void setUpDirs(CuTest *tc, Directory * dir, Directory * aux) {
IndexWriter4Test * writer = NULL;
WhitespaceAnalyzer analyzer;
writer = newWriter(dir, &analyzer, true);
writer->setMaxBufferedDocs(1000);
// add 1000 documents in 1 segment
addDocs(writer, 1000);
assertEquals(1000, writer->docCount());
assertEquals(1, writer->getSegmentCount());
writer->close();
_CLLDELETE(writer);
writer = newWriter(aux, &analyzer, true);
writer->setUseCompoundFile(false); // use one without a compound file
writer->setMaxBufferedDocs(100);
writer->setMergeFactor(10);
// add 30 documents in 3 segments
for (int i = 0; i < 3; i++) {
addDocs(writer, 10);
writer->close();
_CLLDELETE(writer);
writer = newWriter(aux, &analyzer, false);
writer->setUseCompoundFile(false); // use one without a compound file
writer->setMaxBufferedDocs(100);
writer->setMergeFactor(10);
}
assertEquals(30, writer->docCount());
assertEquals(3, writer->getSegmentCount());
writer->close();
_CLLDELETE(writer);
}
// case 0: add self or exceed maxMergeDocs, expect exception
void testAddSelf(CuTest * tc) {
// main directory
Directory * dir = _CLNEW RAMDirectory();
// auxiliary directory
Directory * aux = _CLNEW RAMDirectory();
IndexWriter4Test * writer = NULL;
WhitespaceAnalyzer analyzer;
writer = newWriter(dir, &analyzer, true);
// add 100 documents
addDocs(writer, 100);
assertEquals(100, writer->docCount());
writer->close();
_CLLDELETE(writer);
writer = newWriter(aux, &analyzer, true);
writer->setUseCompoundFile(false); // use one without a compound file
writer->setMaxBufferedDocs(1000);
// add 140 documents in separate files
addDocs(writer, 40);
writer->close();
_CLLDELETE(writer);
writer = newWriter(aux, &analyzer, true);
writer->setUseCompoundFile(false); // use one without a compound file
writer->setMaxBufferedDocs(1000);
addDocs(writer, 100);
writer->close();
_CLLDELETE(writer);
writer = newWriter(dir, &analyzer, false);
try {
// cannot add self
ValueArray<Directory*> dirs(2);
dirs[0] = aux;
dirs[1] = dir;
writer->addIndexesNoOptimize( dirs );
assertTrue(false);
}
catch (CLuceneError&) {
assertEquals(100, writer->docCount());
}
writer->close();
_CLLDELETE(writer);
// make sure the index is correct
verifyNumDocs(tc, dir, 100);
dir->close();
_CLLDELETE(dir);
aux->close();
_CLLDELETE(aux);
}
void unpackCB(LONG id, LONG width, LONG height, BYTE *rgbBuf) {
struct writer_t *w;
struct bmp24_writer_t *bw;
FILE *f;
char fn[256];
if (rgbBuf == NULL)
return;
sprintf(fn, "out/%d.bmp", id);
f = fopen(fn, "wb");
if (f == NULL) {
fprintf(stderr, "Failed to open %s for writing: %s\n", fn, strerror(errno));
exit(1);
}
w = newWriter(f);
bw = newBMP24Writer(w, width, height);
writeBMP24Header(bw);
writeBMP24Bitmap(bw, rgbBuf);
deleteBMP24Writer(bw);
deleteWriter(w);
fclose(f);
if (progress % 10 == 0)
fprintf(stderr, "%d..\n", progress);
progress++;
}
// case 3: tail segments, invariants hold, copy, invariants hold
void testNoMergeAfterCopy(CuTest * tc) {
// main directory
Directory * dir = _CLNEW RAMDirectory();
// auxiliary directory
Directory * aux = _CLNEW RAMDirectory();
WhitespaceAnalyzer an;
setUpDirs(tc, dir, aux);
IndexWriter4Test * writer = newWriter(dir, &an, false);
writer->setMaxBufferedDocs(10);
writer->setMergeFactor(4);
ValueArray<Directory*> dirs(2);
dirs[0] = aux;
dirs[1] = aux;
writer->addIndexesNoOptimize(dirs);
assertEquals(1060, writer->docCount());
assertEquals(1000, writer->getDocCount(0));
writer->close();
_CLLDELETE(writer);
// make sure the index is correct
verifyNumDocs(tc, dir, 1060);
dir->close();
_CLLDELETE(dir);
aux->close();
_CLLDELETE(aux);
}
// case 5: tail segments, invariants not hold
void testMoreMerges(CuTest * tc) {
// main directory
Directory * dir = _CLNEW RAMDirectory();
// auxiliary directory
Directory * aux = _CLNEW RAMDirectory();
Directory * aux2 = _CLNEW RAMDirectory();
WhitespaceAnalyzer an;
setUpDirs(tc, dir, aux);
IndexWriter4Test * writer = newWriter(aux2, &an, true);
writer->setMaxBufferedDocs(100);
writer->setMergeFactor(10);
ValueArray<Directory*> dirs(1);
dirs[0] = aux;
writer->addIndexesNoOptimize(dirs);
assertEquals(30, writer->docCount());
assertEquals(3, writer->getSegmentCount());
writer->close();
_CLLDELETE(writer);
IndexReader * reader = IndexReader::open(aux);
for (int i = 0; i < 27; i++) {
reader->deleteDocument(i);
}
assertEquals(3, reader->numDocs());
reader->close();
_CLLDELETE(reader);
reader = IndexReader::open(aux2);
for (int i = 0; i < 8; i++) {
reader->deleteDocument(i);
}
assertEquals(22, reader->numDocs());
reader->close();
_CLLDELETE( reader );
writer = newWriter(dir, &an, false);
writer->setMaxBufferedDocs(6);
writer->setMergeFactor(4);
ValueArray<Directory*> dirs2(2);
dirs2[0] = aux;
dirs2[1] = aux2;
writer->addIndexesNoOptimize(dirs2);
assertEquals(1025, writer->docCount());
assertEquals(1000, writer->getDocCount(0));
writer->close();
_CLLDELETE( writer );
// make sure the index is correct
verifyNumDocs(tc, dir, 1025);
dir->close();
_CLLDELETE(dir);
aux->close();
_CLLDELETE(aux);
aux2->close();
_CLLDELETE(aux2);
}
void testSimpleCase(CuTest *tc) {
// main directory
Directory * dir = _CLNEW RAMDirectory();
// two auxiliary directories
Directory * aux = _CLNEW RAMDirectory();
Directory * aux2 = _CLNEW RAMDirectory();
IndexWriter4Test * writer = NULL;
WhitespaceAnalyzer analyzer;
writer = newWriter(dir, &analyzer, true);
// add 100 documents
addDocs(writer, 100);
assertEquals(100, writer->docCount());
writer->close();
_CLLDELETE(writer);
writer = newWriter(aux, &analyzer, true);
writer->setUseCompoundFile(false); // use one without a compound file
// add 40 documents in separate files
addDocs(writer, 40);
assertEquals(40, writer->docCount());
writer->close();
_CLLDELETE(writer);
writer = newWriter(aux2, &analyzer, true);
// add 40 documents in compound files
addDocs2(writer, 50);
assertEquals(50, writer->docCount());
writer->close();
_CLLDELETE(writer);
// test doc count before segments are merged
writer = newWriter(dir, &analyzer, false);
assertEquals(100, writer->docCount());
{
ValueArray<Directory*> dirs(2);
dirs[0] = aux;
dirs[1] = aux2;
writer->addIndexesNoOptimize( dirs );
}
assertEquals(190, writer->docCount());
writer->close();
_CLLDELETE(writer);
// make sure the old index is correct
verifyNumDocs(tc, aux, 40);
// make sure the new index is correct
verifyNumDocs(tc, dir, 190);
// now add another set in.
Directory * aux3 = _CLNEW RAMDirectory();
writer = newWriter(aux3, &analyzer, true);
// add 40 documents
addDocs(writer, 40);
assertEquals(40, writer->docCount());
writer->close();
_CLLDELETE(writer);
// test doc count before segments are merged/index is optimized
writer = newWriter(dir, &analyzer, false);
assertEquals(190, writer->docCount());
{
ValueArray<Directory*> dirs(1);
dirs[0] = aux3;
writer->addIndexesNoOptimize( dirs );
}
assertEquals(230, writer->docCount());
writer->close();
_CLLDELETE(writer);
// make sure the new index is correct
verifyNumDocs(tc, dir, 230);
Term t1(_T("content"), _T("aaa"));
Term t2(_T("content"), _T("bbb"));
verifyTermDocs(tc, dir, &t1, 180);
verifyTermDocs(tc, dir, &t2, 50);
// now optimize it.
writer = newWriter(dir, &analyzer, false);
writer->optimize();
writer->close();
_CLLDELETE(writer);
// make sure the new index is correct
verifyNumDocs(tc, dir, 230);
verifyTermDocs(tc, dir, &t1, 180);
verifyTermDocs(tc, dir, &t2, 50);
// now add a single document
Directory * aux4 = _CLNEW RAMDirectory();
writer = newWriter(aux4, &analyzer, true);
addDocs2(writer, 1);
writer->close();
_CLLDELETE(writer);
writer = newWriter(dir, &analyzer, false);
assertEquals(230, writer->docCount());
{
ValueArray<Directory*> dirs(1);
dirs[0] = aux4;
writer->addIndexesNoOptimize( dirs );
}
assertEquals(231, writer->docCount());
writer->close();
_CLLDELETE(writer);
verifyNumDocs(tc, dir, 231);
verifyTermDocs(tc, dir, &t2, 51);
dir->close();
_CLLDELETE(dir);
aux->close();
_CLLDELETE(aux);
aux2->close();
_CLLDELETE(aux2);
aux3->close();
_CLLDELETE(aux3);
aux4->close();
_CLLDELETE(aux4);
}
void doSomeStuff(){
printf("Creating account for Dan Brown\n");
struct Writer danBrown = newWriter(
"Dan Brown",
"danbrown",
"password"
);
addWriter(&danBrown, &eBookStore);
printf("Created account for Dan Brown with ID %d\n\n", danBrown.id);
printf("Adding the book Angels and Demons\n");
struct Book angelsAndDemons = newBook(
"Angels and Demons",
"Dan Brown",
"Science and Mythology and Stuff",
"In not so distant past...",
danBrown.id,
100
);
publishBook(&angelsAndDemons, &eBookStore);
printf("Added the book Angels and Demons with ID %d\n\n", angelsAndDemons.id);
printf("Adding the book Da Vinci Code\n");
struct Book daVinciCode = newBook(
"Da Vinci Code",
"Dan Brown",
"Its about the hipster Leo",
"Once upon a time...",
danBrown.id,
100
);
publishBook(&daVinciCode, &eBookStore);
printf("Added the book Da Vinci Code with ID %d\n\n", daVinciCode.id);
printf("Books published by %s on our platform are,\n", danBrown.name);
for(int i=0;i<danBrown.bookCount;i++){
int bookID = danBrown.bookIDs[i];
struct Book *b = eBookStore.bookDataStore.books[bookID];
printf("#%d:\t%s\n", i+1, b->title);
}
printf("\n");
printf("Finding books similar to %s\n", daVinciCode.title);
assignSimilarBooks(&daVinciCode, &eBookStore);
for(int i=0;i<daVinciCode.similarBookCount;i++){
int bookID = daVinciCode.similarBooks[i];
struct Book *b = eBookStore.bookDataStore.books[bookID];
printf("#%d:\t%s\n", i+1, b->title);
}
printf("\n");
printf("Creating customer account for Robert Langdon\n");
struct Customer robertLangdon = newCustomer("robertlangdon", "password");
addCustomerToDataStore(&robertLangdon, &eBookStore.customerDataStore);
printf("Created customer account for Robert Langdon with user id %d\n\n", robertLangdon.id);
printf("Searching for books with the title 'Angels and Demons'\n");
struct Book* searchResult = searchBooksInDataStore(1, "Angels and Demons", 1, &eBookStore.bookDataStore)[0];
printf("Found one book with the name %s by the author %s\n", searchResult->title, searchResult->author);
printf("Its summary is, %s\n\n", searchResult->summary);
printf("Submitting an excellent review for %s from user name %s\n", searchResult->title, robertLangdon.username);
addReview(searchResult, &robertLangdon, 5, "Awesome read! :)", &eBookStore);
printf("Fetching all reviews for %s,\n", searchResult->title);
for(int i=0;i<searchResult->reviewCount;i++){
printf("#%d: Rating: %d Comments: %s\n", i+1, searchResult->reviews[i]->rating, searchResult->reviews[i]->body);
}
printf("Average rating of %s is %f\n\n", searchResult->title, getAverageRating(searchResult));
printf("That's all for this demo folks! :)\n");
}
