Metanet Programmer Directory

[otters]

Why does === exist?

So you can check for identical types as well as values? O_o

Why doesn't == do that? O_o

[smartalco]

Because more often you would want to know something like if the string '0' is equal to the value zero.

[Tanner]

Because more often you would want to know something like if the string '0' is equal to the value zero.

Wait. So comparing a string of "0" and an int of "0" would return that they're the same? I'm thoroughly opposed to this idea.

[otters]

Because more often you would want to know something like if the string '0' is equal to the value zero.

Wait. So comparing a string of "0" and an int of "0" would return that they're the same? I'm thoroughly opposed to this idea.

Exactly. If you want to figure out of the integer value of "0" is 0, then cast it as a gosh-darned integer or something. This is part of my "babysitting the programmer" idea. Another is:

Code: Select all

<?php # prints out nothing
if("0") print "hello";
if(0) print "hello";
if("") print "hello";
if(0.0) print "hello";

[sidke]

Wait. So comparing a string of "0" and an int of "0" would return that they're the same? I'm thoroughly opposed to this idea.

Exactly. If you want to figure out of the integer value of "0" is 0, then cast it as a gosh-darned integer or something. This is part of my "babysitting the programmer" idea. Another is:

Code: Select all

<?php # prints out nothing
if("0") print "hello";
if(0) print "hello";
if("") print "hello";
if(0.0) print "hello";

php doesn't have type definition... so instead of complaining, you should thank php for being flexible like this, learn its quirks, and get over them.
because if you can't, you use a different language. simple.

[t̷s͢uk̕a͡t͜ư]

php doesn't have type definition... so instead of complaining, you should thank php for being flexible like this, learn its quirks, and get over them.
because if you can't, you use a different language. simple.

The point is that this is bad programming practice, and PHP, by encouraging this kind of bullshit, will build bad programming habits.
But I guess this really isn't that serious of a problem if you're planning on never using any other language, collaborating with others, or working on a large project.

As for the distinction between equality and identity, that's actually a valid issue. If that's the difference between the == and the === operator in PHP, I have nothing against that. In Python, that difference is the == operator versus the is operator. In C, it's the difference between comparing values and comparing memory addresses. In Lisp, it's the difference between equal and eq. It's definitely an important distinction.

[sidke]

The point is that this is bad programming practice, and PHP, by encouraging this kind of bullshit, will build bad programming habits.
But I guess this really isn't that serious of a problem if you're planning on never using any other language, collaborating with others, or working on a large project.

because there've never been large-scale web development projects done in php using - holy shit, get this - a repository that lets teams of developers work on the same project aka collaborating. i think you're funny.

it was inevitable that a programming thread would end up a language bashing thread.

[Geti]

I'm just surprised we haven't got back onto editor wars yet.

@Tsukatu are you opposed to dynamically typed variables or just PHP's implementation? I've always found them both a blessing and a curse, but if you're smart about how you're coding something there's no issue with them. Unless I'm missing your point entirely O_o

[otters]

I'm just surprised we haven't got back onto editor wars yet.

TEXTMATE \m/

[scythe]

C and Lua. Not C++, though it's a decent language that is simply misunderstood.

[smartalco]

Dynamically typed variables are the shit, as the kids say.

[scythe]

Dynamically typed variables are the shit, as the kids say.

And they can be fast, too.

[t̷s͢uk̕a͡t͜ư]

@Tsukatu are you opposed to dynamically typed variables or just PHP's implementation? I've always found them both a blessing and a curse, but if you're smart about how you're coding something there's no issue with them. Unless I'm missing your point entirely O_o

I'm talking about the entirety of PHP's style.

Not C++, though it's a decent language that is simply misunderstood.

http://harmful.cat-v.org/software/c++/coders-at-work

[Geti]

Ah, fair enough. To each his xor her own though. Interesting read in the article. Everyone there's thoughts on complexity seem to be mimicking mine, so that's a good sign.

@scythe: wait, justintime compilation has nothing to do with dynamic typing though, it's just cramming everying into bytecode at runtime so you aren't iterating through code, you're iterating through what it means to the machine. Sure, faster, and yes, marvellous idea, but not really to do with dynamic type.
Actually, most interpreted languages (the ones that benefit from all that JIT goodness) nowadays are dynamically typed, so maybe that's what you were getting at?

TEXTMATE \m/

Pff I think you mean geany for graphical editors (though, if anyone has another suggestion for lightweight *nix IDEs I'm open, geany does its job pretty damn well though). I'm undecided as far as cli editors go.

[t̷s͢uk̕a͡t͜ư]

Anyone else tried out Google's Go?

I am unenthusiastic after my first foray:

[scythe]

Anyone else tried out Google's Go?

I am unenthusiastic after my first foray:

Took a look, wasn't interested. I was hoping it would be like OCaml, only without that horrible abortion of a syntax that OCaml uses (OCaml makes Perl look elegant). Instead it turned out to be compiled Java, only not a total piece of shit, but twice as slow.

[t̷s͢uk̕a͡t͜ư]

[Go] turned out to be compiled Java, only not a total piece of shit, but twice as slow.

You forgot "and 2,500 times the size" (javac'd Hello World was ~400 bytes).

[GamingWolf]

http://harmful.cat-v.org/software/c++/coders-at-work

Oh hey there was other articles on that site that I probably needed to read. Thank you for that.

[t̷s͢uk̕a͡t͜ư]

http://harmful.cat-v.org/software/c++/coders-at-work

Oh hey there was other articles on that site that I probably needed to read. Thank you for that.

Yeah, I've been using the information on that site to look into stuff I should know but have little means of finding out about in the first place.
I'm still slightly skeptical of some of its criticisms, but I'm always interested in gaining perspective, so that list of better practices is something of a "to learn" list for me right now.

[taaveti]

http://harmful.cat-v.org/software/c++/coders-at-work

Oh hey there was other articles on that site that I probably needed to read. Thank you for that.

Yeah, I've been using the information on that site to look into stuff I should know but have little means of finding out about in the first place.
I'm still slightly skeptical of some of its criticisms, but I'm always interested in gaining perspective, so that list of better practices is something of a "to learn" list for me right now.

Yeah, a lot of their criticisms seem to be based on a really narrow world view (e.g. advocating against HTTP or dynamic libraries). There are certainly some valid points made there, but there are a good number of invalid ones too. The fact that they suggest that exchanging tarballs and patches could be better than subversion counts pretty heavily against their credibility (if they had stopped at git I could've called it a matter of opinion or filed it under "narrow world view", but having used CVS, svn, and plain archives I have to say that svn is the clear winner of the three).
That said, the general idea of reducing complexity is definitely one that needs to be stressed. I may not state it as a flat "c++ doubleplusungood", but I would caution someone not to get so absorbed in the OO paradigm that they ignore the obvious and simple answer in pure c.

[mediate]

I'm to learn C++ at university. That would be my first programming language. Right now, I'm learning and using Maple, which is crap-on-wheels.

[scythe]

Not C++, though it's a decent language that is simply misunderstood.

http://harmful.cat-v.org/software/c++/coders-at-work

Qt, though, is awesome. I can agree that C++ is all-too-often terrible, but in the form of Qt it's great.

[t̷s͢uk̕a͡t͜ư]

Extremely relevant to previous conversation:
http://www.reddit.com/r/programming/com ... variables/

And if you're particularly interested in this specific problem, some further reading.

[smartalco]

Extremely relevant to previous conversation:
http://www.reddit.com/r/programming/com ... variables/

And if you're particularly interested in this specific problem, some further reading.

That is fucking awesome. And by 'fucking awesome' I mean 'fucking awesome if you want to write some really fucked up obfuscated code, or just play around with a language'. Not something I would ever consider using in anything I actually wanted to function.

[t̷s͢uk̕a͡t͜ư]

Extremely relevant to previous conversation:
http://www.reddit.com/r/programming/com ... variables/

And if you're particularly interested in this specific problem, some further reading.

That is fucking awesome. And by 'fucking awesome' I mean 'fucking awesome if you want to write some really fucked up obfuscated code, or just play around with a language'. Not something I would ever consider using in anything I actually wanted to function.

"Awesome." There's an interesting word to reach for. I was thinking more along the lines of "horrifying".