor attributes or
font elements at all
?
I don't know enough about the inner workings of the xdocs proposal to
know how easy (2) and (3) would be, though.
Stefan
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
AW: AW: manual with css
I agree it would be better not to rely on modern features. But older browser
just ignore that so it would be just an extra. And IMO this is only an
iterim
because of the autogen-stuff.
Jan
-Ursprüngliche Nachricht-
Von: Peter Reilly [mailto:[EMAIL PROTECTED]
Gesendet am: Montag, 7. März 2005 15:41
An: Ant Developers List
Betreff: Re: AW: manual with css
[EMAIL PROTECTED] wrote:
I found another style:
/* first row */
table tr:first-child td {
background-color: cc;
color:00;
}
which uses a pseudo-class selector [1] of CSS 2.1.
I would rather not use stuff that needs modern versions of
browers.
The table header in
subant
has this special colors, but the old manual (javac) has only a
tabletrtdb so
it´s difficult to select the cells.
We could change the td to th for the first row (by a
script). But the
current bolding by hand should be enough.
Peter
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
AW: extra diagnosis
3. do we want this extra diagnosis in 'ant -diagnosis'? OR something like ant -diagnosis taskname to trigger a task analysis (and nothing else) i like that latter; Me too. That needs the possibility to pass args. ATM the ant -diagnostics taskname sounds to start the target taskname. -help, -version and -diagnostics quit the oata.Main.processArgs() method without any additional parameters so this wouldnt happen. But supplying a parameter for these attributes would be incosistent to the usual invocation syntax ant [options] [targetname] IMO something like ant -diagnostics -- taskname would be more consistent. And it would allow to pass additional parameters (I think DD has done a Bug for that :) Jan
AW: [Patch] modifiedselector, style, remove unused code, slightly more lazy DigestAlgorithm.getValue (now with added source code -doh!)
- removed iterator code fromModifiedSelector Why you use the for-loop instead of iterator? - lazier DigestAlgorithm.getValue (includes JUnit test) - removed unncecessary nesting in EqualsComparator // // - Instantiate the interfaces - // String className = null; String pkg = org.apache.tools.ant.types.selectors.modifiedselector; what do these lines do in ModifiedSelector.configure? Eclipse says that they're never read, and as they're method variables, not class or instance variables (ie not public), I was very tempted to delete them, but I thought I'd better ask first in case they're present to support future functionality That came more from past functionality than for future one :) While developing the modified selector I started with reflection. And in earlier steps I had hardcoded classnames for the three interfaces as default. And so I could simply shorten the text to write. My ascii editor I used there couldn´t check for unneeded variables :) Jan
AW: 1.7 release timetable features.
you know, this summer will mark the two year point since Ant 1,6.0 shipped. Which means time for Ant1.7 is nearing. What do people think is a good release schedule/timetable? Or to put it differently, what do people want in *and stable* before then? Things I must have: - libraries with security docs. - WAR/EAR lib option to flatten libraries pulled in. Makes it trivial to embed libraries defined stuff. Things i'd like - a way of embedding files into propertyset, so you can do file-based loading from inside anything that takes a property file. Or some easy way of going propertyfile file=foo required=true / into Java tasks. I wrote my own element for this for the smartfrog tasks, and it was v. useful. But being LGPL I can't paste it back in :( - a way of embedding libraries inside a classpath declaration. java classname=org.apache.axis.monitor.TCPMon classpath libraries library project=ws-axis artifact=axis version=1.2 / /libraries /classpath /java - setproxy to take a file of proxy settings; if file is absent do nothing, if present read in the props (merging in ant properties) and then set properties from there. - local properties in macros. - various contributed patches that have been on bugzilla for a while and look useful. -http tasks in sandbox to become an antlib. Should we be organised, build up a list of features and then either start doing them or scrub them from the list? -steve - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: 1.7 release timetable features.
Eclipse should is targeted for late June 2005 [1] and it would be good to release Ant 1.7 before, so that (maybe) Eclipse could ship 1.7. Would be also a good distribution :-) Also starting with a new minor version (1.6 - 1.7) we should introduce the antlib subproject, because that is a major step - 1.6 to 1.7 and starting modularization of Ant. Mmh propertyset smartfrog: if you are the author you should be able to change the license, IMO ;-) I would also like the local properties in macrodef. Jan [1] http://www.eclipse.org/eclipse/development/eclipse_project_plan_3_1.html you know, this summer will mark the two year point since Ant 1,6.0 shipped. Which means time for Ant1.7 is nearing. What do people think is a good release schedule/timetable? Or to put it differently, what do people want in *and stable* before then? Things I must have: - libraries with security docs. - WAR/EAR lib option to flatten libraries pulled in. Makes it trivial to embed libraries defined stuff. Things i'd like - a way of embedding files into propertyset, so you can do file-based loading from inside anything that takes a property file. Or some easy way of going propertyfile file=foo required=true / into Java tasks. I wrote my own element for this for the smartfrog tasks, and it was v. useful. But being LGPL I can't paste it back in :( - a way of embedding libraries inside a classpath declaration. java classname=org.apache.axis.monitor.TCPMon classpath libraries library project=ws-axis artifact=axis version=1.2 / /libraries /classpath /java - setproxy to take a file of proxy settings; if file is absent do nothing, if present read in the props (merging in ant properties) and then set properties from there. - local properties in macros. - various contributed patches that have been on bugzilla for a while and look useful. -http tasks in sandbox to become an antlib. Should we be organised, build up a list of features and then either start doing them or scrub them from the list? -steve
AW: Active PMCers and Committers?
That sub project is ok with me - one step closer to modularization. But IMO we should have a look at the AntLets [1], too. Jan [1] http://wiki.apache.org/ant/Proposals/Antlet -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 21. Februar 2005 12:30 An: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Betreff: Active PMCers and Committers? Hi all, I am about to propose a vote on the Antlib subproject proposal[1], which hasn't received much attention. Given that our bylaws state it needs a 2/3 majority of all active committers to get accepted, I'm more than a bit worried that it is going to fail if only because we'll miss the quorum. Should we clean up our roster, since there are certainly people officially listed who are no longer active? I'm CCing the PMC list since I'm unsure whether all PMC members are paying attention to the dev list ATM. Stefan Footnotes: [1] http://wiki.apache.org/ant/Proposals/AntLibSubProject - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: Active PMCers and Committers?
But IMO we should have a look at the AntLets [1], too. My main problem with this was and is, that it is not driven by the Ant developer community. Stefan Ok - no [EMAIL PROTECTED] support no future. But I think the idea behind that is not so bad. Maybe nobody knows that :-) The original author was adsl-49-176.swiftdsl.com.au ... whoever that was ... The suggestion was that additional to the Ant committers Nick Chalko and Adam Jack would work on that. If they want to volunteer, we should have the man power for that. And the AntLibs could be the first projects using the AntLets :-) Jan
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Testcase: testClone2(org.apache.tools.ant.taskdefs.CloneTest): Caused an
ERROR
expected ${ganger} = ; instead was
/home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/clon
e.xml
/home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/clone
.xml:27: expected ${ganger} = ; instead was
/home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/clon
e.xml
at org.apache.tools.ant.taskdefs.Exit.execute(Exit.java:141)
at
org.apache.tools.ant.UnknownElement.execute(UnknownElement.java:275)
at sun.reflect.GeneratedMethodAccessor21.invoke(Unknown Source)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl
.java:25)
at java.lang.reflect.Method.invoke(Method.java:324)
AW: [PROPOSAL] Antlib Subproject
(4) identify the initial set of committers All current Ant PMC members plus the active committers who are not PMC members yet. Maybe not specific enough - active ANT committers or - active APACHE committers? Jan
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Testcase: testNoFileJUnitNoFrames(org.apache.tools.ant.taskdefs.optional.junit.JUnitRe portTest): Caused an ERROR javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found /home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/optio nal/junitreport.xml:14: javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found at org.apache.tools.ant.dispatch.DispatchUtils.execute(DispatchUtils.java:114) at org.apache.tools.ant.Task.perform(Task.java:365) at org.apache.tools.ant.Target.execute(Target.java:341) at org.apache.tools.ant.Target.performTasks(Target.java:369) at org.apache.tools.ant.Project.executeSortedTargets(Project.java:1207) Something in the Gump environment? Jan
AW: IDEA for free!?
nice. Havent they recognized that Ant is a top level project? http://www.jetbrains.com/idea/docs/IDEA45Overview.pdf Jakarta Ant Those of you who depend on Jakarta Ant will find seamless integration of this powerful build tool into IntelliJ IDEA, now with full Ant 1.6 support! But personally I prefer Eclipse 1. I know that :-) 2. It´s OS licensed 3. It´s our company standard But maybe I ask for a license. It´s always good to know the other side... Jan -Ursprüngliche Nachricht- Von: Dominique Devienne [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 8. Februar 2005 23:38 An: Ant Developers List Betreff: FYI: IDEA for free!? JetBrains offers free IntelliJ IDEA licenses By ADT Staff JetBrains will announce this week that it will offer open-source developers no-cost user licenses for its IntelliJ IDEA integrated development environment for Java. The company says the offer is open to developers of qualifying open-source projects, without really specifying what that means, other than saying the project must be recognized by the open-source community. Developers taking part in the JetBrains initiative will be able to use JetBrains' tools with everything they need in one intelligent IDE, dramatically improving their productivity and quality of code, the company says. JetBrains will also provide support. Licenses will be valid for one year and will apply to all product upgrades during that period, with annual renewals required for continued use. More details on the qualification process are available at http://www.jetbrains.com/idea/opensource/application.htm.
AW: IDEA for free!?
Another thing. You are only allowed to use the software for one year. http://www.jetbrains.com/idea/opensource/license.html Upon Your acceptance of this License Agreement (the Agreement), JetBrains grants to You a non-exclusive and non-transferable 1-year license to use the Software, provided that You agree to the following ... Jan -Ursprüngliche Nachricht- Von: Dominique Devienne [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 8. Februar 2005 23:38 An: Ant Developers List Betreff: FYI: IDEA for free!? JetBrains offers free IntelliJ IDEA licenses By ADT Staff JetBrains will announce this week that it will offer open-source developers no-cost user licenses for its IntelliJ IDEA integrated development environment for Java. The company says the offer is open to developers of qualifying open-source projects, without really specifying what that means, other than saying the project must be recognized by the open-source community. Developers taking part in the JetBrains initiative will be able to use JetBrains' tools with everything they need in one intelligent IDE, dramatically improving their productivity and quality of code, the company says. JetBrains will also provide support. Licenses will be valid for one year and will apply to all product upgrades during that period, with annual renewals required for continued use. More details on the qualification process are available at http://www.jetbrains.com/idea/opensource/application.htm.
AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs/option al Rpm.java
Modified:
src/main/org/apache/tools/ant/taskdefs/optional Rpm.java
Log:
return code checking on rpm. How do we test this?
Using a mock object for rpm? Could return the value of a system parameter or
file content or env variable ...
rpm rpmBuildCommand=rpm-mock/
Because rpm runs only on *nix, the test doesnt have to be platform
independant.
Jan
/home/cvs/ant/src/main/org/apache/tools/ant/taskdefs/optional/Rpm.java,v
exe.setCommandline(toExecute.getCommandline());
try {
+int returncode = exe.execute();
+if (returncode != 0) {
+throw new BuildException(' +
+toExecute.getExecutable() +
+' failed with exit code +returncode);
+}
}
AW: AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs/op tion al Rpm.java
Or we really could rpm something up, then exec something to look at it Mmmh, what can the RPM do? 1) add files (e.g. the program files) 2) modify files (e.g. config files) 3) environemt settings (or is that 2?) So the scenario would be: - create the rpm - compare these modifications 1) checked against the content of a (unzipped) zip file 2) check if our settings are inserted 3) dito - clean problem: have to backup the files for 2) before and hope that no files else were modified. Or is there a possibility on *nix to undo an installation? Jan
AW: AW: AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdef s/op tion al Rpm.java
We cant actually install an RPM; you need to be root. But there are some rpm commands to list the contents of a repository; we could parse the output and look for strings. I would prefer such things - it doesnt change the environment ... Jan
AW: DO NOT REPLY [Bug 33354] - input task should hide passwor d entries
http://issues.apache.org/bugzilla/show_bug.cgi?id=33354 To call getpass() we'd need a JNI implementation for every platform that ant runs on. Not necessarily so! See here: http://java.sun.com/developer/technicalArticles/Security/pwordmask/ The problem is known at sun, and a fix pending, but not yet done. The article provides a fix, that albeit clumsy, that is a.) backward compatible b.) (mostly) portable. Because this might be interesting for more than one project: how about placing a class for this into commons io? I use that scenario in a java course, but there is a drawback: _sometimes_ it happens, that the erasure thread does not delete the first character. That´s ok for my course - but not for a tool like Ant. (imho) Jan
AW: cvs commit: ant/docs/manual/CoreTypes selectors.html
Should we place a warning (or note :) in the wrapper scripts if CLASSPATH is
set?
Jan
-Ursprüngliche Nachricht-
Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Gesendet am: Freitag, 14. Januar 2005 00:10
An: [EMAIL PROTECTED]
Betreff: cvs commit: ant/docs/manual/CoreTypes selectors.html
stevel 2005/01/13 15:09:33
Modified:docs/manual running.html
docs/manual/CoreTypes selectors.html
Log:
doc updates; especially command line options, why we hate
CLASSPATH and ant.file.*
I have declared a policy on what we will name new
properties. Is this a good formal policy to have?
Revision ChangesPath
1.30 +47 -13ant/docs/manual/running.html
Index: running.html
===
RCS file: /home/cvs/ant/docs/manual/running.html,v
retrieving revision 1.29
retrieving revision 1.30
diff -u -r1.29 -r1.30
--- running.html19 Nov 2004 09:07:08 - 1.29
+++ running.html13 Jan 2005 23:09:32 - 1.30
@@ -106,6 +106,8 @@
-s lt;filegt; the filesystem and use it
-nice number A niceness value for the main thread:
1 (lowest) to 10 (highest); 5 is
the default
+ -nouserlib Run ant without using the jar
files from ${user.home}/.ant/lib
+ -noclasspath Run ant without using CLASSPATH
/pre
pFor more information about code-logger/code and
code-listener/code see
@@ -122,26 +124,29 @@
${user.home}/.ant/lib. This arrangement allows the Ant
installation to be
shared by many users while still allowing each user to
deploy additional jars.
Such additional jars could be support jars for Ant's
optional tasks or jars
-containing third-party tasks to be used in the build. It
also allows the main Ant
-installation to be locked down which will please system
adminstrators.
+containing third-party tasks to be used in the build. It
also allows the main Ant installation to be locked down which
will please system adminstrators.
/p
p
Additional directories to be searched may be added by
using the -lib option.
The -lib option specifies a search path. Any jars or
classes in the directories
of the path will be added to Ant's classloader. The order
in which jars are
-added to the classpath is as follows
+added to the classpath is as follows:-
/p
ul
li-lib jars in the order specified by the -lib
elements on the command line/li
- lijars from ${user.home}/.ant/lib/li
+ lijars from ${user.home}/.ant/lib (unless -nouserlib
is set)/li
lijars from ANT_HOME/lib/li
/ul
p
Note that the CLASSPATH environment variable is passed to
Ant using a -lib
-option. Ant itself is started with a very minimalistic classpath.
+option. Ant itself is started with a very minimalistic classpath.
+Ant should work perfectly well with an empty CLASSPATH
environment variable,
+something the the -noclasspath option actually enforces.
We get many more support calls related to classpath problems
(especially quoting problems), than
+we like.
+
/p
p
@@ -213,14 +218,14 @@
/ul
h3a name=syspropsJava System Properties/a/h3
-pSome of Ants core classes ant tasks can be configured
via system properties./p
-pSo here the result of a search through the codebase.
Because system properties are
+pSome of Ant's core classes ant tasks can be configured
via system properties./p
+pHere is the result of a search through the codebase.
Because system properties are
available via Project instance, I searched for them with a
pre
grep -r -n getPropert * gt; ..\grep.txt
/pre
command. After that I filtered out the often-used but
not-so-important values (most of them
-read-only values): ipath.separator, ant.home, basedir,
user.dir, os.name, ant.file,
+read-only values): ipath.separator, ant.home, basedir,
user.dir, os.name,
line.separator, java.home, java.version, java.version,
user.home, java.class.path/ibr
And I filtered out the igetPropertyHelper/i access./p
table border=1
@@ -231,11 +236,31 @@
/tr
tr
tdcodeant.executor.class/code/td
- tdclassname; default is
org.apache.tools.ant.helper.DefaultExecutor/td
+ tdclassname; default is org. apache. tools. ant.
helper. DefaultExecutor/td
tdbSince Ant 1.6.3/b Ant will delegate Target
invocation to the
org.apache.tools.ant.Executor implementation specified here.
/td
/tr
+
+tr
+ tdcodeant.file/code/td
+ tdread only: full filename of the build file/td
+ tdThis is set to the name of the build file. In
+ a href=CoreTasks/import.html
+ lt;importgt;-ed/a files, this is set to the
containing build file.
+ /td
+/tr
+
+tr
+
AW: Property name policy (was Re: cvs commit: ant/docs/manual/Cor eTypes selectors.html)
I have declared a policy on what we will name new properties. Is this a good formal policy to have? snip/ If new properties get added (it happens), expect them to appear under the ant. and org.apache.tools.ant prefixes, unless the developers have a very good reason to use another prefix. Accordingly, please avoid using properties that begin with these prefixes, to reduce the risk that future Ant releases break your build file. another +1. we can modify propertyto give a warning when these names are used. maybe we should reserve the build. prefix as well - build.sysclasspath and CVS HEAD's build.clonevm. I wouldnt do that - I cant count the numbers of buildfiles where I have seen build.dir and build.classes ... Jan
AW: local properties
joking or a mixture? dimlet, define-my, local-var, ... or a very personal thing: the class is not typedef´d and the user can choose or we use something like the-type-which-introduces-the-local-only-available-of-properties-which-are-n ot-available-from-a-different-scope and the user uses a presetdef for getting the name ... /joking Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 11. Januar 2005 13:58 An: [EMAIL PROTECTED] Betreff: Re: local properties On Tue, 11 Jan 2005, Kev Jackson [EMAIL PROTECTED] wrote: Sounds good, but is it possible to get a different name than let ? Peter Following the VBness of let, how about dim? ;) Actually it's a Lispness - for those of us who liked the name let, at least 8-) Hmm, borrowing from Emacs lisp, defvar? define-property? Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs Length .java
Log: Oops, used setProperty() instead of setNewProperty(). Thanks to Jan for the example on the userlist reminding me of my error. ;) In former time I used the setProperty() for myself ;) Jan
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Ok, the usual testNoFileJUnitNoFrames(org.apache.tools.ant.taskdefs.optional.junit.JUnitRe portTest) Xalan error. But there are another few ... Jan Testcase: testResolveFile(org.apache.tools.ant.ProjectTest):FAILED expected:lt;C:[\]gt; but was:lt;C:[]gt; junit.framework.ComparisonFailure: expected:lt;C:[\]gt; but was:lt;C:[]gt; at junit.framework.Assert.assertEquals(Assert.java:81) at junit.framework.Assert.assertEquals(Assert.java:87) at org.apache.tools.ant.ProjectTest.testResolveFile(ProjectTest.java:81) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39 ) Testcase: testResolveFile(org.apache.tools.ant.util.FileUtilsTest): FAILED expected:lt;C:[\]gt; but was:lt;C:[]gt; junit.framework.ComparisonFailure: expected:lt;C:[\]gt; but was:lt;C:[]gt; at junit.framework.Assert.assertEquals(Assert.java:81) at junit.framework.Assert.assertEquals(Assert.java:87) at org.apache.tools.ant.util.FileUtilsTest.testResolveFile(FileUtilsTest.java:1 16) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39 ) Testcase: testNormalize(org.apache.tools.ant.util.FileUtilsTest): Caused an ERROR C: is not an absolute path C: is not an absolute path at org.apache.tools.ant.util.FileUtils.normalize(FileUtils.java:787) at org.apache.tools.ant.util.FileUtilsTest.testNormalize(FileUtilsTest.java:198 ) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39 ) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl .java:25)
AW: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs Checks um.java
Indeed, you can see directly from the code that a File is needed. But you dont know where that should point to. And without comment checkstyle will remind us until someone has the time to lock (and understand) the source for documentation. Mmh - on the other hand. Simple setters (for tasks) are documented in the manual. Jan -Ursprüngliche Nachricht- Von: Martijn Kruithof [mailto:[EMAIL PROTECTED] Gesendet am: Donnerstag, 6. Januar 2005 12:30 An: Ant Developers List Betreff: Re: cvs commit: ant/src/main/org/apache/tools/ant/taskdefs Checksum.java I understand the importance of improving the style, but shouldn't we prefer no comments (warnings stay to indicate they are missing) over this kind of comments: Kind Regards, Martijn + * @param file a codeFile/code value + * @param algorithm a codeString/code value + * @param provider a codeString/code value + * @param fileext a codeString/code value + * @param property a codeString/code value + * @param totalproperty a codeString/code value + * @param verifyProperty a codeString/code value + * @param forceOverwrite a codeboolean/code value + * @param size an codeint/code value + * @param e an codeenumerated/code value + * @param p a codeString/code value + * @throws BuildException on error + * @throws BuildException on error + * @return a codeMessageFormat/code object. + * @return an array of values. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Testcase: testNoFileJUnitNoFrames(org.apache.tools.ant.taskdefs.optional.junit.JUnitRe portTest): Caused an ERROR javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found /home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/optio nal/junitreport.xml:14: javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found at org.apache.tools.ant.dispatch.DispatchUtils.execute(DispatchUtils.java:105) at org.apache.tools.ant.Task.perform(Task.java:366) at org.apache.tools.ant.Target.execute(Target.java:341) at org.apache.tools.ant.Target.performTasks(Target.java:369) at org.apache.tools.ant.Project.executeSortedTargets(Project.java:1243) Mmh - configuration problem of Gump? Jan -Ursprüngliche Nachricht- Von: Gump Integration Build [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 4. Januar 2005 10:34 An: [EMAIL PROTECTED] Betreff: [EMAIL PROTECTED]: Project test-ant (in module ant) failed To whom it may engage... This is an automated request, but not an unsolicited one. For more information please visit http://gump.apache.org/nagged.html, and/or contact the folk at [EMAIL PROTECTED]
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Testcase: testNoFileJUnitNoFrames(org.apache.tools.ant.taskdefs.optional.junit.JUnitRe portTest): Caused an ERROR javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found /home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/optio nal/junitreport.xml:14: javax.xml.transform.TransformerFactoryConfigurationError: Provider for javax.xml.transform.TransformerFactory cannot be found at org.apache.tools.ant.dispatch.DispatchUtils.execute(DispatchUtils.java:105) at org.apache.tools.ant.Task.perform(Task.java:366) at org.apache.tools.ant.Target.execute(Target.java:341) at org.apache.tools.ant.Target.performTasks(Target.java:369) at org.apache.tools.ant.Project.executeSortedTargets(Project.java:1243) -Ursprüngliche Nachricht- Von: Gump Integration Build [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 22. Dezember 2004 04:36 An: [EMAIL PROTECTED] Betreff: [EMAIL PROTECTED]: Project test-ant (in module ant) failed
AW: Encryption?
AFAIK Ant doesnt include any encryption algorithms. It uses the standard APIs from java.security (checksum, modified) or java.util.zip (modified) package. Or they use JDK tools (signjar). Jan -Ursprüngliche Nachricht- Von: Magnus Svensson (HF/EAB) [mailto:[EMAIL PROTECTED] Gesendet am: Freitag, 17. Dezember 2004 14:29 An: '[EMAIL PROTECTED]' Betreff: Encryption? Hi, I have a question regarding encryption and export control. Does ANT include any SW encryption that restricts export (i.e. special ECCN code)? Regards, Magnus Svensson - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: PropertyHelper log message
You can change that, but then with your second suggestion
\foo\
because property names could contain blanks. And that would be
hard to read.
Mmh, the override-ignored message doesnt contain the \ neither.
Override ignored for property with blank
What do you think here? (setNewProperty())
Jan
project
property name=with blank value=with blank value/
echo message=${with blank}/
/project
-Ursprüngliche Nachricht-
Von: Matt Benson [mailto:[EMAIL PROTECTED]
Gesendet am: Dienstag, 14. Dezember 2004 21:27
An: Ant Developers List
Betreff: PropertyHelper log message
In PropertyHelper.replaceProperties() a verbose log
message is written when substitution syntax is found
that does not map to a property: Property ${foo} has
not been set. I submit that the message should read
Property foo has not been set or Property \foo\
has not been set as this is the type of misuse of ${}
notation that causes so much trouble in if/unless
attributes. I'd like to know the popular opinion on
the subject.
Thanks,
Matt
__
Do you Yahoo!?
Send a seasonal email greeting and help others. Do good.
http://celebrity.mail.yahoo.com
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
[junit] Testsuite: org.apache.tools.ant.taskdefs.TouchTest [junit] Tests run: 11, Failures: 0, Errors: 1, Time elapsed: 0.75 sec [junit] Testcase: testGoodPattern(org.apache.tools.ant.taskdefs.TouchTest): Caused an ERROR [junit] condition satisfied [junit] /home/gump/workspaces2/public/workspace/ant/src/etc/testcases/taskdefs/touch .xml:106: condition satisfied [junit] at org.apache.tools.ant.taskdefs.Exit.execute(Exit.java:141) (16 hours, 40 minutes ago) by mbenson touch gets nested mapper plus verbose, mkdirs, and pattern attributes. Jan -Ursprüngliche Nachricht- Von: Gump Integration Build [mailto:[EMAIL PROTECTED] Gesendet am: Mittwoch, 15. Dezember 2004 11:05 An: [EMAIL PROTECTED] Betreff: [EMAIL PROTECTED]: Project test-ant (in module ant) failed Project test-ant has an issue affecting its community integration. This issue affects 1 projects. The current state of this project is 'Failed', with reason 'Build Failed'. For reference only, the following projects are affected by this: - test-ant : Java based build tool Full details are available at: http://brutus.apache.org/gump/public/ant/test-ant/index.html
AW: DO NOT REPLY [Bug 32566] - OutOfMemory if filesize are great er than 7 MB
In other words: if you remove any from a please-notify-me field, add it to the CC section :) Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 14. Dezember 2004 10:46 An: [EMAIL PROTECTED] Betreff: Re: DO NOT REPLY [Bug 32566] - OutOfMemory if filesize are greater than 7 MB On Mon, 13 Dec 2004, [EMAIL PROTECTED] wrote: What|Removed |Added -- -- AssignedTo|[EMAIL PROTECTED] |[EMAIL PROTECTED] Martijn, when you do something like this, please add [EMAIL PROTECTED] to the CC list of the bug. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
I think Peter has tested that before, so is there something to do with Gump? Jan http://cvs.apache.org/viewcvs.cgi/ant/src/testcases/org/apache/tools/ant/uti l/ ClasspathUtilsTest.java - 2 days - peterreilly - with the bugzilla change 30161 it is not necessary to place Path.systemClasspat... Testcase: testOnlyOneInstance(org.apache.tools.ant.util.ClasspathUtilsTest): FAILED Should be only one and not 3 jar:file:/home/gump/workspaces2/public/workspace/ant/build/lib/ant.jar!/org/ apache/tools/ant/taskdefs/defaults.properties jar:file:/home/gump/workspaces2/public/workspace/ant/dist/lib/ant.jar!/org/a pache/tools/ant/taskdefs/defaults.properties jar:file:/home/gump/workspaces2/public/workspace/ant/bootstrap/lib/ant.jar!/ org/apache/tools/ant/taskdefs/defaults.properties junit.framework.AssertionFailedError: Should be only one and not 3 jar:file:/home/gump/workspaces2/public/workspace/ant/build/lib/ant.jar!/org/ apache/tools/ant/taskdefs/defaults.properties jar:file:/home/gump/workspaces2/public/workspace/ant/dist/lib/ant.jar!/org/a pache/tools/ant/taskdefs/defaults.properties jar:file:/home/gump/workspaces2/public/workspace/ant/bootstrap/lib/ant.jar!/ org/apache/tools/ant/taskdefs/defaults.properties at junit.framework.Assert.fail(Assert.java:47) at junit.framework.Assert.assertTrue(Assert.java:20) at org.apache.tools.ant.util.ClasspathUtilsTest.testOnlyOneInstance(ClasspathUt ilsTest.java:64) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39 )
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Hhm if I read the descriptor [1] in the right way, we have test-ant depends on project: dist-ant runs: run-tests -- dump-info,compile-tests,probe-offline - compile-tests (creates test2-antlib.jar; depends on build) - build (compiles the sources) If only a jar defines something new to CLASSPATH, here isnt any added. dist-ant depends on project: ant jar: ant.jar -- seems to be the duplicate from ant ant depends on project: bootstrap-ant jar: ant.jar -- any differences to bootstrap? I would think so, but would have to check. bootstrap-ant no dependencies calls the bootstrap shell script jar: ant.jar Please correct me, so I can start learning Gump ;) Jan [1] http://cvs.apache.org/viewcvs.cgi/gump/project/ant.xml?view=markup -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 13. Dezember 2004 10:46 An: [EMAIL PROTECTED] Betreff: Re: [EMAIL PROTECTED]: Project test-ant (in module ant) failed On Mon, 13 Dec 2004, Jan Materne [EMAIL PROTECTED] wrote: so is there something to do with Gump? Yes. test-ant has three versions of ant.jar (those of bootstrap-ant, ant and dist-ant) on the CLASSPATH. I'll skip the test in Gump and then (try to) fix Gump since I don't think there should be more than one ant.jar be present. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
Just think about why three times ant.jar: test-ant depend project=dist-ant inherit=all/ http://gump.apache.org/metadata/project.html sais For both script and ant based builds, any jars defined by the specified project as outputs are added to the CLASSPATH prior to invoking the build operation, unless you nest a noclasspath element into the depend element. therefore all jars of dist-ant are on the classpath for test-ant. -- dist-ant: jar name=lib/ant.jar id=ant/ OK inherit=all Other choices are all which will copy all dependencies [...] into the list of jars exported by this project. So we have to see at dist-ant: dist-ant depend project=ant inherit=all/ jar name=lib/ant.jar id=ant/ -- already discussed ant depend project=bootstrap-ant/ jar name=lib/ant.jar id=ant/ -- earned by inherit=all bootstrap-ant jar name=lib/ant.jar/ -- earned by standard depend behaviour (see above) But three times the ant.jar is registered with id=ant. Why all three are given to test-ant? Shouldnt the id be overriden? Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 13. Dezember 2004 11:15 An: [EMAIL PROTECTED] Betreff: Re: [EMAIL PROTECTED]: Project test-ant (in module ant) failed On Mon, 13 Dec 2004, Jan Materne [EMAIL PROTECTED] wrote: Hhm if I read the descriptor [1] in the right way, we have You do. But the dependency on dist-ant shouldn't drag along the jars from ant and bootstrap-ant. Well, yes, it does, because of the inherit=all. I'll try to fix the inheritance logic inside the descriptor, it's not Gump's fault. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
CVS-Snapshot
Is there some script not working any more? The homepage [1] has a link to cvs/snapshot [2], but there is nothing. Just for comparison ... Forrest has, httpd has, lenya has, maven not, spamassasin has, ... Jan [1] http://ant.apache.org/cvs.html [2] http://cvs.apache.org/snapshots/ant/ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: [GUMP@brutus]: Project test-ant (in module ant) failed
My search scripts meant to find Testcase: testspawn(org.apache.tools.ant.taskdefs.ExecTaskTest): FAILED log file found after spawn junit.framework.AssertionFailedError: log file found after spawn at junit.framework.Assert.fail(Assert.java:47) at junit.framework.Assert.assertTrue(Assert.java:20) at org.apache.tools.ant.taskdefs.ExecTaskTest.testspawn(ExecTaskTest.java:381) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39 ) Jan -Ursprüngliche Nachricht- Von: Gump Integration Build [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 7. Dezember 2004 03:39 An: [EMAIL PROTECTED] Betreff: [EMAIL PROTECTED]: Project test-ant (in module ant) failed Full details are available at: http://brutus.apache.org/gump/public/ant/test-ant/index.html BUILD FAILED /home/gump/workspaces2/public/workspace/ant/build.xml:1604: At least one test has failed. Total time: 7 minutes 44 seconds - To subscribe to this information via syndicated feeds: - RSS: http://brutus.apache.org/gump/public/ant/test-ant/rss.xml - Atom: http://brutus.apache.org/gump/public/ant/test-ant/atom.xml
AW: Questions about commons-launcher and ant Task for java proces s management
Just the the classes from Ant you need (maybe customize them to your needs) and jar them. So you´ll get your own launcher library. Jan -Ursprüngliche Nachricht- Von: Christopher Fitch [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 29. November 2004 17:05 An: [EMAIL PROTECTED] Betreff: Questions about commons-launcher and ant Task for java process management Hello, I asked Stefan a question about the Task class and he suggested that I ask the ant dev list: I have been digging in to the Java class in the org.apache.tools.ant.taskdefs package and I was wondering if the functionality provided in the class is available in a seperate library or if there is another library available that provides the same functionality. I need to start, monitor, and stop seperate JVMs from a another JVM and the ant Task class seems to be a great starting point but I would prefer not to include the ant libraries. I checked out the commons-launcher but it has the same set of dependencies since it extends Task. I like that these classes are focused on spawning JVMs. Is there another java API I could use for this function? Thanks! -- Christopher Fitch [EMAIL PROTECTED] To announce that there must be no criticism of the President, or that we are to stand by the President, right or wrong, is not only unpatriotic and servile, but is morally treasonable to the American public. - Teddy Roosevelt - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: proposed new condition - isfileselected
Sounds good - you make the whole selector framework accessible to the rest of tasks :) Jan -Ursprüngliche Nachricht- Von: Peter Reilly [mailto:[EMAIL PROTECTED] Gesendet am: Freitag, 26. November 2004 15:39 An: Ant Developers List Betreff: proposed new condition - isfileselected Hi, I would like to add a new condition - isfileselected. This is a condition that uses an embedded selector and checks if a file passes the selector: Example usages: isfileselected file=a.xml date datetime=06/28/2000 2:02 pm when=equal/ /isfileselected isfileselected file=a.xml or date datetime=06/28/2000 2:02 pm when=equal/ size value=100 units=M when=more/ /or /isfileselected Peter - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: cvs commit: ant/src/main/org/apache/tools/ant/types/selectors /modifiedselector ModifiedSelector.java
puh - never heard, that such a comment should be wrong ... So that´s an error in xdoclet? Jan -Ursprüngliche Nachricht- Von: Stefan Bodewig [mailto:[EMAIL PROTECTED] Gesendet am: Montag, 22. November 2004 17:53 An: [EMAIL PROTECTED] Betreff: Re: cvs commit: ant/src/main/org/apache/tools/ant/types/selectors/modifiedselector ModifiedSelector.java On 22 Nov 2004, [EMAIL PROTECTED] wrote: xdoclet complains it is illegal to have comments at the end. xdoclet lies 8-) I'm surprised it's only been in one file, though. JDEEs default template adds a comment at the end of a class/interface for example. Stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
AW: task testing style?
Some tests use ant buildfiles for their work some other are coded directly against the java implementation. I had written the tests [1] for modified selector directly against the api of that class. E.g. in doDelayUpdateTest(int) I use several Mock-objects. Jan [1] http://cvs.apache.org/viewcvs.cgi/ant/src/testcases/org/apache/tools/ant/typ es/selectors/ModifiedSelectorTest.java?view=markup -Ursprüngliche Nachricht- Von: Russell Gold [mailto:[EMAIL PROTECTED] Gesendet am: Dienstag, 16. November 2004 18:34 An: Ant Developers List Betreff: task testing style? The tests I have looked at in ant appear mostly to use a semi- functional test style: they use xml to define a task, run it, and then check some results (which may simply be the lack of an error). I am used to a more unit testing style, in which external classes or subsystems are stubbed out. For example, for my dependencies task, I want to confirm that a dependency is downloaded only if it is not already present, which I do by mocking the fetch mechanism. Is this approach being used somewhere in ant? Has there been any discussion of the two approaches to testing? - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
