From 27cf04aeccfe58fc7f58ea315b68b9f103c1114e Mon Sep 17 00:00:00 2001
From: tzutalin <tzu.ta.lin@gmail.com>
Date: Fri, 7 Jul 2017 11:09:00 +0800
Subject: [PATCH] Fix issue #96

---
 .gitignore            |  1 +
 libs/labelFile.py     |  5 +++--
 libs/pascal_voc_io.py | 19 +++++++++++--------
 3 files changed, 15 insertions(+), 10 deletions(-)

diff --git a/.gitignore b/.gitignore
index 64cefe08..278fd72d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -15,3 +15,4 @@ tags
 cscope*
 .ycm_extra_conf.py
 .subvimrc
+.vscode
diff --git a/libs/labelFile.py b/libs/labelFile.py
index 437ee033..509ab9d6 100644
--- a/libs/labelFile.py
+++ b/libs/labelFile.py
@@ -33,14 +33,15 @@ class LabelFile(object):
         imgFolderPath = os.path.dirname(imagePath)
         imgFolderName = os.path.split(imgFolderPath)[-1]
         imgFileName = os.path.basename(imagePath)
-        imgFileNameWithoutExt = os.path.splitext(imgFileName)[0]
+        print imgFileName
+        #imgFileNameWithoutExt = os.path.splitext(imgFileName)[0]
         # Read from file path because self.imageData might be empty if saving to
         # Pascal format
         image = QImage()
         image.load(imagePath)
         imageShape = [image.height(), image.width(),
                       1 if image.isGrayscale() else 3]
-        writer = PascalVocWriter(imgFolderName, imgFileNameWithoutExt,
+        writer = PascalVocWriter(imgFolderName, imgFileName,
                                  imageShape, localImgPath=imagePath)
         writer.verified = self.verified
 
diff --git a/libs/pascal_voc_io.py b/libs/pascal_voc_io.py
index eefb3459..15aac581 100644
--- a/libs/pascal_voc_io.py
+++ b/libs/pascal_voc_io.py
@@ -7,7 +7,7 @@ from lxml import etree
 import codecs
 
 XML_EXT = '.xml'
-
+ENCODE_METHOD = 'utf-8'
 
 class PascalVocWriter:
 
@@ -20,14 +20,16 @@ class PascalVocWriter:
         self.localImgPath = localImgPath
         self.verified = False
 
-
     def prettify(self, elem):
         """
             Return a pretty-printed XML string for the Element.
         """
         rough_string = ElementTree.tostring(elem, 'utf8')
         root = etree.fromstring(rough_string)
-        return etree.tostring(root, pretty_print=True)
+        return etree.tostring(root, pretty_print=True, encoding=ENCODE_METHOD).replace('  ', '\t')
+        # minidom does not support UTF-8
+        '''reparsed = minidom.parseString(rough_string)
+        return reparsed.toprettyxml(indent="\t", encoding=ENCODE_METHOD)'''
 
     def genXML(self):
         """
@@ -48,8 +50,9 @@ class PascalVocWriter:
         filename = SubElement(top, 'filename')
         filename.text = self.filename
 
-        localImgPath = SubElement(top, 'path')
-        localImgPath.text = self.localImgPath
+        if self.localImgPath is not None:
+            localImgPath = SubElement(top, 'path')
+            localImgPath.text = self.localImgPath
 
         source = SubElement(top, 'source')
         database = SubElement(source, 'database')
@@ -112,9 +115,9 @@ class PascalVocWriter:
         out_file = None
         if targetFile is None:
             out_file = codecs.open(
-                self.filename + XML_EXT, 'w', encoding='utf-8')
+                self.filename + XML_EXT, 'w', encoding=ENCODE_METHOD)
         else:
-            out_file = codecs.open(targetFile, 'w', encoding='utf-8')
+            out_file = codecs.open(targetFile, 'w', encoding=ENCODE_METHOD)
 
         prettifyResult = self.prettify(root)
         out_file.write(prettifyResult.decode('utf8'))
@@ -144,7 +147,7 @@ class PascalVocReader:
 
     def parseXML(self):
         assert self.filepath.endswith(XML_EXT), "Unsupport file format"
-        parser = etree.XMLParser(encoding='utf-8')
+        parser = etree.XMLParser(encoding=ENCODE_METHOD)
         xmltree = ElementTree.parse(self.filepath, parser=parser).getroot()
         filename = xmltree.find('filename').text
         try: