Re: RFR (JAXP) 8029955 : AIOB in XMLEntityScanner.scanLiteral upon parsing literals with > 100 LF chars

[Ulf Zibis]

Hi Joe,

I would use ’\t’ instead of 0x9, to stay consistent with ahead code:
1175                 if (c == ’\t’) {
1176                     storeWhiteSpace(fCurrentEntity.position-1);

Lines 1214..1221 could be simpler:
1214         if (whiteSpaceLen >= whiteSpaceLookup.length) {
1215             int [] tmp = new int[whiteSpaceLookup.length*2];
1216             System.arraycopy(whiteSpaceLookup, 0, tmp, 0, 
whiteSpaceLookup.length);
1217             whiteSpaceLookup = tmp;
1218         }
1219         whiteSpaceLookup[whiteSpaceLen++] = whiteSpacePos;

Or even shorter:
1214         if (whiteSpaceLen >= whiteSpaceLookup.length)
1215             whiteSpaceLookup = Arrays.copyOf(whiteSpaceLookup, 
whiteSpaceLookup.length*2);
1216         whiteSpaceLookup[whiteSpaceLen++] = whiteSpacePos;

(please insert spaces around if clause, else and after commas.)

-Ulf


On 16.12.2013 20:31, huizhe wang wrote:

Hi,

This is a quick fix for a whitespace buffer that was not adjusted properly in one of the two cases. The buffer, whiteSpaceLookup, is filled in two cases and adjusted properly the 2nd time. The code is moved into a method storeWhiteSpace so that it's shared for the 1st case as well.

Note at line 1175, there is no need to save character 0x20 since all whitespace characters will later be replaced with character 0x20.

webrevs:
http://cr.openjdk.java.net/~joehw/jdk8/8029955/webrev/

Thanks,
Joe