static void
test_Highlight_Excerpt(TestBatchRunner *runner, Searcher *searcher, Obj *query) {
String *content = (String*)SSTR_WRAP_UTF8("content", 7);
Highlighter *highlighter = Highlighter_new(searcher, query, content, 3);
String *highlighted;
Vector *spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(2, 1, 0.0f));
String *raw_excerpt = (String *)SSTR_WRAP_UTF8("a b c", 5);
highlighted = Highlighter_Highlight_Excerpt(highlighter, spans,
raw_excerpt, 0);
TEST_TRUE(runner,
Str_Equals_Utf8(highlighted, "a <strong>b</strong> c", 22),
"basic Highlight_Excerpt");
DECREF(highlighted);
DECREF(spans);
spans = Vec_new(2);
Vec_Push(spans, (Obj*)Span_new(0, 1, 1.0f));
Vec_Push(spans, (Obj*)Span_new(10, 10, 1.0f));
raw_excerpt = (String *)SSTR_WRAP_UTF8(PHI, 2);
highlighted = Highlighter_Highlight_Excerpt(highlighter, spans,
raw_excerpt, 0);
TEST_TRUE(runner,
Str_Equals_Utf8(highlighted, "<strong>Φ</strong>", 23),
"don't surround spans off end of raw excerpt.");
DECREF(highlighted);
DECREF(spans);
spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(3, 1, 1.0f));
raw_excerpt = (String *)SSTR_WRAP_UTF8(PHI " " PHI " " PHI, 8);
highlighted = Highlighter_Highlight_Excerpt(highlighter, spans,
raw_excerpt, 1);
TEST_TRUE(runner,
Str_Equals_Utf8(highlighted,
"Φ <strong>Φ</strong> Φ", 37),
"Highlight_Excerpt pays attention to offset");
DECREF(highlighted);
DECREF(spans);
spans = Vec_new(4);
Vec_Push(spans, (Obj*)Span_new(2, 10, 1.0f));
Vec_Push(spans, (Obj*)Span_new(2, 4, 1.0f));
Vec_Push(spans, (Obj*)Span_new(8, 9, 1.0f));
Vec_Push(spans, (Obj*)Span_new(8, 4, 1.0f));
raw_excerpt = (String *)SSTR_WRAP_UTF8(PHI " Oook. Urk. Ick. " PHI, 21);
highlighted = Highlighter_Highlight_Excerpt(highlighter, spans,
raw_excerpt, 0);
TEST_TRUE(runner,
Str_Equals_Utf8(highlighted,
"Φ <strong>Oook. Urk. Ick.</strong> Φ",
46),
"Highlight_Excerpt works with overlapping spans");
DECREF(highlighted);
DECREF(spans);
DECREF(highlighter);
}
static void
test_seg_name_and_num(TestBatchRunner *runner) {
Segment *segment_z = Seg_new(35);
String *seg_z_name = Seg_num_to_name(35);
TEST_TRUE(runner, Seg_Get_Number(segment_z) == INT64_C(35), "Get_Number");
TEST_TRUE(runner, Str_Equals_Utf8(Seg_Get_Name(segment_z), "seg_z", 5),
"Get_Name");
TEST_TRUE(runner, Str_Equals_Utf8(seg_z_name, "seg_z", 5),
"num_to_name");
DECREF(seg_z_name);
DECREF(segment_z);
}
QueryParser*
QParser_init(QueryParser *self, Schema *schema, Analyzer *analyzer,
String *default_boolop, Vector *fields) {
QueryParserIVARS *const ivars = QParser_IVARS(self);
// Init.
ivars->heed_colons = false;
ivars->lexer = QueryLexer_new();
// Assign.
ivars->schema = (Schema*)INCREF(schema);
ivars->analyzer = (Analyzer*)INCREF(analyzer);
ivars->default_boolop = default_boolop
? Str_Clone(default_boolop)
: Str_new_from_trusted_utf8("OR", 2);
if (fields) {
ivars->fields = Vec_Clone(fields);
for (uint32_t i = 0, max = Vec_Get_Size(fields); i < max; i++) {
CERTIFY(Vec_Fetch(fields, i), STRING);
}
Vec_Sort(ivars->fields);
}
else {
Vector *all_fields = Schema_All_Fields(schema);
uint32_t num_fields = Vec_Get_Size(all_fields);
ivars->fields = Vec_new(num_fields);
for (uint32_t i = 0; i < num_fields; i++) {
String *field = (String*)Vec_Fetch(all_fields, i);
FieldType *type = Schema_Fetch_Type(schema, field);
if (type && FType_Indexed(type)) {
Vec_Push(ivars->fields, INCREF(field));
}
}
DECREF(all_fields);
}
Vec_Sort(ivars->fields);
// Derive default "occur" from default boolean operator.
if (Str_Equals_Utf8(ivars->default_boolop, "OR", 2)) {
ivars->default_occur = SHOULD;
}
else if (Str_Equals_Utf8(ivars->default_boolop, "AND", 3)) {
ivars->default_occur = MUST;
}
else {
THROW(ERR, "Invalid value for default_boolop: %o", ivars->default_boolop);
}
return self;
}
static void
test_Write_File_and_Read_File(TestBatchRunner *runner) {
RAMFolder *folder = RAMFolder_new(NULL);
Segment *segment = Seg_new(100);
Segment *got = Seg_new(100);
String *meta;
String *flotsam = (String*)SSTR_WRAP_UTF8("flotsam", 7);
String *jetsam = (String*)SSTR_WRAP_UTF8("jetsam", 6);
Seg_Set_Count(segment, 111);
Seg_Store_Metadata_Utf8(segment, "foo", 3, (Obj*)Str_newf("bar"));
Seg_Add_Field(segment, flotsam);
Seg_Add_Field(segment, jetsam);
RAMFolder_MkDir(folder, Seg_Get_Name(segment));
Seg_Write_File(segment, (Folder*)folder);
Seg_Read_File(got, (Folder*)folder);
TEST_TRUE(runner, Seg_Get_Count(got) == Seg_Get_Count(segment),
"Round-trip count through file");
TEST_TRUE(runner,
Seg_Field_Num(got, jetsam) == Seg_Field_Num(segment, jetsam),
"Round trip field names through file");
meta = (String*)Seg_Fetch_Metadata_Utf8(got, "foo", 3);
TEST_TRUE(runner,
meta
&& Str_Is_A(meta, STRING)
&& Str_Equals_Utf8(meta, "bar", 3),
"Round trip metadata through file");
DECREF(got);
DECREF(segment);
DECREF(folder);
}
static void
test_numbers(TestBatchRunner *runner) {
Integer *i64 = Int_new(33);
String *json = Json_to_json((Obj*)i64);
String *trimmed = Str_Trim(json);
TEST_TRUE(runner, Str_Equals_Utf8(trimmed, "33", 2), "Integer");
DECREF(json);
DECREF(trimmed);
Float *f64 = Float_new(33.33);
json = Json_to_json((Obj*)f64);
if (json) {
double value = Str_To_F64(json);
double diff = 33.33 - value;
if (diff < 0.0) { diff = 0.0 - diff; }
TEST_TRUE(runner, diff < 0.0001, "Float");
DECREF(json);
}
else {
FAIL(runner, "Float conversion to json failed.");
}
DECREF(i64);
DECREF(f64);
}
NumericType*
NumType_Load_IMP(NumericType *self, Obj *dump) {
UNUSED_VAR(self);
Hash *source = (Hash*)CERTIFY(dump, HASH);
// Get a Class
String *class_name = (String*)Hash_Fetch_Utf8(source, "_class", 6);
String *type_spec = (String*)Hash_Fetch_Utf8(source, "type", 4);
Class *klass = NULL;
if (class_name != NULL && Obj_is_a((Obj*)class_name, STRING)) {
klass = Class_singleton(class_name, NULL);
}
else if (type_spec != NULL && Obj_is_a((Obj*)type_spec, STRING)) {
if (Str_Equals_Utf8(type_spec, "i32_t", 5)) {
klass = INT32TYPE;
}
else if (Str_Equals_Utf8(type_spec, "i64_t", 5)) {
klass = INT64TYPE;
}
else if (Str_Equals_Utf8(type_spec, "f32_t", 5)) {
klass = FLOAT32TYPE;
}
else if (Str_Equals_Utf8(type_spec, "f64_t", 5)) {
klass = FLOAT64TYPE;
}
else {
THROW(ERR, "Unrecognized type string: '%o'", type_spec);
}
}
CERTIFY(klass, CLASS);
NumericType *loaded = (NumericType*)Class_Make_Obj(klass);
// Extract boost.
Obj *boost_dump = Hash_Fetch_Utf8(source, "boost", 5);
float boost = boost_dump ? (float)Json_obj_to_f64(boost_dump) : 1.0f;
// Find boolean properties.
Obj *indexed_dump = Hash_Fetch_Utf8(source, "indexed", 7);
Obj *stored_dump = Hash_Fetch_Utf8(source, "stored", 6);
Obj *sort_dump = Hash_Fetch_Utf8(source, "sortable", 8);
bool indexed = indexed_dump ? Json_obj_to_bool(indexed_dump) : true;
bool stored = stored_dump ? Json_obj_to_bool(stored_dump) : true;
bool sortable = sort_dump ? Json_obj_to_bool(sort_dump) : false;
return NumType_init2(loaded, boost, indexed, stored, sortable);
}
static void
S_test_local_part(TestBatchRunner *runner, const char *source,
const char *wanted, const char *test_name) {
StackString *source_str = SSTR_WRAP_UTF8(source, strlen(source));
String *got = IxFileNames_local_part((String*)source_str);
TEST_TRUE(runner, Str_Equals_Utf8(got, wanted, strlen(wanted)), test_name);
DECREF(got);
}
static void
test_new(TestBatchRunner *runner) {
static char chars[] = "A string " SMILEY " with a smile.";
{
char *buffer = (char*)MALLOCATE(sizeof(chars));
strcpy(buffer, chars);
String *thief = Str_new_steal_utf8(buffer, sizeof(chars) - 1);
TEST_TRUE(runner, Str_Equals_Utf8(thief, chars, sizeof(chars) - 1),
"Str_new_steal_utf8");
DECREF(thief);
}
{
char *buffer = (char*)MALLOCATE(sizeof(chars));
strcpy(buffer, chars);
String *thief
= Str_new_steal_trusted_utf8(buffer, sizeof(chars) - 1);
TEST_TRUE(runner, Str_Equals_Utf8(thief, chars, sizeof(chars) - 1),
"Str_new_steal_trusted_utf8");
DECREF(thief);
}
{
String *wrapper = Str_new_wrap_utf8(chars, sizeof(chars) - 1);
TEST_TRUE(runner, Str_Equals_Utf8(wrapper, chars, sizeof(chars) - 1),
"Str_new_wrap_utf8");
DECREF(wrapper);
}
{
String *wrapper = Str_new_wrap_trusted_utf8(chars, sizeof(chars) - 1);
TEST_TRUE(runner, Str_Equals_Utf8(wrapper, chars, sizeof(chars) - 1),
"Str_new_wrap_trusted_utf8");
DECREF(wrapper);
}
{
String *smiley_str = Str_new_from_char(smiley_cp);
TEST_TRUE(runner, Str_Equals_Utf8(smiley_str, smiley, smiley_len),
"Str_new_from_char");
DECREF(smiley_str);
}
}
static void
test_Utf8_To_String(TestBatchRunner *runner) {
ByteBuf *bb = BB_new_bytes("foo", 3);
{
String *string = BB_Utf8_To_String(bb);
TEST_TRUE(runner, Str_Equals_Utf8(string, "foo", 3), "Utf8_To_String");
DECREF(string);
}
{
String *string = BB_Trusted_Utf8_To_String(bb);
TEST_TRUE(runner, Str_Equals_Utf8(string, "foo", 3),
"Trusted_Utf8_To_String");
DECREF(string);
}
DECREF(bb);
}
void
TestFH_Run_IMP(TestFileHandle *self, TestBatchRunner *runner) {
TestBatchRunner_Plan(runner, (TestBatch*)self, 2);
FileHandle *fh = S_new_filehandle();
String *foo = SSTR_WRAP_UTF8("foo", 3);
TEST_TRUE(runner, Str_Equals_Utf8(FH_Get_Path(fh), "", 0), "Get_Path");
FH_Set_Path(fh, foo);
TEST_TRUE(runner, Str_Equals(FH_Get_Path(fh), (Obj*)foo), "Set_Path");
DECREF(fh);
}
static Method*
S_find_method(Class *self, const char *name) {
size_t name_len = strlen(name);
for (size_t i = 0; self->methods[i]; i++) {
Method *method = self->methods[i];
if (Str_Equals_Utf8(method->name, name, name_len)) {
return method;
}
}
return NULL;
}
static void
test_threads(TestBatchRunner *runner) {
Err_set_error(Err_new(Str_newf("main")));
HANDLE thread = CreateThread(NULL, 0, S_err_thread, runner, 0, NULL);
TEST_TRUE(runner, thread != NULL, "CreateThread succeeds");
DWORD event = WaitForSingleObject(thread, INFINITE);
TEST_INT_EQ(runner, event, WAIT_OBJECT_0, "WaitForSingleObject succeeds");
CloseHandle(thread);
String *mess = Err_Get_Mess(Err_get_error());
TEST_TRUE(runner, Str_Equals_Utf8(mess, "main", 4),
"thread doesn't clobber global error");
}
static void
test_threads(TestBatchRunner *runner) {
Err_set_error(Err_new(Str_newf("main")));
int err;
pthread_t thread;
err = pthread_create(&thread, NULL, S_err_thread, runner);
TEST_INT_EQ(runner, err, 0, "pthread_create succeeds");
err = pthread_join(thread, NULL);
TEST_INT_EQ(runner, err, 0, "pthread_join succeeds");
String *mess = Err_Get_Mess(Err_get_error());
TEST_TRUE(runner, Str_Equals_Utf8(mess, "main", 4),
"thread doesn't clobber global error");
}
static void
test_metadata_storage(TestBatchRunner *runner) {
Segment *segment = Seg_new(1);
String *got;
Seg_Store_Metadata_Utf8(segment, "foo", 3, (Obj*)Str_newf("bar"));
got = (String*)Seg_Fetch_Metadata_Utf8(segment, "foo", 3);
TEST_TRUE(runner,
got
&& Str_Is_A(got, STRING)
&& Str_Equals_Utf8(got, "bar", 3),
"metadata round trip"
);
DECREF(segment);
}
static void
test_Trim(TestBatchRunner *runner) {
String *ws_smiley = S_smiley_with_whitespace(NULL);
String *ws_foo = S_get_str(" foo ");
String *ws_only = S_get_str(" \t \r\n");
String *trimmed = S_get_str("a b");
String *got;
got = Str_Trim(ws_smiley);
TEST_TRUE(runner, Str_Equals_Utf8(got, smiley, smiley_len), "Trim");
DECREF(got);
got = Str_Trim_Top(ws_foo);
TEST_TRUE(runner, Str_Equals_Utf8(got, "foo ", 5), "Trim_Top");
DECREF(got);
got = Str_Trim_Tail(ws_foo);
TEST_TRUE(runner, Str_Equals_Utf8(got, " foo", 5), "Trim_Tail");
DECREF(got);
got = Str_Trim(ws_only);
TEST_TRUE(runner, Str_Equals_Utf8(got, "", 0), "Trim with only whitespace");
DECREF(got);
got = Str_Trim_Top(ws_only);
TEST_TRUE(runner, Str_Equals_Utf8(got, "", 0),
"Trim_Top with only whitespace");
DECREF(got);
got = Str_Trim_Tail(ws_only);
TEST_TRUE(runner, Str_Equals_Utf8(got, "", 0),
"Trim_Tail with only whitespace");
DECREF(got);
got = Str_Trim(trimmed);
TEST_TRUE(runner, Str_Equals(got, (Obj*)trimmed),
"Trim doesn't change trimmed string");
DECREF(got);
got = Str_Trim_Top(trimmed);
TEST_TRUE(runner, Str_Equals(got, (Obj*)trimmed),
"Trim_Top doesn't change trimmed string");
DECREF(got);
got = Str_Trim_Tail(trimmed);
TEST_TRUE(runner, Str_Equals(got, (Obj*)trimmed),
"Trim_Tail doesn't change trimmed string");
DECREF(got);
DECREF(trimmed);
DECREF(ws_only);
DECREF(ws_foo);
DECREF(ws_smiley);
}
Schema*
Schema_Load_IMP(Schema *self, Obj *dump) {
Hash *source = (Hash*)CERTIFY(dump, HASH);
String *class_name
= (String*)CERTIFY(Hash_Fetch_Utf8(source, "_class", 6), STRING);
Class *klass = Class_singleton(class_name, NULL);
Schema *loaded = (Schema*)Class_Make_Obj(klass);
Hash *type_dumps
= (Hash*)CERTIFY(Hash_Fetch_Utf8(source, "fields", 6), HASH);
Vector *analyzer_dumps
= (Vector*)CERTIFY(Hash_Fetch_Utf8(source, "analyzers", 9), VECTOR);
Vector *analyzers
= (Vector*)Freezer_load((Obj*)analyzer_dumps);
UNUSED_VAR(self);
// Start with a blank Schema.
Schema_init(loaded);
SchemaIVARS *const loaded_ivars = Schema_IVARS(loaded);
Vec_Grow(loaded_ivars->uniq_analyzers, Vec_Get_Size(analyzers));
HashIterator *iter = HashIter_new(type_dumps);
while (HashIter_Next(iter)) {
String *field = HashIter_Get_Key(iter);
Hash *type_dump = (Hash*)CERTIFY(HashIter_Get_Value(iter), HASH);
String *type_str = (String*)Hash_Fetch_Utf8(type_dump, "type", 4);
if (type_str) {
if (Str_Equals_Utf8(type_str, "fulltext", 8)) {
// Replace the "analyzer" tick with the real thing.
Obj *tick
= CERTIFY(Hash_Fetch_Utf8(type_dump, "analyzer", 8), OBJ);
Analyzer *analyzer
= (Analyzer*)Vec_Fetch(analyzers,
(uint32_t)Json_obj_to_i64(tick));
if (!analyzer) {
THROW(ERR, "Can't find analyzer for '%o'", field);
}
Hash_Store_Utf8(type_dump, "analyzer", 8, INCREF(analyzer));
FullTextType *type
= (FullTextType*)S_load_type(FULLTEXTTYPE,
(Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else if (Str_Equals_Utf8(type_str, "string", 6)) {
StringType *type
= (StringType*)S_load_type(STRINGTYPE, (Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else if (Str_Equals_Utf8(type_str, "blob", 4)) {
BlobType *type
= (BlobType*)S_load_type(BLOBTYPE, (Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else if (Str_Equals_Utf8(type_str, "i32_t", 5)) {
Int32Type *type
= (Int32Type*)S_load_type(INT32TYPE, (Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else if (Str_Equals_Utf8(type_str, "i64_t", 5)) {
Int64Type *type
= (Int64Type*)S_load_type(INT64TYPE, (Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else if (Str_Equals_Utf8(type_str, "f32_t", 5)) {
Float32Type *type
= (Float32Type*)S_load_type(FLOAT32TYPE, (Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else if (Str_Equals_Utf8(type_str, "f64_t", 5)) {
Float64Type *type
= (Float64Type*)S_load_type(FLOAT64TYPE, (Obj*)type_dump);
Schema_Spec_Field(loaded, field, (FieldType*)type);
DECREF(type);
}
else {
THROW(ERR, "Unknown type '%o' for field '%o'", type_str, field);
}
}
else {
FieldType *type
= (FieldType*)CERTIFY(Freezer_load((Obj*)type_dump),
FIELDTYPE);
Schema_Spec_Field(loaded, field, type);
DECREF(type);
}
}
DECREF(iter);
DECREF(analyzers);
return loaded;
}
static void
test_tokenizer(TestBatchRunner *runner) {
StandardTokenizer *tokenizer = StandardTokenizer_new();
String *word = SSTR_WRAP_C(
" ."
"tha\xCC\x82t's"
":"
"1,02\xC2\xADZ4.38"
"\xE0\xB8\x81\xC2\xAD\xC2\xAD"
"\xF0\xA0\x80\x80"
"a"
"/");
Vector *got = StandardTokenizer_Split(tokenizer, word);
String *token = (String*)Vec_Fetch(got, 0);
TEST_TRUE(runner,
token
&& Str_is_a(token, STRING)
&& Str_Equals_Utf8(token, "tha\xcc\x82t's", 8),
"Token: %s", Str_Get_Ptr8(token));
token = (String*)Vec_Fetch(got, 1);
TEST_TRUE(runner,
token
&& Str_is_a(token, STRING)
&& Str_Equals_Utf8(token, "1,02\xC2\xADZ4.38", 11),
"Token: %s", Str_Get_Ptr8(token));
token = (String*)Vec_Fetch(got, 2);
TEST_TRUE(runner,
token
&& Str_is_a(token, STRING)
&& Str_Equals_Utf8(token, "\xE0\xB8\x81\xC2\xAD\xC2\xAD", 7),
"Token: %s", Str_Get_Ptr8(token));
token = (String*)Vec_Fetch(got, 3);
TEST_TRUE(runner,
token
&& Str_is_a(token, STRING)
&& Str_Equals_Utf8(token, "\xF0\xA0\x80\x80", 4),
"Token: %s", Str_Get_Ptr8(token));
token = (String*)Vec_Fetch(got, 4);
TEST_TRUE(runner,
token
&& Str_is_a(token, STRING)
&& Str_Equals_Utf8(token, "a", 1),
"Token: %s", Str_Get_Ptr8(token));
DECREF(got);
FSFolder *modules_folder = TestUtils_modules_folder();
if (modules_folder == NULL) {
SKIP(runner, 1372, "Can't locate test data");
}
else {
String *path = Str_newf("unicode/ucd/WordBreakTest.json");
Vector *tests = (Vector*)Json_slurp_json((Folder*)modules_folder, path);
if (!tests) { RETHROW(Err_get_error()); }
for (uint32_t i = 0, max = Vec_Get_Size(tests); i < max; i++) {
Hash *test = (Hash*)Vec_Fetch(tests, i);
String *text = (String*)Hash_Fetch_Utf8(test, "text", 4);
Vector *wanted = (Vector*)Hash_Fetch_Utf8(test, "words", 5);
Vector *got = StandardTokenizer_Split(tokenizer, text);
TEST_TRUE(runner, Vec_Equals(wanted, (Obj*)got), "UCD test #%d", i + 1);
DECREF(got);
}
DECREF(tests);
DECREF(modules_folder);
DECREF(path);
}
DECREF(tokenizer);
}
static void
test_Raw_Excerpt(TestBatchRunner *runner, Searcher *searcher, Obj *query) {
String *content = (String*)SSTR_WRAP_UTF8("content", 7);
Highlighter *highlighter = Highlighter_new(searcher, query, content, 6);
int32_t top;
String *raw_excerpt;
String *field_val = (String *)SSTR_WRAP_UTF8("Ook. Urk. Ick. ", 18);
Vector *spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(0, 18, 1.0f));
HeatMap *heat_map = HeatMap_new(spans, 133);
DECREF(spans);
raw_excerpt = Highlighter_Raw_Excerpt(highlighter, field_val, &top,
heat_map);
TEST_TRUE(runner,
Str_Equals_Utf8(raw_excerpt, "Ook.", 4),
"Raw_Excerpt at top %s", Str_Get_Ptr8(raw_excerpt));
TEST_TRUE(runner,
top == 0,
"top is 0");
DECREF(raw_excerpt);
DECREF(heat_map);
spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(6, 12, 1.0f));
heat_map = HeatMap_new(spans, 133);
DECREF(spans);
raw_excerpt = Highlighter_Raw_Excerpt(highlighter, field_val, &top,
heat_map);
TEST_TRUE(runner,
Str_Equals_Utf8(raw_excerpt, "Urk.", 4),
"Raw_Excerpt in middle, with 2 bounds");
TEST_TRUE(runner,
top == 6,
"top in the middle modified by Raw_Excerpt");
DECREF(raw_excerpt);
DECREF(heat_map);
field_val = (String *)SSTR_WRAP_UTF8("Ook urk ick i.", 14);
spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(12, 1, 1.0f));
heat_map = HeatMap_new(spans, 133);
DECREF(spans);
raw_excerpt = Highlighter_Raw_Excerpt(highlighter, field_val, &top,
heat_map);
TEST_TRUE(runner,
Str_Equals_Utf8(raw_excerpt, ELLIPSIS " i.", 6),
"Ellipsis at top");
TEST_TRUE(runner,
top == 10,
"top correct when leading ellipsis inserted");
DECREF(heat_map);
DECREF(raw_excerpt);
field_val = (String *)SSTR_WRAP_UTF8("Urk. Iz no good.", 17);
spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(6, 2, 1.0f));
heat_map = HeatMap_new(spans, 133);
DECREF(spans);
raw_excerpt = Highlighter_Raw_Excerpt(highlighter, field_val, &top,
heat_map);
TEST_TRUE(runner,
Str_Equals_Utf8(raw_excerpt, "Iz no" ELLIPSIS, 8),
"Ellipsis at end");
TEST_TRUE(runner,
top == 6,
"top trimmed");
DECREF(heat_map);
DECREF(raw_excerpt);
// Words longer than excerpt len
field_val = (String *)SSTR_WRAP_UTF8("abc/def/ghi/jkl/mno", 19);
spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(0, 3, 1.0f));
heat_map = HeatMap_new(spans, 133);
DECREF(spans);
raw_excerpt = Highlighter_Raw_Excerpt(highlighter, field_val, &top,
heat_map);
TEST_TRUE(runner,
Str_Equals_Utf8(raw_excerpt, "abc/d" ELLIPSIS, 8),
"Long word at top");
DECREF(heat_map);
DECREF(raw_excerpt);
spans = Vec_new(1);
Vec_Push(spans, (Obj*)Span_new(8, 3, 1.0f));
heat_map = HeatMap_new(spans, 133);
DECREF(spans);
raw_excerpt = Highlighter_Raw_Excerpt(highlighter, field_val, &top,
heat_map);
TEST_TRUE(runner,
Str_Equals_Utf8(raw_excerpt, ELLIPSIS " f/g" ELLIPSIS, 10),
"Long word in middle");
DECREF(heat_map);
DECREF(raw_excerpt);
DECREF(highlighter);
}
