ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Conor MacNeill <>
Subject Re: jar task bug?? It is..
Date Fri, 21 Dec 2001 10:31:43 GMT
Mallampati, Suresh wrote:

> Yup, there were inconsistencies, which were fixed in jdk1.3. Now, there
> arent any. 

Really? Read the Notes section of the *1.4* documentation of the Jar 
File Spec in the link Magesh has provided

     * Attributes:

       In all cases for all sections, attributes which are not 
understood are ignored.

       Attribute names are case insensitive. Programs which generate 
manifest and signature files should use the cases shown in this 
specification however.

       Attribute names cannot be repeated within a section.

That is a pretty clear. It does not distinguish between the main section 
and the unnamed sections. Sun have chosen to violate this restriction 
for Class-Path headers. They do not mention that in this spec but in 
another spec. It is not clear if they intend to relax this restriction 
for other headers. For other headers it may not make sense to simply 
concatenate them in the way you can with a Class-Path header.

Frankly the specs are a bit of a mess on this point.

> Anyways, can you please tell me if these changes will be part of
> the next version of Ant, i.e., Ant1.4.2 or something. I am not sure what
> 1.5Alpha means, will the next version be 1.5 and such..

There will not be a 1.4.2 release.

I expect there to be a 1.5 release but not for a while. It will be 
subject to a vote by Ant's committers

Ant 1.5alpha is the version that is currently the HEAD revision in CVS. 
It is built nightly and is available for download and use if you want to 
try out new features such as the one Magesh has mentioned. Please follow 
the links from the Ant home page.

> And though it might not seem important, but you said that this is a special
> treatment intended only for Class-Path attribute. Though the other main
> section attributes might not be of more length, but wont it be like "not
> following the JAR Specification" if other "main section" attributes are not
> offered the same treatment..

If you can point out the approriate section of the specification, I 
think we would be happy to follow it.

> Thanks
> -Suresh


To unsubscribe, e-mail:   <>
For additional commands, e-mail: <>

View raw message