https://bz.apache.org/bugzilla/show_bug.cgi?id=69667
Bug ID: 69667
Summary: WriteAccessRecord constructor setUsername throws
IllegalArgumentException
Product: POI
Version: 5.4.1-FINAL
Hardware: Macintosh
Status: NEW
Severity: blocker
Priority: P2
Component: HSSF
Assignee: dev@poi.apache.org
Reporter: jason.sla...@ring-zero.com
Target Milestone: ---
I have an XLS file generated by Data Dynamics' SpreadBuilder.
When I attempt to parse the file, it encounters a WriteAccessRecord and I am
unable to proceed any further.
In WriteAccessRecord constructor:
- nChars is 81 and encoding is UTF-16LE.
- it allocates a buffer of 109 bytes (DATA_SIZE-3)
- it reads the data into the buffer
- because it's UTF-16:
- byteCnt = Math.min(nChars * 2, data.length) => 109, since it's the min of
81*2 (162) and 109
- however, when it goes to create a string from the buffer and trim it:
- String rawValue = new String(data, 0, byteCnt, charset); <--- the resulting
string has some garbage character(s) at the end. "Data Dynamics' SpreadBuilder
�"
- the garbage characters cause the trim() to have no effect,
- so, when it gets to setUsername, the string is 55 characters long. Because
it calculates encodedByteCount as length * 2 for this encoding, it throws an
exception because 110 > 109
I believe the problem originates:
if (nChars <= 109 && (is16BitFlag & 254) == 0) {
data = IOUtils.safelyAllocate((long)in.remaining(), 109);
in.readFully(data);
if (UTF16FLAG.isSet(is16BitFlag)) {
byteCnt = Math.min(nChars * 2, data.length); <-- **HERE**
charset = StandardCharsets.UTF_16LE;
} else {
byteCnt = Math.min(nChars, data.length);
charset = StandardCharsets.ISO_8859_1;
}
} else if (in.isEncrypted()) {
because the byte count for UTF-16LE should always be an even number.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@poi.apache.org
For additional commands, e-mail: dev-h...@poi.apache.org
