From 2fbb69a09361bf1ecf517516f76405c3be494d6d Mon Sep 17 00:00:00 2001 From: Konstantin Ritt Date: Thu, 22 Nov 2012 03:25:05 +0200 Subject: QTBF: Fix issue with no splitting the words at "." (FULL STOP) As of Unicode 5.1, some punctuation marks were mapped to MidLetter and MidNumLet for better URL and abbreviations handling which caused "hi.there" to be treated like if it were just a single word; until we have the Unicode Text Segmentation tailoring mechanism, retain the old behavior by remapping (some of) those characters back to their old values. Change-Id: I49dea6064f2ea40a82fc0b1bc3c4f0b4e803919f Reviewed-by: David Faure Reviewed-by: Lars Knoll --- .../tst_qtextboundaryfinder.cpp | 43 +++++++++++++++++++++- 1 file changed, 41 insertions(+), 2 deletions(-) (limited to 'tests') diff --git a/tests/auto/corelib/tools/qtextboundaryfinder/tst_qtextboundaryfinder.cpp b/tests/auto/corelib/tools/qtextboundaryfinder/tst_qtextboundaryfinder.cpp index eeb22f4c21..b126c62c72 100644 --- a/tests/auto/corelib/tools/qtextboundaryfinder/tst_qtextboundaryfinder.cpp +++ b/tests/auto/corelib/tools/qtextboundaryfinder/tst_qtextboundaryfinder.cpp @@ -330,8 +330,40 @@ void tst_QTextBoundaryFinder::wordBoundaries_manual_data() expectedStartPositions << 0 << 4 << 8 << 14 << 18 << 22; expectedEndPositions << 3 << 7 << 11 << 17 << 21 << 25; - QTest::newRow("data1") << testString << expectedBreakPositions - << expectedStartPositions << expectedEndPositions; + QTest::newRow("words1") << testString << expectedBreakPositions + << expectedStartPositions << expectedEndPositions; + } + { + QString testString(QString::fromUtf8("Hello (sad) world !")); + QList expectedBreakPositions, expectedStartPositions, expectedEndPositions; + expectedBreakPositions << 0 << 5 << 6 << 7 << 10 << 11 << 12 << 17 << 18 << 19; + expectedStartPositions << 0 << 7 << 12; + expectedEndPositions << 5 << 10 << 17; + + QTest::newRow("words2") << testString << expectedBreakPositions + << expectedStartPositions << expectedEndPositions; + } + { + QString testString(QString::fromUtf8("mr.Hamster")); + QList expectedBreakPositions, expectedStartPositions, expectedEndPositions; + expectedBreakPositions << 0 << 2 << 3 << 10; + expectedStartPositions << 0 << 3; + expectedEndPositions << 2 << 10; + + QTest::newRow("words3") << testString << expectedBreakPositions + << expectedStartPositions << expectedEndPositions; + } + { + QString testString(QString::fromUtf8("This is a sample buffer.Please test me . He's don't Le'Clerk.")); + QList expectedBreakPositions, expectedStartPositions, expectedEndPositions; + expectedBreakPositions << 0 << 4 << 5 << 7 << 8 << 9 << 10 << 11 << 12 << 13 << 14 << 20 << 21 << 27 + << 28 << 34 << 35 << 39 << 40 << 42 << 43 << 44 << 45 << 46 << 47 << 48 + << 49 << 53 << 54 << 59 << 60 << 68 << 69; + expectedStartPositions << 0 << 5 << 12 << 14 << 21 << 28 << 35 << 40 << 49 << 54 << 60; + expectedEndPositions << 4 << 7 << 13 << 20 << 27 << 34 << 39 << 42 << 53 << 59 << 68; + + QTest::newRow("words4") << testString << expectedBreakPositions + << expectedStartPositions << expectedEndPositions; } { // text with trailing space @@ -512,6 +544,13 @@ void tst_QTextBoundaryFinder::sentenceBoundaries_manual_data() QTest::newRow("data2") << testString << expectedBreakPositions; } + { + QString testString(QString::fromUtf8("mr.Hamster")); + QList expectedBreakPositions; + expectedBreakPositions << 0 << 3 << 10; + + QTest::newRow("data3") << testString << expectedBreakPositions; + } } void tst_QTextBoundaryFinder::sentenceBoundaries_manual() -- cgit v1.2.3