mbeckerle commented on code in PR #821:
URL: https://github.com/apache/daffodil/pull/821#discussion_r942622355


##########
daffodil-runtime1/src/main/scala/org/apache/daffodil/infoset/JDOMInfosetOutputter.scala:
##########
@@ -48,14 +48,57 @@ class JDOMInfosetOutputter extends InfosetOutputter
     true
   }
 
-  def startSimple(diSimple: DISimple): Boolean = {
+  def startSimple(diSimple: DISimple, xmlOutputStyle: String): Boolean = {
 
     val elem = createElement(diSimple)
+    val correctFormat = new StringBuilder("");
+    val cdataIntro = "<![CDATA["
+    val cdataOutro = "]]>"
+    var charEntMode: Boolean = false
 
     if (diSimple.hasValue) {
       val text =
         if (diSimple.erd.optPrimType.get.isInstanceOf[NodeInfo.String.Kind]) {
-          remapped(diSimple.dataValueAsString)
+          val s = remapped(diSimple.dataValueAsString)
+
+            if(xmlOutputStyle == "prettyPrintSafe"){
+              val newData = s.replaceAll(">","&gt;").replace("\\r\\n", 
"&#xE00D;")
+              val readyForCDATA = newData.replaceAll("0x","&#x")
+              //Figure out what mode you have to be in
+              if(readyForCDATA(0) == '&') {
+                charEntMode = true
+              } else {
+                charEntMode = false
+                correctFormat.append(cdataIntro)
+              }
+
+              //Traverse over the string surrounding correct areas with CDATA 
info
+              for(c <- readyForCDATA) {
+                if(charEntMode) {
+                  correctFormat.append(c)
+                  if(c == ';'){
+                    correctFormat.append(cdataIntro)
+                    charEntMode = false
+                  }
+                } else {
+                  if(c == '&'){
+                    correctFormat.append(cdataOutro)
+                    charEntMode = true
+                  }
+                  correctFormat.append(c)
+                }
+              }
+
+              //You are done with the string. If you are still a non
+                   //char ent then close and finish up.

Review Comment:
   What if the string ends with `abcd&gt` i.e., no ";" at end of entity?
   You will create `<![CDATA[abcd]]>&gt` when really it should have been 
`<![CDATA[abcd&gt]]> `
   I think your loop needs to grab whole char entities in a sub-loop, and only 
decide to end the CDATA with the cdataOutro and then set down the entity IF the 
";" is found. If no ";" is found and end of string is hit, then you just 
include the characters inside the CDATA.
   
   Add a test for "abcd&gt" case.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to