| 1 |
On Wed, Oct 29, 2008 at 7:51 PM, Andrey Vul <andrey.vul@×××××.com> wrote: |
| 2 |
<snip> |
| 3 |
> elif system in ('Linux',): |
| 4 |
> # Linux based systems |
| 5 |
> distname,distversion,distid = dist('') |
| 6 |
> if distname and not terse: |
| 7 |
> platform = _platform(system,release,machine,processor, |
| 8 |
> 'with', |
| 9 |
> distname,distversion,distid) |
| 10 |
> else: |
| 11 |
> # If the distribution name is unknown check for libc vs. glibc |
| 12 |
> libcname,libcversion = libc_ver(sys.executable) |
| 13 |
> platform = _platform(system,release,machine,processor, |
| 14 |
> 'with', |
| 15 |
> libcname+libcversion) |
| 16 |
<snip> |
| 17 |
|
| 18 |
Hrm. I know just enough about python to get myself in trouble here... |
| 19 |
but it looks like a python bug in magicking up the libc name and |
| 20 |
version... but the below is WAY outside my level of practice with |
| 21 |
python (it'll take re-reading and digging elsewhere a good few times |
| 22 |
if I'm ever to make sense of it... |
| 23 |
|
| 24 |
------------------ |
| 25 |
def libc_ver(executable=sys.executable,lib='',version='', |
| 26 |
|
| 27 |
chunksize=2048): |
| 28 |
|
| 29 |
""" Tries to determine the libc version that the file executable |
| 30 |
(which defaults to the Python interpreter) is linked against. |
| 31 |
|
| 32 |
Returns a tuple of strings (lib,version) which default to the |
| 33 |
given parameters in case the lookup fails. |
| 34 |
|
| 35 |
Note that the function has intimate knowledge of how different |
| 36 |
libc versions add symbols to the executable and thus is probably |
| 37 |
only useable for executables compiled using gcc. |
| 38 |
|
| 39 |
The file is read and scanned in chunks of chunksize bytes. |
| 40 |
|
| 41 |
""" |
| 42 |
f = open(executable,'rb') |
| 43 |
binary = f.read(chunksize) |
| 44 |
pos = 0 |
| 45 |
while 1: |
| 46 |
m = _libc_search.search(binary,pos) |
| 47 |
if not m: |
| 48 |
binary = f.read(chunksize) |
| 49 |
if not binary: |
| 50 |
break |
| 51 |
pos = 0 |
| 52 |
continue |
| 53 |
libcinit,glibc,glibcversion,so,threads,soversion = m.groups() |
| 54 |
if libcinit and not lib: |
| 55 |
lib = 'libc' |
| 56 |
elif glibc: |
| 57 |
if lib != 'glibc': |
| 58 |
lib = 'glibc' |
| 59 |
version = glibcversion |
| 60 |
elif glibcversion > version: |
| 61 |
version = glibcversion |
| 62 |
elif so: |
| 63 |
if lib != 'glibc': |
| 64 |
lib = 'libc' |
| 65 |
if soversion > version: |
| 66 |
version = soversion |
| 67 |
if threads and version[-len(threads):] != threads: |
| 68 |
version = version + threads |
| 69 |
pos = m.end() |
| 70 |
f.close() |
| 71 |
return lib,version |
| 72 |
------------------ |
| 73 |
|
| 74 |
It parses the header of an executable and guesses, but... the how is |
| 75 |
too many directions from this that I'm not seeing it with my haphazard |
| 76 |
abuse of grep. I'd presume anything that might care what platform it's |
| 77 |
running on (underneath python itself) would be susceptible, so a word |
| 78 |
thrown in the direction of upstream python would be the main way to |
| 79 |
go... though it looks like emerge didn't used to use that call... |
| 80 |
|
| 81 |
Portage 2.1.4.5 (default/linux/x86/2008.0, gcc-4.1.2, glibc-2.6.1-r0, |
| 82 |
2.6.25-gentoo-r7-mahain i686) |
| 83 |
================================================================= |
| 84 |
System uname: 2.6.25-gentoo-r7-mahain i686 AMD Athlon(tm) MP 2400+ |
| 85 |
|
| 86 |
is my output, based on a call in emerge to "uname -mrp" .. not |
| 87 |
platform.platform() |
| 88 |
|
| 89 |
Looks like gentoo-dev aimed to drop that dependency in newer versions after all. |
| 90 |
|
| 91 |
-- |
| 92 |
Poison [BLX] |
| 93 |
Joshua M. Murphy |
Archives