« first day (386 days earlier)      last day (4095 days later) » 

12:02 AM
@Rojo And now you're there - on the house
 
@Xerxes Got it working, the problem was a few \:200b and \:200c had crept in
way faster than my silly increment thing
 
 
1 hour later…
1:24 AM
@belisarius Oh yeaaah
 
 
2 hours later…
3:20 AM
@j.m. how large was your inbox? :D
 
3:41 AM
@rm-rf Oh, sixty messages and twenty notifications...
 
send to trash
 
Doable, yes. :)
 
btw, have you ever had the need for map/fold/nest/nestwhile/foldlist/etc style functions in MATLAB or would you find them useful?
 
@rm-rf Well, something like Fold[] would have been nice; makes for a compact implementation of Horner's method, for instance.
 
I've been using a custom package for some of these, and was wondering if I should invest the time to make it more robust and idiot proof...
Of course, the implementation is all in MATLAB, so there's no gains performance wise, but it does make for some clean code in quite a few places...
 
3:53 AM
@rm-rf That depends on the probability of an idiot using it? Have you estimated it?
 
@rm-rf "more robust and idiot proof" - for public consumption, most certainly.
@belisarius There are many ways to use any implement dumbly, much more than the proper ways.
 
@J.M. Hey @J.M. ... is that you yourself? Not my imagination?
cheers!
 
Also your "completely stoofid" can very well be somebody else's "seemed like a good idea at the time"...
@belisarius I'm only here for a short while; I need to leave civilization again in a few hours.
 
@J.M. damn! Sounds Tarzanic
 
@belisarius It's not as fun, and I don't get to wear a loincloth too...
 
3:57 AM
@J.M. Bring a few books with you into the wild. You may need them to evangelize beasts
 
@belisarius The humidity won't be kind to the books...
 
@J.M. that depends
 
@belisarius No, nothing like that where I'm going. If there were, then it'd be less unpleasant...
 
@J.M. I'm too curious, but I won't ask here :)
 
@J.M. Which is why belisarius is bringing it to you here :)
 
4:03 AM
@belisarius It's fine, I'm not at liberty to reveal details anyway. :P :)
@rm-rf (Hopefully it's not his...)
 
@J.M. Hmmm ... a mad chemist going into the jungle without women ... Is Werner Von B nearby?
 
@belisarius It's not chemistry-related either. So, a real arse-pain for me, but it has to be done.
 
@J.M. Ok, Ok, just pretending to be funny. Seems the wrong time for that. Hope you set yourself free in a near future
 
@belisarius It's fine, I appreciate the attempt at humor. :)
 
@J.M. We really miss you
2
 
4:08 AM
Just keep me in you guys' thoughts; should be fine for the time being. :)
 
@J.M. may the Force be with you. Or the Energy. Most probably the Entropy.
 
The last one's the most likely, yes.
 
@J.M. Virtual worlds are frustrating somehow. You can't even offer help. Truly inhuman
 
4:53 AM
Well, I should be leaving now. See y'all hopefully soon.
 
5:20 AM
Anyone awake?
 
I'm not anyone, but I am awake
 
@Xerxes Touche. :-) I don't think we've chatted before. Hello.
 
Hello, I'm new. It's been fun so far. Just checking things out.
How's everything?
 
Okay. As you are new I don't know your level of familiarity with Mathematica. I was going to ask how others have handed a problem relating to creating a custom Plot function. Do you have experience with something like this?
Uh-oh, I scared off the new guy. :o)
 
@Mr.Wizard in what way?
I've had quite a bit of experience. What are you looking for?
 
5:35 AM
Hi @rcollyer -- one moment.
 
had to put the baby to bed
 
@rcollyer I made a plot function that uses a custom FrameTicks specification in one direction. I'd like to allow for passing standard FrameTicks options to this function but I want to use the custom part where appropriate, e.g. any time True or Automatic is used for that edge.
I find myself writing convoluted code to catch every form of FrameTicks syntax, and I wonder if there is a better way.
 
Sounds familiar. I think ticks specifications suck, in general.
 
I would probably detect the Automatic or True as a first processing step and just retain it in the final tick specification
 
Try using Position to get the specific point that has Automatic or True.
@Xerxes exactly. Determine if it exists (with Position), and then set it up if needed.
Perfect (and one of the few) uses for MapAt.
 
5:43 AM
I use MapAt a bunch, actually; I rolled my own version of it so that it could take Span specifications and such
 
Hell, don't use MapAt or Position; use ReplaceAll.
@Xerxes Haven't gotten around to fixing it to use Span, yet.
 
Here's what I've got now but it's still missing certain forms:
FrameTicks -> (OptionValue[FrameTicks] /. {
    {{w_, e_}, {s_, n_}} :> {{w, e}, {s, n} /. Automatic -> ticks},
    {ns_, ew_} :> {ns /. Automatic -> ticks, ew},
    dir : Automatic | All | True :> {ticks, dir}
    })
 
Why don't you turn it into a canonical form, such as {{w, e}, {s, n}}, first?
You could then use that functionality elsewhere, too.
 
Good idea. That should simplify things. :-)
 
I did that with one plot I added to my code sample for WRI. Crafted a mini-parser for one of the options, so that it would always be in canonical form.
stdform[{rls:({__}->_).., mthd:(Method-> _):(Method->Automatic)}]:=
Thread@Which[
	Length@mthd[[2]]>1,
	  {First@mthd[[2]],{rls},Sequence@@Rest[mthd[[2]]]},
	Length@mthd[[2]]==1,
	  {First@mthd[[2]], {rls}},
	True,
	  {mthd[[2]], {rls}}
]
stdform[rls:({__}->_)]:= {{Automatic,rls}}
stdform[rls:{(({__}-> _) | {{__}->_, Method->_})..}]:= If[Head[#]=!=List, {Automatic,#}, Flatten@stdform@#]&/@ rls
stdform[{}]:={}
 
5:49 AM
@Mr.Wizard That's probably not sufficient... look at the implementation of ListPlot.
I tried to dive into it for something a while back... got lost in the maze and eventually decided to do something else instead
 
how do you look at the implementation of something like that?
 
@rm-rf as @Xerxes put it, where did you find the implementation of it?
 
uh oh :D
ClearAttributes[ListPlot, {Protected, ReadProtected}]
?? ListPlot
And do the same for symbols inside... essentially, follow the rabbit hole
 
Ah. Now there's locked. :P
(I should write in english ...)
 
ListPlot is not locked
@Xerxes You can also use this fantastic answer:
16
A: What is the most convenient way to read definitions of in-memory symbols when we don't have the source files?

Simon WoodsI have been using this. It's mostly Leonid's code from the stackoverflow question you linked to, but it uses Definition instead of DownValues. Symbol names are printed without any context, but the full symbol name is put into a Tooltip so you can always find out what context a symbol is in. Upda...

 
5:54 AM
@rm-rf: (neo) whoa (/neo)
 
@Xerxes welcome to the real world.
 
if I delve too deep and see a System`Protection`Balrog, I'm outta here
 
Just find the south bridge, and he can't cross.
@Mr.Wizard if you have any questions, ping me. we could even set up a regular chat room, if you wanted.
Night all.
 
hey, there are all kinds of undocumented options down here
did you guys know about PlotTheme?
 
6:57 AM
@rm-rf Sorry, I had to leave quickly before. I didn't mean to be rude.
I'm not seeing top-level definitions for ListPlot in v7. What do you have in mind for parsing FrameTicks?
 
I was reading in the MMA documentation that Manipulate is a higher abstraction based on Manipulate, is there a MMA command to see the source code of an inhouse function, such as Manipulate? I don't expect to be able to find the low level source, but maybe I can see how higher abstractions are programmed?
err on Dynamic
 
 
3 hours later…
9:48 AM
Hi Anyone there?
I have a problem with a function that I am writing with MMA.
I am trying to relate between the ImageSize and the pixelvalue of a text input to figure the size of the text box?
Does anyone know anything about it so far?
 
 
3 hours later…
1:04 PM
hi guys
 
Hi
 
@jVincent how are you?
 
I'm good how are you? :)
 
same to you..
I posted a question about InputFields
 
I can't seem to find it.
 
1:08 PM
oh..kkk
can i send that link
You're around for so long and you still don't understand that without a small, working code example it is generally rather hard to answer questions like this. Also, this must be your umpteenth question about InputField and CDF. Will this ever stop? And, on a final note: is it so difficult to learn to format properly? You still randomly capitalize words and use code format for text that isn't code. — Sjoerd C. de Vries 16 hours ago
 
Ah yes, I think I saw that one. Personally Sjoerd was a bit rude with his remark. Though I did think your question could be improved.
As I recall you didn't post any code making it very hard to intepret what you where asking about. Perhaps your question was closed, which would explain why I can't find it (though I may just be looking the wrong places).
 
but code was big..so I didn't post the code.
I will edit my question
 
Ohh, don't post your actual code.
What you should do is to make a very small segment of code that shows the problem.
 
ok...
where you from?
 
So for isnstance if you have 100 lines of code and get a problem where an inputfield turns black unexpected, you should copy the inputfield out, and try to make it show the error.
Then you post only the small example of the code that creates the problem.
I'm from Denmark, why?
 
1:17 PM
just I am asking
I am from india...I am little bit poor in english..so try to understand
 
If there is something you do not understand, you can ask and I will try to explain.
 
ok...thank you
 
acl
@subbu posting a minimal example showing the problem helps a lot. for instance, you have some code and it doesn't work. start writing a smaller example to do somethign similar, and try to build it up until you hit the same problem. then you've located the problem and can post that code as an example (hopefully it's small enough)
 
@acl I thnk my code was big,so I didn't posted,I will modify my question..
 
acl
@subbu yes that's always a problem; if it's long then most people won't bother trying to understand it. try to find a minimal example that demonstrates the problem
 
1:30 PM
@Szabolcs Linux
 
@jVincent also suggested thi same thing
next time I will modify, generally is there any Option are there for Cursor styles?
 
@subbu Yes, you can change the cursor using for instance: MouseAppearance[Plot[x, {x, 0, 1}], "DragAndDrop"]
 
1:50 PM
Morning
 
@jVincent thank you...I will check it
see you tomorrow guys..
 
@Rojo Hehe.. morning
 
@Rojo Good afternoon
 
@halirutan Hey. Are you already in the acceptance stage of grief for the death of your camel humps?
@jVincent We'll agree on "hello"
 
@Rojo It's unbelievable how I got used to this. Now it takes me even longer to code because I always rely on the expansion..
 
1:56 PM
Almost 11 AM here. Where are you in earth?
@halirutan It was definately a good idea
 
@Rojo I hope John Fultz meant it when he said it's not yet included in the current patch.
 
@halirutan I was thinking the same thing
@halirutan I wonder what we lose by using 9.0.0's front end
 
2:37 PM
camel humps :?
 
3:10 PM
Can someone please review this question? I don't see the justification for the bugs tag... am I missing something obvious?
 
3:47 PM
28
A: Is it possible to improve Mathematica's auto-completion feature?

halirutanCamel-humps auto-completion for Mathematica The answer to my question is yes, we can expand the auto-completion capabilities of Mathematica. For those who want to know, how exactly this can be done, I suggest reading the Reverse engineering section below. There I explained how you can investigat...

@OleksandrR. I think this was a mistake. To relatively new members it is not clear that the bugs tag is for confirmed bugs only.
 
4:13 PM
@halirutan thanks. I was confused because it seemed to me not even a suspected bug, let alone confirmed, so I thought I had misunderstood. But it turns out that David agreed with me and now the tag's been removed.
 
5:10 PM
@halirutan Did he say that? Then maybe there's hope
 
6:03 PM
@Szabolcs yep..
Jan 8 at 12:30, by John Fultz
Incidentally, in the upcoming 9.0.1, some of the behind-the-scenes stuff for completion is changing. It's going to invalidate a few answers, like the one for camel-hump completion. But there are a number of improvements and bug fixes. But no camel-hump completion (yet).
 
 
4 hours later…
10:22 PM
@halirutan Give me my bounty back ;D
 
@Rojo Hehe.. it was a tick I planned from the beginning ;-)
 
Hehe
Quick way to build a square matrix spiraling a list
such as Range[0, 15]
I'm assmuing some kind of Normal@SparseArray[{i_, j_}:>list[[someFunctionOf[i,j]]], Sqrt@Length@list {1, 1}]
 
@Rojo What do you mean "spiraling a list"?
 
spiral[Range@9] giving MatrixForm[ {
{7, 8, 9},
{6, 1, 2},
{5, 4, 3}
} ]
I don't actually mind the starting direction or even if it rotates clockwise or counterclockwise
and not really giving MatrixForm :P
 
Hmm... we had a question on something like that.
 
10:36 PM
Brb, gotta reboot
 
Wasn't it a question by belisarius?
 
10:52 PM
@Szabolcs He has only 17 questions and at a glance it doesn't seem to be any of them
 
@Rojo The he posted an answer, it would be more difficult to fish it out from there
 
@belisarius heeeyyyyyy
 
@Rojo what's up?
 
@Szabolcs Thanks, looking into it
@belisarius Hey
Do you remember answering a question on this?
 
I couldn't find his post either. Maybe be on StackOverflow
 
@Rojo like juat?
 
this==look up half a page
 
11:14 PM
@Rojo The spiral thing?
 
@Rojo I can't find anyone on the first page of the user list sorted by rep who has a Q/A ratio as high as mine ... does that reflect badly on me?
 
@belisarius yea
 
mmm yup. lemme search
 
@Szabolcs On the contrary
I'm always glad to see questions of those high rep users. They tend to be more interesting
 
@Rojo 42
@Rojo Sorry, cant find it
 
11:19 PM
@belisarius Ok, thanks anyway
 
Impressive looking steady growth:
 
spiral[L_List]:=Module[{dim=2Ceiling[(Sqrt[Length[L]]-1)/2]+1},#/.x_?NumericQ:>PadRight[L,dim^2][[Position[Sort[Flatten[#]],x][[1,1]]]‌​]&@Array[2Norm[{##},\[Infinity]]+Mod[ArcTan[##]/\[Pi]+7/4-1/(2 dim),2]&,{dim,dim},{-Floor[dim/2]-1.*^-6,-Floor[dim/2]}]]
 
oops, not quite
 
11:35 PM
@Szabolcs Yeah, not totally helpful. Graph seems to layout in spiral when they become a certain length...
Very weird way of building that graph
Why not just u = UndirectedEdge @@@ Partition[Range@507, 2, 1]
 
spiral[L_List]:=Module[{dim=2Ceiling[(Sqrt[Length[L]]-1)/2]+1,x0},x0=Floor[dim/2];#/.x_?NumericQ:>PadRight[L,dim^2][[Position[Sort[Fla‌​tten[#]],x][[1,1]]]]&@Array[2Norm[{##},\[Infinity]]+Mod[(ArcTan[##]-ArcTan[x0,x0+1])/\[Pi],2]&,{dim,dim},{-x0-1.*^-6,-x0}]]
 
@Xerxes That seems to work for odd sided matrices. Thanks!
 

« first day (386 days earlier)      last day (4095 days later) »