added some exception handling

README.md

@@ -31,7 +31,7 @@ This will produce five PNGs of the form `<*-page-*>-chars,words,lines,paragraphs
 
 ###How to Use
 ------------
-Put `"edu.psu.sagnik.research" %% "pdsimplifyparser" % "<version-number>" exclude("javax.jms", "jms") exclude("com.sun.jdmk", "jmxtools") exclude("com.sun.jmx", "jmxri")` where <version-number> is the latest version number from `Build.sbt`. Current is `0.0.4` (subject to change).
+Put `"edu.psu.sagnik.research" %% "pdsimplifyparser" % "<version-number>" exclude("javax.jms", "jms") exclude("com.sun.jdmk", "jmxtools") exclude("com.sun.jmx", "jmxri")` where <version-number> is the latest version number from `Build.sbt`. Current is `0.0.5` (subject to change).
 
 in your dependencies and make sure you have `https://oss.sonatype.org/content/repositories/releases/` in your resolvers.   
 

build.sbt

@@ -3,9 +3,9 @@
 version in ThisBuild := {
   val major = 0
   val minor = 0
-  val patch = 4
+  val patch = 5
   s"$major.$minor.$patch"
-} //added patch for rasters that we were not getting before. 
+} //added some error handling for font information.
 
 scalaVersion in ThisBuild := "2.11.8"
 

pdsimplifyparser/src/main/scala/edu/psu/sagnik/research/pdsimplify/impl/ProcessDocument.scala

@@ -8,6 +8,8 @@ import edu.psu.sagnik.research.pdsimplify.raster.impl.ProcessRaster
 import edu.psu.sagnik.research.pdsimplify.text.impl.ProcessText
 import org.apache.pdfbox.pdmodel.{PDDocument, PDPage}
 
+import scala.util.Try
+
 /**
   * Created by schoudhury on 7/1/16.
   */
@@ -29,12 +31,14 @@ object ProcessDocument {
       paragraphs=paragraphs,
       gPaths=pdGraphicsPaths,
       rasters=rasters,
-      bb=Rectangle(
-        page.getBBox.getLowerLeftX,
-        page.getBBox.getHeight-page.getBBox.getUpperRightY,
-        page.getBBox.getUpperRightX,
-        page.getBBox.getUpperRightY
-      )
+      bb=
+        Rectangle(
+          page.getBBox.getLowerLeftX,
+          page.getBBox.getHeight-page.getBBox.getUpperRightY,
+          page.getBBox.getUpperRightX,
+          page.getBBox.getUpperRightY
+        )
+
     )
   }
 

pdsimplifyparser/src/main/scala/edu/psu/sagnik/research/pdsimplify/model/DocumentDataModels.scala

@@ -4,6 +4,8 @@ import edu.psu.sagnik.research.pdsimplify.path.model.PDPath
 import edu.psu.sagnik.research.pdsimplify.raster.model.PDRasterImage
 import edu.psu.sagnik.research.pdsimplify.text.model.PDParagraph
 
+import scala.util.Try
+
 /**
   * Created by schoudhury on 7/1/16.
   */

pdsimplifyparser/src/main/scala/edu/psu/sagnik/research/pdsimplify/text/impl/CreateTextStyle.scala

@@ -1,23 +1,37 @@
 package edu.psu.sagnik.research.pdsimplify.text.impl
 
+import java.util.logging.{Level, Logger}
+
 import edu.psu.sagnik.research.pdsimplify.path.impl.CreatePathStyle
 import edu.psu.sagnik.research.pdsimplify.text.model.{PDChar, PDCharStyle, PDFontInfo}
 import org.apache.pdfbox.pdmodel.graphics.state.PDGraphicsState
 import org.apache.pdfbox.text.TextPosition
 
+import scala.util.{Failure, Success, Try}
+
 /**
   * Created by schoudhury on 6/30/16.
   */
 object CreateTextStyle {
+  lazy val logger = Logger.getLogger("pdsimplify.text.impl.CreateTextStyle")
+  logger.setLevel(Level.ALL)
+  //TODO: change these to proper default values
+  val DEFAULT_FONT_NAME="times new roman"
+  val DEFAULT_FONT_SIZE=10f
+  val DEFAULT_FONT_FAMILY="times"
+  val DEFAULT_IS_BOLD=false
+  val DEFAULT_IS_ITALIC=false
+  val DEFAULT_FONT_WEIGHT=100f
+  val DEFAULT_ROTATION=0f
 
   def apply(x:TextPosition,gs:PDGraphicsState):PDCharStyle=PDCharStyle(
     font=PDFontInfo(
-      fontName= x.getFont.getName,
-      fontSize = x.getFontSizeInPt,
-      fontFamily = x.getFont.getFontDescriptor.getFontFamily,
-      isBold=x.getFont.getFontDescriptor.isForceBold, //TODO: text can be made look bold or italic by modifying the text or the text line matrix
-      isItalic=x.getFont.getFontDescriptor.isItalic,
-      fontWeight=x.getFont.getFontDescriptor.getFontWeight
+      fontName= Try(x.getFont.getName) match {case Success(v) => v; case Failure(e) => logger.warning(e.getMessage); DEFAULT_FONT_NAME },
+      fontSize = Try(x.getFontSizeInPt) match {case Success(v) => v; case Failure(e) => logger.warning(e.getMessage); DEFAULT_FONT_SIZE },
+      fontFamily = Try(x.getFont.getFontDescriptor.getFontFamily) match {case Success(v) => v; case Failure(e) => logger.warning(e.getMessage); DEFAULT_FONT_FAMILY },
+      isBold=Try(x.getFont.getFontDescriptor.isForceBold) match {case Success(v) => v; case Failure(e) => logger.warning(e.getMessage); DEFAULT_IS_BOLD }, //TODO: text can be made look bold or italic by modifying the text or the text line matrix
+      isItalic=Try(x.getFont.getFontDescriptor.isItalic) match {case Success(v) => v; case Failure(e) => logger.warning(e.getMessage); DEFAULT_IS_ITALIC },
+      fontWeight=Try(x.getFont.getFontDescriptor.getFontWeight) match {case Success(v) => v; case Failure(e) => logger.warning(e.getMessage); DEFAULT_FONT_WEIGHT }
     ),
     fill = if (gs.getTextState.getRenderingMode.isFill)
       CreatePathStyle.getHexRGB(gs.getNonStrokingColor)

pdsimplifyparser/src/main/scala/edu/psu/sagnik/research/pdsimplify/text/impl/ProcessText.scala

@@ -100,18 +100,16 @@ class ProcessText(page:PDPage) extends PDFTextStripper {
     else None
   }
 
-  /*
   @Override @throws[IOException]
   override protected def writeString(s: String, textPositions: util.List[TextPosition]): Unit = {
     //this has to be done because sometimes the writeLine() method is not calling the writeWords() method at all, especially
     //when the string has space characters.
     val tPs=textPositions.asScala.toList
     tPs.foreach(tP=>{
-      println("<"+tP.getUnicode+"/>")
       if (!" ".equals(tP)){
         currentChars=currentChars :+ PDChar(
           content=tP.getUnicode,
-          bb=TextPositionBB.approximate(tP),
+          bb=TextPositionBB.approximate(tP,page),
           glyphBB=TextPositionBB.glyphBased(tP,page),
           CreateTextStyle(tP,getGraphicsState)
         )
@@ -121,8 +119,8 @@ class ProcessText(page:PDPage) extends PDFTextStripper {
       }
     })
   }
-  */
 
+/*
   @Override @throws[IOException]
   override protected def writeString(s: String, textPositions: util.List[TextPosition]): Unit = {
     //this has to be done because sometimes the writeLine() method is not calling the writeWords() method at all, especially
@@ -148,6 +146,7 @@ class ProcessText(page:PDPage) extends PDFTextStripper {
     )
     super.writeString(s)
   }
+*/
 
   def stripPage(pdPageNum: Int, document: PDDocument): List[PDParagraph] = {
     setStartPage(pdPageNum + 1);

writers/src/main/scala/edu/psu/sagnik/research/pdwriters/impl/ShowResults.scala

@@ -4,17 +4,18 @@ import java.awt.Color
 import java.io.File
 
 import edu.psu.sagnik.research.pdsimplify.model.Rectangle
-import edu.psu.sagnik.research.pdsimplify.path.impl.ProcessPaths
-import edu.psu.sagnik.research.pdsimplify.raster.impl.ProcessRaster
-import edu.psu.sagnik.research.pdsimplify.text.impl.ProcessText
-import edu.psu.sagnik.research.pdwriters.writers.pdf.CreateMarkedPDF
-import edu.psu.sagnik.research.pdwriters.writers.svg.CreateSVG
 import org.apache.pdfbox.pdmodel.PDDocument
 import java.util.logging.{Level, Logger}
-import Level.{FINE, INFO}
 
+import edu.psu.sagnik.research.pdsimplify.impl.ProcessDocument
+import edu.psu.sagnik.research.pdsimplify.path.model.PDPath
+import edu.psu.sagnik.research.pdsimplify.raster.model.PDRasterImage
+import edu.psu.sagnik.research.pdsimplify.text.model.PDParagraph
 import edu.psu.sagnik.research.pdwriters.writers.image.CreateMarkedPNG
 
+
+import scala.util.{Failure, Success}
+
 /**
  * Created by schoudhury on 6/27/16.
  */
@@ -41,14 +42,14 @@ object ShowResults {
     val document = PDDocument.load(new File(pdLoc))
     val page = document.getPage(pageNum)
 
-    val paragraphs=new ProcessText(page).stripPage(pageNum,document)
+    val SimplifiedDocument=ProcessDocument(document)
+
+    val paragraphs=SimplifiedDocument.pages(pageNum).paragraphs
+    val rasters=SimplifiedDocument.pages(pageNum).rasters
+    val graphicsPaths=SimplifiedDocument.pages(pageNum).gPaths
 
-    val imFinder=new ProcessRaster(page)
-    imFinder.getImages()
 
-    val pathFinder=new ProcessPaths(page)
-    pathFinder.getPaths()
-    val segments=pathFinder.paths
+    val segments=graphicsPaths
       .filter(x=> x.doPaint)
       .flatMap(x=>x.subPaths)
       .flatMap(x=>x.segments)
@@ -77,7 +78,7 @@ object ShowResults {
 
     printExtractionResult(pdLoc,pageNum,paragraphs.map(_.bb),Color.CYAN,"paragraphs")
 
-    printExtractionResult(pdLoc,pageNum,imFinder.rasterImages.map(_.bb),Color.MAGENTA,"rasters")
+    printExtractionResult(pdLoc,pageNum,rasters.map(_.bb),Color.MAGENTA,"rasters")
 
     printExtractionResult(pdLoc,pageNum,segments.map(_.bb),Color.ORANGE,"paths")
 

writers/src/main/scala/edu/psu/sagnik/research/pdwriters/writers/json/CreateJSON.scala

@@ -36,7 +36,6 @@ object CreateJSON {
 
   def apply(jsonLoc:String,pDS: PDDocumentSimple)= {
     val jsonContent =
-      (
         "pages" ->
           pDS.pages.map {
             p =>
@@ -94,6 +93,6 @@ object CreateJSON {
                     )
                 )
           }
-        )
+
   }
 }