| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Comments:
<0> hehe
<1> sell me some souls.
<0> I don't like some python syntax
<2> ironfroggy: you got enough of them yesterday...
<0> but it is much better then c/c++
<1> i never have enough
<1> sunnmover: dont take this the wrong way, but DUH
<2> ironfroggy: frog...
<0> :)
<2> sunnmover: well that is just a pure opinion ;)... personally I like both... the only problem with C++ is that you have to do everything the difficult way... the good thing about it is that you get some speeeed....
<0> here comes a stupid question: does a computer need a python installed in order to run programs written in python?
<0> roc: ya
<3> sunnmover: There are packages that transform python programs into monolithic executables.
<3> What don't you like about the Python syntax?
<3> It's okay. Not as beautiful as LISP, but it has its charm.
<0> i am just studying the online tutorials
<4> freezing binaries is quite uncommon however sunnmover, generaly it needs to be installed
<5> dungodung: no. that only defines the cl*** and two methods. it creates no thread. is line 62 the only line with a call to timerr()?
<1> roc: the day comes when C++ isnt worth the trouble anymore.
<0> and some bits don't look appealing
<3> C and C++ are still good for library integration.
<3> There still aren't good bindings for some C libraries.
<4> i think comparing c/c++ with python is silly
<2> ironfroggy: I hope to see the day... when highlevel languages become faster than low level ones...
<4> you use the best tool for the job
<3> roc: Pfft.
<3> Well, maybe not pfft.
<0> see, that is what I am trying to figure out
<1> roc: then help out with PyPy
<6> ChrisLong: no. it's activated in "def on_welcome(self, c, e):"
<4> if you need to do kernel programming or memory manipulation, you use C
<3> But we need _much_ better languages.
<2> oliver: no matter how you do it as long as it gets done ;)
<4> if you want to code a quick socket server
<4> use python
<2> ironfroggy: that project was doomed when it was started...
<3> IMHO, the languages with the greatest potential for optimization are the functional languages.
<0> I want to create an audio sequencer
<1> roc: what are you talking about?
<3> Notwithstanding that you'd need an optimizer several times better than we have today to get decent performance out of them.
<0> I am wondering what high level language to study for that
<2> ironfroggy: maybe if I just shut up...
<7> sunnmover: you won't get unbiased advice in here
<1> roc: did you have an honest critisism?
<8> you won't get unbiased *and* knowledgeable advice anywhere.
<5> dungodung: is on_welcome a method of the cl*** which should tell the thread to die?
<0> so normally, a python app need python installed on a computer to run?
<8> (but, for what it's worth, Ruby's object model is just like Python's)
<7> sunnmover: no.
<1> sunnmover: the same is true for ANY language.
<8> sunnmover: it does, for some value of 'installed'.
<0> I thought python is somethin like .NET
<7> "your app + python.dll"
<2> ironfroggy: almost... for some reason I just don't have faith on pypy...
<7> or whatever
<9> ugh
<8> sunnmover: not particularly, although there's two Python-on-top-of-.NET implementations.
<2> krooh...
<7> the official bittorrent client is pythong and it doesn't need python installed on the machine to run
<7> hah, py-thong
<8> Patrick`: it does, it just installs Python inside the .exe.
<2> Patrick`: LOL
<1> sunnmover: even if it was, what difference does that make? you need .Net installed for any .Net program, just the same.
<7> Yhg1s: er
<7> Yhg1s: installed deparately
<7> *separately
<1> roc: no aolspeek
<10> hrm... I need to output the time it took to run a block of code, in a nicely-formatted way... so far, I'm calling time.time() at the beginning of the block, then again at the end, then subtracting end from start. This gives me the time elapsed in seconds, but I'm finding it difficult to format it nicely... anyone have any ideas on how to do this elegantly?
<0> patrick: oh, so you can make that little python.dll part of your app so users won't need to worry about installing python?
<7> something functionally equivalent, yes
<8> Patrick`: hence 'for some value of installed'. you still need the compiler and interpreter and most of the standard library somewhere on your machine. there's no fundamental difference between 'properly so everyone can use it' and 'restricted to your one application, giving you maintenance hell'.
<1> sunnmover: python.exe is more what you need to include with it.
<7> I'm pulling names out of thin air but yes, you can have one installer package with your application
<7> find out how the bittorrent.com client does it
<0> cool
<7> there's py2exe as well, but that just bundles it all in one thing
<7> and only for windows
<7> similar things exist for macos
<0> user integration is important
<8> it does it using 'py2exe'.
<0> but the cool think about python is easy porting
<0> thing
<7> semantics of what *actually* happens under the hood aside, it is possible to have users install only one installer.
<2> Yhg1s: tried that pretty many times but didn't work out... maybe I'm just so stupid... well anyway you could always code your own program that runs yer python scripts... and include the python dll...
<0> I see
<7> rather than, say, two
<8> roc: or you could just do the smart thing, and install Python once, system-wide.
<0> ok
<7> Yhg1s: that's kind of pointless for end-users
<2> Yhg1s: system-wide?...
<8> Patrick`: perhaps to you, not to me. I prefer my end user installing Python. it doesn't hurt them, and it means they can download as many applications in the form of .py files as they want, and they can all play nicely together.
<7> yeah, it'd be nice
<8> Patrick`: and it just means one (1) installer more, once (1 time.) With proper docs, too.
<7> but my end users aren't computer scientists
<7> they don't care.
<7> they flounce and sigh about the h***le of two installers
<8> if they can double-click on one .exe file, they can double-click on a .exe file and a .py file.
<7> I'm not taking python's portability and turning it into another java disaster
<11> can't the python .msi be set to install python silently?
<3> Will Python gain full closures any time soon? :-)
<11> full closure?
<3> Lexical closures, ala Lisp.
<11> I don't think so
<3> Ahwell. That's the only feature I really miss...
<6> ChrisLong: it's in the main cl***
<0> http://www.ruby-lang.org/en/20020101.html the feature list is interesting
<0> hehe
<8> quotemstr: writable closures, you mean. they're not actively persued by anyone. use a cl***, isntead.
<12> hello, I'm having some problems with validating form data in my cgi script-- especially not having my previous form information erased when I use cgi.FieldStorage()
<0> not to start a versus war
<12> the problematic code can be found here: http://deadbeefbabe.org/paste/733
<12> I would be very grateful for any help
<3> Yhg1s: Yes, I know... It's a shame, though. You're right --- a cl*** has every facility you'd need to implement a lexical closure; the interpreter is willing, but the syntax is weak.
<3> It's just a shame.
<8> quotemstr: no, the syntax is simple, by design. many people prefer to keep it that way.
<13> platen, why are you calling cgi.fieldStorage() twice?
<5> dungodung: so this is a 'no'?
<14> platen: i'd suggest a more pythonic style than has_key(); use if key in dict instead.
<6> ChrisLong: I guess
<11> I don't like the Lisp way myself
<3> Yhg1s: Then why was the abominable decorator syntax adopted?
<12> polpak: otherwise the loop is indefinite
<3> I generally like Python. :-)
<14> ;)
<13> platen, you can't do cgi like that
<5> dungodung: you guess? you don't know if the main cl*** is the cl*** which decides if the thread has to die?
<12> polpak: any suggestions?
<13> platen, someone submits the form, the web server runs your cgi program. the cgi gets that one set of input
<13> platen, if it's not the input you need, then you have to report the error and send them back to the form page
<6> ChrisLong: hehe... well the method that executes that elif cmd.lower() == "die": is in that same cl*** as the method that starts
<12> ohh ok
<12> polpak: thanks
<9> quotemstr: the discussion about it last half an year, and adopted it to end the discussion
<8> quotemstr: you can call it abominable, but it isn't complex. see the python-dev archives.
<13> platen, your program isn't interactive. It just handles one submission at a time
<13> platen, if it's good, it does the "right thing"
<13> platen, if not report the error and exit
<12> platen, yeah that makes sense... must be the fact that this is my first CGI script
<5> dungodung: if it's the same method that starts the thread, then it knows the thread. so why can't it set it t.stoppity where t is the thread?
<6> ChrisLong: it's not the same method... it's the same cl***, though
<5> dungodung: then, in the method creating the thread, store it in an attribute [s.th like self.my_timer_thread = timerr(...)] and in the method trying to stop the thread do self.my_timer_thread.stoppity = True. What is the problem?
<15> is wikipedia coming up empty for anyone else?
<15> i'm getting an empty page on every request
<3> MFen: Same here.
<15> hmm
<16> yeah, it's not well
<6> ChrisLong: I'll try
<2> me too... blank...
<3> Their backend is down.
<3> I'm getting HTTP headers indicating some kind of cache miss.
Return to
#python
or
Go to some related
logs:
#osdev
initrd md genkernel
example pymatlib
pycario
.ICEAuthority ownership mysql
#kde
debian vmware no screens found
#physics
apt-get install ajunta
inetd xwin fedora
|
|