moved text support api into own namespace

Author: holgerbrandl

build.gradle

@@ -49,8 +49,8 @@ task javadocJar(type: Jar, dependsOn: javadoc) {
 
 //http://stackoverflow.com/questions/34377367/why-is-gradle-install-replacing-my-version-with-unspecified
 group = 'de.mpicbg.scicomp'
-version = '1.1.4'
-//version = '1.1-SNAPSHOT'
+//version = '1.1.4'
+version = '1.2-SNAPSHOT'
 
 
 artifacts {

src/main/kotlin/kscript/KscriptUtil.kt

@@ -6,7 +6,7 @@ import kotlin.system.exitProcess
  * @author Holger Brandl
  */
 
-/** Similar to require but without a stacktrace. */
+/** Similar to require but without a stacktrace which makes it more suited for CLIs. */
 inline fun stopIfNot(value: Boolean, lazyMessage: () -> Any) {
     if (!value) {
         System.err.println("[ERROR] " + lazyMessage().toString())

src/main/kotlin/kscript/experimental/OneLinerContext.kt

@@ -1,6 +1,6 @@
 package kscript.experimental
 
-import kscript.resolveArgFile
+import kscript.text.resolveArgFile
 
 /**
  * @author Holger Brandl
@@ -9,7 +9,7 @@ import kscript.resolveArgFile
 abstract class OneLinerContext(args: Array<String>) {
 
     val arg by lazy { resolveArgFile(args) }
-    //    val stdin by lazy {  kscript.stdin }
+    //    val stdin by lazy {  kscript.support.getStdin }
 
     init {
         apply(arg)

src/main/kotlin/kscript/StreamUtil.kt …c/main/kotlin/kscript/text/StreamUtil.ktsrc/main/kotlin/kscript/StreamUtil.kt renamed to src/main/kotlin/kscript/text/StreamUtil.kt src/main/kotlin/kscript/StreamUtil.kt renamed to src/main/kotlin/kscript/text/StreamUtil.kt

@@ -1,13 +1,10 @@
-package kscript
-
-import java.io.*
-import java.util.zip.GZIPOutputStream
+package kscript.text
 
 
 /** a `Sequence<String>` iterator for standard input */
 val stdin by lazy { generateSequence() { readLine() } }
 
-fun linesFrom(file: File) = BufferedReader(FileReader(file)).lineSequence()
+fun linesFrom(file: java.io.File) = java.io.BufferedReader(java.io.FileReader(file)).lineSequence()
 
 // just used for testing and development
 fun linesFrom(vararg lines: String) = lines.asSequence()
@@ -17,24 +14,34 @@ fun linesFrom(vararg lines: String) = lines.asSequence()
  * File argument processor that works similar to awk. If data is available on stdin, use it. If not expect a file argument and read from that one instead.
  * */
 fun resolveArgFile(args: Array<String>, position: Int = 0): Sequence<String> {
-    if (stdin.iterator().hasNext()) return stdin
+    //    if (stdin.iterator().hasNext()) return stdin
+    if (System.`in`.available() > 0) return kscript.text.stdin
 
-    stopIfNot(args.isNotEmpty()) { "Missing file or input input stream" }
-    stopIfNot(args.size >= position) { "arg position ${position} exceeds number of arguments ${args.size} " }
+    kscript.stopIfNot(args.isNotEmpty()) { "Missing file or input input stream" }
+    kscript.stopIfNot(args.size >= position) { "arg position ${position} exceeds number of arguments ${args.size} " }
 
     val fileArg = args[position]
 
     //    stdinNames: List<String> = listOf("-", "stdin")
     //    if (stdinNames.contains(fileArg)) return stdin
 
-    val inputFile = File(fileArg)
+    val inputFile = java.io.File(fileArg)
+
+    kscript.stopIfNot(inputFile.canRead()) { "Can not read from '${fileArg}'" }
+
+    // test for compression and uncompress files automatically
+    val isCompressedInput = inputFile.name.run { endsWith(".zip") || endsWith(".gz") }
+
+    val lineReader = if (isCompressedInput) {
+        java.io.InputStreamReader(java.util.zip.GZIPInputStream(java.io.FileInputStream(inputFile)))
+    } else {
+        java.io.FileReader(inputFile)
+    }
 
-    stopIfNot(inputFile.canRead()) { "Can not read from '${fileArg}'" }
 
     // todo we don't close the buffer with this approach
     //    BufferedReader(FileReader(inputFile )).use { return it }
-
-    return BufferedReader(FileReader(inputFile)).lineSequence()
+    return java.io.BufferedReader(lineReader).lineSequence()
 }
 
 
@@ -48,7 +55,7 @@ fun Iterable<String>.print() = forEach { println(it) }
 
 //https://dzone.com/articles/readingwriting-compressed-and
 /** Save a list of items into a file. Output can be option ally zipped and a the stringifying operation can be changed from toString to custom operation if needed. */
-fun <T> Iterable<T>.saveAs(f: File,
+fun <T> Iterable<T>.saveAs(f: java.io.File,
                            transform: (T) -> String = { it.toString() },
                            separator: Char = '\n',
                            overwrite: Boolean = true,
@@ -57,7 +64,7 @@ fun <T> Iterable<T>.saveAs(f: File,
     // ensure that file is not yet there or overwrite flag is set
     require(!f.isFile || overwrite) { "$f is present already. Use overwrite=true to enforce file replacement." }
 
-    val p = if (!compress) PrintWriter(f) else BufferedWriter(OutputStreamWriter(GZIPOutputStream(FileOutputStream(f))))
+    val p = if (!compress) java.io.PrintWriter(f) else java.io.BufferedWriter(java.io.OutputStreamWriter(java.util.zip.GZIPOutputStream(java.io.FileOutputStream(f))))
 
     toList().forEach { p.write(transform(it) + separator) }
 

src/main/kotlin/kscript/AwkLike.kt src/main/kotlin/kscript/text/Tables.ktsrc/main/kotlin/kscript/AwkLike.kt renamed to src/main/kotlin/kscript/text/Tables.kt

@@ -1,4 +1,4 @@
-package kscript
+package kscript.text
 
 /**
  * Utility methods to allow for awk-like data processing using kscript.
@@ -50,7 +50,7 @@ fun Sequence<Row>.add(rule: (Row) -> String): Sequence<Row> {
 }
 
 
-//@Deprecated("use kscript.awk() instead")
+//@Deprecated("use kscript.support.awk() instead")
 //fun Sequence<String>.splitMap(vararg rules: (Row) -> String, separator: String = "\t", joinWith: String = separator) {
 //    map { it.split(separator).let { splitLine -> rules.map { it(splitLine) } } }.print()
 //}

src/main/kotlin/kscript/DocOpt.kt src/main/kotlin/kscript/util/DocOpt.ktsrc/main/kotlin/kscript/DocOpt.kt renamed to src/main/kotlin/kscript/util/DocOpt.kt

@@ -1,4 +1,4 @@
-package kscript
+package kscript.util
 
 import org.docopt.Docopt
 import java.io.File

src/test/kotlin/kscript/examples/AwkComparison.kt

@@ -1,7 +1,7 @@
 package kscript.examples
 
-import kscript.*
 import kscript.experimental.OneLinerContext
+import kscript.text.*
 
 /**
  * @author Holger Brandl

src/test/kotlin/kscript/examples/OneLinerExample.kt

@@ -1,12 +1,12 @@
 package kscript.examples
 
-import kscript.add
-import kscript.split
 import kscript.experimental.OneLinerContext
-import kscript.print
+import kscript.text.add
+import kscript.text.print
+import kscript.text.split
 
 /**
- * One-liner kscript example. To ease devlopment simply extend [OneLinerContext] as shown, which will provide
+ * One-liner kscript example. To ease development simply extend [OneLinerContext] as shown, which will provide
  * the same context as `kscript` when running in single line mode.
  *
  * @author Holger Brandl
@@ -17,7 +17,8 @@ fun main(args: Array<String>) {
     object : OneLinerContext(args) {
 
         override fun apply(lines: Sequence<String>) {
-            lines.split().filter { it[3] == "UA" }.add { it[3] + ":" + it[3] }.print()
+            lines.split().drop(1).filter { it[9] == "UA" }.add { it[3] + ":" + it[3] }.print()
+            // kscript 'lines.split().drop(1).filter { it[9] == "UA" }.add { it[3] + ":" + it[3] }.print()'
         }
     }
 }

src/test/kotlin/kscript/test/DocOptTest.kt

@@ -2,7 +2,7 @@ package kscript.test
 
 import io.kotlintest.matchers.shouldBe
 import io.kotlintest.specs.StringSpec
-import kscript.DocOpt
+import kscript.util.DocOpt
 import java.io.File
 
 /**

src/test/kotlin/kscript/test/SupportApiTest.kt

@@ -1,24 +1,40 @@
 package kscript.test
 
+import io.kotlintest.matchers.should
+import io.kotlintest.matchers.shouldBe
+import io.kotlintest.matchers.shouldEqual
+import io.kotlintest.matchers.startWith
 import io.kotlintest.specs.StringSpec
-import kscript.*
+import kscript.text.linesFrom
+import kscript.text.print
+import kscript.text.resolveArgFile
+import kscript.text.split
 import java.io.File
 
 /**
  * @author Holger Brandl
  */
 class SupportApiTest : StringSpec() { init {
 
-    "allow to use stdin for filtering and mapping" {
-        stdin.filter { "^de0[-0]*".toRegex().matches(it) }.map { it + "foo:" }.print()
+    //    "allow to use stdin for filtering and mapping" {
+    //        kscript.text.stdin.filter { "^de0[-0]*".toRegex().matches(it) }.map { it + "foo:" }.print()
+    //    }
+
+    "extract field with column filter" {
+        linesFrom(File("src/test/resources/flights_head.txt")).split().
+                filter { it[8] == "N14228" }.
+                map { it[10] }.
+                toList().
+                apply {
+                    size shouldEqual 1
+                    first() shouldEqual "EWR"
+                }
+
     }
 
-    "length should return size of string" {
-        //        "hello".length shouldBe 5
-        //        stopIfNot("FOO"=="BAR"){"condition not met"}
-        println("current dir is " + File(".").absolutePath)
-        linesFrom(File("src/test/resources/flights_head.txt")).map { it.split("\t")[7] }.print()
-        linesFrom(File("src/test/resources/flights_head.txt")).filter { it.split("\t")[7] == "UA" }.print()
+    "compressed lines should be unzipped on the fly"{
+        resolveArgFile(arrayOf("src/test/resources/flights.tsv.gz")).
+                drop(1).first() should startWith("2013")
     }
 }
 }