void RDFParserSerd::doParse(const char *fileName, const char *baseUri, RDFNotation notation, bool ignoreErrors, RDFCallback *callback) {
this->callback = callback;
this->numByte = fileUtil::getSize(fileName);
// Create Base URI and environment
SerdURI base_uri = SERD_URI_NULL;
SerdNode base = serd_node_new_file_uri((const uint8_t *)fileName, NULL, &base_uri, false);
env = serd_env_new(&base);
SerdReader* reader = serd_reader_new(
getParserType(notation), this, NULL,
(SerdBaseSink)hdtserd_basechanged,
(SerdPrefixSink)hdtserd_prefixchanged,
(SerdStatementSink)hdtserd_process_triple,
(SerdEndSink)hdtserd_end);
serd_reader_set_error_sink(reader, hdtserd_error, NULL);
const uint8_t* input=serd_uri_to_path((const uint8_t *)fileName);
FILE *in_fd = fopen((const char*)input, "r");
// TODO: fadvise sequential
if(in_fd==NULL) {
throw "Could not open input file for parsing";
}
SerdStatus status = serd_reader_read_file_handle(reader, in_fd, (const uint8_t *)fileName);
serd_reader_free(reader);
fclose(in_fd);
serd_env_free(env);
serd_node_free(&base);
}
int
main(int argc, char** argv)
{
if (argc < 2) {
return print_usage(argv[0], true);
}
FILE* in_fd = NULL;
SerdSyntax input_syntax = SERD_TURTLE;
SerdSyntax output_syntax = SERD_NTRIPLES;
bool from_file = true;
const uint8_t* in_name = NULL;
int a = 1;
for (; a < argc && argv[a][0] == '-'; ++a) {
if (argv[a][1] == '\0') {
in_name = (const uint8_t*)"(stdin)";
in_fd = stdin;
break;
} else if (argv[a][1] == 'h') {
return print_usage(argv[0], false);
} else if (argv[a][1] == 'v') {
return print_version();
} else if (argv[a][1] == 's') {
in_name = (const uint8_t*)"(string)";
from_file = false;
++a;
break;
} else if (argv[a][1] == 'i') {
if (++a == argc) {
fprintf(stderr, "Missing value for -i\n");
return 1;
}
if (!set_syntax(&input_syntax, argv[a])) {
return 1;
}
} else if (argv[a][1] == 'o') {
if (++a == argc) {
fprintf(stderr, "Missing value for -o\n");
return 1;
}
if (!set_syntax(&output_syntax, argv[a])) {
return 1;
}
} else {
fprintf(stderr, "Unknown option `%s'\n", argv[a]);
return print_usage(argv[0], true);
}
}
if (a == argc) {
fprintf(stderr, "Missing input\n");
return 1;
}
const uint8_t* input = (const uint8_t*)argv[a++];
uint8_t* in_path = NULL;
if (from_file) {
in_name = in_name ? in_name : input;
if (!in_fd) {
in_path = absolute_path(serd_uri_to_path(in_name));
if (!in_path || !(in_fd = fopen((const char*)in_path, "rb"))) {
return 1;
}
}
}
SerdURI base_uri = SERD_URI_NULL;
SerdNode base_uri_node = SERD_NODE_NULL;
if (a < argc) { // Base URI given on command line
base_uri_node = serd_node_new_uri_from_string(
(const uint8_t*)argv[a], NULL, &base_uri);
} else if (from_file) { // Use input file URI
base_uri_node = serd_node_new_file_uri(in_path, NULL, &base_uri, false);
}
if (!base_uri_node.buf) {
fprintf(stderr, "Missing base URI\n");
return 1;
}
SordWorld* world = sord_world_new();
SordModel* sord = sord_new(world, SORD_SPO|SORD_OPS, false);
SerdEnv* env = serd_env_new(&base_uri_node);
SerdReader* reader = sord_new_reader(sord, env, input_syntax, NULL);
const SerdStatus status = (from_file)
? serd_reader_read_file_handle(reader, in_fd, in_name)
: serd_reader_read_string(reader, input);
serd_reader_free(reader);
fprintf(stderr, "Loaded %lu statements\n",
(unsigned long)sord_num_quads(sord));
SerdEnv* write_env = serd_env_new(&base_uri_node);
int output_style = SERD_STYLE_RESOLVED;
if (output_syntax == SERD_NTRIPLES) {
output_style |= SERD_STYLE_ASCII;
} else {
output_style |= SERD_STYLE_CURIED | SERD_STYLE_ABBREVIATED;
}
SerdWriter* writer = serd_writer_new(
output_syntax,
(SerdStyle)output_style,
write_env, &base_uri, serd_file_sink, stdout);
// Write @prefix directives
serd_env_foreach(env,
(SerdPrefixSink)serd_writer_set_prefix,
writer);
// Write statements
sord_write(sord, writer, NULL);
serd_writer_finish(writer);
serd_writer_free(writer);
serd_env_free(env);
serd_env_free(write_env);
serd_node_free(&base_uri_node);
sord_free(sord);
sord_world_free(world);
return (status > SERD_FAILURE) ? 1 : 0;
}
const char*
lilv_uri_to_path(const char* uri)
{
return (const char*)serd_uri_to_path((const uint8_t*)uri);
}
int
main(void)
{
#define MAX 1000000
#define NUM_TESTS 1000
for (int i = 0; i < NUM_TESTS; ++i) {
double dbl = rand() % MAX;
dbl += (rand() % MAX) / (double)MAX;
if (!test_strtod(dbl, 1 / (double)MAX)) {
return 1;
}
}
const double expt_test_nums[] = {
2.0E18, -5e19, +8e20, 2e+34, -5e-5, 8e0, 9e-0, 2e+0
};
const char* expt_test_strs[] = {
"02e18", "-5e019", "+8e20", "2E+34", "-5E-5", "8E0", "9e-0", " 2e+0"
};
for (unsigned i = 0; i < sizeof(expt_test_nums) / sizeof(double); ++i) {
const double num = serd_strtod(expt_test_strs[i], NULL);
const double delta = fabs(num - expt_test_nums[i]);
if (delta > DBL_EPSILON) {
return failure("Parsed `%s' %lf != %lf (delta %lf)\n",
expt_test_strs[i], num, expt_test_nums[i], delta);
}
}
// Test serd_node_new_decimal
const double dbl_test_nums[] = {
0.0, 9.0, 10.0, .01, 2.05, -16.00001, 5.000000005, 0.0000000001, NAN, INFINITY
};
const char* dbl_test_strs[] = {
"0.0", "9.0", "10.0", "0.01", "2.05", "-16.00001", "5.00000001", "0.0", NULL, NULL
};
for (unsigned i = 0; i < sizeof(dbl_test_nums) / sizeof(double); ++i) {
SerdNode node = serd_node_new_decimal(dbl_test_nums[i], 8);
const bool pass = (node.buf && dbl_test_strs[i])
? !strcmp((const char*)node.buf, (const char*)dbl_test_strs[i])
: ((const char*)node.buf == dbl_test_strs[i]);
if (!pass) {
return failure("Serialised `%s' != %s\n",
node.buf, dbl_test_strs[i]);
}
const size_t len = node.buf ? strlen((const char*)node.buf) : 0;
if (node.n_bytes != len || node.n_chars != len) {
return failure("Length %zu,%zu != %zu\n",
node.n_bytes, node.n_chars, len);
}
serd_node_free(&node);
}
// Test serd_node_new_integer
const long int_test_nums[] = {
0, -0, -23, 23, -12340, 1000, -1000
};
const char* int_test_strs[] = {
"0", "0", "-23", "23", "-12340", "1000", "-1000"
};
for (unsigned i = 0; i < sizeof(int_test_nums) / sizeof(double); ++i) {
SerdNode node = serd_node_new_integer(int_test_nums[i]);
if (strcmp((const char*)node.buf, (const char*)int_test_strs[i])) {
return failure("Serialised `%s' != %s\n",
node.buf, int_test_strs[i]);
}
const size_t len = strlen((const char*)node.buf);
if (node.n_bytes != len || node.n_chars != len) {
return failure("Length %zu,%zu != %zu\n",
node.n_bytes, node.n_chars, len);
}
serd_node_free(&node);
}
// Test serd_node_new_blob
for (size_t size = 0; size < 256; ++size) {
uint8_t* data = (uint8_t*)malloc(size);
for (size_t i = 0; i < size; ++i) {
data[i] = (uint8_t)(rand() % 256);
}
SerdNode blob = serd_node_new_blob(data, size, size % 5);
if (blob.n_bytes != blob.n_chars) {
return failure("Blob %zu bytes != %zu chars\n",
blob.n_bytes, blob.n_chars);
}
size_t out_size;
uint8_t* out = (uint8_t*)serd_base64_decode(
blob.buf, blob.n_bytes, &out_size);
if (out_size != size) {
return failure("Blob size %zu != %zu\n", out_size, size);
}
for (size_t i = 0; i < size; ++i) {
if (out[i] != data[i]) {
return failure("Corrupt blob at byte %zu\n", i);
}
}
serd_node_free(&blob);
free(out);
free(data);
}
// Test serd_strlen
const uint8_t str[] = { '"', '5', 0xE2, 0x82, 0xAC, '"', '\n', 0 };
size_t n_bytes;
SerdNodeFlags flags;
size_t len = serd_strlen(str, &n_bytes, &flags);
if (len != 5 || n_bytes != 7
|| flags != (SERD_HAS_QUOTE|SERD_HAS_NEWLINE)) {
return failure("Bad serd_strlen(%s) len=%zu n_bytes=%zu flags=%u\n",
str, len, n_bytes, flags);
}
len = serd_strlen(str, NULL, &flags);
if (len != 5) {
return failure("Bad serd_strlen(%s) len=%zu flags=%u\n",
str, len, flags);
}
// Test serd_strerror
const uint8_t* msg = NULL;
if (strcmp((const char*)(msg = serd_strerror(SERD_SUCCESS)), "Success")) {
return failure("Bad message `%s' for SERD_SUCCESS\n", msg);
}
for (int i = SERD_FAILURE; i <= SERD_ERR_INTERNAL; ++i) {
msg = serd_strerror((SerdStatus)i);
if (!strcmp((const char*)msg, "Success")) {
return failure("Bad message `%s' for (SerdStatus)%d\n", msg, i);
}
}
msg = serd_strerror((SerdStatus)-1);
// Test serd_uri_to_path
const uint8_t* uri = (const uint8_t*)"file:///home/user/foo.ttl";
if (strcmp((const char*)serd_uri_to_path(uri), "/home/user/foo.ttl")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
uri = (const uint8_t*)"file://localhost/home/user/foo.ttl";
if (strcmp((const char*)serd_uri_to_path(uri), "/home/user/foo.ttl")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
uri = (const uint8_t*)"file:illegal/file/uri";
if (serd_uri_to_path(uri)) {
return failure("Converted invalid URI `%s' to path `%s'\n",
uri, serd_uri_to_path(uri));
}
uri = (const uint8_t*)"file:///c:/awful/system";
if (strcmp((const char*)serd_uri_to_path(uri), "c:/awful/system")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
uri = (const uint8_t*)"file:///c:awful/system";
if (strcmp((const char*)serd_uri_to_path(uri), "/c:awful/system")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
uri = (const uint8_t*)"file:///0/1";
if (strcmp((const char*)serd_uri_to_path(uri), "/0/1")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
uri = (const uint8_t*)"C:\\Windows\\Sucks";
if (strcmp((const char*)serd_uri_to_path(uri), "C:\\Windows\\Sucks")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
uri = (const uint8_t*)"C|/Windows/Sucks";
if (strcmp((const char*)serd_uri_to_path(uri), "C|/Windows/Sucks")) {
return failure("Bad path %s for %s\n", serd_uri_to_path(uri), uri);
}
// Test serd_node_new_file_uri and serd_file_uri_parse
SerdURI furi;
const uint8_t* path_str = USTR("C:/My 100%");
SerdNode file_node = serd_node_new_file_uri(path_str, 0, &furi, true);
uint8_t* hostname = NULL;
uint8_t* out_path = serd_file_uri_parse(file_node.buf, &hostname);
if (strcmp((const char*)file_node.buf, "file:///C:/My%20100%%")) {
return failure("Bad URI %s\n", file_node.buf);
} else if (hostname) {
return failure("hostname `%s' shouldn't exist\n", hostname);
} else if (strcmp((const char*)path_str, (const char*)out_path)) {
return failure("path=>URI=>path failure %s => %s => %s\n",
path_str, file_node.buf, out_path);
}
free(out_path);
serd_node_free(&file_node);
path_str = USTR("C:\\Pointless Space");
file_node = serd_node_new_file_uri(path_str, USTR("pwned"), 0, true);
hostname = NULL;
out_path = serd_file_uri_parse(file_node.buf, &hostname);
if (strcmp((const char*)file_node.buf, "file://pwned/C:/Pointless%20Space")) {
return failure("Bad URI %s\n", file_node.buf);
} else if (!hostname || strcmp((const char*)hostname, "pwned")) {
return failure("Bad hostname `%s'\n", hostname);
} else if (strcmp((const char*)out_path, "C:/Pointless Space")) {
return failure("path=>URI=>path failure %s => %s => %s\n",
path_str, file_node.buf, out_path);
}
free(hostname);
free(out_path);
serd_node_free(&file_node);
path_str = USTR("/foo/bar");
file_node = serd_node_new_file_uri(path_str, 0, 0, true);
hostname = NULL;
out_path = serd_file_uri_parse(file_node.buf, &hostname);
if (strcmp((const char*)file_node.buf, "file:///foo/bar")) {
return failure("Bad URI %s\n", file_node.buf);
} else if (hostname) {
return failure("hostname `%s' shouldn't exist\n", hostname);
} else if (strcmp((const char*)path_str, (const char*)out_path)) {
return failure("path=>URI=>path failure %s => %s => %s\n",
path_str, file_node.buf, out_path);
}
free(out_path);
serd_node_free(&file_node);
path_str = USTR("/foo/bar");
file_node = serd_node_new_file_uri(path_str, USTR("localhost"), 0, true);
out_path = serd_file_uri_parse(file_node.buf, &hostname);
if (strcmp((const char*)file_node.buf, "file://localhost/foo/bar")) {
return failure("Bad URI %s\n", file_node.buf);
} else if (strcmp((const char*)hostname, "localhost")) {
return failure("incorrect hostname `%s'\n", hostname);
} else if (strcmp((const char*)path_str, (const char*)out_path)) {
return failure("path=>URI=>path failure %s => %s => %s\n",
path_str, file_node.buf, out_path);
}
free(hostname);
free(out_path);
serd_node_free(&file_node);
path_str = USTR("a/relative path");
file_node = serd_node_new_file_uri(path_str, 0, 0, false);
out_path = serd_file_uri_parse(file_node.buf, &hostname);
if (strcmp((const char*)file_node.buf, "a/relative path")) {
return failure("Bad URI %s\n", file_node.buf);
} else if (hostname) {
return failure("hostname `%s' shouldn't exist\n", hostname);
} else if (strcmp((const char*)path_str, (const char*)out_path)) {
return failure("path=>URI=>path failure %s => %s => %s\n",
path_str, file_node.buf, out_path);
}
free(hostname);
free(out_path);
serd_node_free(&file_node);
if (serd_file_uri_parse(USTR("file://invalid"), NULL)) {
return failure("successfully parsed bogus URI <file://invalid>\n");
}
out_path = serd_file_uri_parse(USTR("file://host/foo/%XYbar"), NULL);
if (strcmp((const char*)out_path, "/foo/bar")) {
return failure("bad tolerance of junk escape: `%s'\n", out_path);
}
free(out_path);
out_path = serd_file_uri_parse(USTR("file://host/foo/%0Abar"), NULL);
if (strcmp((const char*)out_path, "/foo/bar")) {
return failure("bad tolerance of junk escape: `%s'\n", out_path);
}
free(out_path);
// Test serd_node_equals
const uint8_t replacement_char_str[] = { 0xEF, 0xBF, 0xBD, 0 };
SerdNode lhs = serd_node_from_string(SERD_LITERAL, replacement_char_str);
SerdNode rhs = serd_node_from_string(SERD_LITERAL, USTR("123"));
if (serd_node_equals(&lhs, &rhs)) {
return failure("%s == %s\n", lhs.buf, rhs.buf);
}
SerdNode qnode = serd_node_from_string(SERD_CURIE, USTR("foo:bar"));
if (serd_node_equals(&lhs, &qnode)) {
return failure("%s == %s\n", lhs.buf, qnode.buf);
}
if (!serd_node_equals(&lhs, &lhs)) {
return failure("%s != %s\n", lhs.buf, lhs.buf);
}
SerdNode null_copy = serd_node_copy(&SERD_NODE_NULL);
if (!serd_node_equals(&SERD_NODE_NULL, &null_copy)) {
return failure("copy of null node != null node\n");
}
// Test serd_node_from_string
SerdNode node = serd_node_from_string(SERD_LITERAL, (const uint8_t*)"hello\"");
if (node.n_bytes != 6 || node.n_chars != 6 || node.flags != SERD_HAS_QUOTE
|| strcmp((const char*)node.buf, "hello\"")) {
return failure("Bad node %s %zu %zu %d %d\n",
node.buf, node.n_bytes, node.n_chars, node.flags, node.type);
}
node = serd_node_from_string(SERD_URI, NULL);
if (!serd_node_equals(&node, &SERD_NODE_NULL)) {
return failure("Creating node from NULL string failed\n");
}
// Test serd_node_new_uri_from_string
SerdURI base_uri;
SerdNode base = serd_node_new_uri_from_string(USTR("http://example.org/"),
NULL, &base_uri);
SerdNode nil = serd_node_new_uri_from_string(NULL, &base_uri, NULL);
SerdNode nil2 = serd_node_new_uri_from_string(USTR(""), &base_uri, NULL);
if (nil.type != SERD_URI || strcmp((const char*)nil.buf, (const char*)base.buf) ||
nil2.type != SERD_URI || strcmp((const char*)nil2.buf, (const char*)base.buf)) {
return failure("URI %s != base %s\n", nil.buf, base.buf);
}
serd_node_free(&base);
serd_node_free(&nil);
serd_node_free(&nil2);
// Test SerdEnv
SerdNode u = serd_node_from_string(SERD_URI, USTR("http://example.org/foo"));
SerdNode b = serd_node_from_string(SERD_CURIE, USTR("invalid"));
SerdNode c = serd_node_from_string(SERD_CURIE, USTR("eg.2:b"));
SerdEnv* env = serd_env_new(NULL);
serd_env_set_prefix_from_strings(env, USTR("eg.2"), USTR("http://example.org/"));
if (!serd_env_set_base_uri(env, &node)) {
return failure("Set base URI to %s\n", node.buf);
}
if (!serd_node_equals(serd_env_get_base_uri(env, NULL), &node)) {
return failure("Base URI mismatch\n");
}
SerdChunk prefix, suffix;
if (!serd_env_expand(env, &b, &prefix, &suffix)) {
return failure("Expanded invalid curie %s\n", b.buf);
}
SerdNode xnode = serd_env_expand_node(env, &node);
if (!serd_node_equals(&xnode, &SERD_NODE_NULL)) {
return failure("Expanded %s to %s\n", c.buf, xnode.buf);
}
SerdNode xu = serd_env_expand_node(env, &u);
if (strcmp((const char*)xu.buf, "http://example.org/foo")) {
return failure("Expanded %s to %s\n", c.buf, xu.buf);
}
serd_node_free(&xu);
SerdNode badpre = serd_node_from_string(SERD_CURIE, USTR("hm:what"));
SerdNode xbadpre = serd_env_expand_node(env, &badpre);
if (!serd_node_equals(&xbadpre, &SERD_NODE_NULL)) {
return failure("Expanded invalid curie %s\n", badpre.buf);
}
SerdNode xc = serd_env_expand_node(env, &c);
if (strcmp((const char*)xc.buf, "http://example.org/b")) {
return failure("Expanded %s to %s\n", c.buf, xc.buf);
}
serd_node_free(&xc);
if (!serd_env_set_prefix(env, &SERD_NODE_NULL, &SERD_NODE_NULL)) {
return failure("Set NULL prefix\n");
}
const SerdNode lit = serd_node_from_string(SERD_LITERAL, USTR("hello"));
if (!serd_env_set_prefix(env, &b, &lit)) {
return failure("Set prefix to literal\n");
}
int n_prefixes = 0;
serd_env_set_prefix_from_strings(env, USTR("eg.2"), USTR("http://example.org/"));
serd_env_foreach(env, count_prefixes, &n_prefixes);
if (n_prefixes != 1) {
return failure("Bad prefix count %d\n", n_prefixes);
}
SerdNode shorter_uri = serd_node_from_string(SERD_URI, USTR("urn:foo"));
SerdNode prefix_name;
if (serd_env_qualify(env, &shorter_uri, &prefix_name, &suffix)) {
return failure("Qualified %s\n", shorter_uri.buf);
}
// Test SerdReader and SerdWriter
const char* path = "serd_test.ttl";
FILE* fd = fopen(path, "w");
if (!fd) {
return failure("Failed to open file %s\n", path);
}
SerdWriter* writer = serd_writer_new(
SERD_TURTLE, (SerdStyle)0, env, NULL, serd_file_sink, fd);
if (!writer) {
return failure("Failed to create writer\n");
}
serd_writer_chop_blank_prefix(writer, USTR("tmp"));
serd_writer_chop_blank_prefix(writer, NULL);
if (!serd_writer_set_base_uri(writer, &lit)) {
return failure("Set base URI to %s\n", lit.buf);
}
if (!serd_writer_set_prefix(writer, &lit, &lit)) {
return failure("Set prefix %s to %s\n", lit.buf, lit.buf);
}
if (!serd_writer_end_anon(writer, NULL)) {
return failure("Ended non-existent anonymous node\n");
}
if (serd_writer_get_env(writer) != env) {
return failure("Writer has incorrect env\n");
}
uint8_t buf[] = { 0x80, 0, 0, 0, 0 };
SerdNode s = serd_node_from_string(SERD_URI, USTR(""));
SerdNode p = serd_node_from_string(SERD_URI, USTR("http://example.org/pred"));
SerdNode o = serd_node_from_string(SERD_LITERAL, buf);
// Write 3 invalid statements (should write nothing)
const SerdNode* junk[][5] = { { &s, &p, NULL, NULL, NULL },
{ &s, NULL, &o, NULL, NULL },
{ NULL, &p, &o, NULL, NULL },
{ &s, &p, &SERD_NODE_NULL, NULL, NULL },
{ &s, &SERD_NODE_NULL, &o, NULL, NULL },
{ &SERD_NODE_NULL, &p, &o, NULL, NULL },
{ &s, &o, &o, NULL, NULL },
{ &o, &p, &o, NULL, NULL },
{ NULL, NULL, NULL, NULL, NULL } };
for (unsigned i = 0; i < sizeof(junk) / (sizeof(SerdNode*) * 5); ++i) {
if (!serd_writer_write_statement(
writer, 0, NULL,
junk[i][0], junk[i][1], junk[i][2], junk[i][3], junk[i][4])) {
return failure("Successfully wrote junk statement %d\n", i);
}
}
const SerdNode t = serd_node_from_string(SERD_URI, USTR("urn:Type"));
const SerdNode l = serd_node_from_string(SERD_LITERAL, USTR("en"));
const SerdNode* good[][5] = { { &s, &p, &o, NULL, NULL },
{ &s, &p, &o, &SERD_NODE_NULL, &SERD_NODE_NULL },
{ &s, &p, &o, &t, NULL },
{ &s, &p, &o, NULL, &l },
{ &s, &p, &o, &t, &l },
{ &s, &p, &o, &t, &SERD_NODE_NULL },
{ &s, &p, &o, &SERD_NODE_NULL, &l },
{ &s, &p, &o, NULL, &SERD_NODE_NULL },
{ &s, &p, &o, &SERD_NODE_NULL, NULL },
{ &s, &p, &o, &SERD_NODE_NULL, NULL } };
for (unsigned i = 0; i < sizeof(good) / (sizeof(SerdNode*) * 5); ++i) {
if (serd_writer_write_statement(
writer, 0, NULL,
good[i][0], good[i][1], good[i][2], good[i][3], good[i][4])) {
return failure("Failed to write good statement %d\n", i);
}
}
// Write 1 statement with bad UTF-8 (should be replaced)
if (serd_writer_write_statement(writer, 0, NULL,
&s, &p, &o, NULL, NULL)) {
return failure("Failed to write junk UTF-8\n");
}
// Write 1 valid statement
o = serd_node_from_string(SERD_LITERAL, USTR("hello"));
if (serd_writer_write_statement(writer, 0, NULL,
&s, &p, &o, NULL, NULL)) {
return failure("Failed to write valid statement\n");
}
serd_writer_free(writer);
// Test chunk sink
SerdChunk chunk = { NULL, 0 };
writer = serd_writer_new(
SERD_TURTLE, (SerdStyle)0, env, NULL, serd_chunk_sink, &chunk);
o = serd_node_from_string(SERD_URI, USTR("http://example.org/base"));
if (serd_writer_set_base_uri(writer, &o)) {
return failure("Failed to write to chunk sink\n");
}
serd_writer_free(writer);
uint8_t* out = serd_chunk_sink_finish(&chunk);
if (strcmp((const char*)out, "@base <http://example.org/base> .\n")) {
return failure("Incorrect chunk output:\n%s\n", chunk.buf);
}
free(out);
// Rewind and test reader
fseek(fd, 0, SEEK_SET);
ReaderTest* rt = (ReaderTest*)malloc(sizeof(ReaderTest));
rt->n_statements = 0;
rt->graph = NULL;
SerdReader* reader = serd_reader_new(
SERD_TURTLE, rt, free,
NULL, NULL, test_sink, NULL);
if (!reader) {
return failure("Failed to create reader\n");
}
if (serd_reader_get_handle(reader) != rt) {
return failure("Corrupt reader handle\n");
}
SerdNode g = serd_node_from_string(SERD_URI, USTR("http://example.org/"));
serd_reader_set_default_graph(reader, &g);
serd_reader_add_blank_prefix(reader, USTR("tmp"));
serd_reader_add_blank_prefix(reader, NULL);
if (!serd_reader_read_file(reader, USTR("http://notafile"))) {
return failure("Apparently read an http URI\n");
}
if (!serd_reader_read_file(reader, USTR("file:///better/not/exist"))) {
return failure("Apprently read a non-existent file\n");
}
SerdStatus st = serd_reader_read_file(reader, USTR(path));
if (st) {
return failure("Error reading file (%s)\n", serd_strerror(st));
}
if (rt->n_statements != 12) {
return failure("Bad statement count %d\n", rt->n_statements);
} else if (!rt->graph || !rt->graph->buf ||
strcmp((const char*)rt->graph->buf, "http://example.org/")) {
return failure("Bad graph %p\n", rt->graph);
}
if (!serd_reader_read_string(reader, USTR("This isn't Turtle at all."))) {
return failure("Parsed invalid string successfully.\n");
}
serd_reader_free(reader);
fclose(fd);
serd_env_free(env);
printf("Success\n");
return 0;
}
int
main(int argc, char** argv)
{
if (argc < 2) {
return print_usage(argv[0], true);
}
int a = 1;
for (; a < argc && argv[a][0] == '-'; ++a) {
if (argv[a][1] == 'l') {
one_line_errors = true;
} else if (argv[a][1] == 'v') {
return print_version();
} else {
fprintf(stderr, "%s: Unknown option `%s'\n", argv[0], argv[a]);
return print_usage(argv[0], true);
}
}
SordWorld* world = sord_world_new();
SordModel* model = sord_new(world, SORD_SPO|SORD_OPS, false);
SerdEnv* env = serd_env_new(&SERD_NODE_NULL);
SerdReader* reader = sord_new_reader(model, env, SERD_TURTLE, NULL);
for (; a < argc; ++a) {
const uint8_t* input = (const uint8_t*)argv[a];
uint8_t* in_path = absolute_path(serd_uri_to_path(input));
if (!in_path) {
fprintf(stderr, "Skipping file %s\n", input);
continue;
}
SerdURI base_uri;
SerdNode base_uri_node = serd_node_new_file_uri(
in_path, NULL, &base_uri, false);
serd_env_set_base_uri(env, &base_uri_node);
const SerdStatus st = serd_reader_read_file(reader, in_path);
if (st) {
fprintf(stderr, "error reading %s: %s\n",
in_path, serd_strerror(st));
}
serd_node_free(&base_uri_node);
free(in_path);
}
serd_reader_free(reader);
serd_env_free(env);
#define URI(prefix, suffix) \
uris.prefix##_##suffix = sord_new_uri(world, NS_##prefix #suffix)
URIs uris;
URI(foaf, Document);
URI(owl, AnnotationProperty);
URI(owl, Class);
URI(owl, DatatypeProperty);
URI(owl, FunctionalProperty);
URI(owl, InverseFunctionalProperty);
URI(owl, ObjectProperty);
URI(owl, OntologyProperty);
URI(owl, Restriction);
URI(owl, Thing);
URI(owl, cardinality);
URI(owl, equivalentClass);
URI(owl, minCardinality);
URI(owl, onDatatype);
URI(owl, onProperty);
URI(owl, someValuesFrom);
URI(owl, withRestrictions);
URI(rdf, PlainLiteral);
URI(rdf, Property);
URI(rdf, first);
URI(rdf, rest);
URI(rdf, type);
URI(rdfs, Class);
URI(rdfs, Literal);
URI(rdfs, Resource);
URI(rdfs, domain);
URI(rdfs, label);
URI(rdfs, range);
URI(rdfs, subClassOf);
URI(xsd, anyURI);
URI(xsd, decimal);
URI(xsd, maxInclusive);
URI(xsd, minInclusive);
URI(xsd, pattern);
URI(xsd, string);
#ifndef HAVE_PCRE
fprintf(stderr, "warning: Built without PCRE, datatypes not checked.\n");
#endif
const int prop_st = check_properties(model, &uris);
const int inst_st = check_instances(model, &uris);
printf("Found %d errors among %d files (checked %d restrictions)\n",
n_errors, argc - 1, n_restrictions);
sord_free(model);
sord_world_free(world);
return prop_st || inst_st;
}
int
main(int argc, char** argv)
{
if (argc < 2) {
return print_usage(argv[0], true);
}
FILE* in_fd = NULL;
SerdSyntax input_syntax = SERD_TURTLE;
SerdSyntax output_syntax = SERD_NTRIPLES;
bool from_file = true;
bool bulk_read = true;
bool bulk_write = false;
bool full_uris = false;
bool lax = false;
bool quiet = false;
const uint8_t* in_name = NULL;
const uint8_t* add_prefix = NULL;
const uint8_t* chop_prefix = NULL;
const uint8_t* root_uri = NULL;
int a = 1;
for (; a < argc && argv[a][0] == '-'; ++a) {
if (argv[a][1] == '\0') {
in_name = (const uint8_t*)"(stdin)";
in_fd = stdin;
break;
} else if (argv[a][1] == 'b') {
bulk_write = true;
} else if (argv[a][1] == 'e') {
bulk_read = false;
} else if (argv[a][1] == 'f') {
full_uris = true;
} else if (argv[a][1] == 'h') {
return print_usage(argv[0], false);
} else if (argv[a][1] == 'l') {
lax = true;
} else if (argv[a][1] == 'q') {
quiet = true;
} else if (argv[a][1] == 'v') {
return print_version();
} else if (argv[a][1] == 's') {
in_name = (const uint8_t*)"(string)";
from_file = false;
++a;
break;
} else if (argv[a][1] == 'i') {
if (++a == argc) {
return missing_arg(argv[0], 'i');
} else if (!set_syntax(&input_syntax, argv[a])) {
return print_usage(argv[0], true);
}
} else if (argv[a][1] == 'o') {
if (++a == argc) {
return missing_arg(argv[0], 'o');
} else if (!set_syntax(&output_syntax, argv[a])) {
return print_usage(argv[0], true);
}
} else if (argv[a][1] == 'p') {
if (++a == argc) {
return missing_arg(argv[0], 'p');
}
add_prefix = (const uint8_t*)argv[a];
} else if (argv[a][1] == 'c') {
if (++a == argc) {
return missing_arg(argv[0], 'c');
}
chop_prefix = (const uint8_t*)argv[a];
} else if (argv[a][1] == 'r') {
if (++a == argc) {
return missing_arg(argv[0], 'r');
}
root_uri = (const uint8_t*)argv[a];
} else {
SERDI_ERRORF("invalid option -- '%s'\n", argv[a] + 1);
return print_usage(argv[0], true);
}
}
if (a == argc) {
SERDI_ERROR("missing input\n");
return 1;
}
const uint8_t* input = (const uint8_t*)argv[a++];
if (from_file) {
in_name = in_name ? in_name : input;
if (!in_fd) {
input = serd_uri_to_path(in_name);
if (!input || !(in_fd = serd_fopen((const char*)input, "r"))) {
return 1;
}
}
}
SerdURI base_uri = SERD_URI_NULL;
SerdNode base = SERD_NODE_NULL;
if (a < argc) { // Base URI given on command line
base = serd_node_new_uri_from_string(
(const uint8_t*)argv[a], NULL, &base_uri);
} else if (from_file && in_fd != stdin) { // Use input file URI
base = serd_node_new_file_uri(input, NULL, &base_uri, false);
}
FILE* out_fd = stdout;
SerdEnv* env = serd_env_new(&base);
int output_style = 0;
if (output_syntax == SERD_NTRIPLES) {
output_style |= SERD_STYLE_ASCII;
} else {
output_style |= SERD_STYLE_ABBREVIATED;
if (!full_uris) {
output_style |= SERD_STYLE_CURIED;
}
}
if (input_syntax != SERD_NTRIPLES || (output_style & SERD_STYLE_CURIED)) {
// Base URI may change and/or we're abbreviating URIs, so must resolve
output_style |= SERD_STYLE_RESOLVED; // Base may chan
}
if (bulk_write) {
output_style |= SERD_STYLE_BULK;
}
SerdWriter* writer = serd_writer_new(
output_syntax, (SerdStyle)output_style,
env, &base_uri, serd_file_sink, out_fd);
SerdReader* reader = serd_reader_new(
input_syntax, writer, NULL,
(SerdBaseSink)serd_writer_set_base_uri,
(SerdPrefixSink)serd_writer_set_prefix,
(SerdStatementSink)serd_writer_write_statement,
(SerdEndSink)serd_writer_end_anon);
serd_reader_set_strict(reader, !lax);
if (quiet) {
serd_reader_set_error_sink(reader, quiet_error_sink, NULL);
serd_writer_set_error_sink(writer, quiet_error_sink, NULL);
}
SerdNode root = serd_node_from_string(SERD_URI, root_uri);
serd_writer_set_root_uri(writer, &root);
serd_writer_chop_blank_prefix(writer, chop_prefix);
serd_reader_add_blank_prefix(reader, add_prefix);
SerdStatus status = SERD_SUCCESS;
if (!from_file) {
status = serd_reader_read_string(reader, input);
} else if (bulk_read) {
status = serd_reader_read_file_handle(reader, in_fd, in_name);
} else {
status = serd_reader_start_stream(reader, in_fd, in_name, false);
while (!status) {
status = serd_reader_read_chunk(reader);
}
serd_reader_end_stream(reader);
}
serd_reader_free(reader);
if (from_file) {
fclose(in_fd);
}
serd_writer_finish(writer);
serd_writer_free(writer);
serd_env_free(env);
serd_node_free(&base);
return (status > SERD_FAILURE) ? 1 : 0;
}
int
main(int argc, char** argv)
{
if (argc < 2) {
return print_usage(argv[0], true);
}
FILE* in_fd = NULL;
SerdSyntax input_syntax = SERD_TURTLE;
SerdSyntax output_syntax = SERD_NTRIPLES;
bool from_file = true;
bool bulk_write = false;
bool full_uris = false;
const uint8_t* in_name = NULL;
const uint8_t* add_prefix = NULL;
const uint8_t* chop_prefix = NULL;
const uint8_t* root_uri = NULL;
int a = 1;
for (; a < argc && argv[a][0] == '-'; ++a) {
if (argv[a][1] == '\0') {
in_name = (const uint8_t*)"(stdin)";
in_fd = stdin;
break;
} else if (argv[a][1] == 'b') {
bulk_write = true;
} else if (argv[a][1] == 'f') {
full_uris = true;
} else if (argv[a][1] == 'h') {
return print_usage(argv[0], false);
} else if (argv[a][1] == 'v') {
return print_version();
} else if (argv[a][1] == 's') {
in_name = (const uint8_t*)"(string)";
from_file = false;
++a;
break;
} else if (argv[a][1] == 'i') {
if (++a == argc || !set_syntax(&input_syntax, argv[a])) {
return bad_arg(argv[0], 'i');
}
} else if (argv[a][1] == 'o') {
if (++a == argc || !set_syntax(&output_syntax, argv[a])) {
return bad_arg(argv[0], 'o');
}
} else if (argv[a][1] == 'p') {
if (++a == argc) {
return bad_arg(argv[0], 'p');
}
add_prefix = (const uint8_t*)argv[a];
} else if (argv[a][1] == 'c') {
if (++a == argc) {
return bad_arg(argv[0], 'c');
}
chop_prefix = (const uint8_t*)argv[a];
} else if (argv[a][1] == 'r') {
if (++a == argc) {
return bad_arg(argv[0], 'r');
}
root_uri = (const uint8_t*)argv[a];
} else {
fprintf(stderr, "%s: Unknown option `%s'\n", argv[0], argv[a]);
return print_usage(argv[0], true);
}
}
if (a == argc) {
fprintf(stderr, "%s: Missing input\n", argv[0]);
return 1;
}
const uint8_t* input = (const uint8_t*)argv[a++];
if (from_file) {
in_name = in_name ? in_name : input;
if (!in_fd) {
input = serd_uri_to_path(in_name);
if (!input || !(in_fd = serd_fopen((const char*)input, "r"))) {
return 1;
}
}
}
SerdURI base_uri = SERD_URI_NULL;
SerdNode base = SERD_NODE_NULL;
if (a < argc) { // Base URI given on command line
base = serd_node_new_uri_from_string(
(const uint8_t*)argv[a], NULL, &base_uri);
} else if (from_file && in_fd != stdin) { // Use input file URI
base = serd_node_new_file_uri(input, NULL, &base_uri, false);
}
FILE* out_fd = stdout;
SerdEnv* env = serd_env_new(&base);
int output_style = 0;
if (output_syntax == SERD_NTRIPLES) {
output_style |= SERD_STYLE_ASCII;
} else {
output_style |= SERD_STYLE_ABBREVIATED;
if (!full_uris) {
output_style |= SERD_STYLE_CURIED;
}
}
if (input_syntax != SERD_NTRIPLES // Base URI may change (@base)
|| (output_syntax == SERD_TURTLE)) {
output_style |= SERD_STYLE_RESOLVED;
}
if (bulk_write) {
output_style |= SERD_STYLE_BULK;
}
SerdWriter* writer = serd_writer_new(
output_syntax, (SerdStyle)output_style,
env, &base_uri, serd_file_sink, out_fd);
SerdReader* reader = serd_reader_new(
input_syntax, writer, NULL,
(SerdBaseSink)serd_writer_set_base_uri,
(SerdPrefixSink)serd_writer_set_prefix,
(SerdStatementSink)serd_writer_write_statement,
(SerdEndSink)serd_writer_end_anon);
SerdNode root = serd_node_from_string(SERD_URI, root_uri);
serd_writer_set_root_uri(writer, &root);
serd_writer_chop_blank_prefix(writer, chop_prefix);
serd_reader_add_blank_prefix(reader, add_prefix);
const SerdStatus status = (from_file)
? serd_reader_read_file_handle(reader, in_fd, in_name)
: serd_reader_read_string(reader, input);
serd_reader_free(reader);
if (from_file) {
fclose(in_fd);
}
serd_writer_finish(writer);
serd_writer_free(writer);
serd_env_free(env);
serd_node_free(&base);
return (status > SERD_FAILURE) ? 1 : 0;
}