Merge pull request #469 from olivierdalang/SanitizeEntryNames

Warning about invalid entry names

python/core/qgsproject.sip

@@ -150,7 +150,7 @@ class QgsProject : QObject
       keys would be the familiar QSettings-like '/' delimited entries, implying
       a hierarchy of keys and corresponding values
 
-      @note The key string <em>must</em> include '/'s.  E.g., "/foo" not "foo".
+      @note The key string must be valid xml tag names in order to be saved to the file.
     */
     //@{
     //! @note available in python bindings as writeEntryBool
@@ -167,8 +167,6 @@ class QgsProject : QObject
         keys would be the familiar QSettings-like '/' delimited entries,
         implying a hierarchy of keys and corresponding values
 
-
-        @note The key string <em>must</em> include '/'s.  E.g., "/foo" not "foo".
     */
     //@{
     QStringList readListEntry( const QString & scope, const QString & key, const QStringList& def = QStringList(), bool *ok = 0 ) const;

src/core/qgsproject.cpp

@@ -27,6 +27,7 @@
 #include "qgslayertreeregistrybridge.h"
 #include "qgslogger.h"
 #include "qgsmaplayerregistry.h"
+#include "qgsmessagelog.h"
 #include "qgspluginlayer.h"
 #include "qgspluginlayerregistry.h"
 #include "qgsprojectfiletransform.h"
@@ -66,6 +67,24 @@ QStringList makeKeyTokens_( QString const &scope, QString const &key )
   // be sure to include the canonical root node
   keyTokens.push_front( "properties" );
 
+  //check validy of keys since an unvalid xml name will will be dropped upon saving the xml file. If not valid, we print a message to the console.
+  for (int i = 0; i < keyTokens.size(); ++i){
+    QString keyToken = keyTokens.at(i);
+
+    //invalid chars in XML are found at http://www.w3.org/TR/REC-xml/#NT-NameChar
+    //note : it seems \x10000-\xEFFFF is valid, but it when added to the regexp, a lot of unwanted characters remain
+    QString nameCharRegexp = QString( "[^:A-Z_a-z\\xC0-\\xD6\\xD8-\\xF6\\xF8-\\x2FF\\x370-\\x37D\\x37F-\\x1FFF\\x200C-\\x200D\\x2070-\\x218F\\x2C00-\\x2FEF\\x3001-\\xD7FF\\xF900-\\xFDCF\\xFDF0-\\xFFFD\\-\\.0-9\\xB7\\x0300-\\x036F\\x203F-\\x2040]" );
+    QString nameStartCharRegexp = QString( "^[^:A-Z_a-z\\xC0-\\xD6\\xD8-\\xF6\\xF8-\\x2FF\\x370-\\x37D\\x37F-\\x1FFF\\x200C-\\x200D\\x2070-\\x218F\\x2C00-\\x2FEF\\x3001-\\xD7FF\\xF900-\\xFDCF\\xFDF0-\\xFFFD]" );
+
+    if( keyToken.contains( QRegExp(nameCharRegexp) ) || keyToken.contains( QRegExp(nameStartCharRegexp) ) ){
+
+      QString errorString = QObject::tr("Entry token invalid : '%1'. The token will not be saved to file.").arg(keyToken);
+      QgsMessageLog::logMessage( errorString, QString::null, QgsMessageLog::CRITICAL );
+
+    }
+
+  }
+
   return keyTokens;
 } // makeKeyTokens_
 

src/core/qgsproject.h

@@ -197,7 +197,7 @@ class CORE_EXPORT QgsProject : public QObject
       keys would be the familiar QSettings-like '/' delimited entries, implying
       a hierarchy of keys and corresponding values
 
-      @note The key string <em>must</em> include '/'s.  E.g., "/foo" not "foo".
+      @note The key string must be valid xml tag names in order to be saved to the file.
     */
     //@{
     //! @note available in python bindings as writeEntryBool
@@ -214,8 +214,6 @@ class CORE_EXPORT QgsProject : public QObject
         keys would be the familiar QSettings-like '/' delimited entries,
         implying a hierarchy of keys and corresponding values
 
-
-        @note The key string <em>must</em> include '/'s.  E.g., "/foo" not "foo".
     */
     //@{
     QStringList readListEntry( const QString & scope, const QString & key, const QStringList& def = QStringList(), bool *ok = 0 ) const;

tests/src/python/CMakeLists.txt

@@ -30,6 +30,7 @@ ADD_PYTHON_TEST(PyQgsDistanceArea test_qgsdistancearea.py)
 ADD_PYTHON_TEST(PyQgsEditWidgets test_qgseditwidgets.py)
 ADD_PYTHON_TEST(PyQgsExpression test_qgsexpression.py)
 ADD_PYTHON_TEST(PyQgsFeature test_qgsfeature.py)
+ADD_PYTHON_TEST(PyQgsProject test_qgsproject.py)
 ADD_PYTHON_TEST(PyQgsFeatureIterator test_qgsfeatureiterator.py)
 ADD_PYTHON_TEST(PyQgsField test_qgsfield.py)
 ADD_PYTHON_TEST(PyQgsFontUtils test_qgsfontutils.py)

tests/src/python/test_qgsproject.py

@@ -0,0 +1,113 @@
+# -*- coding: utf-8 -*-
+"""QGIS Unit tests for QgsProject.
+
+.. note:: This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2 of the License, or
+(at your option) any later version.
+"""
+__author__ = 'Sebastian Dietrich'
+__date__ = '19/11/2015'
+__copyright__ = 'Copyright 2015, The QGIS Project'
+# This will get replaced with a git SHA1 when you do a git archive
+__revision__ = '$Format:%H$'
+
+import qgis
+
+from qgis.core import QgsProject, QgsMessageLog
+
+from utilities import getQgisTestApp, TestCase, unittest
+
+QGISAPP, CANVAS, IFACE, PARENT = getQgisTestApp()
+
+
+class TestQgsProject(TestCase):
+
+    def __init__(self, methodName):
+        """Run once on class initialisation."""
+        unittest.TestCase.__init__(self, methodName)
+        self.messageCaught = False
+
+    def test_makeKeyTokens_(self):
+        # see http://www.w3.org/TR/REC-xml/#d0e804 for a list of valid characters
+
+        invalidTokens = []
+        validTokens = []
+
+        # all test tokens will be generated by prepending or inserting characters to this token
+        validBase = u"valid";
+
+        # some invalid characters, not allowed anywhere in a token
+        # note that '/' must not be added here because it is taken as a separator by makeKeyTokens_()
+        invalidChars = u"+*,;<>|!$%()=?#\x01";
+
+        # generate the characters that are allowed at the start of a token (and at every other position)
+        validStartChars = u":_";
+        charRanges = [
+            (ord(u'a'), ord(u'z')),
+            (ord(u'A'), ord(u'Z')),
+            (0x00F8, 0x02FF),
+            (0x0370, 0x037D),
+            (0x037F, 0x1FFF),
+            (0x200C, 0x200D),
+            (0x2070, 0x218F),
+            (0x2C00, 0x2FEF),
+            (0x3001, 0xD7FF),
+            (0xF900, 0xFDCF),
+            (0xFDF0, 0xFFFD),
+            #(0x10000, 0xEFFFF),   while actually valid, these are not yet accepted by makeKeyTokens_()
+            ]
+        for r in charRanges:
+            for c in range(r[0], r[1]):
+                validStartChars += unichr(c)
+
+        # generate the characters that are only allowed inside a token, not at the start
+        validInlineChars = u"-.\xB7";
+        charRanges = [
+            (ord(u'0'), ord(u'9')),
+            (0x0300, 0x036F),
+            (0x203F, 0x2040),
+            ]
+        for r in charRanges:
+            for c in range(r[0], r[1]):
+                validInlineChars += unichr(c)
+
+        # test forbidden start characters
+        for c in invalidChars + validInlineChars:
+            invalidTokens.append(c + validBase)
+
+        # test forbidden inline characters
+        for c in invalidChars:
+            invalidTokens.append(validBase[:4] + c + validBase[4:])
+
+        # test each allowed start character
+        for c in validStartChars:
+            validTokens.append(c + validBase)
+
+        # test each allowed inline character
+        for c in validInlineChars:
+            validTokens.append(validBase[:4] + c + validBase[4:]);
+
+        logger = QgsMessageLog.instance()
+        logger.messageReceived.connect(self.catchMessage)
+        prj = QgsProject.instance()
+
+        for token in validTokens:
+            self.messageCaught = False
+            prj.readEntry("test", token)
+            myMessage = "valid token '%s' not accepted" % (token)
+            assert not self.messageCaught, myMessage
+
+        for token in invalidTokens:
+            self.messageCaught = False
+            prj.readEntry("test", token)
+            myMessage = "invalid token '%s' accepted" % (token)
+            assert self.messageCaught, myMessage
+
+        logger.messageReceived.disconnect(self.catchMessage)
+
+    def catchMessage(self):
+        self.messageCaught = True
+
+if __name__ == '__main__':
+    unittest.main()