Issue1599900

classification

Title:	Incorrect usage of String.getBytes()
Type:		Severity:	normal
Components:	Core	Versions:
		Milestone:

process

Status:	closed	Resolution:	fixed
Dependencies:		Superseder:
Assigned To:		Nosy List:	akuchling, cgroves
Priority:	normal	Keywords:

Created on 2006-11-20.18:46:38 by akuchling, last changed 2007-12-02.20:38:29 by cgroves.

Messages
msg1286 (view)	Author: A.M. Kuchling (akuchling)	Date: 2006-11-20.18:46:38
I submitted a patch that fixes problems with the SHA module. One problem was the use of java.lang.String.getBytes() without specifying some encoding; this resulted in bad results where the first character of a hash value was >256. Presumably the encoding on my Mac is UTF-8. The fix is to do getBytes("ISO-8859-1"); this in turn requires catching an UnsupportedEncodingException. Other portions of code that just call .getBytes() may also need changing, wherever strings are being treated as containing 8-bit characters. Possible files that need fixing: FilelikeInputStream.java, parser.java, Py.java, PyArray, PyFile, binascii.java, and of course PyString. zlib.py's adler32() function also uses getBytes and I think it needs to be fixed, too.
msg1287 (view)	Author: Charlie Groves (cgroves)	Date: 2006-12-31.21:31:54
Use the method to_bytes(String) added to PyString in r3038 to do the conversion. This also handles when the String has values greater than 255 in it unlike getBytes("ISO-8859-1")
msg1288 (view)	Author: Charlie Groves (cgroves)	Date: 2007-12-02.20:38:29
The rest of the places using getBytes were fixed in 2.2.1.

History
Date	User	Action	Args
2006-11-20 18:46:38	akuchling	create