Merge branch 'main' into 67-shortest-path-bounded

Makefile

@@ -43,7 +43,7 @@ EXTENSION_FLAGS=\
 -DDUCKDB_EXTENSION_${EXTENSION_NAME}_TEST_PATH="$(PROJ_DIR)test/sql"
 
 #### Add more of the DuckDB in-tree extensions here that you need (also feel free to remove them when not needed)
-EXTRA_EXTENSIONS_FLAG=-DBUILD_EXTENSIONS="tpch;visualizer"
+EXTRA_EXTENSIONS_FLAG=-DBUILD_EXTENSIONS="tpch"
 
 BUILD_FLAGS=-DEXTENSION_STATIC_BUILD=1 $(EXTENSION_FLAGS) ${EXTRA_EXTENSIONS_FLAG} $(OSX_BUILD_FLAG) $(TOOLCHAIN_FLAGS)
 CLIENT_FLAGS:=

README.md

@@ -11,7 +11,7 @@ git clone --recurse-submodules [email protected]:cwida/duckpgq-extension.git
 ```
 Note that `--recurse-submodules` will ensure the correct version of DuckDB is pulled allowing you to get started right away.
 
-This repository uses a modified version of DuckDB (currently a modified v0.8.1) and is not yet easily installed from a standard DuckDB (e.g. `pip install duckdb`) installation. 
+This repository uses a modified version of DuckDB (Currently up-to-date with v0.10.0) and is not yet easily installed from a standard DuckDB (e.g. `pip install duckdb`) installation. 
 If you want to use the SQL/PGQ syntax, you will have to build this repository from the source. 
 In the future, we aim to have an easily installed and loaded DuckDB extension. 
 

duckpgq/src/duckpgq/functions/scalar/csr_creation.cpp

@@ -36,7 +36,7 @@ static void CsrInitializeVertex(DuckPGQState &context, int32_t id,
     csr->initialized_v = true;
     context.csr_list[id] = std::move(csr);
   } catch (std::bad_alloc const &) {
-    throw Exception("Unable to initialize vector of size for csr vertex table "
+    throw Exception(ExceptionType::INTERNAL, "Unable to initialize vector of size for csr vertex table "
                     "representation");
   }
 
@@ -55,7 +55,7 @@ static void CsrInitializeEdge(DuckPGQState &context, int32_t id, int64_t v_size,
     csr_entry->second->e.resize(e_size, 0);
     csr_entry->second->edge_ids.resize(e_size, 0);
   } catch (std::bad_alloc const &) {
-    throw Exception("Unable to initialize vector of size for csr edge table "
+    throw Exception(ExceptionType::INTERNAL, "Unable to initialize vector of size for csr edge table "
                     "representation");
   }
   for (auto i = 1; i < v_size + 2; i++) {
@@ -82,7 +82,7 @@ static void CsrInitializeWeight(DuckPGQState &context, int32_t id,
       throw NotImplementedException("Unrecognized weight type detected.");
     }
   } catch (std::bad_alloc const &) {
-    throw Exception("Unable to initialize vector of size for csr weight table "
+    throw Exception(ExceptionType::INTERNAL, "Unable to initialize vector of size for csr weight table "
                     "representation");
   }
 

duckpgq/src/duckpgq/functions/scalar/reachability.cpp

@@ -245,7 +245,7 @@ static void ReachabilityFunction(DataChunk &args, ExpressionState &state,
           break;
         }
         default:
-          throw Exception("Unknown mode encountered");
+          throw Exception(ExceptionType::INTERNAL, "Unknown reachability mode encountered");
         }
       } else {
         exit_early = BfsWithoutArray(exit_early, csr, input_size, seen, visit,

duckpgq/src/duckpgq/functions/tablefunctions/create_property_graph.cpp

@@ -7,15 +7,15 @@ void CreatePropertyGraphFunction::CheckPropertyGraphTableLabels(
     const shared_ptr<PropertyGraphTable> &pg_table, TableCatalogEntry &table) {
   if (!pg_table->discriminator.empty()) {
     if (!table.ColumnExists(pg_table->discriminator)) {
-      throw BinderException("Column %s not found in table %s",
-                            pg_table->discriminator, pg_table->table_name);
+      throw Exception(ExceptionType::INVALID, "Column " + pg_table->discriminator +
+        " not found in table " + pg_table->table_name);
     }
     auto &column = table.GetColumn(pg_table->discriminator);
     if (!(column.GetType() == LogicalType::BIGINT ||
           column.GetType() == LogicalType::INTEGER)) {
-      throw BinderException("The discriminator column %s for table %s should "
-                            "be of type BIGINT or INTEGER",
-                            pg_table->discriminator, pg_table->table_name);
+      throw Exception(ExceptionType::INVALID, "The discriminator column " +
+        pg_table->discriminator + " of table " +
+        pg_table->table_name + " should be of type BIGINT or INTEGER");
     }
   }
 }
@@ -29,8 +29,8 @@ void CreatePropertyGraphFunction::CheckPropertyGraphTableColumns(
   if (pg_table->all_columns) {
     for (auto &except_column : pg_table->except_columns) {
       if (!table.ColumnExists(except_column)) {
-        throw BinderException("Except column %s not found in table %s",
-                              except_column, pg_table->table_name);
+        throw Exception(ExceptionType::INVALID, "Except column " + except_column +
+          " not found in table " + pg_table->table_name);
       }
     }
 
@@ -49,24 +49,25 @@ void CreatePropertyGraphFunction::CheckPropertyGraphTableColumns(
 
   for (auto &column : pg_table->column_names) {
     if (!table.ColumnExists(column)) {
-      throw BinderException("Column %s not found in table %s", column,
-                            pg_table->table_name);
+      throw Exception(ExceptionType::INVALID, "Column " + column +
+        " not found in table " + pg_table->table_name);
     }
   }
 }
 
-duckdb::unique_ptr<FunctionData>
+unique_ptr<FunctionData>
 CreatePropertyGraphFunction::CreatePropertyGraphBind(
     ClientContext &context, TableFunctionBindInput &input,
     vector<LogicalType> &return_types, vector<string> &names) {
   names.emplace_back("Success");
   return_types.emplace_back(LogicalType::BOOLEAN);
   auto lookup = context.registered_state.find("duckpgq");
   if (lookup == context.registered_state.end()) {
-    throw BinderException("Registered DuckPGQ state not found");
+    throw Exception(ExceptionType::INVALID,
+                    "Registered DuckPGQ state not found");
   }
-  auto duckpgq_state = (DuckPGQState *)lookup->second.get();
-  auto duckpgq_parse_data =
+  const auto duckpgq_state = (DuckPGQState *)lookup->second.get();
+  const auto duckpgq_parse_data =
       dynamic_cast<DuckPGQParseData *>(duckpgq_state->parse_data.get());
 
   if (!duckpgq_parse_data) {
@@ -79,12 +80,10 @@ CreatePropertyGraphFunction::CreatePropertyGraphBind(
       duckpgq_state->registered_property_graphs.find(info->property_graph_name);
 
   if (pg_table != duckpgq_state->registered_property_graphs.end()) {
-    throw BinderException("Property graph table with name %s already exists",
-                          info->property_graph_name);
+    throw Exception(ExceptionType::INVALID, "Property graph table with name " + info->property_graph_name + " already exists");
   }
 
   auto &catalog = Catalog::GetCatalog(context, info->catalog);
-
   case_insensitive_set_t v_table_names;
   for (auto &vertex_table : info->vertex_tables) {
     auto &table = catalog.GetEntry<TableCatalogEntry>(context, info->schema,
@@ -108,53 +107,47 @@ CreatePropertyGraphFunction::CreatePropertyGraphBind(
 
     if (v_table_names.find(edge_table->source_reference) ==
         v_table_names.end()) {
-      throw BinderException("Referenced vertex table %s does not exist.",
-                            edge_table->source_reference);
+      throw Exception(ExceptionType::INVALID, "Referenced vertex table " + edge_table->source_reference + " does not exist.");
     }
 
     auto &pk_source_table = catalog.GetEntry<TableCatalogEntry>(
         context, info->schema, edge_table->source_reference);
     for (auto &pk : edge_table->source_pk) {
       if (!pk_source_table.ColumnExists(pk)) {
-        throw BinderException("Primary key %s does not exist in table %s", pk,
-                              edge_table->source_reference);
+        throw Exception(ExceptionType::INVALID, "Primary key " + pk + " does not exist in table " + edge_table->source_reference);
       }
     }
 
     if (v_table_names.find(edge_table->source_reference) ==
         v_table_names.end()) {
-      throw BinderException("Referenced vertex table %s does not exist.",
-                            edge_table->source_reference);
+      throw Exception(ExceptionType::INVALID, "Referenced vertex table " + edge_table->source_reference + " does not exist");
     }
 
     auto &pk_destination_table = catalog.GetEntry<TableCatalogEntry>(
         context, info->schema, edge_table->destination_reference);
 
     for (auto &pk : edge_table->destination_pk) {
       if (!pk_destination_table.ColumnExists(pk)) {
-        throw BinderException("Primary key %s does not exist in table %s", pk,
-                              edge_table->destination_reference);
+        throw Exception(ExceptionType::INVALID,"Primary key " + pk + " does not exist in table " + edge_table->destination_reference);
       }
     }
 
     for (auto &fk : edge_table->source_fk) {
       if (!table.ColumnExists(fk)) {
-        throw BinderException("Foreign key %s does not exist in table %s", fk,
-                              edge_table->table_name);
+        throw Exception(ExceptionType::INVALID,"Foreign key " + fk + " does not exist in table " + edge_table->table_name);
       }
     }
 
     for (auto &fk : edge_table->destination_fk) {
       if (!table.ColumnExists(fk)) {
-        throw BinderException("Foreign key %s does not exist in table %s", fk,
-                              edge_table->table_name);
+        throw Exception(ExceptionType::INVALID,"Foreign key " + fk + " does not exist in table " + edge_table->table_name);
       }
     }
   }
   return make_uniq<CreatePropertyGraphBindData>(info);
 }
 
-duckdb::unique_ptr<GlobalTableFunctionState>
+unique_ptr<GlobalTableFunctionState>
 CreatePropertyGraphFunction::CreatePropertyGraphInit(
     ClientContext &context, TableFunctionInitInput &input) {
   return make_uniq<CreatePropertyGraphGlobalData>();
@@ -167,7 +160,7 @@ void CreatePropertyGraphFunction::CreatePropertyGraphFunc(
   auto pg_info = bind_data.create_pg_info;
   auto lookup = context.registered_state.find("duckpgq");
   if (lookup == context.registered_state.end()) {
-    throw BinderException("Registered DuckPGQ state not found");
+    throw Exception(ExceptionType::INVALID,"Registered DuckPGQ state not found");
   }
   auto duckpgq_state = (DuckPGQState *)lookup->second.get();
   auto pg_lookup = duckpgq_state->registered_property_graphs.find(
@@ -176,8 +169,7 @@ void CreatePropertyGraphFunction::CreatePropertyGraphFunc(
     duckpgq_state->registered_property_graphs[pg_info->property_graph_name] =
         pg_info->Copy();
   } else {
-    throw BinderException("A property graph with name %s already exists.",
-                          pg_info->property_graph_name);
+    throw Exception(ExceptionType::INVALID,"A property graph with name " + pg_info->property_graph_name + " already exists.");
   }
 }
 }; // namespace duckdb

duckpgq/src/duckpgq/functions/tablefunctions/match.cpp

@@ -31,8 +31,8 @@ PGQMatchFunction::FindGraphTable(const string &label,
                                  CreatePropertyGraphInfo &pg_table) {
   const auto graph_table_entry = pg_table.label_map.find(label);
   if (graph_table_entry == pg_table.label_map.end()) {
-    throw BinderException("The label %s is not registered in property graph %s",
-                          label, pg_table.property_graph_name);
+    throw Exception(ExceptionType::BINDER, "The label " + label +
+      " is not registered in property graph " + pg_table.property_graph_name);
   }
 
   return graph_table_entry->second;

duckpgq/src/duckpgq_extension.cpp

@@ -85,7 +85,7 @@ ParserExtensionParseResult duckpgq_parse(ParserExtensionInfo *info,
   parser.ParseQuery((query[0] == '-') ? query.substr(1, query.length())
                                       : query);
   if (parser.statements.size() != 1) {
-    throw ParserException(
+    throw Exception(ExceptionType::PARSER,
         "More than 1 statement detected, please only give one.");
   }
   return {make_uniq_base<ParserExtensionParseData, DuckPGQParseData>(
@@ -97,7 +97,7 @@ BoundStatement duckpgq_bind(ClientContext &context, Binder &binder,
                             SQLStatement &statement) {
   auto lookup = context.registered_state.find("duckpgq");
   if (lookup == context.registered_state.end()) {
-    throw BinderException("Registered state not found");
+    throw Exception(ExceptionType::BINDER, "Registered state not found");
   }
 
   auto duckpgq_state = (DuckPGQState *)lookup->second.get();
@@ -107,7 +107,7 @@ BoundStatement duckpgq_bind(ClientContext &context, Binder &binder,
   if (duckpgq_parse_data) {
     return duckpgq_binder->Bind(*(duckpgq_parse_data->statement));
   }
-  throw BinderException("Unable to find DuckPGQ Parse Data");
+  throw Exception(ExceptionType::BINDER, "Unable to find DuckPGQ Parse Data");
 }
 
 void duckpgq_find_match_function(TableRef *table_ref,
@@ -135,7 +135,7 @@ duckpgq_handle_statement(SQLStatement *statement, DuckPGQState &duckpgq_state) {
     const auto select_node =
         dynamic_cast<SelectNode *>(select_statement->node.get());
     duckpgq_find_match_function(select_node->from_table.get(), duckpgq_state);
-    throw Exception("use duckpgq_bind instead");
+    throw Exception(ExceptionType::BINDER, "use duckpgq_bind instead");
   }
   if (statement->type == StatementType::CREATE_STATEMENT) {
     const auto &create_statement = statement->Cast<CreateStatement>();
@@ -170,7 +170,7 @@ duckpgq_handle_statement(SQLStatement *statement, DuckPGQState &duckpgq_state) {
     const auto select_node =
         dynamic_cast<SelectNode *>(copy_statement.select_statement.get());
     duckpgq_find_match_function(select_node->from_table.get(), duckpgq_state);
-    throw Exception("use duckpgq_bind instead");
+    throw Exception(ExceptionType::BINDER, "use duckpgq_bind instead");
   }
   if (statement->type == StatementType::INSERT_STATEMENT) {
     const auto &insert_statement = statement->Cast<InsertStatement>();
@@ -180,8 +180,8 @@ duckpgq_handle_statement(SQLStatement *statement, DuckPGQState &duckpgq_state) {
 
   // Preferably throw NotImplementedExpection here, but only BinderExceptions
   // are caught properly on MacOS right now
-  throw BinderException("%s has not been implemented yet for DuckPGQ queries",
-                        StatementTypeToString(statement->type));
+  throw Exception(ExceptionType::NOT_IMPLEMENTED,
+    StatementTypeToString(statement->type) + "has not been implemented yet for DuckPGQ queries");
 }
 
 ParserExtensionPlanResult
@@ -201,7 +201,7 @@ duckpgq_plan(ParserExtensionInfo *, ClientContext &context,
       dynamic_cast<DuckPGQParseData *>(duckpgq_state->parse_data.get());
 
   if (!duckpgq_parse_data) {
-    throw BinderException("No DuckPGQ parse data found");
+    throw Exception(ExceptionType::BINDER, "No DuckPGQ parse data found");
   }
 
   auto statement = duckpgq_parse_data->statement.get();

test/sql/altering_table.test

@@ -64,3 +64,4 @@ statement error
     WHERE a.name = 'Daniel'
     COLUMNS (a.jd)
     ) study;
+----

test/sql/create_pg/create_property_graph.test

@@ -47,7 +47,7 @@ EDGE TABLES (
             PROPERTIES ( createDate ) LABEL Knows
     )
 ----
-Binder Error: Property graph table with name pg already exists
+Invalid Error: Property graph table with name pg already exists
 
 # Alias for the vertex table
 statement ok
@@ -130,7 +130,7 @@ EDGE TABLES (
             PROPERTIES ( createDate ) LABEL Knows
     )
 ----
-Binder Error: Referenced vertex table Student does not exist.
+Invalid Error: Referenced vertex table Student does not exist.
 
 
 # Should fail since the edge table references vertex tables that do not exist
@@ -145,4 +145,4 @@ EDGE TABLES (
             PROPERTIES ( createDate ) LABEL Knows
     );
 ----
-Binder Error: Referenced vertex table Student does not exist.
+Invalid Error: Referenced vertex table Student does not exist.

test/sql/create_pg/drop_property_graph.test

@@ -65,6 +65,8 @@ FROM GRAPH_TABLE (pg
     (a:Person)
     COLUMNS (a.id)
     ) study;
+----
+Binder Error: Property graph pg does not exist
 
 statement ok
 -CREATE PROPERTY GRAPH pg

test/sql/create_pg/optional_edge_table_clause.test

@@ -20,4 +20,6 @@ statement error
 -FROM GRAPH_TABLE (snb
     MATCH (p:Person)-[k:Knows]->(p2:Person)
     COLUMNS (*)
-    ) tmp
+    ) tmp
+----
+Binder Error: The label knows is not registered in property graph snb

test/sql/path-finding/complex_matching.test

@@ -109,18 +109,19 @@ query III
     MATCH o = ANY SHORTEST (p:Person)-[w:knows]-> {1,3}(p2:Person)-[i:hasInterest]->(t:Tag)
     COLUMNS (p.id as p_id, p2.id as p2_id, t.id)
     ) tmp
+    ORDER BY p_id, p2_id
     limit 10;
 ----
-14	32985348833329	3
-14	32985348833329	139
-14	30786325577731	196
-14	28587302322196	280
-14	28587302322180	294
-14	24189255811081	295
-14	28587302322196	448
-14	32985348833329	470
-14	28587302322196	540
-14	24189255811109	543
+14	10995116277782	598
+14	10995116277782	805
+14	10995116277782	1174
+14	10995116277782	1183
+14	10995116277782	1527
+14	10995116277782	1676
+14	10995116277782	1998
+14	10995116277782	798
+14	10995116277782	1031
+14	10995116277782	1986
 
 query IIIII
 WITH CTE1 AS (SELECT  CREATE_CSR_EDGE(

test/sql/path-finding/shortest_path.test

@@ -86,6 +86,8 @@ statement error
     WHERE a.name = 'Daniel'
     COLUMNS (p, a.name as name, b.name as b_name)
     ) study;
+----
+Binder Error: Referenced column "p" not found in FROM clause!
 
 
 query III