/*
* Export data out of GPDB.
* invoked by GPDB, be careful with C++ exceptions.
*/
Datum s3_export(PG_FUNCTION_ARGS) {
/* Must be called via the external table format manager */
if (!CALLED_AS_EXTPROTOCOL(fcinfo))
elog(ERROR, "extprotocol_import: not called by external protocol manager");
/* Get our internal description of the protocol */
GPWriter *gpwriter = (GPWriter *)EXTPROTOCOL_GET_USER_CTX(fcinfo);
/* last call. destroy writer */
if (EXTPROTOCOL_IS_LAST_CALL(fcinfo)) {
thread_cleanup();
if (!writer_cleanup(&gpwriter)) {
ereport(ERROR,
(0, errmsg("Failed to cleanup S3 extension: %s", s3extErrorMessage.c_str())));
}
EXTPROTOCOL_SET_USER_CTX(fcinfo, NULL);
PG_RETURN_INT32(0);
}
/* first call. do any desired init */
if (gpwriter == NULL) {
const char *url_with_options = EXTPROTOCOL_GET_URL(fcinfo);
const char *format = get_format_str(fcinfo);
thread_setup();
gpwriter = writer_init(url_with_options, format);
if (!gpwriter) {
ereport(ERROR, (0, errmsg("Failed to init S3 extension, segid = %d, "
"segnum = %d, please check your "
"configurations and net connection: %s",
s3ext_segid, s3ext_segnum, s3extErrorMessage.c_str())));
}
EXTPROTOCOL_SET_USER_CTX(fcinfo, gpwriter);
}
char *data_buf = EXTPROTOCOL_GET_DATABUF(fcinfo);
int32 data_len = EXTPROTOCOL_GET_DATALEN(fcinfo);
if (!writer_transfer_data(gpwriter, data_buf, data_len)) {
ereport(ERROR,
(0, errmsg("s3_export: could not write data: %s", s3extErrorMessage.c_str())));
}
PG_RETURN_INT32(data_len);
}
void cleanup_context(PG_FUNCTION_ARGS, gphadoop_context* context)
{
pfree(context->uri.data);
pfree(context->write_file_name.data);
pfree(context);
EXTPROTOCOL_SET_USER_CTX(fcinfo, NULL);
}
/*
* Import data into GPDB.
*/
Datum
gphdfsprotocol_import(PG_FUNCTION_ARGS)
{
char *data;
int datlen;
size_t nread = 0;
gphdfs_import_t *myData;
/* Must be called via the external table format manager */
if (!CALLED_AS_EXTPROTOCOL(fcinfo))
elog(ERROR, "cannot execute gphdfsprotocol_import outside protocol manager");
/* Get our internal description of the protocol */
myData = (gphdfs_import_t*) EXTPROTOCOL_GET_USER_CTX(fcinfo);
/* =======================================================================
* DO LAST CALL
* Nothing to be done if it has already been closed
* ======================================================================= */
if (EXTPROTOCOL_IS_LAST_CALL(fcinfo))
{
if (myData != NULL && !myData->importDone)
url_fclose(myData->importFile, false, "gphdfs protocol");
PG_RETURN_INT32(0);
}
/* =======================================================================
* DO OPEN
* ======================================================================= */
if (myData == NULL)
{
myData = palloc(sizeof(gphdfs_import_t));
myData->importFile = gphdfs_fopen(fcinfo, false);
myData->importDone = false;
EXTPROTOCOL_SET_USER_CTX(fcinfo, myData);
}
/* =======================================================================
* DO THE IMPORT
* ======================================================================= */
data = EXTPROTOCOL_GET_DATABUF(fcinfo);
datlen = EXTPROTOCOL_GET_DATALEN(fcinfo);
if (datlen > 0 && !myData->importDone)
nread = url_execute_fread(data, datlen, myData->importFile, NULL);
/* =======================================================================
* DO CLOSE
* close early to raise error early
* ======================================================================= */
if (nread == 0)
{
myData->importDone = true;
url_fclose(myData->importFile, true, "gphdfs protocol");
}
PG_RETURN_INT32((int)nread);
}
gphadoop_context* create_context(PG_FUNCTION_ARGS)
{
gphadoop_context* context = NULL;
context = palloc0(sizeof(gphadoop_context));
/* first thing we do, store the context */
EXTPROTOCOL_SET_USER_CTX(fcinfo, context);
initStringInfo(&context->uri);
initStringInfo(&context->write_file_name);
return context;
}
/*
* Export data out of GPDB.
*/
Datum
gphdfsprotocol_export(PG_FUNCTION_ARGS)
{
URL_FILE *myData;
char *data;
int datlen;
size_t wrote = 0;
static char ebuf[512] = {0};
int ebuflen = 512;
/* Must be called via the external table format manager */
if (!CALLED_AS_EXTPROTOCOL(fcinfo))
elog(ERROR, "cannot execute gphdfsprotocol_export outside protocol manager");
/* Get our internal description of the protocol */
myData = (URL_FILE *) EXTPROTOCOL_GET_USER_CTX(fcinfo);
/* =======================================================================
* DO CLOSE
* ======================================================================= */
if (EXTPROTOCOL_IS_LAST_CALL(fcinfo))
{
if (myData)
url_fclose(myData, true, "gphdfs protocol");
PG_RETURN_INT32(0);
}
/* =======================================================================
* DO OPEN
* ======================================================================= */
if (myData == NULL)
{
myData = gphdfs_fopen(fcinfo, true);
EXTPROTOCOL_SET_USER_CTX(fcinfo, myData);
/* add schema info to pipe */
StringInfo schema_data = makeStringInfo();
Relation relation = FORMATTER_GET_RELATION(fcinfo);
ExtTableEntry *exttbl = GetExtTableEntry(relation->rd_id);
if (fmttype_is_avro(exttbl->fmtcode) || fmttype_is_parquet(exttbl->fmtcode) )
{
int relNameLen = strlen(relation->rd_rel->relname.data);
appendIntToBuffer(schema_data, relNameLen);
appendBinaryStringInfo(schema_data, relation->rd_rel->relname.data, relNameLen);
int ncolumns = relation->rd_att->natts;
appendIntToBuffer(schema_data, ncolumns);
int i = 0;
for (; i< ncolumns; i++)
{
Oid type = relation->rd_att->attrs[i]->atttypid;
/* add attname,atttypid,attnotnull,attndims to schema_data filed */
int attNameLen = strlen(relation->rd_att->attrs[i]->attname.data);
appendIntToBuffer(schema_data, attNameLen);
appendBinaryStringInfo(schema_data, relation->rd_att->attrs[i]->attname.data, attNameLen);
appendIntToBuffer(schema_data, type);
bool notNull = relation->rd_att->attrs[i]->attnotnull;
appendInt1ToBuffer(schema_data, notNull?1:0);
appendIntToBuffer(schema_data, relation->rd_att->attrs[i]->attndims);
/* add type delimiter, for udt, it can be anychar */
char delim = 0;
int16 typlen;
bool typbyval;
char typalien;
Oid typioparam;
Oid func;
get_type_io_data(type, IOFunc_input, &typlen, &typbyval, &typalien, &delim, &typioparam, &func);
appendInt1ToBuffer(schema_data, delim);
}
StringInfo schema_head = makeStringInfo();
appendIntToBuffer(schema_head, schema_data->len + 2);
appendInt2ToBuffer(schema_head, 2);
url_execute_fwrite(schema_head->data, schema_head->len, myData, NULL);
url_execute_fwrite(schema_data->data, schema_data->len, myData, NULL);
pfree(schema_head->data);
pfree(schema_data->data);
}
}
/* =======================================================================
* DO THE EXPORT
* ======================================================================= */
data = EXTPROTOCOL_GET_DATABUF(fcinfo);
datlen = EXTPROTOCOL_GET_DATALEN(fcinfo);
if (datlen > 0)
wrote = url_execute_fwrite(data, datlen, myData, NULL);
if (url_ferror(myData, wrote, ebuf, ebuflen))
{
ereport(ERROR,
(errcode_for_file_access(),
strlen(ebuf) > 0 ? errmsg("could not write to external resource:\n%s",ebuf) :
errmsg("could not write to external resource: %m")));
}
PG_RETURN_INT32((int)wrote);
}
/*
* Import data into GPDB.
*/
Datum
demoprot_import(PG_FUNCTION_ARGS)
{
extprotocol_t *myData;
char *data;
int datlen;
size_t nread = 0;
/* Must be called via the external table format manager */
if (!CALLED_AS_EXTPROTOCOL(fcinfo))
elog(ERROR, "extprotocol_import: not called by external protocol manager");
/* Get our internal description of the protocol */
myData = (extprotocol_t *) EXTPROTOCOL_GET_USER_CTX(fcinfo);
if(EXTPROTOCOL_IS_LAST_CALL(fcinfo))
{
/* we're done receiving data. close our connection */
if(myData && myData->file)
if(fclose(myData->file))
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not close file \"%s\": %m",
myData->filename)));
PG_RETURN_INT32(0);
}
if (myData == NULL)
{
/* first call. do any desired init */
const char *p_name = "demoprot";
DemoUri *parsed_url;
char *url = EXTPROTOCOL_GET_URL(fcinfo);
myData = palloc(sizeof(extprotocol_t));
myData->url = pstrdup(url);
parsed_url = ParseDemoUri(myData->url);
myData->filename = pstrdup(parsed_url->path);
if(strcasecmp(parsed_url->protocol, p_name) != 0)
elog(ERROR, "internal error: demoprot called with a different protocol (%s)",
parsed_url->protocol);
FreeDemoUri(parsed_url);
/* open the destination file (or connect to remote server in other cases) */
myData->file = fopen(myData->filename, "r");
if (myData->file == NULL)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("demoprot_import: could not open file \"%s\" for reading: %m",
myData->filename),
errOmitLocation(true)));
EXTPROTOCOL_SET_USER_CTX(fcinfo, myData);
}
/* =======================================================================
* DO THE IMPORT
* ======================================================================= */
data = EXTPROTOCOL_GET_DATABUF(fcinfo);
datlen = EXTPROTOCOL_GET_DATALEN(fcinfo);
if(datlen > 0)
{
nread = fread(data, 1, datlen, myData->file);
if (ferror(myData->file))
ereport(ERROR,
(errcode_for_file_access(),
errmsg("demoprot_import: could not write to file \"%s\": %m",
myData->filename)));
}
PG_RETURN_INT32((int)nread);
}
/*
* Import data into GPDB.
* invoked by GPDB, be careful with C++ exceptions.
*/
Datum s3_import(PG_FUNCTION_ARGS) {
S3ExtBase *myData;
char *data;
int data_len;
size_t nread = 0;
/* Must be called via the external table format manager */
if (!CALLED_AS_EXTPROTOCOL(fcinfo))
elog(ERROR,
"extprotocol_import: not called by external protocol manager");
/* Get our internal description of the protocol */
myData = (S3ExtBase *)EXTPROTOCOL_GET_USER_CTX(fcinfo);
if (EXTPROTOCOL_IS_LAST_CALL(fcinfo)) {
if (myData) {
thread_cleanup();
if (!myData->Destroy()) {
ereport(ERROR, (0, errmsg("Failed to cleanup S3 extention")));
}
delete myData;
}
/*
* Cleanup function for the XML library.
*/
xmlCleanupParser();
PG_RETURN_INT32(0);
}
if (myData == NULL) {
/* first call. do any desired init */
curl_global_init(CURL_GLOBAL_ALL);
thread_setup();
const char *p_name = "s3";
char *url_with_options = EXTPROTOCOL_GET_URL(fcinfo);
char *url = truncate_options(url_with_options);
char *config_path = get_opt_s3(url_with_options, "config");
if (!config_path) {
// no config path in url, use default value
// data_folder/gpseg0/s3/s3.conf
config_path = strdup("s3/s3.conf");
}
bool result = InitConfig(config_path, "");
if (!result) {
free(config_path);
ereport(ERROR, (0, errmsg("Can't find config file, please check")));
} else {
ClearConfig();
free(config_path);
}
InitLog();
if (s3ext_accessid == "") {
ereport(ERROR, (0, errmsg("ERROR: access id is empty")));
}
if (s3ext_secret == "") {
ereport(ERROR, (0, errmsg("ERROR: secret is empty")));
}
if ((s3ext_segnum == -1) || (s3ext_segid == -1)) {
ereport(ERROR, (0, errmsg("ERROR: segment id is invalid")));
}
myData = CreateExtWrapper(url);
if (!myData ||
!myData->Init(s3ext_segid, s3ext_segnum, s3ext_chunksize)) {
if (myData) delete myData;
ereport(ERROR, (0, errmsg("Failed to init S3 extension, segid = "
"%d, segnum = %d, please check your "
"configurations and net connection",
s3ext_segid, s3ext_segnum)));
}
EXTPROTOCOL_SET_USER_CTX(fcinfo, myData);
free(url);
}
/* =======================================================================
* DO THE IMPORT
* =======================================================================
*/
data = EXTPROTOCOL_GET_DATABUF(fcinfo);
data_len = EXTPROTOCOL_GET_DATALEN(fcinfo);
uint64_t readlen = 0;
if (data_len > 0) {
readlen = data_len;
if (!myData->TransferData(data, readlen))
ereport(ERROR, (0, errmsg("s3_import: could not read data")));
nread = (size_t)readlen;
// S3DEBUG("read %d data from S3", nread);
}
PG_RETURN_INT32((int)nread);
}