f = open("shoot.wav")
s = f.read()
f.close()
print len(s)
print len(repr(s))
print len(eval(repr(s)))
217
675
217
On Thu, Feb 28, 2008 at 7:11 PM, Douglas Bagnall <[EMAIL PROTECTED]>
wrote:
> PyMike wrote:
>
> > After doing that, this is the output:
> >
> > snd = 'RIFFZB\x00\x00WAVEfmt
> >
> \x10\x00\x00\x00\x01\x00\x01\x00D\xac\x00\x00\x88X\x01\x00\x02\x00\x10\x00data6B\x00\x00y\x18\x90\x17\xaf\x16\xd7\x15\x07\x15>\x14}\x13\xc3\x12\x10\x12d\x11\xbe\x10\x1f\x10\x85\x0f\xf1\x0ec\x0e\xd9\rU\r\xd6\x0c\\\x0c\xe6\x0bu\x0b\x07\x0b\x9e\n\xd0\xfd\\\xd9\xcc\xda/\xdc\x84\xdd\xcd\xde\t\xe0:\xe1`\xe2z\xe3\x8a\xe4\x90\xe5\x8c\xe6\x7f\xe7h\xe8I\xe9"\xea\xf2\xea\xbb\xeb|\xec6\xed\xea\xed\x96\xee<\xef\xdc\xefv\xf0\n\xf1\x99\xf1"\xf2\xa6\xf2%\xf3\xa0\xf3\x16\xf4\x88\xf4\xf5\xf4^\xf5\xc4\xf5%\xf6\x83\xf6\xde\xf65\xf7\x89\xf7\xd9\xf7\'\xf8r\xf8\xba\xf8\xff\xf8B\xf9\x82\xf9\xc0\xf9\xb0\x12<*\xa9(&\'\xb0%I$\xef"\xa2!a
> > -\x1f\x03\x1e\xe5\x1c\xd2\x1b\xc8'
>
> That's way too short. Are you sure shoot.wav is complete? What is
> len(s), len(repr(s)), and len(eval(repr(s))? The last should be the
> same as the first, and the same as the file size reported by the OS. The
> middle one will be about 3 times as long.
>
> Also, just to be sure you got this, where you go:
>
> wav = StringIO(snd)
>
> That snd shouldn't be the same as repr(s), but the same as s. repr(s) is
> just a convenient way of getting it into a source file. You *don't* want
> to be going snd = repr(s).
>
> douglas
>
> > But I still get the same error.
> >
> > On Thu, Feb 28, 2008 at 5:20 PM, Douglas Bagnall <
> [EMAIL PROTECTED]>
> > wrote:
> >
> >> PyMike wrote:
> >>
> >>> I tried that out, and it got close to working. But I get an
> >> "unrecognized
> >>> file-type"' error
> >>> Code is below. What did I do wrong?
> >>>
> >>> f = open("shoot.wav")
> >>> s = f.read()
> >>> snd = repr(s)
> >>> print snd
> >>
> >> Sorry, I wasn't clear. You don't need to repr() the string if you are
> >> using it directly. You would only do that if you were writing it to a
> >> file in the form of a string, like so:
> >>
> >> f = open("sound.py", "w")
> >> f.write("snd = " + repr(s))
> >>
> >> Then sound.py becomes the basis of your decoding script. The string you
> >> pass to mixer.Sound should look like lots of these: "\x00", not lots of
> >> these "\\x00".
> >>
> >> BTW, if you are concerned about the length, as other threads would
> >> suggest, you would be better off using base64 encoding, possibly with a
> >> bz2 stage.
> >>
> >> import binascii #or base64 (same functions with different names)
> >>
> >> f = open("shoot.wav")
> >> snd = f.read()
> >>
> >> b64 = binascii.b2a_base64(snd)
> >>
> >> # save b64 to file like above, then
> >>
> >> snd = binascii.a2b_base64(b64)
> >> #etc
> >>
> >> To compress the wav before base64 encoding it, try bz2.compress().
> >>
> >>
> >> douglas
> >>
> >>
> >>
> >>
> >
> >
>
>
>
--
- PyMike
