On Wed, Sep 10, 2008 at 2:58 PM, Gilles Scokart <[EMAIL PROTECTED]> wrote:
> If I take the example,
> I <-\
> / \ |
>/ \ |
> B C|
> / \ |
> / \ |
> file.txt A--/
>
> and I start with /, with the pattern /**/*.txt,
2008/9/10 Stefan Bodewig <[EMAIL PROTECTED]>:
> On Mon, 8 Sep 2008, Dominique Devienne <[EMAIL PROTECTED]> wrote:
>
>> On Mon, Sep 8, 2008 at 5:01 AM, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
>> > Does that sound OK?
>>
>> Sounds good to me.
>
> Of course we'd first check when we hit the name for
On Wed, 10 Sep 2008, Dominique Devienne <[EMAIL PROTECTED]> wrote:
> On Wed, Sep 10, 2008 at 3:20 AM, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> >> Similarly, maybe an internal flag in DS should be gettable to
> >> possibly have client code remove duplicate entries if it wants to???
> >
> > Not s
Code is in, documentation is not. Please help me find good names and
default values 8-)
On Mon, 08 Sep 2008, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> OK, combining suggestions by Gilles and Dominique here is what we
> could do:
>
> * add a new sttribute for maxSymlinkRecursionDepth (better n
On Wed, Sep 10, 2008 at 3:20 AM, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
>> Similarly, maybe an internal flag in DS should be gettable to
>> possibly have client code remove duplicate entries if it wants to???
>
> Not sure what you mean here.
Have an new accessor on DirectoryScanner to find out
On Mon, 8 Sep 2008, Dominique Devienne <[EMAIL PROTECTED]> wrote:
> On Mon, Sep 8, 2008 at 5:01 AM, Stefan Bodewig <[EMAIL PROTECTED]> wrote:
> > Does that sound OK?
>
> Sounds good to me.
Of course we'd first check when we hit the name for the (MAX + 1)th
time not for the MAXth time since the f
On Wed, 10 Sep 2008, Kevin Jackson <[EMAIL PROTECTED]> wrote:
>> Prior to the patch if you said
>>
>>
>>
>> and foo happened to be a symbolic link pointing nowhere, Ant would
>> not delete anything (and just note that there was no such file in
>> verbose mode).
>>
>> would even silently ignore i
