Stuart,
This is a significant improvement over the original str-utils library,
and goes a long way towards making "string processing kick ass in
Clojure". I like the fact that you made some design decisions for the
library, and did everything you could to stick with them. That makes
the library
Okay, I'm not sure what the correct thing do for the entire library
is, but I think I've got a convincing argument for some functions.
The following functions share a name with core functions
butlast
contains?
drop
get
partition
repeat
reverse
take
These functions should follow their correspon
On Aug 19, 2:16 pm, Sean Devlin wrote:
> First, I would change the names of functions functions that collide
> with core to str-take, str-drop, etc. It's just as much to type, and
> it is safe to use these names. Also, it would make it easier for Rich
> to promote the library to the standard li
On Aug 20, 8:26 am, Stuart Sierra wrote:
> Seems like opinion is pretty evenly divided here. I'll leave the
> library as-is for now, give it some time to see how things play out.
>
> In the mean time, as a compromise, I've added str-utils2/partial,
> which is like clojure.core/partial for functi
Seems like opinion is pretty evenly divided here. I'll leave the
library as-is for now, give it some time to see how things play out.
In the mean time, as a compromise, I've added str-utils2/partial,
which is like clojure.core/partial for functions that take their
primary argument first.
(str2/
On Wed, 2009-08-19 at 23:29 -0700, Meikel Brandmeyer wrote:
> Hi,
>
> Disclaimer: personal opinion following...
>
> I'm sorry. I don't get the elegance of point-free style.
>
> In mathematics f denotes the function, while f(x) denotes the value f
> takes over x. This is actually a nice and easy
On Aug 20, 2009, at 2:29 AM, Meikel Brandmeyer wrote:
> Hi,
>
> Disclaimer: personal opinion following...
I think that's all we have when it comes to matters of style :-)
> I'm sorry. I don't get the elegance of point-free style.
>
> In mathematics f denotes the function, while f(x) denotes the
Hi,
Disclaimer: personal opinion following...
I'm sorry. I don't get the elegance of point-free style.
In mathematics f denotes the function, while f(x) denotes the value f
takes over x. This is actually a nice and easy to understand notation.
But why do I have to clutter my clojure code with `
On Thu, Aug 20, 2009 at 12:45 AM, samppi wrote:
>
> For me, I'd like it if the core functions had the "data" as the first
> argument, but have a special function—I can't come up with a better
> name than "partial-2"—so that (partial-2 function opt1 opt2 opt3) is
> equivalent to (fn [data] (functi
For me, I'd like it if the core functions had the "data" as the first
argument, but have a special function—I can't come up with a better
name than "partial-2"—so that (partial-2 function opt1 opt2 opt3) is
equivalent to (fn [data] (function data opt1 opt2 opt3)). That way, I
could do things like
+1
On Aug 19, 11:02 pm, Stuart Sierra
wrote:
> On Aug 19, 9:56 pm, CuppoJava wrote:
>
> > If I were to have my way, I would redefine all the clojure.core
> > functions to assume the "data" is the last argument instead of the
> > first. (this includes ->) This way they would play nice with both
On Aug 19, 9:56 pm, CuppoJava wrote:
> If I were to have my way, I would redefine all the clojure.core
> functions to assume the "data" is the last argument instead of the
> first. (this includes ->) This way they would play nice with both
> partial and ->.
That's a really interesting idea. Wha
I'm also looking for a satisfactory answer to this problem.
So far I'm slightly in favor of putting the "data" (ie. the sequence/
collection/object ...) first in the argument list and the "parameters"
following.
This is because there's so many core functions that take a function
and arguments an
Hmmm... that's pretty clever. Well done.
Well, if we're gonna play golf :)
(def & comp)
(def p partial)
;;I like this because the amount of white spaces tells me something
;;Almost Pythonesque
(map (& (p map (& (p str2/drop 2)
(p str2/take 5)))
(p str2/split #"\t"))
On Aug 19, 5:16 pm, Sean Devlin wrote:
> However, over time I found this signature did not work well with my
> code. Often I would write something like this
>
> (map (comp (partial map (comp #(str2/drop % 2)
> #(str2/take % 5)))
> #(str2/split %
On Aug 19, 5:16 pm, Sean Devlin wrote:
> I suspect I am in the minority with my next concern. The library
> takes the string as the first argument, so that it works well with the
> -> macro. When I originally wrote my string library, I favored this
> type of signature too.
>
> However, over tim
On Aug 19, 3:09 pm, Howard Lewis Ship wrote:
> Have you considered splitting the str-utils2 into two namespaces, one
> that can be imported, and another that needs to be required with a
> namespace?
Hi Howard,
Hadn't thought of that, actually. There are 9 conflicts, out of 32
definitions:
tak
Have you considered splitting the str-utils2 into two namespaces, one
that can be imported, and another that needs to be required with a
namespace?
On Wed, Aug 19, 2009 at 11:22 AM, Chouser wrote:
>
> On Wed, Aug 19, 2009 at 1:59 PM, Vagif Verdi wrote:
>>
>> I'm using str-utils2 for a couple of m
On Aug 19, 2009, at 2:22 PM, Chouser wrote:
> I use (require '[clojure.contrib.str-utils2 :as str2]) for
> now and would recommend just 'str' if the lib name changes.
Except, of course, since there is already a str function, 'str' would
be a bad alias.
'strutils' or 'str-utils' sound fine t
On Wed, Aug 19, 2009 at 1:59 PM, Vagif Verdi wrote:
>
> I'm using str-utils2 for a couple of months now. Do not care about the
> old library.
Me too. I think it would be helpful to have a recommended
namespace alias to help keep different people's code a bit
more uniform.
I use (require '[cloju
I'm using str-utils2 for a couple of months now. Do not care about the
old library.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note t
21 matches
Mail list logo
