Make Python scripts work with older versions of Python 3

Generation/fluency/CreateLM/sfr2frames.py

@@ -6,7 +6,7 @@
 #
 # adapted by GvN: now undoes MWU, because that is how the model is applied
 # in the generator
-import sys
+import io, sys
 
 def is_mwu(tag):
     result=False
@@ -27,8 +27,10 @@ def is_mwu(tag):
     curId = ''
     frames = []
 
-    sys.stdin.reconfigure(encoding='utf-8')
-    sys.stdout.reconfigure(encoding='utf-8')
+    # sys.stdin.reconfigure(encoding='utf-8')
+    # sys.stdout.reconfigure(encoding='utf-8')
+    sys.stdin = io.TextIOWrapper(sys.stdin.detach(), encoding='utf-8', newline=None)
+    sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8', newline=None, line_buffering=True)
 
     for line in sys.stdin:
         lineParts = line.split('|')

Generation/fluency/CreateLM/sfr2words.py

@@ -5,14 +5,16 @@
 #
 #
 
-import sys
+import io, sys
 
 if __name__ == '__main__':
     curId = ''
     frames = []
 
-    sys.stdin.reconfigure(encoding='utf-8')
-    sys.stdout.reconfigure(encoding='utf-8')
+    # sys.stdin.reconfigure(encoding='utf-8')
+    # sys.stdout.reconfigure(encoding='utf-8')
+    sys.stdin = io.TextIOWrapper(sys.stdin.detach(), encoding='utf-8', newline=None)
+    sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8', newline=None, line_buffering=True)
 
     for line in sys.stdin:
         lineParts = line.split('|')

Generation/fluency/CreateMaxent/evaluate.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python3
 
 import argparse
+import io
 import random
 import re
 import sys
@@ -165,8 +166,11 @@ def printModelScores(scores, bestScores):
     # Make results reproducable.
     random.seed(13)
 
-    sys.stdin.reconfigure(encoding='utf-8')
-    sys.stdout.reconfigure(encoding='utf-8')
+    # sys.stdin.reconfigure(encoding='utf-8')
+    # sys.stdout.reconfigure(encoding='utf-8')
+    sys.stdin = io.TextIOWrapper(sys.stdin.detach(), encoding='utf-8', newline=None)
+    sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8', newline=None, line_buffering=True)
+
 
     with open(options.features, encoding='utf-8') as featureFh:
         featureWeights = readFeatureWeights(featureFh)

Generation/fluency/CreateMaxent/filter_features.py

@@ -4,6 +4,7 @@
 #
 
 import argparse
+import io
 import sys
 
 VELLDAL_FEATURES = ['lds', 'ldsb', 'lds_dl', 'lds_skew', 'tngram', 'tngramw']
@@ -38,8 +39,10 @@
     tagNgramFeatures = options.tagNgramFeatures
     velldalFeatures = options.velldalFeatures
 
-    sys.stdin.reconfigure(encoding='utf-8')
-    sys.stdout.reconfigure(encoding='utf-8')
+    # sys.stdin.reconfigure(encoding='utf-8')
+    # sys.stdout.reconfigure(encoding='utf-8')
+    sys.stdin = io.TextIOWrapper(sys.stdin.detach(), encoding='utf-8', newline=None)
+    sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8', newline=None, line_buffering=True)
 
     for line in sys.stdin:
         line = line.strip()

TreebankTools/bin/dtcanonicalize.py

@@ -68,7 +68,7 @@
 
 from xml.dom import minidom
 from xml.parsers.expat import ExpatError
-import sys, os
+import io, os, sys
 import tempfile
 import html
 
@@ -204,13 +204,11 @@ def dump_content(self):
             print("comment:", comment)
 
 
-    def write_xml(self, outfile, encoding='UTF-8', version='1.5'):
-        """write the dependency structure to xml in encoding"""
-
-        outfile.reconfigure(encoding=encoding)
+    def write_xml(self, outfile, version='1.5'):
+        """write the dependency structure to xml"""
 
         # de xml-header
-        outfile.write('<?xml version="1.0" encoding="%s"?>\n' % encoding)
+        outfile.write('<?xml version="1.0" encoding="UTF-8"?>\n')
 
         # de root node
         if version == "":
@@ -766,12 +764,12 @@ def write_xml_data(outfile, data):
         parser.print_help(sys.stderr)
         sys.exit(1)
 
-
     dtparser = DTParser()
 
     # de bestanden canoniek maken
     error_occurred = False
     if options.stdin:
+        sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8', newline=None, line_buffering=True)
         try:
             cn = Canonicalizer(dtparser, sys.stdin)
             cn.canonicalize()
@@ -796,7 +794,7 @@ def write_xml_data(outfile, data):
                                                   os.path.dirname(file))
 
                 #  - we schrijven de nieuwe file naar die tempfile
-                tempfileobj = os.fdopen(tempfd, "w")
+                tempfileobj = os.fdopen(tempfd, "w", encoding="utf-8")
                 cn.write_xml(tempfileobj)
                 tempfileobj.close()
 

webdemo.py

@@ -7,7 +7,7 @@
 from urllib import parse
 
 #sys.stdin.reconfigure(encoding='utf-8')
-sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8')
+sys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding='utf-8', newline=None, line_buffering=True)
 form = {}
 for line in sys.stdin:
     for key, val in parse.parse_qsl(line,