babelfish-for-postgresql / babelfish_extensions / 9155067827

76%

BABEL_5_X_DEV: 77%

push

github

Fix the logic of calculating the length of the named procedure to consider characters position instead of bytes (#2582)

Previously, string.length() method was being used to calculate the length of named procedure inside
makeExecuteProcedure. This method calculates length in terms of bytes. For example, length of
"ZERO WIDTH SPACE" with unicode point U+200B will be computed as 3 which was causing certain
assertion failure makeExecuteProcedure resulting in crash. But as far as parser and token's position
concerns, we should really care of character position instead of bytes position.

This commit aims to fix this issue by fixing logic of calculating the length of named procedure by using
getStartIndex() and getStopIndex() of the procedure token.

Task: BABEL-4961
Signed-off-by: Dipesh Dhameliya <dddhamel@amazon.com>

3 of 3 new or added lines in 1 file covered. (100.0%)

41834 of 57474 relevant lines covered (72.79%)

103158.69 hits per line