Hi all,
I'm very pleased you catch mine idea and take this great discussion!
There are many places where inter-task communication would be useful.
Having
the context of what is going on around a task is very interesting. I don't
Yes! I think is is _needed_ in some cases. See requests about
Hi all,
Here are patches to enable all stuff we discussed. I haven't any test cases
and docs are missing in most places as well. If you like the patch I'll add
that!.
Now script like this:
fileset id=1
includes name=1/
includes name=3/
/fileset
fileset id=1 mode=replace
includes
I haven't looked at in detail yet Martin but I will. Somthing to
consider is that a solution that performs the merge upon the raw xml of
the multiple definitions would/should work for all datatypes without the
derived class having to implement MergeWith. Thats what we discussed
previously but
Can you add remove and intersect modes for completeness?
This way we'd have all operators found in set theory.
Examples:
Assuming f1 contains: [1,4]
mode=replace with arguments [1,2] produces [1,2]
mode=append with arguments [2,3,4] produces [1,2,3,4] (A u B)
mode=intersect with arguments
I definitely think we ought to consider using xml merging here, as the way
Martin implemented it will definitely cause problems, eg. when you change
the base directory in the redefined fileset, you can't just add the files of
redefined fileset to the original fileset definition ...
when we can
I just committed a small change to ScriptTask to allow the definition of
custom functions inside a script task instance - see example below:
This is quite nice as it means people can easily prototype new functions
without having to compile and load a dedicated task/function assembly.
Note that
Nice.
I also though about it some time ago. Wouldn't it be cleaner to write it
like the following?
script language=C# definefunctions=true namespace=test
[Function(test-func)]
public string Testfunc( ) {
return some result ;
}
/script
Most parts of the script code
that should be doable. The class would automatically have the
FunctionSet attribute added if definefunctions was true.
but what if you actually wanted to use the constructor to do somthing ?
With the longer format it would be easier to copy your function code out
when you get around to making
Can you add remove and intersect modes for completeness?
This way we'd have all operators found in set theory.
Examples:
Assuming f1 contains: [1,4]
mode=replace with arguments [1,2] produces [1,2]
mode=append with arguments [2,3,4] produces [1,2,3,4] (A u B)
mode=intersect with arguments
I'd lean the other way and require that the script be basically a full
namespace/class def. (or no function decl. at all).
script language=C#
namespace test {
[FunctionSet(test, Test)]
class myFuncs : FunctionSetBase {
[Function(test-func)]
public string Testfunc( ) {
script language=C# function=Testfunc
return some result ;
/script
That can't be done because you need the type information for formal
parameters and returned value.
Jarek
---
This SF.net email is sponsored by: Perforce
Which conflicts with what is there right now. As it there is
automatically a class and the minimum you need to add is a ScriptMain as
in the example at:
http://nant.sourceforge.net/nightly/help/tasks/script.html
forcing the user to *have* to specify class and namespace is going too
far I
It might be worth investigating what is possible with items in msbuild
and what use cases we can apply using this XPathObjectNavigator solution.
John Lam posted some of his experiences using output items with msbuild at:
http://www.iunknown.com/000379.html
Ian
Yep - each task can survive on its
Hi,
I keep getting the following message every time I try to post an error to
this forum:
You message from [EMAIL PROTECTED] to [EMAIL PROTECTED]
has been DISCARDED due to a word or phrase thought to be objectionable by our
filters. Please review the message and try again.
Are there certain
- Original Message -
From: Tim Dallmann [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Wednesday, January 07, 2004 8:49 PM
Subject: [nant-dev] can't post a problem to this list
Hi,
I keep getting the following message every time I try to post an error to
this forum:
You message
I've finally made it into the office today and was able to test out the quick
changes I made. I fixed a missing attribute and missing #endregion directive
(this is the version without the dynamic attribute).
BTW, thanks for taking the time to looks at this stuff - I really appreciate it.
James.
I see a bunch of unreadable ( maybe unicode ) characters after your
name - is your sig somthing exotic that might be causing this ?
However this message got thru so who knows ?
Ian
Tim Dallmann wrote:
Hi,
I keep getting the following message every time I try to post an error to
this forum:
James,
I've made a few changes (please let me know if it still works for you) and
committed it to NAntContrib. Would you like cvs write access to NAntContrib
repository on sf.net? :) If so, just send me your sf.net userid.
Thanks for the work :)
- Original Message -
From: James C. Papp
18 matches
Mail list logo
