1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
|
/***************************************************************************
* Copyright (C) 2013 by Albert Astals Cid <aacid@kde.org> *
* *
* This program is free software; you can redistribute it and/or modify *
* it under the terms of the GNU General Public License as published by *
* the Free Software Foundation; either version 2 of the License, or *
* (at your option) any later version. *
***************************************************************************/
#include <qtest_kde.h>
#include "../core/document.h"
#include "../core/page.h"
#include "../core/textpage.h"
#include "../settings_core.h"
Q_DECLARE_METATYPE(Okular::Document::SearchStatus)
class SearchFinishedReceiver : public QObject
{
Q_OBJECT
private slots:
void searchFinished(int id, Okular::Document::SearchStatus status)
{
m_id = id;
m_status = status;
}
public:
int m_id;
Okular::Document::SearchStatus m_status;
};
class SearchTest : public QObject
{
Q_OBJECT
private slots:
void initTestCase();
void testNextAndPrevious();
void test311232();
void test323262();
void test323263();
void testDottedI();
void testHyphenAtEndOfLineWithoutYOverlap();
void testHyphenWithYOverlap();
void testHyphenAtEndOfPage();
void testOneColumn();
void testTwoColumns();
};
void SearchTest::initTestCase()
{
qRegisterMetaType<Okular::Document::SearchStatus>();
Okular::SettingsCore::instance( "searchtest" );
}
static void createTextPage(const QVector<QString>& text, const QVector<Okular::NormalizedRect>& rect,
Okular::TextPage*& tp, Okular::Page*& page) {
tp = new Okular::TextPage();
for (int i = 0; i < text.size(); i++) {
tp->append(text[i], new Okular::NormalizedRect(rect[i]));
}
//The Page::setTextPage method invokes the layout analysis algorithms tested by some tests here
//and also sets the tp->d->m_page field (the latter was used in older versions of Okular by
//TextPage::stringLengthAdaptedWithHyphen).
//Note that calling "delete page;" will delete the TextPage as well.
page = new Okular::Page(1, 100, 100, Okular::Rotation0);
page -> setTextPage(tp);
}
#define CREATE_PAGE \
QCOMPARE(text.size(), rect.size()); \
Okular::Page* page; \
Okular::TextPage* tp; \
createTextPage(text, rect, tp, page);
#define TEST_NEXT_PREV(searchType, expectedStatus) \
{ \
Okular::RegularAreaRect* result = tp->findText(0, searchString, searchType, Qt::CaseSensitive, NULL); \
QCOMPARE(!!result, expectedStatus); \
delete result; \
}
//The test testNextAndPrevious checks that
//a) if one starts a new search, then the first or last match is found, depending on the search direction
// (2 cases: FromTop/FromBottom)
//b) if the last search has found a match,
// then clicking the "Next" button moves to the next occurrence an "Previous" to the previous one
// (if there is any). Altogether there are four combinations of the last search and new search
// direction: Next-Next, Previous-Previous, Next-Previous, Previous-Next; the first two combination
// have two subcases (the new search may give a match or not, so altogether 6 cases to test).
//This gives 8 cases altogether. By taking into account the cases where the last search has given no match,
//we would have 4 more cases (Next (no match)-Next, Next (no match)-Previous, Previous (no match)-Previous,
//Previous (no match)-Next), but those are more the business of Okular::Document::searchText rather than
//Okular::TextPage (at least in the multi-page case).
// We have four test situations: four documents and four corresponding search strings.
// The first situation (document="ababa", search string="b") is a generic one where the
//two matches are not side-by-side and neither the first character nor the last character of
//the document match. The only special thing is that the search string has only length 1.
// The second situation (document="abab", search string="ab") is notable for that the two occurrences
//of the search string are side-by-side with no characters in between, so some off-by-one errors
//would be detected by this test. As the first match starts at the beginning at the document the
//last match ends at the end of the document, it also detects off-by-one errors for finding the first/last match.
// The third situation (document="abababa", search string="aba") is notable for it shows whether
//the next match is allowed to contain letters from the previous one: currently it is not
//(as in the majority of browsers, viewers and editors), and therefore "abababa" is considered to
//contain not three but two occurrences of "aba" (if one starts search from the beginning of the document).
// The fourth situation (document="a ba b", search string="a b") demonstrates the case when one TinyTextEntity
//contains multiple characters that are contained in different matches (namely, the middle "ba" is one TinyTextEntity);
//in particular, since these matches are side-by-side, this test would detect some off-by-one
//offset errors.
void SearchTest::testNextAndPrevious()
{
#define TEST_NEXT_PREV_SITUATION_COUNT 4
QVector<QString> texts[TEST_NEXT_PREV_SITUATION_COUNT] = {
QVector<QString>() << "a" << "b" << "a" << "b" << "a",
QVector<QString>() << "a" << "b" << "a" << "b",
QVector<QString>() << "a" << "b" << "a" << "b" << "a" << "b" << "a" ,
QVector<QString>() << "a" << " " << "ba" << " " << "b"
};
QString searchStrings[TEST_NEXT_PREV_SITUATION_COUNT] = {
"b",
"ab",
"aba",
"a b"
};
for (int i = 0; i < TEST_NEXT_PREV_SITUATION_COUNT; i++) {
const QVector<QString>& text = texts[i];
const QString& searchString = searchStrings[i];
QVector<Okular::NormalizedRect> rect; \
for (int i = 0; i < text.size(); i++) {
rect << Okular::NormalizedRect(0.1*i, 0.0, 0.1*(i+1), 0.1); \
}
CREATE_PAGE;
//Test 3 of the 8 cases listed above:
//FromTop, Next-Next (match) and Next-Next (no match)
TEST_NEXT_PREV(Okular::FromTop, true);
TEST_NEXT_PREV(Okular::NextResult, true);
TEST_NEXT_PREV(Okular::NextResult, false);
//Test 5 cases: FromBottom, Previous-Previous (match), Previous-Next,
//Next-Previous, Previous-Previous (no match)
TEST_NEXT_PREV(Okular::FromBottom, true);
TEST_NEXT_PREV(Okular::PreviousResult, true);
TEST_NEXT_PREV(Okular::NextResult, true);
TEST_NEXT_PREV(Okular::PreviousResult, true);
TEST_NEXT_PREV(Okular::PreviousResult, false);
delete page;
}
}
void SearchTest::test311232()
{
Okular::Document d(0);
SearchFinishedReceiver receiver;
QSignalSpy spy(&d, SIGNAL(searchFinished(int,Okular::Document::SearchStatus)));
QObject::connect(&d, SIGNAL(searchFinished(int,Okular::Document::SearchStatus)), &receiver, SLOT(searchFinished(int,Okular::Document::SearchStatus)));
const QString testFile = KDESRCDIR "data/file1.pdf";
const KMimeType::Ptr mime = KMimeType::findByPath( testFile );
d.openDocument(testFile, KUrl(), mime);
const int searchId = 0;
d.searchText(searchId, " i ", true, Qt::CaseSensitive, Okular::Document::NextMatch, false, QColor());
QTime t;
t.start();
while (spy.count() != 1 && t.elapsed() < 500)
qApp->processEvents();
QCOMPARE(spy.count(), 1);
QCOMPARE(receiver.m_id, searchId);
QCOMPARE(receiver.m_status, Okular::Document::MatchFound);
d.continueSearch( searchId, Okular::Document::PreviousMatch );
t.start();
while (spy.count() != 2 && t.elapsed() < 500)
qApp->processEvents();
QCOMPARE(spy.count(), 2);
QCOMPARE(receiver.m_id, searchId);
QCOMPARE(receiver.m_status, Okular::Document::NoMatchFound);
}
void SearchTest::test323262()
{
QVector<QString> text;
text << "a\n";
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(1, 2, 3, 4);
CREATE_PAGE;
Okular::RegularAreaRect* result = tp->findText(0, "a", Okular::FromBottom, Qt::CaseSensitive, NULL);
QVERIFY(result);
delete result;
delete page;
}
void SearchTest::test323263()
{
QVector<QString> text;
text << "a" << "a" << "b";
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(0, 0, 1, 1)
<< Okular::NormalizedRect(1, 0, 2, 1)
<< Okular::NormalizedRect(2, 0, 3, 1);
CREATE_PAGE;
Okular::RegularAreaRect* result = tp->findText(0, "ab", Okular::FromTop, Qt::CaseSensitive, NULL);
QVERIFY(result);
Okular::RegularAreaRect expected;
expected.append(rect[1]);
expected.append(rect[2]);
expected.simplify();
QCOMPARE(*result, expected);
delete result;
delete page;
}
void SearchTest::testDottedI()
{
//Earlier versions of okular had the bug that the letter "İ" (capital dotter i) did not match itself
//in case-insensitive mode (this was caused by an unnecessary call of toLower() and the fact that
//QString::fromUtf8("İ").compare(QString::fromUtf8("İ").toLower(), Qt::CaseInsensitive) == FALSE,
//at least in Qt 4.8).
//In the future it would be nice to add support for matching "İ"<->"i" and "I"<->"ı" in case-insensitive
//mode as well (QString::compare does not match them, at least in non-Turkish locales, since it follows
//the Unicode case-folding rules http://www.unicode.org/Public/6.2.0/ucd/CaseFolding.txt).
QVector<QString> text;
text << QString::fromUtf8("İ");
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(1, 2, 3, 4);
CREATE_PAGE;
Okular::RegularAreaRect* result = tp->findText(0, QString::fromUtf8("İ"), Okular::FromTop, Qt::CaseInsensitive, NULL);
QVERIFY(result);
delete result;
delete page;
}
void SearchTest::testHyphenAtEndOfLineWithoutYOverlap()
{
QVector<QString> text;
text << "super-"
<< "cali-\n"
<< "fragilistic" << "-"
<< "expiali" << "-\n"
<< "docious";
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(0.4, 0.0, 0.9, 0.1)
<< Okular::NormalizedRect(0.0, 0.1, 0.6, 0.2)
<< Okular::NormalizedRect(0.0, 0.2, 0.8, 0.3) << Okular::NormalizedRect(0.8, 0.2, 0.9, 0.3)
<< Okular::NormalizedRect(0.0, 0.3, 0.8, 0.4) << Okular::NormalizedRect(0.8, 0.3, 0.9, 0.4)
<< Okular::NormalizedRect(0.0, 0.4, 0.7, 0.5);
CREATE_PAGE;
Okular::RegularAreaRect* result = tp->findText(0, "supercalifragilisticexpialidocious",
Okular::FromTop, Qt::CaseSensitive, NULL);
QVERIFY(result);
Okular::RegularAreaRect expected;
for (int i = 0; i < text.size(); i++) {
expected.append(rect[i]);
}
expected.simplify();
QCOMPARE(*result, expected);
delete result;
delete page;
}
#define CREATE_PAGE_AND_TEST_SEARCH(searchString, matchExpected) \
{ \
CREATE_PAGE; \
\
Okular::RegularAreaRect* result = tp->findText(0, searchString, \
Okular::FromTop, Qt::CaseSensitive, NULL); \
\
QCOMPARE(!!result, matchExpected); \
\
delete result; \
delete page; \
}
void SearchTest::testHyphenWithYOverlap()
{
QVector<QString> text;
text << "a-"
<< "b";
QVector<Okular::NormalizedRect> rect(2);
//different lines (50% y-coordinate overlap), first rectangle has larger height
rect[0] = Okular::NormalizedRect(0.0, 0.0, 0.9, 0.35);
rect[1] = Okular::NormalizedRect(0.0, 0.3, 0.2, 0.4);
CREATE_PAGE_AND_TEST_SEARCH("ab", true);
//different lines (50% y-coordinate overlap), second rectangle has larger height
rect[0] = Okular::NormalizedRect(0.0, 0.0, 0.9, 0.1);
rect[1] = Okular::NormalizedRect(0.0, 0.05, 0.2, 0.4);
CREATE_PAGE_AND_TEST_SEARCH("ab", true);
//same line (90% y-coordinate overlap), first rectangle has larger height
rect[0] = Okular::NormalizedRect(0.0, 0.0, 0.4, 0.2);
rect[1] = Okular::NormalizedRect(0.4, 0.11, 0.6, 0.21);
CREATE_PAGE_AND_TEST_SEARCH("ab", false);
CREATE_PAGE_AND_TEST_SEARCH("a-b", true);
//same line (90% y-coordinate overlap), second rectangle has larger height
rect[0] = Okular::NormalizedRect(0.0, 0.0, 0.4, 0.1);
rect[1] = Okular::NormalizedRect(0.4, 0.01, 0.6, 0.2);
CREATE_PAGE_AND_TEST_SEARCH("ab", false);
CREATE_PAGE_AND_TEST_SEARCH("a-b", true);
}
void SearchTest::testHyphenAtEndOfPage()
{
//Tests for segmentation fault that would occur if
//we tried look ahead (for determining whether the
//next character is at the same line) at the end of the page.
QVector<QString> text;
text << "a-";
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(0, 0, 1, 1);
CREATE_PAGE;
{
Okular::RegularAreaRect* result = tp->findText(0, "a",
Okular::FromTop, Qt::CaseSensitive, NULL);
QVERIFY(result);
delete result;
}
{
Okular::RegularAreaRect* result = tp->findText(0, "a",
Okular::FromBottom, Qt::CaseSensitive, NULL);
QVERIFY(result);
delete result;
}
delete page;
}
void SearchTest::testOneColumn()
{
//Tests that the layout analysis algorithm does not create too many columns.
//Bug 326207 was caused by the fact that if all the horizontal breaks in a line
//had the same length and were smaller than vertical breaks between lines then
//the horizontal breaks were treated as column separators.
//(Note that "same length" means "same length after rounding rectangles to integer pixels".
//The resolution used by the XY Cut algorithm with a square page is 1000 x 1000,
//and the horizontal spaces in the example are 0.1, so they are indeed both exactly 100 pixels.)
QVector<QString> text;
text << "Only" << "one" << "column"
<< "here";
//characters and line breaks have length 0.05, word breaks 0.1
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(0.0, 0.0, 0.2, 0.1)
<< Okular::NormalizedRect(0.3, 0.0, 0.5, 0.1)
<< Okular::NormalizedRect(0.6, 0.0, 0.9, 0.1)
<< Okular::NormalizedRect(0.0, 0.15, 0.2, 0.25);
CREATE_PAGE;
Okular::RegularAreaRect* result = tp->findText(0, "Only one column",
Okular::FromTop, Qt::CaseSensitive, NULL);
QVERIFY(result);
delete result;
delete page;
}
void SearchTest::testTwoColumns()
{
//Tests that the layout analysis algorithm can detect two columns.
QVector<QString> text;
text << "This" << "text" << "in" << "two"
<< "is" << "set" << "columns.";
//characters, word breaks and line breaks have length 0.05
QVector<Okular::NormalizedRect> rect;
rect << Okular::NormalizedRect(0.0, 0.0, 0.20, 0.1)
<< Okular::NormalizedRect(0.25, 0.0, 0.45, 0.1)
<< Okular::NormalizedRect(0.6, 0.0, 0.7, 0.1)
<< Okular::NormalizedRect(0.75, 0.0, 0.9, 0.1)
<< Okular::NormalizedRect(0.0, 0.15, 0.1, 0.25)
<< Okular::NormalizedRect(0.15, 0.15, 0.3, 0.25)
<< Okular::NormalizedRect(0.6, 0.15, 1.0, 0.25);
CREATE_PAGE;
Okular::RegularAreaRect* result = tp->findText(0, "This text in",
Okular::FromTop, Qt::CaseSensitive, NULL);
QVERIFY(!result);
delete result;
delete page;
}
QTEST_KDEMAIN( SearchTest, GUI )
#include "searchtest.moc"
|