Skip to content

Support for STRUCT type #669

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
JelteF merged 21 commits into from
Apr 15, 2025
Merged

Support for STRUCT type #669

Changes from 1 commit
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
6ed554c
identified oid; branch still in draft state
destrex271 Mar 16, 2025
893b1d0
replaced with DuckdbRowOid
destrex271 Mar 18, 2025
9097d5e
Merge branch 'main' into feature/struct_support
destrex271 Mar 18, 2025
13c5665
replaced with DuckdbRowOid
destrex271 Mar 19, 2025
0a78578
resolved merge conflicts
destrex271 Mar 19, 2025
d969f93
fixed return type in Struct logical type
destrex271 Mar 19, 2025
4d2b349
added duckdb.struct type
destrex271 Mar 19, 2025
28825b0
added duckdb.struct type
destrex271 Mar 19, 2025
4a5b7ba
added duckdb_struct_in & duckdb_struct_out functions
destrex271 Mar 22, 2025
8935463
merged with main
destrex271 Mar 22, 2025
cab89be
added subscript functions, same as duckdb.row
destrex271 Mar 22, 2025
7f75168
fixed syntax in sql file
destrex271 Mar 25, 2025
b4f9969
rough demo of duckdb_struct
destrex271 Mar 25, 2025
96ef0de
able to read and subscript struct type
destrex271 Mar 25, 2025
129a2ad
fixed formatting
destrex271 Mar 26, 2025
d3ef369
fixed struct tests in json
destrex271 Mar 26, 2025
1f42730
Merge branch 'main' into feature/struct_support
destrex271 Apr 2, 2025
6c5470c
removed unecessary functions and
destrex271 Apr 2, 2025
aa40d50
Make some improvements
JelteF Apr 15, 2025
b25db1a
Merge branch 'main' into feature/struct_support
JelteF Apr 15, 2025
53f753f
Remove now unnecessary comment
JelteF Apr 15, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 20 additions & 0 deletions src/pgduckdb_types.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -250,6 +250,12 @@ ConvertTimeTzDatum(const duckdb::Value &value) {
return TimeTzADTPGetDatum(result);
}

inline Datum
ConvertDuckRowDatum(const duckdb::Value &value) {
Copy link
Copy Markdown
Collaborator

@JelteF JelteF Mar 16, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should really have a separate duckdb.struct type and not reuse the duckdb.row type (but you can copy and slightly modify the implementation of duckdb.row to create this new duckdb.struct type). We need this for one important reason (and possibly a few other less important ones): When duckdb.row is part of a select list, we replace it with a * before sending it to duckdb. We don't want that for this struct type. Apart from that they can basically behave the same, but we need two separate types to make the distinction between a type that is really a multiple columns under the hood (and which thus needs * expansion) versus a type that actually is a single column with a composite type in it.

Copy link
Copy Markdown
Contributor Author

@destrex271 destrex271 Mar 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@JelteF do we also need to give the ability to create this type from postgres? like -

CREATE TABLE tester(a duckdb.struct);
INSERT INTO TABLE tester(a) VALUES('{a:12, b:12, c:Akshat}');

Copy link
Copy Markdown
Contributor Author

@destrex271 destrex271 Mar 25, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also correct me if I am wrong but we are handling the star exapnsion in pgduckdb_ddl::DuckdbHandleDDL, right?
As far as I understood, pgduckdb_target_list_contains_unresolved_type_or_row function helps us in this decision, right?

auto val = value.GetValue<duckdb::StructValue>();
return pgduckdb::PGDUCKDB_DUCK_TIMESTAMP_OFFSET;
}

inline Datum
ConvertTimestampDatum(const duckdb::Value &value) {
// Extract raw int64_t value of timestamp
Expand Down Expand Up @@ -875,6 +881,11 @@ ConvertDuckToPostgresValue(TupleTableSlot *slot, duckdb::Value &value, idx_t col
slot->tts_values[col] = ConvertTimestampDatum(value);
break;
}
case 17496: {
elog(LOG, "ROW TYPEE!!!!!!!!");
slot->tts_values[col] = ConvertDuckRowDatum(value);
break;
}
case TIMESTAMPTZOID: {
duckdb::timestamp_tz_t timestamp = value.GetValue<duckdb::timestamp_tz_t>();
slot->tts_values[col] = timestamp.value - pgduckdb::PGDUCKDB_DUCK_TIMESTAMP_OFFSET;
Expand Down Expand Up @@ -1220,6 +1231,10 @@ GetPostgresDuckDBType(const duckdb::LogicalType &type) {
return UUIDOID;
case duckdb::LogicalTypeId::VARINT:
return NUMERICOID;
case duckdb::LogicalTypeId::STRUCT:
elog(LOG, "hi returning the appropriate type");
return 17496;
/* return pgduckdb::DuckdbRowOid(); */
case duckdb::LogicalTypeId::LIST: {
const duckdb::LogicalType *duck_type = &type;
while (duck_type->id() == duckdb::LogicalTypeId::LIST) {
Expand Down Expand Up @@ -1519,6 +1534,11 @@ ConvertPostgresToDuckValue(Oid attr_type, Datum value, duckdb::Vector &result, i
data[offset] = duckdb::StringVector::AddStringOrBlob(result, s);
break;
}
case duckdb::LogicalTypeId::STRUCT: {
elog(LOG, "Hi in Struct type");

break;
}
case duckdb::LogicalTypeId::LIST: {
// Convert Datum to ArrayType
auto array = DatumGetArrayTypeP(value);
Expand Down
Loading