The h Bar

4:01 PM

@BernardoMeurer there's a couple of approaches. I would probably read the whole file into a character array in on block, then step through the array changing the '\n' to '\0' and storing the offset to the start of each word in a separate array.

Bernardo Meurer

@JohnRennie Can't I just do fread or something?

Slereah

Ah well

Let's edit it

John Rennie

@BernardoMeurer yes. Get the file size, allocate an array that size +1 and fread the whole file into your array.

Then tack on '\0'

Slereah

Wait

it was tex'd

But

The bracket notation has < instead of langle

Seems a bit flimsy to edit

John Rennie

Ooh, ooh, my advance copy of Borne by Jeff VanderMeer has just arrived.

That's tonight's reading sorted then :-)

Bernardo Meurer

4:05 PM

FILE *wordlist = fopen(argv[1], "r");
fseek(wordlist, 0, SEEK_END);
long size = ftell(wordlist);
fseek(wordlist, 0, SEEK_SET);

@JohnRennie Then I malloc with size+1?

John Rennie

@BernardoMeurer Can't you just stat the file to get the size?

Bernardo Meurer

@JohnRennie Not portable, that only works on POSIX compliant os's

John Rennie

stat will work on every OS that supports the C run time libraries. It even works on Windows for god's sake!

But, yes, you can calculate the size by seeking to the end of the file.

Bernardo Meurer

@JohnRennie Hmm? Isn't stat from sys/stat.h?

John Rennie

Everything supports that.

Bernardo Meurer

4:09 PM

I'm not betting my horses on that kind of lib :P

I will use the stdlib, it's safer

John Rennie

OK :-) Your method is fine. Get the size, malloc that size + 1 and fread the whole file into it.

Bernardo Meurer

@JohnRennie Hmm, but then I won't be able to reference this with array syntax, will I?

words[0] and so on

John Rennie

Allocate an array of *char called words and set each member of the array to the address of the start of the word. So words[1] is a *char pointing to the start of the second word, and so on.

Bernardo Meurer

Wait, let me process this

So the same old array of pointers :P

Can't I

Hmm

John Rennie

It's not hard. Allocate an array called raw_data or whatever, then allocate a second array called words.

Then words[0] = array, words[1] = array+n where n is the offset to the second word and so on.

The only thing is you won't know in advance how many words there are. But you can always parse the raw_data array twice, the first time to get the number of words in it.

Bernardo Meurer

4:17 PM

Hmm

I think I'm better offloading this into a function

Slereah

"The probability model of quantum mechanics is different from the Kolmogorov model, is due to von Neumann, and actually predates the Kolmogorov model."

Dammit :V

Why does everything have to be so cryptic

Bernardo Meurer

@JohnRennie Brr, what a PITA exercise

John Rennie

@BernardoMeurer this sort of thing is quite common though. In Windows you'd use a memory mapped file, which automatically treats the whole file as an array.

i.e. maps the file into memory.

Bernardo Meurer

Meh, I hate file handling in C

it's bad enough in C++

Slereah

Files are plenty easy in C

Bernardo Meurer

4:32 PM

@Slereah How would you do this then

I have a file with one word per line

I want to load this file into an array words[], each cell of the array has a word in it.

Slereah

Make a char* buffer, read the file char by char, copy the buffer into a new char* when you encounter a newline

Bernardo Meurer

@Slereah But the the memory holding each word won't be dynamically allocated will it?

Slereah

That's why you use a buffer

Make it large so that it's unlikely to overflow

Then you can create a new char array of the appropriate size to put in the array

Bernardo Meurer

@Slereah I can just count the size of the file and allocate accordingly, no?

Slereah

Well, if you're worried, you can do a first scan, calculate the size of the largest word, and allocate appropriately

Bernardo Meurer

4:40 PM

    fseek(wordfile, 0, SEEK_END);
    *length = (int)ftell(wordfile);
    fseek(wordfile, 0, SEEK_SET);


    char *buffer = malloc((size_t)length);

Slereah

Then just give the buffer that size

Bernardo Meurer

@Slereah Why not just that?

Slereah

Making the buffer the size of the whole file is a bit much, maybe

Bernardo Meurer

@Slereah Why? It's just a txt file, the buffer will have all of it in it, no?

Slereah

The buffer is to put a single word in

Then you can copy the buffer into the appropriate string of fixed size

Jaime Gallego

4:41 PM

If you're not comfortable with allocating a big, static chunk of memory, another solution would be to implement a linked list structure for each word that grows appropriately until the word ends

Bernardo Meurer

@JaimeGallego That still involves caching the whole file into memory

Unless I did some trick with gets

0celouvskyopoulos

$\mathfrak o$

$\mathfrak v$

why does that one look like an o with a trump haircut

paracetamol

in The Periodic Table, 3 mins ago, by CowperKettle

> Why do you drink vodka?
Because it is liquid. If it were solid, I would have gnawed on it.

Jaime Gallego

DON'T ever use gets

Anonymous

@paracetamol And if it were a gas?

dmckee

4:47 PM

@BernardoMeurer You can also build a growable array structure, that allows writes in O(1) time on average (though any particular write may be O(n)).

It's not hard, but is more complexity than a simple project justifies.

paracetamol

@blue Breathe

Bernardo Meurer

@JaimeGallego Why not? I use gets sometimes, it's alright

dmckee

As a side effect you get 'array' bounds checking automatically.

But you must not access it with the [] operator.

Jaime Gallego

Read the manpage

Bernardo Meurer

@dmckee I'm going to read the file into a buffer, an

dmckee

4:49 PM

@BernardoMeurer If you are on linux (and this is not an assignment that disallows such APIs) it is faster to memory map the file.

Bernardo Meurer

Hmm

John Rennie

For heavens sake people go easy with the flagging. That was a joke.

Jaime Gallego

0celouvskyopoulos

@JohnRennie Then a mod can unban him, no?

ACuriousMind

@JohnRennie Still...not an appropriate thing to another user if someone takes offense. Alas, I would have preferred the flagger voiced their offense instead of immediately flagging.

dmckee

4:51 PM

@BernardoMeurer gets really is broken for almost all uses. getline is in POSIX and is much better.

(And it uses that growable array trick under the hood, come to think of it.)

John Rennie

@0celouvskyopoulos It depends on who flagged. Mods will generally not undo bans imposed by other mods as a mark of mutual respect. If it was jut 10K users then an ameanable mod may be prepared to lift the ban.

But Bernardo's comment was a bit close to the edge.

0celouvskyopoulos

Hmm, what does linear growth mean on a manifold?

Jaime Gallego

I did not take offense on what Bernardo said.

0celouvskyopoulos

Probably $Hf=0$.

@JaimeGallego Other people might have.

ACuriousMind

Well, someone flagged it. I would not have flagged it or acted on it because it was rather obviously meant as a joke, and I'd be interested in hearing the flagger's perspective here.

Slereah

4:56 PM

So is there any fundamental difference between the Kolmogorov probability and the von Neumann one

According to one site, it is

THE BELL INEQUALITY

Is that correct

Bernardo Meurer

Hello back

Thanks @mods

John Rennie

Back from the dead!!! :-)

ACuriousMind

@BernardoMeurer You can talk again, but apparently you should watch your language a bit more.

0celouvskyopoulos

@ACuriousMind I don't think we should be threatening people with physical violence in chat.

Bernardo Meurer

@ACuriousMind I will try not to threaten people with violence

0celouvskyopoulos

4:57 PM

@BernardoMeurer Thank you.

Slereah

I will threaten u all

With my rippling musculature

ACuriousMind

@0celouvskyopoulos And you feel you don't know Bernardo well enough to politely ask him to not do that instead of flagging?

Bernardo Meurer

@dmckee @JaimeGallego Come to think of it, that was bad language, when I said "some gets trickery" I meant something from the gets "family" of functions

0celouvskyopoulos

^ inappropriate

It has been made very clear that accusations of trolling go agains the Be Nice policy.

@BernardoMeurer Please remove that.

John Rennie

→ 1 message moved to Trash

Slereah

4:59 PM

@ACuriousMind So, after Morte and Fall from Grace

Who will be your next avatar

John Rennie

Settle down children

2

Bernardo Meurer

@0celouvskyopoulos Go back to your cave

Slereah

Will it be Nordom

0celouvskyopoulos

@BernardoMeurer What does that even mean?

Slereah

Will it be Ignus

Secret

5:00 PM

Anyone here have experience with PBS Professional?

Slereah

@0celouvskyopoulos Plato reference, probably

ACuriousMind

@Slereah I was Ignus already, that's where ocelot's idea of me being Alicia Keys comes from :P

Bernardo Meurer

@JohnRennie Alright, so back at, it, what's the best way of doing this?

user228700

Pretty sure there are at least 10 ghosts in my vicinity at the moment :-/

Bernardo Meurer

Should I load it all into a buffer

count the number of words as the number of \n

John Rennie

5:01 PM

@Kaumudi.H why? What's happening?

ACuriousMind

@Slereah I can't remember exactly, but I think I also used Nordom for a while

Bernardo Meurer

What the hell, why am I having such a hard time with such a silly thing

user228700

@JohnRennie Oh, nothing at all. I just happen to be inside a forest (at night).

John Rennie

@Kaumudi.H remember what happened to Red Riding Hood!

If you see any grandmothers then run away fast. No, hang on ...

user228700

@JohnRennie Ah, u'd make a fine grandpa indeed!

AccidentalFourierTransform

5:02 PM

what does that mean?

Anonymous

@Kaumudi.H Look back. You might see someone following you ;).....

Slereah

The Lie algebra of a Lie group is the tangent space at the identity element

user228700

@blue Nah, I'm on the front porch; the house is behind me :-P

Slereah

Then you can regain the whole group via the exponential map

So basically they are vectors

John Rennie

@BernardoMeurer: reading in the whole file is always a bit risky unless you're sure it's always small enough to fit in memory. But in this case I don't see what the alternative is. If you want an array of words they'll all be in memory anyway.

AccidentalFourierTransform

5:04 PM

yeah, but what do they mean by "SL seen as an abelian vector group"?

it is most certainly not an abelian vector group

Anonymous

@Kaumudi.H Your eyes are deceiving you :P Have a nice time with the ghosts.....

AccidentalFourierTransform

or is it?

John Rennie

@BernardoMeurer: if you're going to read the whole file into memory anyway the quickest and simplest method is the one I described. It's one malloc and one file I/O operation for the whole data.

user228700

@blue :-/ Thanks(?)

Anonymous

Good time to watch Conjuring 2 actually :D

Jaime Gallego

5:06 PM

@BernardoMeurer How are you going to use the array words? Depending on what the operations you are going to perform are, there might be more exotic ways to store the data

user228700

@blue No, no, I'm good.

Slereah

How... how does wavefunction collapse work for formalisms without wavefunctions

How do you perform a wavefunction collapse on a path integral

AccidentalFourierTransform

anyone? :-(

John Rennie

@Kaumudi.H that's what they always say in the horror films just before the giant spider with the dripping fangs attacks!

ACuriousMind

@AccidentalFourierTransform It sounds like total non-sense to me, where did you get that from?

user228700

5:07 PM

@JohnRennie -_- Thanks.

John Rennie

Don't worry though, the giant spider is no danger.
The giant rat ate it! :-)

AccidentalFourierTransform

from arxiv.org/abs/1610.08526

page 87

Slereah

>product of a group and an algebra

what

AccidentalFourierTransform

right lol

didnt notice that one

Slereah

Yeah i'm not sure what that's about

Maybe $\mathfrak{sl}$ is also a group?

I dunno

AccidentalFourierTransform

5:10 PM

I think they explained the notation before

user228700

@JohnRennie I'm not scared of giant animals (other than snakes!). I'm afraid of the quintessential lady in the wide dress.

AccidentalFourierTransform

let me see

Slereah

I kinda hate products on math structures because they're always a bit ambiguous

John Rennie

@Kaumudi.H lady in the wide dress?

Slereah

You don't know how much structure the product means

ACuriousMind

5:11 PM

@AccidentalFourierTransform Ohhhhh

I know what they mean

Slereah

Do tell us!

user228700

@JohnRennie Yes. Around here, a ghost is represented by a lady in a white dress with anklets on that alert you when she comes around to eff with you.

AccidentalFourierTransform

yes pls

user228700

White Lady (ghost)

A White Lady is a type of female ghost dressed in all white reportedly seen in rural areas and associated with some local legend of tragedy. While White Lady legends are found in many countries around the world, they are most prominent in parts of the United States, Ireland and Great Britain. Most famous is the lady in white in Rochester New York. Common to many of these legends is the theme of loss of a daughter or betrayal of a husband, boyfriend or fiancé. == White Ladies in different parts of the world == === United Kingdom === In popular medieval legend, a White Lady is fabled to a...

user228700

Huh. You should be knowing about this...

John Rennie

5:13 PM

@Kaumudi.H if a ghost appears just start explaining electrodynamics to it and it will flee in terror.

Bernardo Meurer

@JohnRennie Okay, i think I'm closer now

user228700

@JohnRennie Hehe, alright! x'D I'll tell u if it works.

ACuriousMind

@AccidentalFourierTransform We have $\mathfrak{sl}(2,\mathbb{R}) = \mathbb{R}^3$ as vector spaces. Now, the action of the $\mathrm{SL}(2,\mathbb{R})$ in that semidirect product on the $\mathbb{R}^3$ is precisely such as if the components of the $\mathbb{R}^3$-vector were the coefficients of the standard basis of $\mathfrak{sl}(2,\mathbb{R})$ and the SL were acting by the adjoint action

Bernardo Meurer

FILE *wordfile = fopen(path, "r");
    //Count lines in file
    int ch, lines = 0;
    do {
        ch = fgetc(wordfile);
        if (ch == '\n' || ch == '\0') ++lines;
    } while (ch != EOF);
    rewind(wordfile);
    //Get file size
    fseek(wordfile, 0, SEEK_END);
    *length = (int) ftell(wordfile);
    rewind(wordfile);
    //Allocate array
    char **words = calloc(lines, sizeof(char*));
    if(words == NULL) catchError("Memory allocation failed");
    for(int i = 0; i < lines; ++i){
        words[i] =

@JohnRennie Now I gotta figure out how to allocate the words with the right size

so I can use getline() after

AccidentalFourierTransform

@ACuriousMind hmm but what is the addition rule on $\mathfrak{sl}(2,\mathbb R)$?

how is it even defined?

Bernardo Meurer

5:15 PM

Ah!

I don't need to!

ACuriousMind

@AccidentalFourierTransform Every Lie algebra is a vector space, no?

Bernardo Meurer

getline allocates it!

beauty

AccidentalFourierTransform

how do you add two $\mathfrak{sl}$ vectors? and what is the multiplication by scalar?

oh fuck

yes, I was being stupid

ACuriousMind

You just add them. The reason why the algebra is easier than the group is precisely that it's a vector space :P

Anonymous

@Kaumudi.H These ghosts should start wearing colorful dresses. White dress ghosts are too common. Ugh, these ghosts have no dressing sense. If you spot one lend them a dress or two :P

Anonymous

5:16 PM

Also feed them well

Anonymous

They must be hungry

AccidentalFourierTransform

yes, I was thinking that given $a,b\in\mathrm{GL}$, $a+b$ need not be in $\mathrm{GL}$. But its about the algebra, not the group :: facepalm ::

thanks!

John Rennie

@BernardoMeurer is getline in C?

user228700

@blue I agree! x'D I'll try...if I get the chance before they go into my body/laugh etc. (Jesus Christ, their laugh scares the crap out of me)

Bernardo Meurer

@JohnRennie yes

I hope

Yes

Slereah

5:17 PM

did u remember to thank Muḥammad ibn Mūsā al-Khwārizmī for inventing algebra

ACuriousMind

@AccidentalFourierTransform Still, that section is written rather weirdly, it could have been expressed a lot better.

user228700

Yeah, no, I should stop talking about them :-P I really am in the middle of a forest.

DanielSank

Hi, everybody.

Slereah

Hey @DanielSank

user228700

Yello!

Slereah

5:18 PM

I'm afraid I'll have to read a bit of your thesis

John Rennie

@BernardoMeurer ah yes, it is.

AccidentalFourierTransform

15 mins ago, by AccidentalFourierTransform

Slereah

Apparently there's some quantum measurement stuff in it!

DanielSank

@Slereah Why afraid?

AccidentalFourierTransform

$\mathfrak{ls}$ is the abelian vector group, not $\mathrm{SL}$

DanielSank

5:18 PM

@Slereah Perhaps. What exactly are you looking for?

Slereah

There might be ghosts in it

@DanielSank What the hell a measurement is

What count as a measurement for quantum systems

DanielSank

The measurement stuff in my thesis is somewhat "engineeringy". If you are interested in more quantumy things, I can direct you to interesting resources.

@Slereah Ah.

Slereah

Is it any interaction between the two systems?

Something that make an entanglement develop?

DanielSank

@Slereah No.

You need entanglement.

Slereah

What interaction produces no entanglement?

DanielSank

5:19 PM

(which usually, but not always, happens when you couple things)

@Slereah Well, lots of couplings lead to periodic motion, so there are, for example, times when the entanglement is zero.

Slereah

for instance, take two static charges moving through spacetime

DanielSank

In general, coupling --> entanglement.

Slereah

Do they get entangled once they enter into each other's light cones?

0celouvskyopoulos

I am tired of people saying "maximum principle" and then writing some insane inequality

DanielSank

@Slereah Ufff, relativity...

I dunno, probably.

Slereah

5:21 PM

Why do you care about the maximum principle anyway, doesn't it involve the axiom of choice?

Yeah a lot of my questions involve relativity and QFT :p

0celouvskyopoulos

@Slereah wot

Slereah

A lot of quantum measurement stuff is with point particles

DanielSank

I am unfortunately poorly trained in relativity.

Slereah

@0celouvskyopoulos Oh wait, was thinking of another maximum thing

AccidentalFourierTransform

according to recent conversations, you only need to know that $\mathrm ds^2=-\mathrm dt^2+\mathrm d\boldsymbol x^2$

Slereah

5:23 PM

Also a lot of measurement stuff involves splitting the Hilbert space in twain

0celouvskyopoulos

it's magic!

Slereah

Since you consider products of wavefunctions for the whole system

but I'm not sure that works out so great in QFT

Also I'm not sure how the wavefunction collapses work with relativity

Does that work with any spacelike foliation???

DanielSank

@Slereah I have no idea and my thesis won't help.

My thesis tells you how to measure a superconducting qubit really fast and with reasonably high accuracy.

I didn't go into this kind of stuff at all.

Slereah

yeah it's pretty hard to find for some reason

0celouvskyopoulos

I'm convinced people write papers to be purposefully obfuscatory.

Slereah

5:28 PM

A lot of measurement stuff is only for non-relativistic QM

DanielSank

@0celouvskyopoulos Some times.

@Slereah Yeah because the really good experiments are all in that domain.

Slereah

Or if it's for QFT as well, it does not really explain well how it's supposed to work there

@DanielSank Probably!

DanielSank

One of the delights of superconducting qubits is that you can do really good experiments on exactly what happens when you measure.

Slereah

But I'm sure there's plenty of gedanke experiments you can do with it

at least

DanielSank

There was recent work with simultaneously measuring $\sigma_x$ and $\sigma_z$ on a two level quantum system.

@Slereah Oh, for that kind of thing, look up Bell violation papers.

That's a case where people care about space-time separation and all that business.

In fact, I will ask my colleague, who did a famous Bell violation, for references.

Slereah

5:30 PM

thx

AccidentalFourierTransform

IIRC, DeWitt does some analysis on quantum measurements in his book

Slereah

I'm also curious to know how it's supposed to work for fields, too

AccidentalFourierTransform

and the book is about (general) relativistic qm

Slereah

How do you pick an "observer" in a field?

Technically it fills the entire space

or a system, really

It's fairly simple in QM since it's just a point particle and the total system is just a product, but conceptually it seems harder in QFT

DanielSank

@Slereah I have indicated your plight to Dr. Marissa Giustina.

Now I must go.

Slereah

5:33 PM

Thx

Later

DanielSank

Adieu

Slereah

tips hat

AccidentalFourierTransform

ciao

Slereah

If we take a machine and an electron

They are both technically part of the same Hilbert space of electrons

Can you split it to make sense of the tensor product usually used?

Bernardo Meurer

@JohnRennie Here's a nice one, I have a char array containing a string terminated with a newline

How do I remove that trailing newline?

I was thinking of doing a realloc with size-1

Slereah

5:36 PM

Replace the newline with \0

Bernardo Meurer

Then my array is larger than necessary

mathvc_

hey guys. good evening! i've a question. i wanted to calculate the partition function of the hydrogen atom but it diverges. wtf is this?

Slereah

your body is larger than necessary

@mathvc_ QFT?

also via what process

Path integral or operators

mathvc_

statistical mechanics not in QFT

Slereah

Oh

then I don't know

mathvc_

5:38 PM

\sum exp^{\beta E_0/n^2} diverges

it's actually \sum n^2 exp^{\beta E_0/n^2}

AccidentalFourierTransform

10

Q: Partition function of a hydrogen gas

Hi I have a doubt (I'm not very expert in statistical mechanics, so sorry for this question). We consider a gas of hydrogen atoms with no interactions between them. The partition function is: $$ Z=\frac{Z_s^N}{N!} $$ where $Z_s$ is the partition function of one atom. So we write $$ Z_s =Tr\{e^...

@mathvc_ ^

0celouvskyopoulos

Hmm

I can't directly apply the maximum principle

The maximum could be at infinity

mathvc_

@AccidentalFourierTransform hmm it has introduced a revised partition function but i guess i can always find a system (maybe with ridiculous energy levels) that its partition function diverges.

Slereah

The one I was thinking about was Hausdorff's maximal principle, btw

which is ~ AC

0celouvskyopoulos

6:04 PM

Ah

I've needed that before

For the ol Besicovich covering lemma

Slereah

obviously wrong, since it uses the axiom of choice

0celouvskyopoulos

those 1900s covering lemmas are pretty magical

bolbteppa

6:34 PM

spectral density functions, mass has a continuum of states after $2m$ because of relative momenta not contributing to overall momentum, PS 7.1, Sred sec 5. Yeah that makes sense, but how do you see it mathematically?