update the fixedWidth intrinsic to inherit any type information

Author: cwensel

tessellate-main/src/main/antora/modules/reference/pages/transforms.adoc

@@ -83,7 +83,7 @@ NOTE: Many more functions are planned.
 
 Built-in functions on fields can be applied to one or more fields in every tuple in the tuple stream.
 
-`tsid`:: create a unique id as a long or string (using https://github.com/f4b6a3/tsid-creator)
+`tsid`:: Create a unique id as a long or string (using https://github.com/f4b6a3/tsid-creator).
 Def:::
 `^tsid{node:...,nodeCount:...,epoch:...,format:...,counterToZero:...} +> intoField|type`
 `type`:::: must be `string` or `long`, defaults to `long`.
@@ -105,7 +105,8 @@ Placeholders:::::
 - `%z`: base-62
 `counterToZero`:::: Resets the counter portion when the millisecond changes, defaults to `false`.
 
-`fixedWidth`:: pads a row/tuple to a fixed width by inserting nulls at a given index
+`fixedWidth`:: Pads a row/tuple to a fixed width by inserting nulls at a given index.
+If any, type information is inherited from the current fields.
 Def:::
 - `^fixedWidth{width:...,insertAt:...} ->` - replace ALL fields with the fixed with result
 - `^fixedWidth{width:...,insertAt:...} -> new_field|type + new_field2|type + etc` - name the new fields
@@ -114,7 +115,7 @@ Params:::
 `width`:::: The width of the row/tuple, defaults to size of the result fields.
 `insertAt`:::: The index to begin inserting the null padding, defaults to `-1` (last element).
 
-`toJson`:: converts a row/tuple to a JSON string
+`toJson`:: Converts a row/tuple to a JSON string.
 Def:::
 - `^toJson{} ->` - replace ALL fields with the JSON string with the default field name `json`
 - `from_field1 + from_field2 ^toJson{} +> node` - add the arguments to a new json object named `node`
@@ -123,7 +124,7 @@ Def:::
 Def:::
 - `^fromJson{} -> to_field1 + to_field2` - the fields names must match the json properties at the root node
 
-`formatFields`:: reformats all the argument fields to a new format.
+`formatFields`:: Reformats all the argument fields to a new format.
 This is especially useful for remaining compatible with Apache Parquet.
 Def:::
 - `^formatFields{format:...} ->` - replace ALL fields with the formatted result

tessellate-main/src/main/java/io/clusterless/tessellate/pipeline/intrinsic/FixedWidthIntrinsic.java

@@ -49,6 +49,11 @@ public Result create(Fields currentFields, Operation operation) {
             throw new IllegalArgumentException("result fields width must match fixed width");
         }
 
+        // currentFields is equivalent to Field.ALL so we can safely copy over the type information
+        if (!toFields.hasTypes() && currentFields.hasTypes()) {
+            toFields = toFields.applyTypes(currentFields.getTypes());
+        }
+
         FixedWidthFunction function = new FixedWidthFunction(toFields, width, insertAt);
 
         return new Result(Fields.ALL, function, toFields);

tessellate-main/src/test/java/io/clusterless/tessellate/pipeline/PipelineTest.java

@@ -21,6 +21,7 @@
 import io.clusterless.tessellate.util.Format;
 import io.clusterless.tessellate.util.URIs;
 import io.clusterless.tessellate.util.json.JSONUtil;
+import org.jetbrains.annotations.NotNull;
 import org.junit.jupiter.api.Test;
 import org.junit.jupiter.api.extension.ExtendWith;
 
@@ -92,7 +93,7 @@ void badWidth(@PathForResource("/data/delimited-variable-width.csv") URI input,
                 "^fixedWidth{ width:5, insertAt:3 } ->"
         );
 
-        fixedWidthBase(input, output, transform);
+        fixedWidthBase(input, output, transform, List.of());
     }
 
     @Test
@@ -101,19 +102,30 @@ void badWidthWithFields(@PathForResource("/data/delimited-variable-width.csv") U
                 "^fixedWidth{ width:5, insertAt:3 } -> _0+_1+_2+_3+_4"
         );
 
-        fixedWidthBase(input, output, transform);
+        fixedWidthBase(input, output, transform, List.of());
     }
 
-    private static void fixedWidthBase(URI input, URI output, Transform transform) throws IOException {
+    @Test
+    void badWidthWithFieldsAndTypes(@PathForResource("/data/delimited-variable-width.csv") URI input, @PathForOutput URI output) throws IOException {
+        Transform transform = new Transform(
+                "^fixedWidth{ width:5, insertAt:3 } -> _0+_1+_2+_3+_4"
+        );
+
+        fixedWidthBase(input, output, transform, Field.asField("a|string", "b|string", "c|string", "d|string", "e|string"));
+    }
+
+    private static void fixedWidthBase(URI input, URI output, Transform transform, List<@NotNull Field> declared) throws IOException {
         PipelineOptions pipelineOptions = new PipelineOptions();
 
         PipelineDef def = PipelineDef.builder()
                 .withName("test")
                 .withSource(Source.builder()
                         .withInputs(List.of(input))
                         .withSchema(Schema.builder()
+                                .withDeclared(declared)
                                 .withFormat(Format.csv)
                                 .withEmbedsSchema(false)
+                                .withStrictParsing(declared.isEmpty())
                                 .build())
                         .build())
                 .withTransform(transform)