#25 Add metadata for numeric fields exposed as strings.

Author: yruslan

pom.xml

@@ -110,7 +110,7 @@
         <!-- Frameworks and libraries -->
         <scala.version>2.12.20</scala.version>
         <scala.compat.version>2.12</scala.compat.version>
-        <spark.version>3.4.4</spark.version>
+        <spark.version>3.5.2</spark.version>
         <scalatest.version>3.2.14</scalatest.version>
         <specs.version>2.4.16</specs.version>
         <guava.version>15.0</guava.version>

spark-cobol/pom.xml

@@ -57,6 +57,12 @@
 			<artifactId>cobol-parser_${scala.compat.version}</artifactId>
 			<version>${project.version}</version>
 		</dependency>
+
+		<dependency>
+			<groupId>org.slf4j</groupId>
+			<artifactId>slf4j-log4j12</artifactId>
+			<scope>test</scope>
+		</dependency>
 	</dependencies>
 
     <build>

spark-cobol/src/main/scala/za/co/absa/cobrix/spark/cobol/schema/CobolSchema.scala

@@ -192,6 +192,8 @@ class CobolSchema(copybook: Copybook,
           case _         => StringType
         }
       case dt: Integral  if isDisplayAlwaysString  =>
+        if (metadataPolicy != MetadataPolicy.NoMetadata)
+          addIntegralStringMetadata(metadata, dt)
         StringType
       case dt: Integral  if strictIntegralPrecision  =>
         DecimalType(precision = dt.precision, scale = 0)
@@ -232,6 +234,11 @@ class CobolSchema(copybook: Copybook,
     metadataBuilder.putLong(MAX_LENGTH, a.length)
   }
 
+  private def addIntegralStringMetadata(metadataBuilder: MetadataBuilder, i: Integral): MetadataBuilder = {
+    val maxLength = if (i.signPosition.isDefined) i.precision + 1 else i.precision
+    metadataBuilder.putLong(MAX_LENGTH, maxLength)
+  }
+
   private def addExtendedMetadata(metadataBuilder: MetadataBuilder, s: Statement): MetadataBuilder = {
     metadataBuilder.putLong("level", s.level)
     if (s.originalName.nonEmpty && s.originalName != s.name)

spark-cobol/src/test/scala/za/co/absa/cobrix/spark/cobol/source/Test01DisplayPicAsStrings.scala

@@ -18,6 +18,7 @@ package za.co.absa.cobrix.spark.cobol.source
 
 import org.scalatest.funsuite.AnyFunSuite
 import org.slf4j.{Logger, LoggerFactory}
+import za.co.absa.cobrix.spark.cobol.parameters.MetadataFields.MAX_LENGTH
 import za.co.absa.cobrix.spark.cobol.source.base.{SimpleComparisonBase, SparkTestBase}
 import za.co.absa.cobrix.spark.cobol.source.fixtures.BinaryFileFixture
 import za.co.absa.cobrix.spark.cobol.utils.SparkUtils
@@ -120,6 +121,9 @@ class Test01DisplayPicAsStrings extends AnyFunSuite with SparkTestBase with Bina
       val actualSchema = df.schema.treeString
       val actualData = SparkUtils.prettyJSON(df.toJSON.collect().mkString("[", ",", "]"))
 
+      assert(df.schema.fields.head.metadata.getLong(MAX_LENGTH) == 4)
+      assert(df.schema.fields(1).metadata.getLong(MAX_LENGTH) == 5)
+
       assertEqualsMultiline(actualSchema, expectedSchema)
       assertEqualsMultiline(actualData, expectedData)
     }