Fix Python 3.6.0 #5
Conversation
usbinfo/darwin.py
Outdated
| @@ -30,6 +31,38 @@ | |||
| OSX_VERSION_MICRO_INT = [int(part) for part in OSX_VERSION_STR.split('.')] | |||
|
|
|||
|
|
|||
| def _sanitize_xml(data): | |||
| pattern = re.compile(r'^(\s*?\<string\>)(.*?)(\<\/string\>.*?)$') | |||
There was a problem hiding this comment.
Can this be compiled once instead of re-compiling on every function call?
| @@ -38,6 +71,10 @@ def _ioreg(nodename): | |||
| """Run ioreg command on specific node name""" | |||
| cmd = ['ioreg', '-a', '-l', '-r', '-n', nodename] | |||
| output = subprocess.check_output(cmd) | |||
| # ST-Link devices (and possibly others?) erroneously store binary data | |||
| # in the <string> serial number, which causes plistlib to blow up. | |||
| # This will convert that to hex and preserve contents otherwise. | |||
There was a problem hiding this comment.
Please add this comment to the function definition.
| to crash. | ||
|
|
||
| Returns the same document, with any mis-encoded <string> values converted | ||
| to ascii hex.""" |
There was a problem hiding this comment.
We should follow the indentation style for other docstrings.
|
|
||
| data = data.decode('utf-8') | ||
|
|
||
| for i, line in enumerate(data.split('\n')): |
There was a problem hiding this comment.
No need to enumerate if you're not using the index.
There was a problem hiding this comment.
You can also use splitlines().
| to ascii hex.""" | ||
| output = [] | ||
|
|
||
| data = data.decode('utf-8') |
There was a problem hiding this comment.
You are mutating data at the caller here. You could just encode it ad the for loop itself.
| output.append(line) | ||
| output = '\n'.join([line for line in output]) | ||
|
|
||
| return output.encode('utf-8') |
| output.append(line) | ||
| else: | ||
| output.append(line) | ||
| output = '\n'.join([line for line in output]) |
There was a problem hiding this comment.
This can simply be:
return '\n'.join(output)
| data = data.decode('utf-8') | ||
|
|
||
| for i, line in enumerate(data.split('\n')): | ||
| chunk = line |
There was a problem hiding this comment.
Temporary assignment not necessary.
| for byte in byte_list: | ||
| if byte < 32: | ||
| needs_patch = True | ||
| if needs_patch: |
There was a problem hiding this comment.
Could this all be reduced to just
if any([ord(byte) not in range(32, 128) for byte in middle]):
...
|
|
||
| sanitize_pattern = re.compile(r'^(\s*?\<string\>)(.*?)(\<\/string\>.*?)$') | ||
|
|
||
| def _sanitize_xml(data): |
There was a problem hiding this comment.
It might be worth investigating doing all of this in an XML library. This function is assuming that ioreg will always return data in a specific format.
Python 3.6.0's plistlib has removed readPlistFromString in favor of loads. There isn't a backwards compatible upgrade path, so feature detection was necessary instead.
This patch fixes Python 3.6.0 by trying plistlib.loads when AttributeError is thrown from attempting to call plistlib.readPlistFromString.
Additionally, dict.iteritems() is deprecated in favor of dict.items() in Python 3. This also works in Python 2.7.
Tested both Python 2.7 (standard install) and Python 3.6.0 (Anaconda) on macOS Sierra 10.12.4 (16E195).