Why this no work? :( (#?OddQ) & /@ Range[10]

incorrect syntax. Try (# // OddQ) & /@ Range[10]

@MikeHoneychurch Thank you mike.

In my old c++, that would be, for(int i=0; i<10; ++i) { myList.push_back( IsOdd(i) ); }

i'm loving functional programming :D

OddQ as you know tests for odd. So the syntax is e.g. OddQ[number], OddQ@number, number//OddQ. In other words it is a function. I think with your use of ? you might have been getting mixed up with PatternTest in which OddQ is used to test an element in a list etc.

e.g. Cases[Range[10], _?OddQ]

Exactly

Sup rm -rf

@AdamDreaver OddQ is Listable, so you can simply do OddQ@Range@10

Oooh nice syntax, ty

rm rf is a good name. 4 syllables, I just included that in a freestyle rap

ok, time to go and socialize at work, take guys/gals

yes forgot it was Listable

.

?

bah... the questions we get these days

@rm-rf I couldn't chose which close reason was appropriate for this

choose not chose...

I'm told there are some planned changes that will allow mods to close with a custom reason... if and when that comes up, we can close a lot of these with a link to the appropriate doc tutorial page

@rm-rf Bad questions these days? I haven't been too active

@Rojo You decide :)

Haha

that link takes me to a 404

It's deleted

@MikeHoneychurch probably better this way :)

I guess I'll take your word for it. Most questions are asked and answered by the time i log on in the morning but have noticed a deterioration of question quality over the last few weeks. Maybe subjective on my part but my 2 cents

(am I really the only one who thinks asking "please plot me the higgs potential" isn't an appropriate question? are we expected to know what the "higgs potential" is now? and anyway it's a 3d plot, just look it up in the docs...)

@MikeHoneychurch no I think that is an accurate observation. that particular question was more or less someone posting some sort of email soliciting job applications

Not that my own questions are of broad interest :)

@acl sounds like something that was happening on the linkedin mathematica group last year

@MikeHoneychurch I don't think I am a member. but this one was, literally, a copy-and-pasted email

Oh, it's not true that I wasn't active. I just did a long one ;D

Had forgot

Anyway, how are you all doing, people?

@acl 3 close votes already

@Rojo I was asking a rhetorical question to encourage others to act :)

@acl Well, you did it wrongly, because I answered

@Rojo (fine, and you? :) )

@acl Haha, I'm good. Watching the Mentalist (I don't like it much but watch it anyway), and probably not going to work tomorrow :)

@Rojo sick?

@acl Not at all. I have an unclear arrangement with my boss, that gives me some flexibility

At the end of the day, it's all good as long as the stuff gets done

@Rojo murky job situations can be useful

(or "underscpecified" job situations)

@acl Exactly :)

I am greatly enjoying the Legends functionality, as I am preparing plots for a paper. looking at my plots, I think I ended up putting way too many separate lines just so I could enjoy adding legends...

now we just need undo

@acl The first days after v9 was released I read complaints here, but I found it great. In any case I'm no purist with visuals

@acl Did you sign?

Hehe

@Rojo no they just need to be presentable. legends, as they currently are, are fine for me. but I have low standards

@Rojo Every release will change something that annoys someone

@Rojo ha

@rm-rf The issue I've had with legends was that i've used them everywhere, but still share with people with v8

So I had to do an ugly If[$VersionNumber<9, Legended[#, "blablablabla"]... etc

in dynamic

several times

@rm-rf but 9.0 is teeming with front-end bugs...

But its stronger than me, I like to use the new stuff even when it isn't needed

I also use 9 even though, at the moment, copying to the clipboard causes a beep and stuff to be spat out to the console

Uh... that doesn't happen here

Weird

Not here

and I use the clipboard to simulate undo, so...

lol

@rm-rf @rojo I contacted support, they also couldn't work it out

It's very high on my todo list but I never actually got to try the undo posted as an answer to my highly voted question

Did you? @rm-rf, @acl

Nope

Hiya!

I didn't read it fully, but from what I remember, I think it was more like a version control than a proper undo

@belisarius Hi!

@belisarius Ola

already complaining about v9?

@Rojo manola

@belisarius If MrWizard asks what you meant, you'll explain

@Rojo Is Mr. still on V 5.2?

@belisarius I think he is

good! 5.2 is very stable

Interesting that the king of the site is 2 versions behind

@belisarius Oh, sorry, I had Unprotect[Dot];Dot=Plus

@Rojo You're so clever that makes me cry

@Rojo not really

@belisarius yes it's painful

@acl I thought it was an intimate feeling. Mmmm ... perhaps his cleverness arises telepathic abilities.

Hahah

@rm-rf He's a genius :)

I've never worked with a language that uses pattern matching. I can see how powerful it is, but it's a little overwhelming for me :D Different levels and deliminating the head and all that.

oh well, i'm putting in the time and I'm now starting to get it. I look forward to when I can use Position/Extract in some useful way.

dude is using project euler to learn mma. so smart.

@Xerxes Thanks @Xerxes but I am looking for a solution that is internal to ReplacePart.

If there is any...

@IstvánZachar Hey, why do you want that?

@Rojo I have a function that replaces parts of an expression (via ReplacePart) that are found by using Position. Now Position returns {} if the whole of the expression matches the pattern, but ReplacePart cannot deal with such a position specification.

So I have to do such external gymnastics as the one Xerxes has suggested.

I see

It seems reasonable that ReplacePart should deal with {}, but actually it can not.

Agree

@IstvánZachar This is disgusting, but another option may be to parse the rules

and then

ReplacePart["S"[2], ({} :> "A") /. rpRule]

ReplacePart["S"[2], ({1} :> "A") /. rpRule]

Less intrusive than unprotecting

Yeah, you're right

Thanks anyway for the help!

@IstvánZachar maybe something like this

myRules /: ReplacePart[expr_, myRules[Except[_List, rspec_]]] :=
  ReplacePart[expr, myRules[{rspec}]];
myRules /: ReplacePart[expr_, myRules[{{}~(Rule | RuleDelayed)~rep_}]] :=
  rep;
myRules /: ReplacePart[expr_, myRules[rspec_]] :=
  ReplacePart[expr, rspec];

expr = f[f[f[]]];
rule = f[x___] :> g[x];
pos = Reverse@Position[expr, First@rule]
ReplacePart[expr, myRules[# -> g[]]] & /@ pos

(* -> {{}, {1}, {1, 1}} *)

(* -> {g[], f[g[]], f[f[g[]]]} *)

Quite complicated, but it works : )

@Rojo yes, many improvements are possible. Since the total replacement overrides everything else I didn't think it was meaningful to have it combined with other rules so I deliberately didn't address that case.

Handling of options would also need to be added. I doubt if I would use this approach myself but @István seems wedded to using ReplacePart for what I assume are good reasons.

If you are interested, I can make it a proper question so that you can post your answers.

It would give room for further improvements

I don't mind. Do it if it would be helpful to you but I don't think either of these "answers" is really anything to be proud of

@IstvánZachar If so, you will have to be more specific as to what you are trying to achieve

@Rojo Sure, I have a very specific problem. If I'll have time later today, I'll try to write it up properly.

@OleksandrR. Hey, perhaps you konw how to handle these issues

I have a continuous piecewise function that has a discontinuous derivative, that I use in NDSolve, and get errors for its Indeterminate derivative

@Rojo sorry, I don't. :) I've never used NDSolve seriously. It's on my list to learn about it.

Ugh, doc center freezes

@Rojo tutorial/NDSolveWhenEvents#1688891301 ?

@OleksandrR. Let's see, seems promising, thanks

@JacobAkkerboom You probably have to localize all the variables you use. Try Block[{pos=1,...},...].

Ah, thanks

So I put the Block outside the Compile ?

or inside I guess

Hmm... Still no luck :(

@JacobAkkerboom post the Compile

hi @halirutan, did you see my comment on the mathlink/librarylink question?

Huh

that works!? :D

@Rojo, @Ajasja tnx!

Yeah, making one variable not local introduces a bunch of uses of MainEvaluate. That makes sense..

I voted close as too localized on this one, as there's not enough info to reproduce and the OP hasn't written back.

@Szabolcs yup

@Szabolcs I saw them, yes, but had no time to answer appropriately. Therefore, I left it until I have a bit time to think about it.

I voted close on this one too because of Jens's (correct) comment. I'm not sure others agree this should be closed, but I don't see a reason to post an answer. It's not of help to future visitors, it's a math mistake.

The fire filter is a per pixel filter that averages pixels over time from a set of "hot" pixels choosen at the bottom of the screen. I wonder if there's a way to replicate that famous demo program in mathematica? I'm so weak on animations and real time graphic capabilities of mma. I don't want the code for this demo, I just want to know, is it possible?

the hot pixels are randomly lit every few dozen ms or so.

Sorry if there've been a lot of those today :-)

I bet that's gonna be a heart :D

@robjohn Happy valentine's day robjohn

@AdamDreaver and to you, too

@AdamDreaver At least here I can be pretty sure that everyone can render it. On math.SE, I had to post a rendering, as well.

I'm trying ot make it 3d right now

now I give up, I realize I have no idear how to make a 2d parametric plot into a 3d parametric

@AdamDreaver It is hard because the two circles (spheres) need to lose the crease part way down the heart.

@AdamDreaver @robjohn:

@rm-rf oh wow, VERY NICE!