Sorry I'm so late to this particular discussion :-/
Why was to-byte-array left behind? It is a bit of an oddball
function, insofar as it doesn't fit into IOFactory or Coercions, but
seems to me like a natural cousin to copy.
- Chas
On May 11, 2010, at 12:16 PM, Stuart Halloway wrote:
No one spoke up for it.
Sorry I'm so late to this particular discussion :-/
Why was to-byte-array left behind? It is a bit of an oddball function,
insofar as it doesn't fit into IOFactory or Coercions, but seems to me like a
natural cousin to copy.
- Chas
On May 11, 2010, at 12:16
2010/5/11 Stuart Halloway stuart.hallo...@gmail.com
Assembla Ticket #311 [1] calls for the promotion of clojure.contrib.iointo
clojure (as
clojure.java.io). I have attached a patch, and am requesting comments and
code review from the community.
I think I have found a bug in the
On 14 May 2010 04:32, Stuart Halloway stuart.hallo...@gmail.com wrote:
On 13 May 2010 03:02, Stuart Halloway stuart.hallo...@gmail.com wrote:
* Decidedly, I have bad feelings when I read about the magic of
coercing a String first as a URL, and if not possible, fall back and
consider it a
2010/5/14 Michael Wood esiot...@gmail.com:
On 14 May 2010 04:32, Stuart Halloway stuart.hallo...@gmail.com wrote:
On 13 May 2010 03:02, Stuart Halloway stuart.hallo...@gmail.com wrote:
* Decidedly, I have bad feelings when I read about the magic of
coercing a String first as a URL, and if
So if program runs from a particular directory and references files as
file:///some/file, then if someone can create a directory called file:
in that directory with some/file inside that, the program will
suddenly try to access the wrong thing? Seems suspicious to me.
Two points:
(1) This is
On 14 May 2010 13:50, Stuart Halloway stuart.hallo...@gmail.com wrote:
So if program runs from a particular directory and references files as
file:///some/file, then if someone can create a directory called file:
in that directory with some/file inside that, the program will
suddenly try to
Hi
On 13 May 2010 03:02, Stuart Halloway stuart.hallo...@gmail.com wrote:
* Decidedly, I have bad feelings when I read about the magic of
coercing a String first as a URL, and if not possible, fall back and
consider it a local absolute/relative path. I'm mitigated in the
sense that either
2010/5/13 Stuart Halloway stuart.hallo...@gmail.com:
* I also have bad feelings when I see the Coercions protocol defined
but not used in conjunction with IOFactory for default behaviours
There are very few examples of this, not worth making changing IMO.
I'll try to do this as an exercise
On 11 May 2010 23:16, Peter Schuller peter.schul...@infidyne.com wrote:
Following symlinks in delete-file-recursively sounds like a recipe for
disaster to me.
Sorry, I should have been more explicit. The (delete-file-recursively
...) in the patch *does* follow symlinks (implicitly, because
I would like to stop following symlinks, but It doesn't appear that
the File API has any ability to detect symlinks. How were you thinking
about doing this?
Following symlinks in delete-file-recursively sounds like a recipe
for
disaster to me.
Sorry, I should have been more explicit.
On 2010-05-12, Stuart Halloway stuart.hallo...@gmail.com wrote:
I would like to stop following symlinks, but It doesn't appear that
the File API has any ability to detect symlinks.
It sounds as if Java7 will finally support dealing with symlinks via
NIO.2
I would like to stop following symlinks, but It doesn't appear that the File
API has any ability to detect symlinks. How were you thinking about doing
this?
Argh, you are right of course. I somehow believed that *testing* for
whether a path is a symlink was part of File's API, but I failed to
Using an option for symlinks raises the possibility of using options
to let delete-file handle everything, e.g.
(delete-file foo :recursive true)
instead of
(delete-file-recursively foo)
Stu
Assembla Ticket #311 [1] calls for the promotion of
clojure.contrib.io into
clojure (as
Stuart,
I think it's a great idea to ask the community for input.
I'm new to working with Assembla (and still a little new to working
with git and github.) So, if I pull down the master branch of Clojure
from http://github.com/richhickey/clojure and apply the patch file I
find on Assembla, I'll
Yes, that is right. git apply or git am should work.
Stu
Stuart,
I think it's a great idea to ask the community for input.
I'm new to working with Assembla (and still a little new to working
with git and github.) So, if I pull down the master branch of Clojure
from
Using an option for symlinks raises the possibility of using options to let
delete-file handle everything, e.g.
(delete-file foo :recursive true)
instead of
(delete-file-recursively foo)
Yes. A potential argument for not conflating them might be that
delete-file is a pretty simple
On 11 May 2010 18:37, Peter Schuller peter.schul...@infidyne.com wrote:
Assembla Ticket #311 [1] calls for the promotion of clojure.contrib.io into
clojure (as clojure.java.io). I have attached a patch, and am requesting
comments and code review from the community.
Should
Hi Stuart,
Nice piece of job !
I will nonetheless try to be constructive and give you some remarks I
made to myself while reading the patch:
* Shouldn't IOFactory protocol have a docstring to describe the
possible options ? (So that it is possible to extend the protocol even
further, e.g. when
Following symlinks in delete-file-recursively sounds like a recipe for
disaster to me.
Sorry, I should have been more explicit. The (delete-file-recursively
...) in the patch *does* follow symlinks (implicitly, because
isDirectory() returns true for symlinks pointing to diretories).
The
20 matches
Mail list logo
