a.lang.IllegalArgumentException: Don't
know how to create ISeq from: Symbol (NO_SOURCE_FILE:8)
What do I do wrong here?
Per
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To post to thi
Thanks, that makes a lot of sense!
Per
On May 20, 10:36 am, Chouser wrote:
> On Wed, May 20, 2009 at 9:59 AM, Michael Reid wrote:
>
> > On Wed, May 20, 2009 at 8:44 AM, pmf wrote:
>
> >> On May 20, 4:47 am, Per wrote:
> >>> ;; The macro
&
mbol: :name in this context
(NO_SOURCE_FILE:4)
Per
On May 20, 7:59 am, Michael Reid wrote:
> On Wed, May 20, 2009 at 8:44 AM, pmf wrote:
>
> > On May 20, 4:47 am, Per wrote:
> >> ;; The macro
> >> (defmacro def-fields [name tgs]
> >>
entries by whatever means (e.g. calls to
user-provided functions in the case of mapping), and return that.
Everything is purely functional from the exterior.
-Per
On Tue, Mar 9, 2010 at 2:57 AM, Jonathan Shore wrote:
> Hi,
> I'm still trying to work out the best way to deal with opera
By the way, I also noticed your logic is wrong. It should be (+ (* i
ncol) j) rather than (* i j).
-Per
On Tue, Mar 9, 2010 at 2:57 AM, Jonathan Shore wrote:
> Hi,
> I'm still trying to work out the best way to deal with operations on
> heterogenous data in clojure. I
e thus easy to
distinguish. An article isn't a terminal since it has multiple
alternatives.
-Per
On Wed, Mar 10, 2010 at 12:47 PM, Mike K wrote:
> In PAIP section 2.3, Norvig gives an example of a generative grammar:
>
> ;; common lisp
> (defparameter *simple-grammar*
> '((s
The way to go is definitely symlinks or hard links, not classpath
hijinks. The only problem is that Java's filesystem API has no support
for links, so you'll probably need to use OS-specific command line
calls. Not too bad though.
-Per
On Thu, Mar 11, 2010 at 11:07 AM, Brent Mill
problems.
-Per
On Thu, Mar 11, 2010 at 1:45 PM, Alex Osborne wrote:
> Brent Millare writes:
>
>> Since leiningen downloads everything to a local repo, can't we do away
>> with copies and use symlinks if they are supported by the filesystem?
>> I feel there should be
jars in its lib/ directory. A bigger project can
easily have well over a hundred megabytes of lib/ jars. Now multiply
that by the number of active clone branches (let's say a few dozen)
and you're looking at a few gigabytes of waste.
-Per
On Thu, Mar 11, 2010 at 2:26 PM, Alex Osborne w
a compile-time reduce:
([x y & more]
(reduce (fn [a b] `(+ ~a ~b)) (cons x (cons y more
-Per
--
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 that posts from new me
ed only when calls to the function are
resolvable at compile time. While nicer than plain macros (as you can
still treat the function as a first-class value at run time) they are
still not semantically transparent because of the possibility of var
rebinding.
-Per
--
You received this message because yo
loads.
-Per
On Sat, Mar 13, 2010 at 3:23 PM, Per Vognsen wrote:
> On Sat, Mar 13, 2010 at 3:03 PM, Konrad Hinsen
> wrote:
>> On 13 Mar 2010, at 03:32, Jonathan Shore wrote:
>>
>>> Could + and other operators be changed to a macro mapping rather than the
>>> cur
That implementation of partitions feels really low level. If you
implement the monadic version of partition-when (which I call
partition-where in my own code), it looks as simple as this:
(defn partitions [xs]
(run-seq (m-partition-where (const [false true]) xs)))
-Per
On Wed, Mar 17, 2010 at
Is there any reason why a .method occurrence in non-operator position
doesn't just do the closure wrapping automagically?
-Per
On Thu, Mar 18, 2010 at 9:50 PM, Stuart Halloway
wrote:
> memfn is from the depths of time and should be deprecated -- it is idiomatic
> to write an anonymou
s))
(map (with-args (.getRGB image 10 10)) [.getRed .getGreen .getBlue])
-Per
On Fri, Mar 19, 2010 at 4:30 AM, Michael Gardner wrote:
> On Mar 18, 2010, at 10:55 AM, Per Vognsen wrote:
>
>> Is there any reason why a .method occurrence in non-operator position
>> doesn
My guess from looking at the API documentation for ResultSet is that
it doesn't know which of the several one-parameter overloads of
getObject to choose. Presumably you want the integer one, so try
(.getObject rs (int i)).
-Per
On Fri, Mar 19, 2010 at 12:08 PM, cageface wrote:
> I
al objects in nursery).
In a Platonic paradise, the cache could even be allocated and freed
without any GC involvement whatosever due to the magic of HotSpot's
escape analysis. But that seldom works as widely in practice as one
might hope and I expect it wouldn't in this example.
-Per
On
On Fri, Mar 19, 2010 at 2:46 PM, Konrad Hinsen
wrote:
> On 18 Mar 2010, at 16:55, Per Vognsen wrote:
>
>> Is there any reason why a .method occurrence in non-operator position
>> doesn't just do the closure wrapping automagically?
>
> There is two reasons I can think
I don't think passing symbols around and having special case behavior
scattered around different functions is going to help beginners. If
anything, it would confuse them when they try to treat .methods as
first-class functions in their own code and discover that it doesn't
work.
-Per
O
is to the `(do ...) block in deftype's
implementation:
(defn ~(symbol (str name "?")) [x#]
(instance? ~classname x#))
It would also be nice if the class itself was exposed under some
standardized name, e.g. (symbol (str ~name "-class"))). Any reason
th
Thanks, I had seen that ::Foo use and was a bit confused. Now it all
makes sense.
It would still be nice to have an auto-generated name?-style predicate
in deftype, I think.
-Per
On Fri, Mar 19, 2010 at 7:44 PM, Meikel Brandmeyer wrote:
> Hi,
>
> On Mar 19, 12:09 pm, Per Vogns
It's pretty common in the Lisp world at large. For example, Common
Lisp's (defstruct foo ...) automatically defines a foo-p predicate.
-Per
On Fri, Mar 19, 2010 at 7:53 PM, Konrad Hinsen
wrote:
> On 19.03.2010, at 13:50, Per Vognsen wrote:
>
>> It would still be nice to
strength of the language compared
to, say, Haskell.
-Per
On Thu, Mar 18, 2010 at 3:13 AM, David Nolen wrote:
> On Wed, Mar 17, 2010 at 4:09 PM, Konrad Hinsen
> wrote:
>>
>> On 17 Mar 2010, at 20:54, David Nolen wrote:
>>
>>> But seriously, in my personal opinion Mo
Learn to love scan: http://gist.github.com/338682
-Per
On Sat, Mar 20, 2010 at 12:13 PM, Douglas Philips wrote:
> Hello all,
> I'm new to clojure, but not lisp.
> I'm looking for a functional way to index/number only some items of a
> list.
>
> For example, I
Aha! I Googled for scan in seq-utils and didn't find anything. It
would be nice if people stuck to standard terminology that has a
continuous history going back to the early 60s.
-Per
On Sat, Mar 20, 2010 at 9:40 PM, Steve Purcell wrote:
> Which looks the same as clojure.contrib.seq/re
ns. I find
recognizability and memorability is usually a more important metric.
Anyway, I don't want to derail this thread any further. :)
-Per
On Sat, Mar 20, 2010 at 10:16 PM, Meikel Brandmeyer wrote:
> Hi,
>
> On Sat, Mar 20, 2010 at 09:50:12PM +0700, Per Vognsen wrote:
>&g
One last thing:
On Sat, Mar 20, 2010 at 10:16 PM, Meikel Brandmeyer wrote:
> Hi,
>
> On Sat, Mar 20, 2010 at 09:50:12PM +0700, Per Vognsen wrote:
>> Aha! I Googled for scan in seq-utils and didn't find anything. It
>> would be nice if people stuck to standard terminolog
hierarchy. Maybe someone more familiar with deftype
can figure this one out. For now, what I have is good enough to get on
with my immediate coding.
-Per
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email
On Mon, Mar 22, 2010 at 1:43 PM, Meikel Brandmeyer wrote:
> Hi,
>
> On Mar 22, 7:13 am, Per Vognsen wrote:
>
>> As a solution, I factored the reify out into a deftype:
>>
>> http://gist.github.com/339834
>
> A short note: you don't have to use :keyword
tool here.
-Per
On Mon, Mar 22, 2010 at 5:45 PM, alux wrote:
> Hi Andrzej,
>
> I'm not a Petri net specialist too, but I dont see how one could
> simulate the view of a Clojure programmer onto STM, without simulation
> too the stuff programmers doesnt see: Richs under-the
Java developer-oriented application with a good
out-of-box experience on all platforms. Shell scripts for UNIX and
installers for Windows and OS X would go a long way towards improving
that for Clojure.
-Per
On Mon, Mar 22, 2010 at 6:31 PM, Luc Préfontaine
wrote:
> Is my first impression right
Note that I didn't propose an installer except for OS X and Windows.
Only a DWIM shell script.
-Per
On Mon, Mar 22, 2010 at 7:04 PM, Ramakrishnan Muthukrishnan
wrote:
> On Mon, Mar 22, 2010 at 5:13 PM, Per Vognsen wrote:
>> good example of a Java developer-oriented applicati
ult implementations in terms of the core protocol
functions via mixins. This is how Scala's Seq trait is designed:
http://www.scala-lang.org/docu/files/api/scala/Seq.html
-Per
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this
ull off this code transformation
entirely in-language with CL-style compiler macros while leaving alone
references to protocol-function in non-operator position.
-Per
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group
ining an :expected-protocol field and
comparing its value to protocol-function's protocol.
-Per
On Mon, Mar 22, 2010 at 11:24 PM, Per Vognsen wrote:
>>>> lookup (find-protocol-method) over the method invocation itself. Is that
>>>> the
>>>> right
jure reads 0xFF as
the _signed_ integer 255. I would expect it to be -1. For example,
(every? #(= % (bit-and -1 %)) (range -128 128)) evaluates to true.
This looks like a genuine bug.
-Per
On Tue, Mar 23, 2010 at 7:08 PM, Konrad Hinsen
wrote:
> On 23.03.2010, at 08:55, Raph wrote:
>
>>
Sorry, I didn't put that right. 0xFF would only be -1 as a signed
byte. What I'm saying is that the interaction between the type system
of integers and the reader's hexadecimal notation is pretty surprising
to me. In particular, (byte 0xFF) throws an error.
-Per
On Tue, Mar 23,
coll (reductions #(if
(pred %2) (rest %1) %1) src coll)))
-Per
On Tue, Mar 23, 2010 at 6:53 PM, Douglas Philips wrote:
> Last week, Per Vognsen answered my first version of this question, how can I
> number just the vowels in a string:
> http://groups.google.com/group/clojure/msg/22186113b36041f1
and consider whether you're sure you really need
references. You haven't supplied enough context for us to make that
call.
-Per
On Tue, Mar 23, 2010 at 8:35 PM, WoodHacker wrote:
> I understand how conj works. But how do you add a value to a
> persistent vector? You have to
Interesting. It's Clojure 1.2.0-master-SNAPSHOT as of last week:
Clojure 1.2.0-master-SNAPSHOT
user=> 0xff
255
user=> (byte 0xff)
java.lang.IllegalArgumentException: Value out of range for byte: 255
(NO_SOURCE_FILE:0)
So, this looks like a new issue. Rich?
-Per
On Tue, Mar 23, 2010
It doesn't seem very accidental: the namespace binding for 'trees' is
retaining the head. You probably want to wrap it in a function:
(defn trees []
...)
-Per
On Tue, Mar 23, 2010 at 9:40 PM, aria42 wrote:
> Hi,
> I was experimenting with some code and I had an largis
of any higher generations.
-Per
On Tue, Mar 23, 2010 at 9:49 PM, aria42 wrote:
> Whoops duh. That was silly, far to early on the west coast.
>
> On Mar 23, 7:46 am, Per Vognsen wrote:
>> It doesn't seem very accidental: the namespace binding for 'trees' is
>> r
Stuart's book is by all accounts excellent, but I'm not sure we want
to be in the situation that Ruby once was in, where buying a book
(PragProg's Pickaxe book) was virtually a prerequisite for getting
started.
-Per
On Tue, Mar 23, 2010 at 10:11 PM, Brian Hurt wrote:
>
>
&
This is awesome, Stuart. With the live web server, a great addition
would be if the embedded code snippets would be interactively runnable
and tweakable right there in the web page. I'll see about hacking that
in myself.
-Per
On Tue, Mar 23, 2010 at 9:13 PM, Stuart Halloway
wrote:
> The
pers. One of my favorite Lisp books
is Christian Quiennec's Lisp in Small Pieces. Though it's long out of
print, you can probably find used copies.
I won't mention the usual suspects like SICP and PAIP.
-Per
On Tue, Mar 23, 2010 at 10:37 PM, Sean Devlin wrote:
> Hey folks,
> I&
like range are not really lazy but only
chunky-lazy. You can verify this by replacing (range 10) by (range
100) and noting that the trace stops at the 32nd element (chunks are
currently 32 elements wide). So, everything is as lazy as can be,
given the input sequences.
-Per
On Wed, Mar 24, 2010 a
ojure but there seem to be a few floating around
in Java.
-Per
On Wed, Mar 24, 2010 at 3:35 AM, kotor wrote:
> On Mar 23, 1:02 pm, Glen Rubin wrote:
>> Does anyone know of any existing libraries for clojure that has code
>> which is optimized to list all of the factors of any gi
t do you think? I hacked this into my local version of core.clj's
destructure and it feels very natural to me.
-Per
--
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 that
version of LuaJIT 2.0, but he had already done LuaJIT 1.0, and he
is arguably one of the world's top experts in this area.
Besides, one of the Clojure's main raisons d'etre is the vast pool of
existing libraries in the JVM (and now CLR) world to draw from.
-Per
On Wed, Mar 24,
I never said automatic overflowing is a good thing per se.
The problem is that the expectations set by hexadecimal notation are
violated. When I write 0xFF, I don't mean the abstract mathematical
number 255; I mean the number whose bit representation (e.g. one's
complement for signe
ndex. But supposing you did, this is an instance of a more general
problem: determining the index (if any) of a number in a sorted
sequence. Something like this:
(defn index-in-sorted-seq [x s]
(loop [s s, i 0]
(when (seq s)
(cond
(= x (first s)) i
(> x (first s)) (recur
On Wed, Mar 24, 2010 at 8:39 PM, Per Vognsen wrote:
> On Wed, Mar 24, 2010 at 8:21 PM, Glen Rubin wrote:
>> I wrote the following code to produce a lazy sequence of the triangle
>> numbers. (triangle numbers are the series of numbers: 1, 1+2, 1+2+3,
>> etc...)
It would be nice if the compiler handled these things with greater
cleverness. In the mean time, some macros might be in order.
I'd also like to add that clojure.core is not generally an exemplar of style. :)
-Per
On Wed, Mar 24, 2010 at 11:58 PM, Sean Devlin wrote:
> It'
def cannot produce that error. You probably mean that it
originates with some reference to savedColors that is expecting a
sequence. You must explicitly dereference savedColors by prefixing it
with @ to get at the contained value:
user> (def saved-colors (atom [1, 2]))
#'user/saved-colors
Are the writes commutative? Since you are using pmap, I presume so. In
that case, you could funnel the writes through an agent serving as a
queue.
-Per
On Thu, Mar 25, 2010 at 9:59 PM, Scott wrote:
> id prefer best practices if possible
>
> typically, cheating has consequences down
123123"))
java.math.BigInteger
Of course, it might also pose a bit of a security threat:
user> (read-string "#=(println \"I OWN YOU NOW!\")")
I OWN YOU NOW!
nil
:)
-Per
On Fri, Mar 26, 2010 at 5:47 AM, Chas Emerick wrote:
> Glen,
>
> You want (java.math.BigIntege
In those hills yonder in the lands of Common Lisp, it's usually
considered good practice to blast the entire read table save for what
you need when you deal with untrusted data. Barring that, a better
option might be a more modular reader: read-number, read-symbol, etc.
-Per
On Fri, Mar 26,
dom-number generator."
Look at java.util.Random for local RNGs. In the long term, it might
make sense for core's rand to refer to a *random-number-generator* var
that defaults to a shared thread-safe RNG but can be rebound per
thread in cases like yours to reduce contention. In the short term
Look at clojure.contrib.repl-utils/add-break-thread!:
http://richhickey.github.com/clojure-contrib/repl-utils-api.html
-Per
On Fri, Mar 26, 2010 at 10:20 AM, Douglas Philips wrote:
> I've been using clojure 1.1.0 (via MacPorts clojure+rlwrap) and every time I
> type Control-C, it ki
arate pieces of functionality that you happen to use the
build tool to invoke. Why this obsession with integration and unified
configuration?
-Per
On Fri, Mar 26, 2010 at 1:55 AM, Chas Emerick wrote:
> I published a blog post earlier today, along with a short screencast that
> might be
You can put a user.clj file in your class path.
-Per
On Fri, Mar 26, 2010 at 11:42 AM, Douglas Philips wrote:
> On 2010 Mar 25, at 11:29 PM, Per Vognsen wrote:
>
>> Look at clojure.contrib.repl-utils/add-break-thread!:
>>
>> http://richhickey.github.com/clojure-con
n the Unix world) that does that.
That's it. It doesn't need to be part of the build system and
shouldn't be. I can call functions in the build system to supply the
relevant arguments. If the build tool and library is well designed,
the added value of integration in this example sh
On Fri, Mar 26, 2010 at 12:59 PM, Chas Emerick wrote:
>
> On Mar 26, 2010, at 12:59 AM, Per Vognsen wrote:
>
>> On Fri, Mar 26, 2010 at 11:50 AM, Chas Emerick
>> wrote:
>>>
>>> Because they're common processes that are ideally built once, and then
>
ber it generated. That would mean you could initialize your own
instance of the algorithm with the last number you saw it output and
immediately duplicate all its future output.
-Per
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this gr
8 9 10 11 12 13 14 15 8 9 10 11 12 13 14 15 16 17)
user=> (0 `cons` 2 `cons` [1 2] `(map +)` [3 4])
(0 2 4 6)
I gisted my diff, in case anyone wants to play around with this:
http://gist.github.com/350244
I withhold judgement on its ultimate merits. :)
-Per
--
You received this message be
oes that work by "accident" (or by undocumented permissiveness)?
>> Is there someplace else that I should for an answer to this?
>
> The documentation is authoritative.
Are you serious? It is neither complete nor consistent. How can it be
authoritative?
-Per
--
You rece
ind, it is very strange to me that you
would suggest treating it as authoritative for the sake of writing
code. A more realistic answer would be that much about Clojure is
still in flux and that you shouldn't count too strongly on things
staying the same, even if it's in the documentation.
It doesn't feel right only if you still think you are programming in
an Algol-style language where , is a separator token.
I can't imagine this is going to change.
-Per
On Fri, Apr 2, 2010 at 12:37 PM, Frank Siebenlist
wrote:
> Even though the specs clearly say that commas are wh
.
It feels more natural for a line-oriented REPL.
-Per
On Fri, Apr 2, 2010 at 2:23 PM, Michael Jaaka
wrote:
> Hi!
>
> I think that I've found inconsistency in REPL behavior.
> Just press ENTER in REPL and you got new command prompt.
> This shouldn't work like this. New line for
Common Lisp, PLT Scheme's [
and ] as synonyms for ( and ), etc.
-Per
On Fri, Apr 2, 2010 at 6:10 PM, Mark J. Reed wrote:
> try this one:
> (list,1,2,3)
> :)
> Per: I'd say it's also weird if you're coming from a Lisp background - just
> weird in the opposite direc
1.2.0-master-SNAPSHOT
user=> (+ 1 1)
2
user=> 1
1
user=>
user=>
user=>
user=> 1
1
user=> 1 2 3
1
2
3
user=>
By the way, these transcripts were generated with the program
'script', which is available by default on most Unix systems.
-Per
On Sat, Apr 3, 2010 at 9:
types
with an identity-based weak-key hash table.
Anyway, the way things stand, your best kept is probably to wrap your
data in some data structure. It could be as simple as [:frequency x]
or {:tag :frequency, :value x}.
-Per
On Sat, Apr 3, 2010 at 5:47 AM, strattonbrazil wrote:
> What's the
rewrap the results in the newtype wrapper.
(Clojure's arithmetic tower isn't expressible via protocol-style
single dispatch, so this is vastly simplified, but you get the idea.)
-Per
On Sat, Apr 3, 2010 at 3:18 PM, Per Vognsen wrote:
> Unfortunately you can only attach metad
atom trick was inspired by E's implementation of
revocable capabilities. IMO, revocable capabilities is one of the most
convincing demonstrations of the cross-cutting power of stateful side
effects. Of course, that is also what makes them so dangerous.
-Per
--
You received this message becau
ation N without triggering the card-marking write
barrier and forcing the generation 0 pieces to be hoisted into
generation N.
-Per
On Sun, Apr 4, 2010 at 7:53 AM, verec
wrote:
> The two issues are orthogonal.
>
> Even if Clojure had been going for full copies to preserve
> immutability, th
You talk of transactions, "persistent in-memory" and garbage
collection. Are you sure you understand what persistence means here?
It's a matter of efficient structural sharing of data structures
rather than persistence in the database sense.
-Per
On Sat, Apr 3, 2010 at 9:13 PM,
(map #(fn? (when-let [x (resolve (symbol %))] @x)) ["map", "first",
"nofun"]) should do the trick. But before you go ahead and do this,
make sure it's what you actually need.
-Per
On Sun, Apr 4, 2010 at 3:06 PM, Manfred Lotz wrote:
> Hi there,
>
> I c
vided
transformation function. A lazy-ref-set function would impose a much
stricter discipline on the user in order to maintain laziness.
All the usual caveats about full laziness apply: beware of space
leaks, delayed side effects, etc.
-Per
On Sun, Apr 4, 2010 at 9:35 PM, Bob Hutchison wrote:
&
would give the false illusion that the user is dealing
with a uniform abstraction when the semantics are really very
different. Certainly lazy vs non-lazy semantics are pragmatically very
different in a language like Clojure and the difference shouldn't be
swept under the rug.
-Per
On Mon, Apr
nd-int (- n m)
(defn knuth-shuffle [xs]
(let [v (vec xs)]
(reduce #(swap-elts %1 %2 (rand-range %2 (count %1))) v (range (count v)
-Per
On Mon, Apr 5, 2010 at 3:14 AM, Linus Ericsson
wrote:
> Hello Clojure!
>
> Is there any straight-forward way to randomly re
huffle [xs]
(let [v (transient (vec xs)), n (count v)]
(persistent! (reduce #(swap-entries! %1 %2 (rand-range %2 n)) v
(range n)
-Per
On Mon, Apr 5, 2010 at 10:19 AM, Lee Spector wrote:
>
> Since we're having a shuffle-a-thon, here's a version I wrote that I kind of
Wow, you're right. The partial laziness of his code was foiling my benchmark.
-Per
On Mon, Apr 5, 2010 at 11:05 AM, Mark Engelberg
wrote:
> On my system, knuth-shuffle performs several times faster than Spector's
> recursive functional shuffle on smallish lists, and the differe
version of my code. The code
itself is only three lines; the rest consists of very general purpose
utilities that I find myself using again and again.
http://gist.github.com/356035
-Per
> -Lee
>
> On Apr 5, 2010, at 12:11 AM, Per Vognsen wrote:
>
>> Wow, you're right. The p
, mutability (possibly
implicit) is required. Here's an example of implicit mutation via
delay/force:
(deftype Account [owner balance])
(deftype Person [name accounts])
(declare per)
(def per-account1 (Account (delay per) 42))
(def per-account2 (Account (delay per) 9000))
(def per (Person "
comparison, the time to sum that many elements with
#(reduce + %) is about 50 ms.
-Per
On Mon, Apr 5, 2010 at 5:56 PM, Sean Devlin wrote:
> If this is significantly faster than c.c.seq/shuffle, you should
> submit a patch. I know Rich was complaining about the speed of that
> fn in th
relational database there are no direct pointers, only
primary keys. A primary key is nothing more than an identifier; it
becomes a specific reference only in the context of a table.
So despite initial appearances it's actually a very familiar programming model.
-Per
--
You received this messa
imply isn't a persistent data
structure.
-Per
On Mon, Apr 5, 2010 at 8:06 PM, ineol wrote:
> It has nothing to do with Clojure but your colleague can look at the
> Google collection library that contains an ImmutableSet.
> http://guava-libraries.googlecode.com/svn/trunk/javado
ject graph traverser to find all references to a given object (e.g.
gc.get_referrers() in Python) and then rewire those references on the
fly.
-Per
On Mon, Apr 5, 2010 at 9:17 PM, Douglas Philips wrote:
> On 2010 Apr 5, at 9:43 AM, Per Vognsen wrote:
>>
>> I already mentioned this in my or
osted.
It's fair to say that while Clojure supports single assignment
variables in the form of promises, they are not fundamental
language-level constructs the way they are in languages like Prolog or
Oz. You have to explicitly dereference them like any other IDeref.
-Per
On Tue, Apr 6, 20
le code is a pair of mutually
cyclic lazy-seqs.
http://gist.github.com/336461
-Per
--
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 that posts from new members are moderated - plea
rses on functional programming. A good paper to read is
Cycle Therapy:
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.64.2027
-Per
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@g
t's okay.
I'd say it's the kind of thing that's convenient during development
but will likely need to change later on. As such, I'd recommend he
goes through the var with explicit dereferencing, or a functional
interface with a similar level of indirection, so he can easily swa
The second case is equivalent to (into [] [{:a 1 :b 2}]). You're
passing a singleton list containing a single element which happens to
be a map.
-Per
On Thu, Apr 8, 2010 at 7:35 AM, Base wrote:
> Hi All -
>
> I have a question about into.
>
> When you run the into against a
Indeed, and in the absence of single stepping (manual or automatic),
some simple trace macros can be very useful. Here's what I use right
now:
http://gist.github.com/360102
-Per
On Thu, Apr 8, 2010 at 8:38 PM, Sean Devlin wrote:
> The REPL is you best friend whenever you have a quest
You can easily code positional keyword parameters yourself. It takes
only a few minutes for a basic version. Here's an admittedly not very
pretty example:
http://gist.github.com/360145
-Per
On Thu, Apr 8, 2010 at 9:13 PM, Sophie wrote:
> On Apr 7, 7:56 am, David Nolen wrote:
>&g
Or you can separate concerns a bit more:
(defn transpose [xs]
(apply map vector xs))
Now Nurullah's original suggestion applies:
(map #(apply max %) (transpose xs))
-Per
On Fri, Apr 9, 2010 at 12:38 AM, James Reeves
wrote:
> On Apr 8, 1:13 pm, John Sanda wrote:
>> [
>>
symbol isn't bound in this way, the compiler will call (resolve
...) on the symbol at compile time. This yields some var v. The code
will then be evaluated as (deref v) at run time.
-Per
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
On Fri, Apr 9, 2010 at 10:35 AM, Douglas Philips wrote:
> On 2010 Apr 8, at 11:16 PM, Per Vognsen wrote:
>>
>> It's not a defn thing. If you write (do (a) (b) (c)) at the REPL, you
>> should see the same exception.
>
> Yes, I would expect that since at
On Fri, Apr 9, 2010 at 11:26 AM, Douglas Philips wrote:
> On 2010 Apr 8, at 11:48 PM, Per Vognsen wrote:
>>
>> The body of fn is still compiled in an expression context. When the
>> compiler sees (fn [...] ...), it will introduce the bindings into the
>> local environm
Nice! Towards a similar purpose, I wrote a little Emacs hack last week
that web-scrapes clojars.org and inserts the artifact declaration at
the cursor. Since I had to rely on their search feature, I couldn't do
proper regular expression matching.
-Per
On Fri, Apr 9, 2010 at 11:23 AM, He
1 - 100 of 165 matches
Mail list logo
