V1 3

build.sbt

@@ -1,5 +1,5 @@
 ThisBuild / organization := "com.phasmidsoftware"
-ThisBuild / version := "1.2.6"
+ThisBuild / version := "1.3.0"
 ThisBuild / scalaVersion := "2.13.17"
 ThisBuild / scalacOptions ++= Seq("-encoding", "UTF-8", "-unchecked", "-deprecation")
 ThisBuild / scalacOptions ++= Seq("-java-output-version", "17")
@@ -54,6 +54,26 @@ lazy val cats = project.dependsOn(core).settings(
   )
 )
 
+lazy val parquet = project.dependsOn(core).settings(
+name := "tableparser-parquet",
+  libraryDependencies ++= Seq(
+    "org.apache.parquet" % "parquet-column" % "1.15.2",
+    "org.apache.parquet" % "parquet-hadoop" % "1.15.2",
+    "org.apache.hadoop"  % "hadoop-common"  % "3.4.1" % "provided",
+    "org.apache.hadoop"  % "hadoop-mapreduce-client-core" % "3.4.1" % Test,
+    "org.scalatest"     %% "scalatest"      % scalaTestVersion % Test
+  )
+)
+
+lazy val spark = project.dependsOn(core).settings(
+  name := "tableparser-spark",
+  libraryDependencies ++= Seq(
+    "org.apache.spark" %% "spark-sql" % sparkVersion % "provided",
+    "org.slf4j" % "slf4j-simple" % "2.0.17" % Test,
+    "org.scalatest" %% "scalatest" % scalaTestVersion % Test
+  )
+)
+
 lazy val zio = project.dependsOn(core).settings(
   name := "tableparser-zio",
   libraryDependencies ++= Seq(
@@ -68,17 +88,8 @@ lazy val zio = project.dependsOn(core).settings(
   )
 )
 
-lazy val spark = project.dependsOn(core).settings(
-  name := "tableparser-spark",
-  libraryDependencies ++= Seq(
-    "org.apache.spark" %% "spark-sql" % sparkVersion % "provided",
-    "org.slf4j" % "slf4j-simple" % "2.0.17" % Test,
-    "org.scalatest" %% "scalatest" % scalaTestVersion % Test
-  )
-)
-
 lazy val root = (project in file("."))
-        .aggregate(core, cats, zio, spark)
+        .aggregate(core, cats, parquet, spark, zio)
         .settings(
           name := "TableParser",
           publish / skip := true

cats/src/main/scala/com/phasmidsoftware/tableparser/cats/table/TableCrypt.scala

@@ -25,6 +25,7 @@ object TableCrypt {
    * @tparam A the cipher algorithm (for which there must be evidence of HexEncryption[A]).
    * @param csvAttributes implicit value of CsvAttributes.
    */
+  @deprecated("Use writeCSVFileEncrypted(Path) instead", "1.3.0")
   def writeCSVFileEncrypted[A: HexEncryption, Row](table: Table[Row])(file: File)(implicit renderer: CsvRenderer[Row], generator: CsvGenerator[Row], hasKey: HasKey[Row], csvAttributes: CsvAttributes): Unit =
     CsvTableEncryptedFileRenderer[Row, A](file).render(table)
 }

core/src/main/scala/com/phasmidsoftware/tableparser/core/examples/crime/Crime.scala

@@ -7,7 +7,7 @@ import scala.util.Try
 
 /**
  * This example of table parsing is based on the Kaggle data set:
- * [[https://www.kaggle.com/datasets/marshuu/crimes-in-uk-2023?select=2023-01-metropolitan-street.csv]]
+ * [[https://www.kaggle.com/datasets/marshuu/crimes-in-uk-2023?select=2023-01-metropolitan-street.csv Metropolitan Crime Data]]
  *
  * The file under resources is an edited version of the Metropolitan Crime Statistics 2023-01 (only the first 5,000 rows)
  *

core/src/main/scala/com/phasmidsoftware/tableparser/core/parse/ColumnHelper.scala

@@ -73,5 +73,9 @@ object ColumnHelper {
   /**
    * Precede each upper case letter (or digit) with _.
    */
-  val camelToSnakeCaseColumnNameMapper: String => String = _.replaceAll("([A-Z\\d])", "_$1")
+  val camelToSnakeCaseColumnNameMapper: String => String =
+    _.replaceAll("([A-Z\\d])", "_$1")
+
+  val camelToSnakeCaseColumnNameMapperLower: String => String =
+    camelToSnakeCaseColumnNameMapper andThen (_.toLowerCase)
 }

core/src/main/scala/com/phasmidsoftware/tableparser/core/parse/Parseable.scala

@@ -6,6 +6,7 @@ package com.phasmidsoftware.tableparser.core.parse
 
 import java.io.File
 import java.net.URL
+import java.nio.file.Path
 import org.joda.time.LocalDate
 import scala.annotation.implicitNotFound
 import scala.util.parsing.combinator.JavaTokenParsers
@@ -290,6 +291,23 @@ object Parseable {
 
   implicit object ParseableURL extends ParseableURL
 
+  /**
+   * A trait representing a path that can be parsed from a String.
+   *
+   * This type class extends `Parseable[java.nio.file.Path]`, enabling the parsing of `String` inputs
+   * into `java.nio.file.Path` objects. It provides a default implementation of the `parse`
+   * method, leveraging utility methods such as `lift` and `parseAndRecover` to handle parsing
+   * and recovery in case of errors.
+   */
+  trait ParseablePath extends Parseable[Path] {
+    def parse(s: String, optModifier: Option[String]): Try[Path] =
+      parseAndRecover(s)(lift(java.nio.file.Paths.get(_)))(
+        w => s"ParseablePath: cannot interpret '$w' as a Path"
+      )
+  }
+
+  implicit object ParseablePath extends ParseablePath
+
   /**
    * Parser of File.
    */
@@ -452,6 +470,8 @@ object ParseableOption {
 
   implicit object ParseableOptionFile extends ParseableOption[File]
 
+  implicit object ParseableOptionPath extends ParseableOption[Path]
+
 }
 
 /**

core/src/main/scala/com/phasmidsoftware/tableparser/core/parse/TableBuilder.scala

@@ -0,0 +1,43 @@
+package com.phasmidsoftware.tableparser.core.parse
+
+import com.phasmidsoftware.tableparser.core.parse.TableParser.includeAll
+import com.phasmidsoftware.tableparser.core.table.Header
+import scala.util.Try
+
+/**
+ * Minimal trait expressing the table-building contract, independent
+ * of how the input source is read.
+ *
+ * TableParser extends this for string-based sources.
+ * ParquetTableParser extends this directly for Parquet sources.
+ * Future source types (JDBC ResultSet, JSON, etc.) can do the same.
+ *
+ * @tparam Table the table type to be built.
+ */
+trait TableBuilder[Table] {
+
+  /**
+   * The row type.
+   */
+  type Row
+
+  /**
+   * Method to construct a Table based on the given iterator of rows and the given header.
+   *
+   * @param rows   an iterator of Row objects representing the data rows.
+   * @param header a Header object representing the table's column headers.
+   * @return the constructed Table based on the input rows and header.
+   */
+  protected def builder(rows: Iterator[Row], header: Header): Table
+
+  /**
+   * If true, individual row failures are logged but do not
+   * cause the overall parse to fail.
+   */
+  protected val forgiving: Boolean = false
+
+  /**
+   * Predicate to filter rows. Defaults to including all rows.
+   */
+  protected val predicate: Try[Row] => Boolean = includeAll
+}

core/src/main/scala/com/phasmidsoftware/tableparser/core/parse/TableParser.scala

@@ -4,7 +4,6 @@
 
 package com.phasmidsoftware.tableparser.core.parse
 
-import com.phasmidsoftware.tableparser.core.parse.TableParser.includeAll
 import com.phasmidsoftware.tableparser.core.table._
 import com.phasmidsoftware.tableparser.core.util.{Joinable, TeeIterator, TryUsing}
 import org.slf4j.{Logger, LoggerFactory}
@@ -19,24 +18,18 @@ import scala.util.{Failure, Random, Success, Try}
  * @tparam Table the Table type.
  */
 @implicitNotFound(msg = "Cannot find an implicit instance of TableParser[${Table}]. Typically, you should define an instance of StringTableParser or StringsTableParser.")
-trait TableParser[Table] {
-
-  /**
-   * The row type.
-   */
-  type Row
-
+trait TableParser[Table] extends TableBuilder[Table] {
   /**
    * The input type, typically `String` or `Strings`.
    */
   type Input
 
   /**
-   * This variable determines if there is a programmed, i.e., fixed, header for the parser.
-   * If its value is None, it signifies that we must look to the first line(s) of data
-   * for an appropriate header.
+   * Method to define a row parser.
+   *
+   * @return a RowParser[Row, Input].
    */
-  protected val maybeHeader: Option[Header] = None
+  val rowParser: RowParser[Row, Input]
 
   /**
    * This indicates the number of header rows which must be read from the input.
@@ -48,20 +41,11 @@ trait TableParser[Table] {
   val headerRowsToRead: Int = 1
 
   /**
-   * Method to construct a Table based on the provided rows and header.
-   *
-   * @param rows   an iterator of Row objects representing the data rows.
-   * @param header a Header object representing the table's column headers.
-   * @return the constructed Table based on the input rows and header.
-   */
-  protected def builder(rows: Iterator[Row], header: Header): Table
-
-  /**
-   * Method to determine how errors are handled.
-   *
-   * @return true if individual errors are logged but do not cause parsing to fail.
+   * This variable determines if there is a programmed, i.e., fixed, header for the parser.
+   * If its value is None, it signifies that we must look to the first line(s) of data
+   * for an appropriate header.
    */
-  protected val forgiving: Boolean = false
+  protected val maybeHeader: Option[Header] = None
 
   /**
    * Value to determine whether it is acceptable to have a quoted string span more than one line.
@@ -71,20 +55,7 @@ trait TableParser[Table] {
   protected val multiline: Boolean = false
 
   /**
-   * Function to determine whether or not a row should be included in the table.
-   * Typically used for random sampling.
-   */
-  protected val predicate: Try[Row] => Boolean = includeAll
-
-  /**
-   * Method to define a row parser.
-   *
-   * @return a RowParser[Row, Input].
-   */
-  val rowParser: RowParser[Row, Input]
-
-  /**
-   * Method to parse a table based on a sequence of Inputs.
+   * Method to parse a table based on an iterator of Inputs.
    *
    * @param xs the sequence of Inputs, one for each row
    * @param n  the number of rows to drop (length of the header).

core/src/main/scala/com/phasmidsoftware/tableparser/core/render/CsvRenderer.scala

@@ -4,6 +4,7 @@ import com.phasmidsoftware.tableparser.core.parse.{StringList, Strings}
 import com.phasmidsoftware.tableparser.core.table._
 import com.phasmidsoftware.tableparser.core.write.Writable
 import java.io.{File, FileWriter}
+import java.nio.file.Path
 import org.joda.time.LocalDate
 import scala.reflect.ClassTag
 import scala.util.Try
@@ -167,8 +168,14 @@ case class CsvTableStringRenderer[T]()(implicit z1: CsvRenderer[T], z2: CsvGener
  *
  * TODO merge this with CsvTableEncryptedFileRenderer to avoid duplicate code.
  *
- * @param file          the file to which the table will be written.
+ * @param path the path to which the table will be written.
  * @param csvAttributes implicit instance of CsvAttributes.
  * @tparam T the type of object to be rendered, must provide evidence of CsvRenderer[T] amd CsvGenerator[T].
  */
-case class CsvTableFileRenderer[T: CsvRenderer : CsvGenerator](file: File)(implicit csvAttributes: CsvAttributes) extends CsvTableRenderer[T, FileWriter]()(implicitly[CsvRenderer[T]], implicitly[CsvGenerator[T]], Writable.fileWritable(file))
+case class CsvTableFileRenderer[T: CsvRenderer : CsvGenerator](path: Path)(implicit csvAttributes: CsvAttributes) extends CsvTableRenderer[T, FileWriter]()(implicitly[CsvRenderer[T]], implicitly[CsvGenerator[T]], Writable.fileWritable(path.toFile))
+
+object CsvTableFileRenderer {
+  @deprecated("Use apply(file: File) instead.", "1.3.0")
+  def apply[T: CsvRenderer : CsvGenerator](file: File)(implicit csvAttributes: CsvAttributes): CsvTableFileRenderer[T] =
+    CsvTableFileRenderer(file.toPath)(implicitly[CsvRenderer[T]], implicitly[CsvGenerator[T]], csvAttributes)
+}

core/src/main/scala/com/phasmidsoftware/tableparser/core/table/Analysis.scala

@@ -202,6 +202,7 @@ object Statistics {
  * resource file path before running the application.
  */
 object Main extends App {
+//  doMain(FP.resource[Crime]("2023-01-metropolitan-street.csv"))
   doMain(FP.resource[Crime]("2023-01-metropolitan-street-sample.csv"))
 
   /**