Better error message for copy

src/include/function/cast/functions/cast_string_non_nested_functions.h

@@ -191,7 +191,7 @@ template<class T, bool IS_SIGNED = true>
 inline void simpleIntegerCast(
     const char* input, uint64_t len, T& result, LogicalTypeID typeID = LogicalTypeID::ANY) {
     if (!trySimpleIntegerCast<T, IS_SIGNED>(input, len, result)) {
-        throw ConversionException(stringFormat("Cast failed. {} is not in {} range.",
+        throw ConversionException(stringFormat("Cast failed. Could not convert \"{}\" to {}.",
             std::string{input, len}, LogicalTypeUtils::toString(typeID)));
     }
 }

src/processor/operator/persistent/reader/csv/driver.cpp

@@ -35,9 +35,13 @@ void ParsingDriver::addValue(
             stringFormat("Error in file {}, on line {}: expected {} values per row, but got more.",
                 reader->fileInfo->path, reader->getLineNumber(), reader->numColumns));
     }
-
-    function::CastString::copyStringToVector(
-        chunk.getValueVector(columnIdx).get(), rowNum, value, &reader->option);
+    try {
+        function::CastString::copyStringToVector(
+            chunk.getValueVector(columnIdx).get(), rowNum, value, &reader->option);
+    } catch (ConversionException& e) {
+        throw CopyException(stringFormat("Error in file {} on line {}: {}", reader->fileInfo->path,
+            reader->getLineNumber(), e.what()));
+    }
 }
 
 bool ParsingDriver::addRow(uint64_t /*rowNum*/, common::column_id_t columnCount) {

test/test_files/csv/errors.test

@@ -30,10 +30,10 @@ Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/csv-error-tests/too
 
 -STATEMENT LOAD FROM "${KUZU_ROOT_DIRECTORY}/dataset/csv-error-tests/union-no-conversion.csv" (HEADER=TRUE) RETURN *
 ---- error
-Conversion exception: Could not convert to union type UNION(u:UINT8, s:INT8): a.
+Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/csv-error-tests/union-no-conversion.csv on line 2: Conversion exception: Could not convert to union type UNION(u:UINT8, s:INT8): a.
 
 # Test that errors in serial mode don't hang the database.
 # File is large so the window for the race is large enough.
 -STATEMENT LOAD FROM "${KUZU_ROOT_DIRECTORY}/dataset/csv-error-tests/large-conversion-failure.csv" (HEADER=TRUE, PARALLEL=FALSE) RETURN *
 ---- error
-Conversion exception: Cast failed. a is not in INT64 range.
+Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/csv-error-tests/large-conversion-failure.csv on line 1002: Conversion exception: Cast failed. Could not convert "a" to INT64.

test/test_files/exceptions/copy/invalid_row.test

@@ -21,3 +21,10 @@
 -STATEMENT COPY watch FROM "${KUZU_ROOT_DIRECTORY}/dataset/copy-fault-tests/invalid-row/eWatches.csv"
 ---- error
 Runtime exception: Unable to find primary key value 5.
+
+-CASE InvalidHeader
+-STATEMENT create node table tableOfTypes (id INT64, int64Column INT64, doubleColumn DOUBLE, booleanColumn BOOLEAN, dateColumn DATE, timestampColumn TIMESTAMP, stringColumn STRING, listOfInt INT64[], PRIMARY KEY (id));
+---- ok
+-STATEMENT COPY tableOfTypes FROM "${KUZU_ROOT_DIRECTORY}/dataset/copy-test/node/csv/types_50k.csv"
+---- error
+Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/copy-test/node/csv/types_50k.csv on line 1: Conversion exception: Cast failed. Could not convert "id" to INT64.

test/test_files/exceptions/copy/wrong_header.test

@@ -23,7 +23,7 @@ False
 True
 -STATEMENT COPY person2 FROM "${KUZU_ROOT_DIRECTORY}/dataset/copy-fault-tests/wrong-header/vPerson.csv" (HEADER=true)
 ---- error
-Conversion exception: Cast failed. Guodong is not in INT64 range.
+Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/copy-fault-tests/wrong-header/vPerson.csv on line 2: Conversion exception: Cast failed. Could not convert "Guodong" to INT64.
 -STATEMENT COPY person3 FROM "${KUZU_ROOT_DIRECTORY}/dataset/copy-fault-tests/wrong-header/vPerson.csv" (HEADER=true)
 ---- error
 Binder exception: Number of columns mismatch. Expected 1 but got 2.
@@ -39,7 +39,7 @@ Binder exception: Number of columns mismatch. Expected 2 but got 1.
 Binder exception: Number of columns mismatch. Expected 3 but got 4.
 -STATEMENT COPY knows FROM "${KUZU_ROOT_DIRECTORY}/dataset/copy-fault-tests/wrong-header/eKnowsMissingColumn.csv" (HEADER=true)
 ---- error
-Conversion exception: Error occurred during parsing interval. Field name is missing.
+Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/copy-fault-tests/wrong-header/eKnowsMissingColumn.csv on line 2: Conversion exception: Error occurred during parsing interval. Field name is missing.
 
 -CASE ParquetHeaderMismatch
 -STATEMENT CREATE NODE TABLE User(name STRING, age INT64, PRIMARY KEY (name));
@@ -85,7 +85,7 @@ Binder exception: Number of columns mismatch. Expected 3 but got 2.
 ---- ok
 -STATEMENT COPY person FROM "${KUZU_ROOT_DIRECTORY}/dataset/tinysnb/vPerson.csv" (HEADER=true)
 ---- error
-Conversion exception: Cast failed. Alice is not in INT64 range.
+Copy exception: Error in file ${KUZU_ROOT_DIRECTORY}/dataset/tinysnb/vPerson.csv on line 2: Conversion exception: Cast failed. Could not convert "Alice" to INT64.
 
 -CASE HeaderError
 -STATEMENT create node table person (ID INT64, fName STRING, PRIMARY KEY (ID))

test/test_files/tck/expressions/typeConversion/TypeConversion2.test

@@ -26,7 +26,7 @@
            RETURN to_int64(foo_string) AS foo, to_int64(empty_string) AS empty;
 ## Outcome: the result should be, in any order:
 ---- error
-Conversion exception: Cast failed. foo is not in INT64 range.
+Conversion exception: Cast failed. Could not convert "foo" to INT64.
 
 #  `toInteger()` handling mixed number types
 -CASE Scenario3