| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Comments:
<0> it uses threading, not thread
<1> gzl: What's the source of your list of tuples?
<0> * Pythy, that doesnt help
<0> :(
<0> can someone help with thread, not threading?
<0> http://pastebin.com/624101
<0> thats my code
<0> thread.start_new_thread(def, tuple) doesnt work
<2> what error are you getting?
<0> sysfault, no error
<0> it just doesnt run
<3> Whats the function to turn a float into an int?
<0> eugman, int(float)
<2> int(float)
<3> thanks
<0> sysfault, you can see i added 5 lines that print "asdf"
<0> which it doesnt
<2> r00723r0: is serverfile empty after the readline()?
<4> is there a way to make ConfigParser not lowercase my keys and add spaces around the '='?
<0> sysfault, no
<2> how do you know what serverfile contains, how are you ensuring it isnt?
<2> dict insuring*
<0> sysfault, because i did .readline() every line before
<0> and it worked
<0> sysfault, the problem is not reading the file
<0> the problem is making the thread start
<0> the "asdf"'s dont print
<0> you see what i mean?
<1> Kosh: And if there was a chance where a more-general solution than what was requested might ever be needed (I'm curious to see gzl's answer, to learn if there's even a remote chance of that here), my earlier solution is as general as yours.
<0> sysfault, ?
<0> ...?
<0> sysfault?
<5> Pythy: your imap one I am fine with , just not the most recent weird one
<5> Pythy: I have just found from experience that you can't know that in 5 years the fields will need to change
<0> can anyone help with thread?
<1> Kosh: We'll have to wait on gzl's answer to find out.
<0> goddamnit if someone helps me with thread ill leave quicker
<5> I don't use threads, I can't help you
<0> anyone else?
<0> want to get me out of here faster?
<5> it doesn't look like anyone on right now knows the answer
<5> most people don't use threads
<5> they usually just make the problems worse
<0> are you kidding me?
<5> and it won't perform any better
<0> most people dont use threads?
<0> lol.
<5> no I am not kidding you
<0> kosh, well then most people dont write applications
<6> kosh: that's a slightly inflamatory statement about threading, even if I tend to agree that they make things more complicated in most cases.
<7> threads == trouble
<0> threads ARE troublesome but threads are important
<6> with an async framework like Twisted, threads become a joke
<1> Kosh: For example, if those are key/value-pairs from a dictionary, I don't expect them to one day soon grow a third value. And if his design changes that radically, then, swapping one solution for another here is the least of is his troubles. (Also, note my "*if* speed is an issue..."-qualifier. He gets to chose the more appropriate answer.)
<5> danderson: I would say about 5% of cases that threads are used that they are good idea for
<0> danderson, because Twisted is threaded....
<5> umm twisted is not threaded last I checked
<6> same here.
<0> quite basically anything that is async is threaded
<0> because its fork()ed
<6> Twisted is specifically *not* threaded
<6> that's the whole point of twisted
<5> r00723r0: you are confusing things
<6> a single reactor, blocking on all events from the whole app
<5> threads mean something very specific
<6> and handing events out to callbacks when needed
<0> threads are VERY broad
<6> (and with a great Deferred system to handle all that gracefully)
<5> twisted is an async framework
<5> no threads are not broad, they mean one particular thing
<0> how could it be async without threads?
<6> clever design.
<0> please explain, im listening
<6> The way people used to do it *before* threads
<5> threads do not refer to multiprocess apps for instance and they are most certainly async and scale
<6> r00723r0: you know select(), for blocking over multiple sockets?
<0> ok, fine, multiprocess
<0> yes, thats fork()ed
<5> actually I don't know of any good way to use threads with imperative style languages
<6> Twisted is the same, for more general events
<0> and fork() basically means threading
<0> it starts up a new thread
<6> no it most certainly doesn't
<5> fork DOES NOT MEAN THREADING
<6> it involves a host of different problems and solutions
<5> they are entirely different
<0> ok, ok, i dont care WHAT threads are
<0> i need a solution to my problem
<5> they have entirely seperate address spaces, how memory is used etc
<0> thread.start_new_thread(def, tuple) does not really...
<0> work
<0> even though it works in the python cli
<5> to be honest I am surprised you can put def in there and have it work
<5> since def is a builtin to create functions
<0> yeah.
<0> but i wanted to separate it
<0> it makes reading code easier
<5> no the point is that you should not use the word def
<0> kosh, it requires a def
<5> you should not use tuple either
<0> im sorry, arg1 and arg2
<0> please stop being so pedantic
<8> r00723r0: it matters
<0> MFen, you know what i mean though
<8> r00723r0: it seems pedantic until you've seen people make this mistake time and again and come in here asking for help fixing syntax errors
<5> r00723r0: most bugs come from these kinds of issues and I did not know what you meant
<2> it was an example.
<8> r00723r0: yes, and nobody blames you for typing that, but don't, because it confuses the issue
<0> i dont have a syntax error, though...
<2> he's not using 'def' within the start_new_thread() method.
<0> im using a def
<0> but not "def"
<2> a function, but not 'def'
<0> yes, thanks sysfaul
<2> im looking at your code.
<0> t
<0> sysfault, ahh, ok
<2> too vague.
<9> hey if I have a string that represents a member variable in a cl*** instance, what's the proper way to get the value of that member variable?
<1> getattr(p, 'mtd')
<5> getattr(self, stringnameofvar)
<0> whoa...
<9> ah right, of course, thanks.
<0> python is cool.
<4> how can i make ConfigParser not lowercase my keys and add spaces around '='s?
<5> remember the purpose of source code is to communicate with other people and yourself later so make it as easy to read as possible
<0> kosh, me?
<5> just in general
<0> right
<5> longer variable names make it clearer what you mean and easier to debug and figure out later
<0> kosh, my variable names arent vague to the least bit
<4> i've got things like self.applications.visible_tree, self.apps.vtree would be less typing but harder to figure out later
<4> :)
<2> r00723r0: sure they're.
<0> really...?
<2> write, and read are file object methods..
<2> and tons of other modules.
<2> pretty ubiquitous, not really distinguishable.
<0> so any luck?
<2> process of elimination may help, write a small script with just the threads, use the same method, and see if it works.
<0> can i write inline defs?
<10> good morning
<8> r00723r0: if your function can be made a single expression, you can use a lambda for your 'inline def'
<8> r00723r0: if not, you can put a def statement anywhere you want
<2> evening parks.
<0> awesome
<8> and it will exist inside that scope
<8> r00723r0: ditto cl***es
<2> last game of americas-army for at least a month.
Return to
#python
or
Go to some related
logs:
#python
supaplex ubuntu
linux-k8 ubuntu
named.conf dlz
pydev UTF-8 greek
troubleshooting mupen64 gentoo
cartoon strength of ten bears
kdelibs missing keyword gentoo
#perl
#linux
|
|