Stu-
> (2) +1 on hiding private APIs. We use private APIs carefully and
> deliberately in Clojure, and I would hate to facilitate people
> writing code that will break later.
That's +4 on this one by my tally, I'll pull the private vars.
> (3) +1 on making it very easy to see which version of a
On Jun 23, 5:17 pm, Michał Marczyk wrote:
> Hi,
>
> just noticed that
>
> (defprotocol Foo (foo [self]))
> (defrecord Bar [x] Foo (foo [self] x))
>
> gives areflectionwarning-- "call to contains can't be resolved".
>
> My initial guess is that this is due to a missing hint on line 215 in
> core_de
On Jul 10, 2010, at 4:10 , David Nolen wrote:
> Some benchmarks thoughts on various databases + aleph.
>
> http://dosync.posterous.com/22516635
>
> Cheers,
> David
Out of curiosity, mind to give it a try with stupiddb? I am very interested how
much the saving of stuff from memory makes things
I'm not sure how relevant the benchmark is, since all it tests in
throughput of database insert operations. The combination of insert,
update, delete and selects with more even distribution of operations
will be much more helpful, but much harder to test.
> If the whole premise of the blog post tu
Is it safe to alter an atom while iterating that atom's value? It seems
to work as I would expect, but am I just getting lucky (and are there
concurrency issues)?
user=> (let [a (atom #{1 2 3 4 5})]
(doseq [x @a] (prn @a) (swap! a disj x)))
#{1 2 3 4 5}
#{2 3 4 5}
#{3 4 5}
#{4 5}
#{5}
Hello
Please could someone tell me the preferred way of attaching field
(key?) specific metadata in a defrecord definition?
And how to I extract this metadata on a field by field basis?
Thank you
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To p
On 11 July 2010 12:38, Jeffrey Schwab wrote:
> Is it safe to alter an atom while iterating that atom's value? It seems to
> work as I would expect, but am I just getting lucky (and are there
> concurrency issues)?
>
> user=> (let [a (atom #{1 2 3 4 5})]
> (doseq [x @a] (prn @a) (swap! a d
On 7/11/10 8:43 AM, James Reeves wrote:
On 11 July 2010 12:38, Jeffrey Schwab wrote:
Is it safe to alter an atom while iterating that atom's value?
user=> (let [a (atom #{1 2 3 4 5})]
(doseq [x @a] (prn @a) (swap! a disj x)))
Whenever you deref an atom (using @), you're gettin
> Thanks. What I'm having trouble with is not the reference vs. value
> semantics, but the order of evaluation. Clojure seems to be all over the map
> on this; it's probably not as bad as it looks, but I'm accustomed to either
> purely strict or purely lazy evaluation. I looked at the documen
On 11 July 2010 15:18, Jeffrey Schwab wrote:
> Which instance in time? Just once, up front, or could the deref happen anew
> on each pass? Is there any way to know, in general, whether a deref is
> guaranteed to happen only once, or might a macro like doseq move the
> expression @a into the midd
PS. Ah, clearly the "except the leftmost one" part of the above is
incorrect/unnecessary -- it's just that the leftmost expression gets
the outermost "loop level", so is evaluated only once. Anyway, this
behaviour is hinted at by for's docstring only in that it mentions
that the inner expressions (
On Sat, Jul 10, 2010 at 11:23 PM, zkim wrote:
>> but what do you think about using Justin's codebase, or an Aleph-based
>> server to
>> host the thing instead of Ruby/Rails? (see the link above for more details)
>
> I'm inclined to move forward with the Ruby / Rails for now. The
> reason I went
On 11 July 2010 14:30, Quzanti wrote:
> Please could someone tell me the preferred way of attaching field
> (key?) specific metadata in a defrecord definition?
I'm not sure if I understand your intention correctly, but if I do,
then there is no such thing. If the fields of a particular instance o
Thanks Michał
You did understand my intention, and it was the former non-existent
case I was referring to.
I wanted to attach metadata about type, field arity etc in the
definition of the record type so that I could write some tools to do
intelligent conversion between record structures.
Charles
hi --
i've seen paul standig's work with clojure + terracotta. wondering if
anyone has continued his work?
--
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
To All,
I was wondering if it was possible to utilize Clojure for safety
critical software development. This would include development for
guidelines within both DO-178B level A and DO-254 certification. I am
also wondering if it was possible to utilize Clojure for space-borne
applications?? A
Hi All,
Has anybody considered implementing Clojure on BEAM. Are there any
works or current attempts currently available?
In your view where are the real trip ups for this to happen.
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this gr
Looks very nice!
Hoping for the categories to arrive to the core namespace. This really
saves a lot of time for a clojure beginner like me.
Additionally, I'd prefer if the source was wasn't exposed by default
(It's quite distracting when one is looking for the examples), some
kind of "source" but
I seem to suck at searching for information and the api because a
recursive get should be there already but I can't find it, ie the
following has probably already been implemented as a macro what will
be prettier to call but what is it called?:
(defn get2 [el lst]
(if (= 1 (count lst))
(g
On Sun, Jul 11, 2010 at 11:45 AM, Quzanti wrote:
> Thanks Michał
>
> You did understand my intention, and it was the former non-existent
> case I was referring to.
>
> I wanted to attach metadata about type, field arity etc in the
> definition of the record type so that I could write some tools to
On 11 July 2010 16:15, hsarvell wrote:
> I seem to suck at searching for information and the api because a
> recursive get should be there already but I can't find it, ie the
> following has probably already been implemented as a macro what will
> be prettier to call but what is it called?:
It's
Hi Pedro
That's one idea isn't it. And I suppose you could write a macro for a
'metarecord' to generate both Foo and FooDefinition from a single
concise and convenient definition?
At the other end of the scale there is always the option of a naming
convention - ducks-Set-Of-Birds etc!!
Charles.
On Sun, Jul 11, 2010 at 8:14 AM, Phil Hagelberg wrote:
> On Sat, Jul 10, 2010 at 11:23 PM, zkim wrote:
>>> but what do you think about using Justin's codebase, or an Aleph-based
>>> server to
>>> host the thing instead of Ruby/Rails? (see the link above for more details)
>>
>> I'm inclined to mo
>Also is it possible to utilize Clojure for programming
> microcontrollers??
this i a question, that i am also interested in:
- what are todays main-objectives for using clojure
in the (higher ends) realtime-programming space ?
- what would be the main-benefits of using clojure
in the (higher en
On a note related to your example, it would be more idiomatic to
keywords in preference to symbols as keys in your map (e.g., {:hej
"hej" :svejs {:igen "hej2"}}) and literal vectors in preference to
quoted literal lists (e.g., [:svejs :igen]).
On Jul 11, 7:15 am, hsarvell wrote:
> I seem to suc
Phil-
A little background on my clojure web experience: I've got a intranet
web tool written using ring / enlive / moustache (system health /
monitoring for a personal project) as well as dabbled in compojure
before the 0.4.0 rev, and more recently I've been looking at sandbar /
carte. The timefra
Lukasz-
> Hoping for the categories to arrive to the core namespace. This really
> saves a lot of time for a clojure beginner like me.
The way Justin did this is great (http://clojure-examples.appspot.com/
clojure.core). I'm hoping to add this in the near future.
> Additionally, I'd prefer if th
Rich-
> It sounded to me like he was only saying that he's more familiar with
> Ruby/Rails than he is with Clojure.
Correct. I think it was a fair question, and I'm happy to provide
insights into my decision making process.
-Zack
On Jul 11, 12:37 pm, Richard Lyman wrote:
> On Sun, Jul 11, 201
Hi, Clojure destructuring is really awesome, but I came up with a use
case for maps that isn't supported right now as far as I know.
Currently you can get the keys that you want using the :keys keyword
and the whole map using :as.
user> (let [{:keys [a b c] :as all} {:a 1 :b 2 :c 3 :d 4}]
(print
We use Rails for all our GUIs,
we need simple CRUD most of
the time and we ActiveScaffold.
Most of our controllers are less than
70 lines and we seldom write forms.
We use the ActiveScaffold partial
forms almost every were sometimes
custpmizing them.
Clojure Web Frameworks are not
there yet.
We
On 7/11/10 9:52 AM, Stuart Halloway wrote:
Thanks. What I'm having trouble with is not the reference vs. value semantics, but the order of evaluation.
Clojure seems to be all over the map on this; it's probably not as bad as it looks, but I'm accustomed to
either purely strict or purely lazy
On 12 July 2010 02:37, Jeffrey Schwab wrote:
> I did not know whether the sequence expressions (the right halves of the
> binding-forms) were evaluated strictly, or re-evaluated on each pass (like
> the body-expr). They are apparently evaluated a priori, when the
> comprehension is defined:
Actu
PS. Note that the inner seq expression cannot be evaluated up front
(without the whole comprehension being strict, at any rate) because
they can refer to outer bindings.
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email
As a Clojure novice I'm happy to see the documentation side of the
Clojure ecosystem growing regardless of the stack used for a doc site.
Having a resource like a user-edited php.net will be a great help. I
am excited about submitting a few examples of my own.
On Jul 11, 9:14 am, Phil Hagelberg w
On 7/11/10 9:08 PM, Michał Marczyk wrote:
On 12 July 2010 02:37, Jeffrey Schwab wrote:
I did not know whether the sequence expressions (the right halves of the
binding-forms) were evaluated strictly, or re-evaluated on each pass (like
the body-expr). They are apparently evaluated a priori, whe
call me crazy but I think any organization planning to use clojure as such
should be able to independently evaluate it's readiness for the proposed
task(s).
On Jul 10, 2010, at 10:55 PM, David Blubaugh
wrote:
> To All,
>
>
> I was wondering if it was possible to utilize Clojure for safety
>
I've been following this too, and have not heard anything in a while. I
know that Luc Prefontaine was trying to get terracotta to work with Clojure
as well. I have not heard any news recently. A potentially more
straightforward solution for distributed processing right now is swarmiji.
With thi
that's too bad. i'm actually more interested in terracotta as a
persistence solution than as a distributed computing solution.
On Jul 11, 9:02 pm, rob levy wrote:
> I've been following this too, and have not heard anything in a while. I
> know that Luc Prefontaine was trying to get terracotta t
Maybe terrastore would be a good solution then?
http://clojars.org/search?q=terrastore
http://github.com/sbtourist/terrastore-cloj
On Sun, Jul 11, 2010 at 10:10 PM, scx wrote:
> that's too bad. i'm actually more interested in terracotta as a
> persistence solution than as a distributed computin
We worked on it but realize at some point that it involved many
redefinitions of some parts of the Clojure runtime in a TIM.
And this had to be maintainefd while Clojure evolved.
We opted instead to share stuff using Java classes from Clojure rather than
directly in Clojure.
However as Clojure mo
Hmm.. I would give it a try for a critical app.
We have sever apps in Clojure running for months non-stop now without
unscheduled down time. Clojure has not been a source of problems, we
had more problems with the customer networks and other site specific
conditions.
We run redundant components by
On Tue, Jul 6, 2010 at 11:44 AM, George Jahad
wrote:
> My experiment with the JDI:
>
>
> http://georgejahad.com/clojure/cdt.html
>
Hi,
Thanks for this great project and notes.
For others who are trying this out, I found a minor issue. I am
running openJDK under Debian GNU/Linux amd64 arch, with
42 matches
Mail list logo
