Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
35 changes: 32 additions & 3 deletions datafusion-postgres/src/pg_catalog.rs
Original file line number Diff line number Diff line change
Expand Up @@ -4,9 +4,9 @@ use std::sync::Arc;

use async_trait::async_trait;
use datafusion::arrow::array::{
as_boolean_array, ArrayRef, BooleanBuilder, RecordBatch, StringArray, StringBuilder,
as_boolean_array, ArrayRef, AsArray, BooleanBuilder, RecordBatch, StringArray, StringBuilder,
};
use datafusion::arrow::datatypes::{DataType, Field, SchemaRef};
use datafusion::arrow::datatypes::{DataType, Field, Int32Type, SchemaRef};
use datafusion::arrow::ipc::reader::FileReader;
use datafusion::catalog::streaming::StreamingTable;
use datafusion::catalog::{MemTable, SchemaProvider, TableFunctionImpl};
Expand Down Expand Up @@ -1109,7 +1109,7 @@ pub fn create_pg_get_statisticsobjdef_columns_udf() -> ScalarUDF {
let args = ColumnarValue::values_to_arrays(args)?;
let oid = &args[0];

let mut builder = BooleanBuilder::new();
let mut builder = StringBuilder::new();
for _ in 0..oid.len() {
builder.append_null();
}
Expand All @@ -1128,6 +1128,34 @@ pub fn create_pg_get_statisticsobjdef_columns_udf() -> ScalarUDF {
)
}

pub fn create_pg_encoding_to_char_udf() -> ScalarUDF {
let func = move |args: &[ColumnarValue]| {
let args = ColumnarValue::values_to_arrays(args)?;
let encoding = &args[0].as_primitive::<Int32Type>();

let mut builder = StringBuilder::new();
for i in 0..encoding.len() {
if encoding.value(i) == 6 {
builder.append_value("UTF-8");
} else {
builder.append_null();
}
}

let array: ArrayRef = Arc::new(builder.finish());

Ok(ColumnarValue::Array(array))
};

create_udf(
"pg_encoding_to_char",
vec![DataType::Int32],
DataType::Utf8,
Volatility::Stable,
Arc::new(func),
)
}

/// Install pg_catalog and postgres UDFs to current `SessionContext`
pub fn setup_pg_catalog(
session_context: &SessionContext,
Expand Down Expand Up @@ -1169,6 +1197,7 @@ pub fn setup_pg_catalog(
session_context.register_udf(create_pg_get_partkeydef_udf());
session_context.register_udf(create_pg_relation_is_publishable_udf());
session_context.register_udf(create_pg_get_statisticsobjdef_columns_udf());
session_context.register_udf(create_pg_encoding_to_char_udf());

Ok(())
}
Expand Down
39 changes: 31 additions & 8 deletions datafusion-postgres/src/pg_catalog/pg_database.rs
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,10 @@ use std::collections::HashMap;
use std::sync::atomic::{AtomicU32, Ordering};
use std::sync::Arc;

use datafusion::arrow::array::{ArrayRef, BooleanArray, Int32Array, RecordBatch, StringArray};
use datafusion::arrow::datatypes::{DataType, Field, Schema, SchemaRef};
use datafusion::arrow::array::{
ArrayRef, BooleanArray, Int32Array, ListArray, RecordBatch, StringArray,
};
use datafusion::arrow::datatypes::{DataType, Field, Int32Type, Schema, SchemaRef};
use datafusion::error::Result;
use datafusion::execution::{SendableRecordBatchStream, TaskContext};
use datafusion::physical_plan::stream::RecordBatchStreamAdapter;
Expand Down Expand Up @@ -36,16 +38,23 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
Field::new("datname", DataType::Utf8, false), // Database name
Field::new("datdba", DataType::Int32, false), // Database owner's user ID
Field::new("encoding", DataType::Int32, false), // Character encoding
Field::new("datlocprovider", DataType::Utf8, false),
Field::new("datcollate", DataType::Utf8, false), // LC_COLLATE for this database
Field::new("datctype", DataType::Utf8, false), // LC_CTYPE for this database
Field::new("datctype", DataType::Utf8, false), // LC_CTYPE for this database
Field::new("datistemplate", DataType::Boolean, false), // If true, database can be used as a template
Field::new("datallowconn", DataType::Boolean, false), // If false, no one can connect to this database
Field::new("datconnlimit", DataType::Int32, false), // Max number of concurrent connections (-1=no limit)
Field::new("datlastsysoid", DataType::Int32, false), // Last system OID in database
Field::new("datfrozenxid", DataType::Int32, false), // Frozen XID for this database
Field::new("datminmxid", DataType::Int32, false), // Minimum multixact ID
Field::new("dattablespace", DataType::Int32, false), // Default tablespace for this database
Field::new("datacl", DataType::Utf8, true), // Access privileges
Field::new("daticulocale", DataType::Utf8, true),
Field::new("daticurules", DataType::Utf8, true),
Field::new(
"datacl",
DataType::List(Arc::new(Field::new("item", DataType::Int32, true))),
true,
), // Access privileges
]));

Self {
Expand All @@ -63,6 +72,7 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
let mut datnames = Vec::new();
let mut datdbas = Vec::new();
let mut encodings = Vec::new();
let mut datlocproviders = Vec::new();
let mut datcollates = Vec::new();
let mut datctypes = Vec::new();
let mut datistemplates = Vec::new();
Expand All @@ -72,7 +82,9 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
let mut datfrozenxids = Vec::new();
let mut datminmxids = Vec::new();
let mut dattablespaces = Vec::new();
let mut datacles: Vec<Option<String>> = Vec::new();
let mut daticulocales: Vec<Option<String>> = Vec::new();
let mut daticurules: Vec<Option<String>> = Vec::new();
let mut datacls: Vec<Option<Vec<Option<i32>>>> = Vec::new();

// to store all schema-oid mapping temporarily before adding to global oid cache
let mut catalog_oid_cache = HashMap::new();
Expand All @@ -93,6 +105,7 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
datnames.push(catalog_name.clone());
datdbas.push(10); // Default owner (assuming 10 = postgres user)
encodings.push(6); // 6 = UTF8 in PostgreSQL
datlocproviders.push("libc".to_string());
datcollates.push("en_US.UTF-8".to_string()); // Default collation
datctypes.push("en_US.UTF-8".to_string()); // Default ctype
datistemplates.push(false);
Expand All @@ -102,7 +115,9 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
datfrozenxids.push(1); // Simplified transaction ID
datminmxids.push(1); // Simplified multixact ID
dattablespaces.push(1663); // Default tablespace (1663 = pg_default in PostgreSQL)
datacles.push(None); // No specific ACLs
daticulocales.push(None);
daticurules.push(None);
datacls.push(None); // No specific ACLs
}

// Always include a "postgres" database entry if not already present
Expand All @@ -121,6 +136,7 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
datnames.push(default_datname);
datdbas.push(10);
encodings.push(6);
datlocproviders.push("libc".to_string());
datcollates.push("en_US.UTF-8".to_string());
datctypes.push("en_US.UTF-8".to_string());
datistemplates.push(false);
Expand All @@ -130,7 +146,9 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
datfrozenxids.push(1);
datminmxids.push(1);
dattablespaces.push(1663);
datacles.push(None);
daticulocales.push(None);
daticurules.push(None);
datacls.push(None);
}

// Create Arrow arrays from the collected data
Expand All @@ -139,6 +157,7 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
Arc::new(StringArray::from(datnames)),
Arc::new(Int32Array::from(datdbas)),
Arc::new(Int32Array::from(encodings)),
Arc::new(StringArray::from(datlocproviders)),
Arc::new(StringArray::from(datcollates)),
Arc::new(StringArray::from(datctypes)),
Arc::new(BooleanArray::from(datistemplates)),
Expand All @@ -148,7 +167,11 @@ impl<C: CatalogInfo> PgDatabaseTable<C> {
Arc::new(Int32Array::from(datfrozenxids)),
Arc::new(Int32Array::from(datminmxids)),
Arc::new(Int32Array::from(dattablespaces)),
Arc::new(StringArray::from_iter(datacles.into_iter())),
Arc::new(StringArray::from(daticulocales)),
Arc::new(StringArray::from(daticurules)),
Arc::new(ListArray::from_iter_primitive::<Int32Type, _, _>(
datacls.into_iter(),
)),
];

// Create a full record batch
Expand Down
13 changes: 13 additions & 0 deletions datafusion-postgres/tests/psql.rs
Original file line number Diff line number Diff line change
Expand Up @@ -92,6 +92,19 @@ const PSQL_QUERIES: &[&str] = &[
FROM pg_catalog.pg_class c, pg_catalog.pg_inherits i
WHERE c.oid = i.inhrelid AND i.inhparent = '16384'
ORDER BY pg_catalog.pg_get_expr(c.relpartbound, c.oid) = 'DEFAULT', c.oid::pg_catalog.regclass::pg_catalog.text;",

r#"SELECT
d.datname as "Name",
pg_catalog.pg_get_userbyid(d.datdba) as "Owner",
pg_catalog.pg_encoding_to_char(d.encoding) as "Encoding",
CASE d.datlocprovider WHEN 'b' THEN 'builtin' WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS "Locale Provider",
d.datcollate as "Collate",
d.datctype as "Ctype",
d.daticulocale as "Locale",
d.daticurules as "ICU Rules",
CASE WHEN pg_catalog.array_length(d.datacl, 1) = 0 THEN '(none)' ELSE pg_catalog.array_to_string(d.datacl, E'\n') END AS "Access privileges"
FROM pg_catalog.pg_database d
ORDER BY 1;"#,
];

#[tokio::test]
Expand Down
Loading