| |
| |
| |
|
Comments:
<0> how come x = ['a', 'b'] then x.append('c') does not provide x = ['a', 'b', 'c'] ?
<0> nevermind...
<1> Uh, it does.
<0> :D
<0> ok, this is what was screwing me up:
<0> x = ['a', 'b']
<0> print x.append('c')
<2> yes
<2> mutating methods return NOne
<0> fair enough
<0> mutating meaning it changes the object
<2> yes
<0> good stuff, thanks :)
<3> gaa
<3> how to print integer with leading 0? e.g. 01 instead of 1
<2> '%02d' % x
<3> ah, was doing %01, thanks xihr
<3> hmm
<3> is there no way to "freeze" a list so that I can hash it?
<3> I have a list of lists
<2> convert it to a tuple
<2> (recursively)
<4> is there a way I can do a dns lookup without blocking or starting a new thread?
<2> it's certainly possible but I don't know if there's anything available in the standard library for that
<4> there doesn't seem to be
<4> I'd be happy with an optional library
<4> I can't find any documentation for any python dns libraries
<2> I don't know of one offhand
<2> the calls in socket are just wrappers around the C library routines, which also definitely block
<4> yeah
<4> I've managed to move to nonblocking sockets, but that doesn't help the dns
<2> typically in C apps it's done in a separate thread or tasked for later, etc.
<2> or a separate process
<4> I'm doing the separate thread thing
<4> but apparently threads take a lot of overhead..
<2> in what sense
<4> writeable memory
<4> whatever that means
<4> I'm told it's important
<2> well you can measure the memory increase but it's unlikely to be catastrophic
<2> if one thread is going to kill you then you shouldn't be using Python in the first place :-)
<4> it's more like 6 threads
<2> still, no big deal
<4> because of doing the dns lookups at the same time
<4> it's 8 MB of writeable memory for every thread open at the same time
<4> and it never goes down when the threads are done
<4> if I can save 8 MB of memory, I should :p
<2> well I'm not sure where you're getting your numbers
<4> I'm not either
<2> I have an app that spawns up to 10 concurrent threads and I see its memory footprint range from 14 MB to 30 MB up and down as threads are created and destroyed
<2> a bare Python interpreter takes 5 MB
<4> pmap apparently
<2> like I said, if you're really this concerned about memory, it's likely that Python (or any other high-level language for that matter) is the wrong choice in the first place
<4> things like dynamic behaviour are much more important than memory
<4> also, now that I'm in the threads-killing game, no amount of evidence or logic will convince me that it's ok to leave the threads
<2> I have no idea what you're talking about at this poitn
<4> I'm talking about the fact that I'm now more out to stop using threads out of spite than actual performance anyway :p
<4> even though the performance concerns still exist afaik
<2> I just proved that they don't exist to the extent you thought they did
<2> it's not like starting a separate process for each DNS lookup is going to be less expensive
<4> true
<4> but there's no reason I should need to do either of those
<2> then you'll have to write your own DNS lookup library
<4> how are you gauging the memory usage on your program?
<4> "gauging" may be misspelled..
<2> top
<4> what does top tell you?
<4> physical memory
<4> the problem is I don't know what any of this stuff means :p
<4> I just go by what other people tell me
<2> wheeeeee
<5> hello
<5> hi
<5> hey
<5> woo
<5> someone has a great link to a nice tutorial to glade + python?
<5> google is not showing me love
<6> thy shalt loveth the Google before it loveth you.
<5> i found it i think
<5> you are lucky you have looks
<6> heh
<7> is there any module in python that talking with PPT ?
<6> PPT? parallel port?
<7> PowerPoint
<7> MS PowerPoint
<6> no idea, i don't use powerpoint
<7> i see.
<5> omg omg it works
Return to
#python
or
Go to some related
logs:
gentoo freeze.py
mrskin wordlist
delphi String.Format
#beginner
florecent light max
#freebsd
#nhl
dweckl chicago
pc400 system board m925lr
illegal mexican crossing the boader
|
|