add recursive tagging option

src/main/groovy/ua/net/nlp/tools/tag/TagOptions.groovy

@@ -12,6 +12,9 @@ public class TagOptions extends OptionsBase {
 
     @Parameters(index = "0", description = "Input files. Default: stdin", arity="0..")
     List<String> inputFiles
+    @Option(names = ["-r", "--recursive"], description = "Tag all files recursively in the given directories")
+    boolean recursive
+
     @Option(names = ["--lemmaOnly"], description = "Prints only lemmas, implies: --outputFormat=txt --disambiguate=true")
     boolean lemmaOnly
 

src/main/groovy/ua/net/nlp/tools/tag/TagTextCore.groovy

@@ -3,6 +3,10 @@
 package ua.net.nlp.tools.tag
 
 import java.math.RoundingMode
+import java.nio.file.Files
+import java.nio.file.Path
+import java.nio.file.Paths
+import java.nio.file.attribute.BasicFileAttributes
 import java.util.concurrent.ExecutorService
 import java.util.concurrent.Executors
 import java.util.concurrent.TimeUnit
@@ -676,30 +680,50 @@ class TagTextCore {
             println("dict_uk version: ${dictUkversion}")
         }
 
-        // TODO: quick hack to support multiple files
-        if( options.inputFiles && options.inputFiles != ["-"] ) {
+        // TODO: quick hack to support recursive processing
+        if( options.recursive ) {
+            def dirs = options.inputFiles ?: ["."]
 
-            ExecutorService executors = Executors.newWorkStealingPool()
-            options.singleThread = true
-            options.inputFiles.forEach{ filename ->
-                options.output = ""
-                options.input = filename
-                nlpUk.setInputOutput(options)
-                IOFiles files = TextUtils.prepareInputOutput(options)
-
-                executors.submit({
-                    nlpUk.process(files)
-                } as Runnable)
+            List<File> files = []
+            dirs.collect { d -> 
+                new File(d).eachFileRecurse { f -> 
+                      if( f.name.toLowerCase().endsWith(".txt") ) {
+                          files << f
+                      }
+                }
             }
 
-            executors.shutdown()
-            executors.awaitTermination(1, TimeUnit.DAYS)
-            nlpUk.postProcess()
+            println "Found ${files.size()} files with .txt extension" // + files
+            options.singleThread = true
+            processFilesParallel(nlpUk, options, files.collect { it.path })
+        }
+        // TODO: quick hack to support multiple files
+        else if( options.inputFiles && options.inputFiles != ["-"] ) {
+            options.singleThread = true
+            processFilesParallel(nlpUk, options, options.inputFiles)
         }
         else {
             nlpUk.process()
             nlpUk.postProcess()
         }
     }
 
+    static processFilesParallel(TagTextCore nlpUk, TagOptions options, List<String> inputFiles) {
+        ExecutorService executors = Executors.newWorkStealingPool()
+        inputFiles.forEach{ filename ->
+            options.output = ""
+            options.input = filename
+            nlpUk.setInputOutput(options)
+            IOFiles files = TextUtils.prepareInputOutput(options)
+
+            executors.submit({
+                nlpUk.process(files)
+            } as Runnable)
+        }
+
+        executors.shutdown()
+        executors.awaitTermination(1, TimeUnit.DAYS)
+        nlpUk.postProcess()
+    }
+
 }

src/main/python/tag_text.py

@@ -2,7 +2,7 @@
 
 # This script allows to tag Ukrinian text
 # by invoking TagText.groovy that uses LanguageTool API
-# groovy >= 3.0 (http://www.groovy-lang.org) needs to be installed and in the path
+# JDK > 17 and groovy >= 4.0 (http://www.groovy-lang.org) needs to be installed and in the path
 # Usage: tag_text.py <inputfile>
 
 import os

src/main/python/tag_text_recursive.py

@@ -0,0 +1,66 @@
+#!/usr/bin/python3
+
+# This script allows to tag Ukrinian text
+# by invoking TagText.groovy that uses LanguageTool API
+# JDK > 17 and groovy >= 4.0 (http://www.groovy-lang.org) needs to be installed and in the path
+# Usage: tag_text.py <inputfile>
+
+import os
+import sys
+import subprocess
+import threading
+import argparse
+
+
+ENCODING='utf-8'
+SCRIPT_PATH=os.path.dirname(__file__) + '/../groovy/ua/net/nlp/tools'
+
+in_txt = None
+
+parser = argparse.ArgumentParser()
+parser.add_argument("-v", help="Verbose",  action="store_true")
+parser.add_argument("-g", help="Disambiguate and print first token only",  action="store_true")
+parser.add_argument("dir", default=".", type=str, help="Directory to look for txt files in")
+
+args = parser.parse_args()
+
+def print_output(p):
+
+    print("output: ", p.stdout.read().decode(ENCODING))
+
+def print_error(p):
+
+    error_txt = p.stderr.read().decode(ENCODING)
+    if error_txt:
+        print("stderr: ", error_txt, "\n", file=sys.stderr)
+
+
+# technically only needed on Windows
+my_env = os.environ.copy()
+my_env["JAVA_TOOL_OPTIONS"] = "-Dfile.encoding=UTF-8"
+
+
+groovy_cmd = 'groovy.bat' if sys.platform == "win32" else 'groovy'
+cmd = [groovy_cmd, SCRIPT_PATH + '/TagText.groovy']
+
+if args.g:
+    cmd.append('-g')
+    cmd.append('-t1')
+
+cmd.append('-r')
+cmd.append(args.dir)
+
+if args.v:
+    print('Running: ' + str(cmd))
+else:
+    cmd.append('-q')
+
+
+p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stdin=subprocess.PIPE, stderr=subprocess.PIPE, env=my_env)
+
+threading.Thread(target=print_output, args=(p,)).start()
+threading.Thread(target=print_error, args=(p,)).start()
+
+
+p.stdin.close()
+