Repository: qpid-proton-j Updated Branches: refs/heads/master a9988721e -> 589504d76
PROTON-1963 Refactor string encoding size calculation Assume ASCII by default when calculating the string size in UTF8 to allow for inlining of this method in most cases and fallback to a multi-byte calculation only when non-ASCII chars are detected. Project: http://git-wip-us.apache.org/repos/asf/qpid-proton-j/repo Commit: http://git-wip-us.apache.org/repos/asf/qpid-proton-j/commit/391af064 Tree: http://git-wip-us.apache.org/repos/asf/qpid-proton-j/tree/391af064 Diff: http://git-wip-us.apache.org/repos/asf/qpid-proton-j/diff/391af064 Branch: refs/heads/master Commit: 391af064e2d50dfd29700f160166e86d8f4cba19 Parents: a998872 Author: Timothy Bish <tabish...@gmail.com> Authored: Tue Nov 13 14:53:13 2018 -0500 Committer: Timothy Bish <tabish...@gmail.com> Committed: Tue Nov 13 14:53:13 2018 -0500 ---------------------------------------------------------------------- .../apache/qpid/proton/codec/StringType.java | 32 +++++++++++++++----- 1 file changed, 24 insertions(+), 8 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/qpid-proton-j/blob/391af064/proton-j/src/main/java/org/apache/qpid/proton/codec/StringType.java ---------------------------------------------------------------------- diff --git a/proton-j/src/main/java/org/apache/qpid/proton/codec/StringType.java b/proton-j/src/main/java/org/apache/qpid/proton/codec/StringType.java index 7687ca4..85448b3 100644 --- a/proton-j/src/main/java/org/apache/qpid/proton/codec/StringType.java +++ b/proton-j/src/main/java/org/apache/qpid/proton/codec/StringType.java @@ -84,26 +84,42 @@ public class StringType extends AbstractPrimitiveType<String> static int calculateUTF8Length(final String s) { - int len = s.length(); - final int length = len; - for (int i = 0; i < length; i++) + final int stringLength = s.length(); + + // ASCII Optimized length case + int utf8len = stringLength; + int processed = 0; + for (; processed < stringLength && s.charAt(processed) < 0x80; processed++) {} + + if (processed < stringLength) + { + // Non-ASCII length remainder + utf8len = extendedCalculateUTF8Length(s, processed, stringLength, utf8len); + } + + return utf8len; + } + + static int extendedCalculateUTF8Length(final String s, int index, int length, int utf8len) { + for (; index < length; index++) { - int c = s.charAt(i); + int c = s.charAt(index); if ((c & 0xFF80) != 0) /* U+0080.. */ { - len++; + utf8len++; if(((c & 0xF800) != 0)) /* U+0800.. */ { - len++; + utf8len++; // surrogate pairs should always combine to create a code point with a 4 octet representation if ((c & 0xD800) == 0xD800 && c < 0xDC00) { - i++; + index++; } } } } - return len; + + return utf8len; } @Override --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@qpid.apache.org For additional commands, e-mail: commits-h...@qpid.apache.org
