Return-Path: Mailing-List: contact ant-user-help@jakarta.apache.org; run by ezmlm Delivered-To: mailing list ant-user@jakarta.apache.org Received: (qmail 42361 invoked from network); 28 Dec 2000 20:56:30 -0000 Received: from pop.eidea.com (HELO portcullis.eidea.com) (63.83.122.149) by h29.sny.collab.net with SMTP; 28 Dec 2000 20:56:30 -0000 Received: by PORTCULLIS.eidea.com with Internet Mail Service (5.5.2650.21) id ; Thu, 28 Dec 2000 10:09:42 -0500 Message-ID: <63CE4D0F20CDD311A0F8009027DC81CF1F6F28@PORTCULLIS.eidea.com> From: Marvin Greenberg To: "'ant-user@jakarta.apache.org'" Subject: RE: uptodate and more general dependencies Date: Thu, 28 Dec 2000 10:09:37 -0500 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C070E0.347938DE" X-Spam-Rating: h29.sny.collab.net 1.6.2 0/1000/N This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C070E0.347938DE Content-Type: text/plain; charset="iso-8859-1" I'd guess that it doesn't really matter, since if you test Fred.java against Fred.class, and it is up to date, then Fred$Inner.class is up to date as well. So javac is fine. The problem with IDL and some related kinds of things is that if I don't necessarily know the name of the generated files from the name of the source file. So if I want a general rule, I need to have the target defined in terms of the directory where I know the files will be generated. -----Original Message----- From: Jason Rosenberg [mailto:jason@squaretrade.com] Sent: Thursday, December 28, 2000 6:41 AM To: ant-user@jakarta.apache.org Subject: Re: uptodate and more general dependencies An even more basic case of the one to many mapping is the case of a .java file which has local private classes declared. This will produce multiple .class files. Does the current task handle that properly? ------_=_NextPart_001_01C070E0.347938DE Content-Type: text/html; charset="iso-8859-1"

I'd guess that it doesn't really matter, since if you test Fred.java against Fred.class, and it is

up to date, then Fred$Inner.class is up to date as well. So javac is fine.

The problem with IDL and some related kinds of things is that if I don't necessarily know

the name of the generated files from the name of the source file. So if I want a general rule, I

need to have the target defined in terms of the directory where I know the files will be generated.

-----Original Message-----
From: Jason Rosenberg [mailto:jason@squaretrade.com]
Sent: Thursday, December 28, 2000 6:41 AM
To: ant-user@jakarta.apache.org
Subject: Re: uptodate and more general dependencies

An even more basic case of the one to many mapping

is the case of a .java file which has local private classes

declared. This will produce multiple .class files. Does

the current <javac> task handle that properly?

------_=_NextPart_001_01C070E0.347938DE--