duckdb 0.7.1-dev157.0 → 0.7.1-dev180.0

package/package.json

@@ -2,7 +2,7 @@
   "name": "duckdb",
   "main": "./lib/duckdb.js",
   "types": "./lib/duckdb.d.ts",
-  "version": "0.7.1-dev157.0",
+  "version": "0.7.1-dev180.0",
   "description": "DuckDB node.js API",
   "gypfile": true,
   "dependencies": {

package/src/duckdb/src/common/hive_partitioning.cpp

@@ -150,6 +150,7 @@ HivePartitionedColumnData::HivePartitionedColumnData(const HivePartitionedColumn
 void HivePartitionedColumnData::ComputePartitionIndices(PartitionedColumnDataAppendState &state, DataChunk &input) {
 	Vector hashes(LogicalType::HASH, input.size());
 	input.Hash(group_by_columns, hashes);
+	hashes.Flatten(input.size());
 
 	for (idx_t i = 0; i < input.size(); i++) {
 		HivePartitionKey key;

package/src/duckdb/src/execution/operator/persistent/buffered_csv_reader.cpp

@@ -893,6 +893,12 @@ vector<LogicalType> BufferedCSVReader::SniffCSV(const vector<LogicalType> &reque
 	DetectCandidateTypes(type_candidates, format_template_candidates, info_candidates, original_options, best_num_cols,
 	                     best_sql_types_candidates, best_format_candidates, best_header_row);
 
+	if (best_format_candidates.empty() || best_header_row.size() == 0) {
+		throw InvalidInputException(
+		    "Error in file \"%s\": CSV options could not be auto-detected. Consider setting parser options manually.",
+		    original_options.file_path);
+	}
+
 	// #######
 	// ### header detection
 	// #######

package/src/duckdb/src/function/table/version/pragma_version.cpp

@@ -1,8 +1,8 @@
 #ifndef DUCKDB_VERSION
-#define DUCKDB_VERSION "0.7.1-dev157"
+#define DUCKDB_VERSION "0.7.1-dev180"
 #endif
 #ifndef DUCKDB_SOURCE_ID
-#define DUCKDB_SOURCE_ID "525abf5063"
+#define DUCKDB_SOURCE_ID "88290f2c1c"
 #endif
 #include "duckdb/function/table/system_functions.hpp"
 #include "duckdb/main/database.hpp"

package/src/duckdb/src/include/duckdb/common/hive_partitioning.hpp

@@ -52,7 +52,15 @@ struct HivePartitionKey {
 
 	struct Equality {
 		bool operator()(const HivePartitionKey &a, const HivePartitionKey &b) const {
-			return a.values == b.values;
+			if (a.values.size() != b.values.size()) {
+				return false;
+			}
+			for (idx_t i = 0; i < a.values.size(); i++) {
+				if (!Value::NotDistinctFrom(a.values[i], b.values[i])) {
+					return false;
+				}
+			}
+			return true;
 		}
 	};
 };

package/src/duckdb/src/include/duckdb/common/radix_partitioning.hpp

@@ -78,13 +78,13 @@ protected:
 		case 2:
 		case 3:
 		case 4:
-			return GetBufferSize(1);
+			return GetBufferSize(1 << 1);
 		case 5:
-			return GetBufferSize(2);
+			return GetBufferSize(1 << 2);
 		case 6:
-			return GetBufferSize(3);
+			return GetBufferSize(1 << 3);
 		default:
-			return GetBufferSize(4);
+			return GetBufferSize(1 << 4);
 		}
 	}
 	void InitializeAppendStateInternal(PartitionedColumnDataAppendState &state) const override;

package/src/duckdb/src/optimizer/statistics/operator/propagate_join.cpp

@@ -83,12 +83,8 @@ void StatisticsPropagator::PropagateStatistics(LogicalComparisonJoin &join, uniq
 						*node_ptr = std::move(cross_product);
 						return;
 					}
-					case JoinType::INNER:
-					case JoinType::LEFT:
-					case JoinType::RIGHT:
-					case JoinType::OUTER: {
-						// inner/left/right/full outer join, replace with cross product
-						// since the condition is always true, left/right/outer join are equivalent to inner join here
+					case JoinType::INNER: {
+						// inner, replace with cross product
 						auto cross_product =
 						    LogicalCrossProduct::Create(std::move(join.children[0]), std::move(join.children[1]));
 						*node_ptr = std::move(cross_product);

package/src/duckdb/src/planner/binder/statement/bind_copy.cpp

@@ -97,17 +97,20 @@ BoundStatement Binder::BindCopyTo(CopyStatement &stmt) {
 	for (auto &option : original_options) {
 		auto loption = StringUtil::Lower(option.first);
 		if (loption == "use_tmp_file") {
-			use_tmp_file = option.second[0].CastAs(context, LogicalType::BOOLEAN).GetValue<bool>();
+			use_tmp_file =
+			    option.second.empty() || option.second[0].CastAs(context, LogicalType::BOOLEAN).GetValue<bool>();
 			user_set_use_tmp_file = true;
 			continue;
 		}
 		if (loption == "allow_overwrite") {
-			allow_overwrite = option.second[0].CastAs(context, LogicalType::BOOLEAN).GetValue<bool>();
+			allow_overwrite =
+			    option.second.empty() || option.second[0].CastAs(context, LogicalType::BOOLEAN).GetValue<bool>();
 			continue;
 		}
 
 		if (loption == "per_thread_output") {
-			per_thread_output = option.second[0].CastAs(context, LogicalType::BOOLEAN).GetValue<bool>();
+			per_thread_output =
+			    option.second.empty() || option.second[0].CastAs(context, LogicalType::BOOLEAN).GetValue<bool>();
 			continue;
 		}
 		if (loption == "partition_by") {

package/src/duckdb/third_party/fmt/include/fmt/core.h

@@ -359,12 +359,11 @@ using wstring_view = basic_string_view<wchar_t>;
 #if FMT_HAS_FEATURE(__cpp_char8_t)
 typedef char8_t fmt_char8_t;
 #else
-typedef unsigned char fmt_char8_t;
+typedef char fmt_char8_t;
 #endif
 
 /** Specifies if ``T`` is a character type. Can be specialized by users. */
 template <typename T> struct is_char : std::false_type {};
-template <> struct is_char<char> : std::true_type {};
 template <> struct is_char<wchar_t> : std::true_type {};
 template <> struct is_char<fmt_char8_t> : std::true_type {};
 template <> struct is_char<char16_t> : std::true_type {};