Full_Name: Version: OS: URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (81.240.252.176) The JLDAP com.novell.ldap.LDIFWriter class uses the UTF-8 character set when writing LDIF content to output streams. However, the com.novell.ldap.LDIFReader class uses the US-ASCII character set to read LDIF content from input streams. The LDIFReader class should also use UTF-8, so LDIF files containing special characters can be read successfully. Proposed patch for http://www.openldap.org/devel/cvsweb.cgi/~checkout~/com/novell/ldap/util/LDIFReader.java?rev=1.42&cvsroot=JLDAP&hideattic=1&sortbydate=0 -----------------------/snip/----------------------- --- LDIFReader.java.ori 2009-10-23 10:56:00.000000000 +0200 +++ LDIFReader.java 2009-10-23 10:56:35.000000000 +0200 @@ -117,7 +117,7 @@ } setVersion( version ); - InputStreamReader isr = new InputStreamReader(in, "US-ASCII"); + InputStreamReader isr = new InputStreamReader(in, "UTF-8"); bufReader = new BufferedReader(isr); // In order to determine if it is a LDIF content file or LDIF change -----------------------/snip/-----------------------
moved from Incoming to Contrib
Would this make it so that non-ascii strings ar enot base 64 encoded?
(In reply to drmuey+github from comment #2) > Would this make it so that non-ascii strings ar enot base 64 encoded? Changing how the reader works has no bearing on whether the writer does base 64 encoding. Your question makes no sense. The LDIF spec says the input values may be in UTF8 or base64 encoded, so this is a legitimate bug that should be fixed.