DEV Community: Drazan-Jarak

How to delete empty rows from tables in a Word document

Drazan-Jarak — Fri, 11 Sep 2020 21:00:05 +0000

Recently I have received a request from a customer that asked: ''How to delete empty rows from tables in a Word document?'' At first, this seemed like an easy task that can be resolved with a simple code snippet. For this assignment we will use Aspose.Words API. Here is the solution:

       //load the document
       Document doc = new Document(path+ "DocumentWithEmpty Rows.docx");
       //get all rows
       Node[] rows = doc.GetChildNodes(NodeType.Row, true).ToArray();
       //iterate throughthe rows and check if all the cells are empty
       foreach (Row row in rows)
       {
           bool removeRow = true;
           foreach (Cell cell in row.Cells)
           {
               if (cell.FirstParagraph != null)
                   removeRow = !cell.FirstParagraph.HasChildNodes;
           }
           //if all the cells in the row are empty remove the row
           if (removeRow)
               row.Remove();
       }
       doc.Save(MyDir + "DocumentWithoutEmptyRows.docx");

This snippet works well in simpler DOC(X) documents and satisfies basic needs.
But if there are bookmark nodes placed on a block (within the document) then the provided snippet throws an exception. To avoid this we will add two more lines of code before loading the document and make a correction in the first line:

       LoadOptions options = new LoadOptions();
       options.AnnotationsAtBlockLevel = false;
       //load the document
       Document doc = new Document(path+ "DocumentWithEmpty Rows.docx", options);

The other potential problems would be the following: "What if some cells contain empty spaces, new line characters/carriage returns, or what if some rows are nested within a cell and so on?" The following code snippet removes all the unnecessary symbols (spaces and new line characters) at the end of the cells and if all the cells within a row are empty then it deletes the row:

       LoadOptions options = new LoadOptions();
       options.AnnotationsAtBlockLevel = false;
        //load the document
       Document doc = new Document(path+ "DocumentWithEmpty Rows.docx", options);
       Words.Document doc = new Words.Document(MyDir + "Original_MultiRow_Columns - Copy.docx");
       Words.SectionCollection sections = doc.Sections;
       foreach (Words.Section section in sections)
           if (section != null)
               foreach (Words.Tables.Table t in section.Body.Tables)
                   foreach (Row r in t.Rows)
                   {
                       bool removeRow = true;
                       foreach (Cell c in r.Cells)
                       {
                           Node[] paragraphs = c.GetChildNodes(NodeType.Paragraph, true).ToArray();
                           for (int i = paragraphs.Length - 1; i >= 0; i--)
                           {
                               Words.Paragraph p = (Words.Paragraph)paragraphs[i];
                               if (p.ToString(SaveFormat.Text).Trim() == "")
                               {
                                   p.Remove();
                               }
                               else
                               {
                                   removeRow = false;
                                        break;
                               }
                           }
                       }
                       if (removeRow)
                           r.Remove();
                   }
       doc.Save(MyDir + "output20.3v1.docx")

I hope this short article can help someone.

Converting HTML file to PDF using Aspose.Words API

Drazan-Jarak — Thu, 20 Aug 2020 17:42:27 +0000

Introduction:

In this article, you will be able to observe and test how to convert an HTML file with images to PDF.
Actually, I got a request from a customer and he wanted to use Aspose.Words library for this task.

Prerequisites:

Our Aspose.Words libraries work on any platform and include – Aspose.Words for .NET, Aspose.Words for Java, Aspose.Words for C++, Aspose.Words for Android via Java, Aspose.Words for SharePoint, Aspose.Words for Reporting Services and Aspose.Words for Jasper Reports. For this article, we will use Aspose.Words for .NET which can be downloaded here.

Simple conversion:

Usually, the conversion from HTML to PDF is simple like this:

   Aspose.Words.LoadOptions options = new Aspose.Words.LoadOptions();
   options.LoadFormat = Aspose.Words.LoadFormat.Html;
   Aspose.Words.Document document = new Aspose.Words.Document(inputFile, options);
   document.Save(outputFile, Aspose.Words.SaveFormat.Pdf);

Potential problem:

A potential problem here could be that the images from the source document don't fit the page width in the PDF output.

Solution:

   //loading the HTML file
   Words.LoadOptions options = new Words.LoadOptions();
           options.LoadFormat = LoadFormat.Html;
   Words.Document doc = new Words.Document("source.htm", options);
   //iterate through the sections in the doc Document
   foreach (Section section in doc.Sections)
   {
      //change document margins – the numbers can be adjusted for your needs
      Words.PageSetup ps = section.PageSetup;
      double margin = ps.TopMargin;
      ps.TopMargin = margin * 0.35;
      ps.BottomMargin = margin * 0.35;
      double targetWidth = ps.PageWidth - ps.LeftMargin - ps.RightMargin;

      //finding all the shapes in the document
      NodeCollection shapes = doc.GetChildNodes(NodeType.Shape, true);
      foreach (Words.Drawing.Shape shape in    shapes.OfType<Words.Drawing.Shape>())
      {
         if (shape.HasImage)
         {
            //checking if the image fits the page 
            double imgWith = ConvertUtil.PixelToPoint(shape.Width);
            if (targetWidth < imgWith)
            {
                shape.Width = targetWidth;
            }
         }

      }
   }
   doc.Save("output.pdf", SaveFormat.Pdf);

Conclusion: The Aspose.HTML and Aspose.PDF APIs could also be used for the conversion, but in this request, the customer wanted to use Aspose.Words. You can find more examples on our Documentation page.

How to lock a text in shape in a PowerPoint presentation

Drazan-Jarak — Wed, 12 Aug 2020 11:30:56 +0000

Introduction and the problem

The tool used for doing this is Aspose.Slides.
The issue here is that after locking the shape, (when double click on it) the text within the shape is still editable.
This is the code that should usually work to lock a shape:
C#:

//Applying shapes locks
AutoShapeLock.PositionLocked = true;
AutoShapeLock.SelectLocked = true;
AutoShapeLock.SizeLocked = true;
AutoShapeLock.AdjustHandlesLocked = true;
AutoShapeLock.ArrowheadsLocked = true;
AutoShapeLock.AspectRatioLocked = true;
AutoShapeLock.EditPointsLocked = true;
AutoShapeLock.GroupingLocked = true;
AutoShapeLock.PositionLocked = true;
AutoShapeLock.RotateLocked = true;
AutoShapeLock.ShapeTypeLocked = true;
AutoShapeLock.SizeLocked = true;
AutoShapeLock.TextLocked = true;

The main problem is in the fact that it is not possible to do this even when using MS PowerPoint.

Solution/workaround

So, the idea behind this article is to explore a workaround and an approach from a different perspective or think outside of the box. Theoretically, using a transparent rectangle shape makes it an easy solution by overlaying it in front of the text box or shape and lock it. Programmatically, it would look like this code snippet:

  Presentation pr = new Presentation(path + "Before.pptx");
  //Traversing through all the shapes in the slides
  ISlide slide = pr.Slides[0];
  for (int slideCount = 0; slideCount < pr.Slides.Count; slideCount++)
  {
      slide = pr.Slides[slideCount];
      IShape shapeEx;
      int shapesCount = slide.Shapes.Count;
      for (int count = 0; count < shapesCount; count++)
      {
          shapeEx = slide.Shapes[count];
          //if shape is autoshape
          if (shapeEx is AutoShape)
          {
              //Type casting to Auto shape and getting auto shape lock
              AutoShape Ashp = shapeEx as AutoShape;
              IAutoShapeLock AutoShapeLock = Ashp.ShapeLock;
              //Applying shapes locks
              AutoShapeLock.PositionLocked = true;
              AutoShapeLock.SelectLocked = true;
              AutoShapeLock.SizeLocked = true;
              AutoShapeLock.AdjustHandlesLocked = true;
              AutoShapeLock.ArrowheadsLocked = true;
              AutoShapeLock.AspectRatioLocked = true;
              AutoShapeLock.EditPointsLocked = true;
              AutoShapeLock.GroupingLocked = true;
              AutoShapeLock.PositionLocked = true;
              AutoShapeLock.RotateLocked = true;
              AutoShapeLock.ShapeTypeLocked = true;
              AutoShapeLock.SizeLocked = true;
              AutoShapeLock.TextLocked = true;
              //Workaround for text locking: create a transparent rectangle shape, overlay it in front of the text box and lock it
              IAutoShape transpRect = slide.Shapes.AddAutoShape(ShapeType.Rectangle, Ashp.X - 5, Ashp.Y - 5, Ashp.Width + 10, Ashp.Height + 10);
              transpRect.FillFormat.FillType = FillType.Solid;
              transpRect.FillFormat.SolidFillColor.Color = System.Drawing.Color.Transparent;
              transpRect.LineFormat.FillFormat.FillType = FillType.NoFill;

              transpRect.ShapeLock.PositionLocked = true;
              transpRect.ShapeLock.SelectLocked = true;
              transpRect.ShapeLock.SizeLocked = true;
          }
          //if shape is GraphicalObjectEx
          if (shapeEx is GraphicalObject)
          {
              //Type casting to Auto shape and getting auto shape lock
              GraphicalObject Ashp = shapeEx as GraphicalObject;
              IGraphicalObjectLock AutoShapeLock = Ashp.ShapeLock;

              //Applying shapes locks
              AutoShapeLock.PositionLocked = true;
              AutoShapeLock.SelectLocked = true;
              AutoShapeLock.SizeLocked = true;
              AutoShapeLock.AspectRatioLocked = true;
              AutoShapeLock.GroupingLocked = true;
              AutoShapeLock.PositionLocked = true;
              AutoShapeLock.SizeLocked = true;
          }

          //if shape is group shape
          else if (shapeEx is GroupShape)
          {
              //Type casting to group shape and getting group shape loc
              GroupShape Group = shapeEx as GroupShape;
              IGroupShapeLock groupShapeLock = Group.ShapeLock;

              //Applying shapes locks
              groupShapeLock.PositionLocked = true;
              groupShapeLock.SelectLocked = true;
              groupShapeLock.SizeLocked = true;
              groupShapeLock.AspectRatioLocked = true;
              groupShapeLock.GroupingLocked = true;
              groupShapeLock.PositionLocked = true;
              groupShapeLock.SizeLocked = true;
          }

          //if shape is Connector shape
          else if (shapeEx is Connector)
          {
              //Type casting to connector shape and getting connector shape lock
              Connector Conn = shapeEx as Connector;
              IConnectorLock ConnLock = Conn.ShapeLock;
              //Applying shapes locks
              ConnLock.PositionMove = false;
              ConnLock.SelectLocked = true;
              ConnLock.SizeLocked = true;
              ConnLock.AdjustHandlesLocked = true;
              ConnLock.ArrowheadsLocked = true;
              ConnLock.AspectRatioLocked = true;
              ConnLock.EditPointsLocked = true;
              ConnLock.GroupingLocked = true;
              ConnLock.RotateLocked = true;
              ConnLock.ShapeTypeLocked = true;
              ConnLock.SizeLocked = true;
          }

          //if shape is picture frame
          else if (shapeEx is PictureFrame)

              //Type casting to picture frame shape and getting picture frame shape lock
              PictureFrame Pic = shapeEx as PictureFrame;
              IPictureFrameLock PicLock = Pic.ShapeLock;
              //Applying shapes locks
              PicLock.PositionLocked = true;
              PicLock.SelectLocked = true;
              PicLock.SizeLocked = true;
              PicLock.AdjustHandlesLocked = true;
              PicLock.ArrowheadsLocked = true;
              PicLock.AspectRatioLocked = true;
              PicLock.EditPointsLocked = true;
              PicLock.GroupingLocked = true;
              PicLock.PositionLocked = true;
              PicLock.ShapeTypeLocked = true;
              PicLock.SizeLocked = true;
          }
      }
  }
  pr.Save(path + @"output.pptx", Aspose.Slides.Export.SaveFormat.Pptx);

This example shows a simple and easy workaround on how to lock all shapes in a presentation as well as text within shapes using Aspose.Slides.
I hope this article will be useful for others encountering the same problem.

DOCX Comments to PDF Annotations Conversion

Drazan-Jarak — Tue, 21 Jul 2020 21:07:53 +0000

Intro

This article will explain how to programmatically convey DOCX comments to PDF annotations. First, you need some tools for doing it. Currently, I don’t know any tool which can be used to make this possible other than Aspose.Words and Aspose.PDF.
So, the first part of the conversion will be handled by using Aspose.Words product.
This part contains a few steps:

Extract/read comments from the DOCX.
Mark the beginning and the end of the comments.
And save the document in memory stream (PDF format) with marked comments.

Aspose.PDF API is responsible for the following actions:

Iterate through every page and text fragment to find the start and end mark of the comments.
When finds the marked text fragments it splits them into segments and adds highlight annotation for each fragment
If there were replies in the Word comments in the next step Aspose.PDF adds replies to their original annotations.
In the end, we simply save the document.

Code explanation

Extract/read comments from the DOCX:

Open document with comments:

       String filename = @" source.docx";
       Aspose.Words.Document d = new 
       Aspose.Words.Document(filename);

Extract/read comments:

Get child nodes

            NodeCollection comments = doc.GetChildNodes(NodeType.Comment, true);
            int numOfComments = comments.Count;
            foreach (Aspose.Words.Comment comment in comments)
            {
                Comment comm = readComment(comment);
                wordComments.Add(comm);
                if (comment.Ancestor == null)
signComment(doc, new DocumentBuilder(doc), comment); // only main comment
            }
            comments.Clear();

Read comments and replies:

        private static Comment readComment(Aspose.Words.Comment comment)
        {
            Comment com = new Comment();
            com.Id = comment.Id;
            com.Contents = comment.GetText();
            com.Author = comment.Author;
            com.DateTime = comment.DateTime;
            com.Replies = new List<Reply>();

            foreach (Aspose.Words.Comment rpl in comment.Replies)
            {
                Reply r = new Reply();
                r.Contents = rpl.GetText();
                r.DateTime = rpl.DateTime;
                r.Author = rpl.Author;
                r.Id = rpl.Id;
                com.Replies.Add(r);
            }
            return com;
         }

Sign comments/Mark the beginning and the end of the comments:

        private static void signComment(Aspose.Words.Document doc, DocumentBuilder builder, Aspose.Words.Comment comment)
        {
            /*
             * insert tempTExt with tempFontSize before and after commented text
             */
            // start Run
            Run run = new Run(doc);
            run.Text = tempText;
            run.Font.Name = tempFont;
            run.Font.Color = tempColor;
            run.Font.Size = fontSize;
            // end Run
            Run run2 = new Run(doc);
            run2.Text = tempText;
            run2.Font.Name = tempFont;
            run2.Font.Color = tempColor;
            run2.Font.Size = fontSize;

            CommentRangeStart commentStart = (CommentRangeStart)doc.GetChild(NodeType.CommentRangeStart, comment.Id, true);
            CommentRangeEnd commentEnd = (CommentRangeEnd)doc.GetChild(NodeType.CommentRangeEnd, comment.Id, true);
            builder.MoveTo(commentStart);
            builder.InsertNode(run);
            builder.MoveTo(commentEnd);
            builder.InsertNode(run2);
        }

Save the document in memory stream (PDF format)

d.Save(mms, Aspose.Words.SaveFormat.Pdf);

Add highlight annotation:

Find text fragments:

        private static int addMarkupAnnotation(MemoryStream mms)
        {
            _document = new Aspose.Pdf.Document(mms);
            var pdfContentEditor = new PdfContentEditor(_document);
            int pagenum = 0;
            bool start = false;
            bool end = false;
            TextFragment tf = null;
            foreach (Page page in _document.Pages)
            {
                pagenum++;
                int numOfCommmentsPerPage = 0;
                TextFragmentAbsorber tfa = new TextFragmentAbsorber();
                tfa.Visit(page);

                int tfnum = 0;
                foreach (TextFragment textFragment in tfa.TextFragments)
                {
                    tfnum++;
                    foreach (var ts in textFragment.Segments)
                    {
                        if (!end && start && ts.TextState.FontSize == fontSize && ts.Text.Equals(tempText))
                        {
                            end = true;
                            ts.Text = "";
                            start = false;
                        }
                        if (ts.TextState.FontSize == fontSize && ts.Text.Equals(tempText))
                        {
                            start = true;
                            ts.Text = "";
                            if (tf == null)
                            {
                                tf = new TextFragment();
                                tf.Position = textFragment.Position;
                            }
                        }
                        if (start && tf == null)
                        {
                            tf = new TextFragment();
                            tf.Position = textFragment.Position;
                        }
                        if (start && ts.TextState.FontSize != fontSize)
                        {
                            tf.Segments.Add(ts);
                        }
                    }
                    if (tf != null && end)
                    {
                        Comment comm = (Comment)wordComments.Where(comment => comment.Id == (numOfCommments)).First();
                        var highlightAnnotation = HighLightTextFragment(page, tf, comm.Contents, comm.Author, comm.DateTime);
                        Console.WriteLine("main comment " + numOfCommments);
                        Aspose.Pdf.Rectangle rect = highlightAnnotation.Rect;
                        page.Annotations.Add(highlightAnnotation);
                        numOfCommments++;
                        numOfCommmentsPerPage++;
                        if (comm.Replies.Count > 0)
                        {
                            foreach (Reply item in comm.Replies)
                            {
                                addReply(page, rect, item.Contents, item.Author, item.DateTime, numOfCommmentsPerPage);
                                numOfCommmentsPerPage++;
                                numOfCommments++;
                            }
                        }
                        tf = null;
                        start = false;
                        end = false;
                    }
                } // end of fragments
                if (start && !end)
                {
                    Console.WriteLine("Comment appears across two pages");
                    Comment comm = (Comment)wordComments.Where(comment => comment.Id == (numOfCommments)).First();

                    var highlightAnnotation = HighLightTextFragment(page, tf, comm.Contents, comm.Author, comm.DateTime);
                    Aspose.Pdf.Rectangle rect = highlightAnnotation.Rect;
                    page.Annotations.Add(highlightAnnotation);
                    Console.WriteLine();
                    numOfCommmentsPerPage++;
                    if (comm.Replies.Count > 0)
                    {
                        foreach (Reply item in comm.Replies)
                        {
                            addReply(page, rect, item.Contents, item.Author, item.DateTime, numOfCommmentsPerPage);
                            numOfCommmentsPerPage++;
                        }
                    }
                    tf = null;
                    start = true;
                    end = false;
                }
            }
            return numOfCommments;
        }

Add highlight annotations

        private static HighlightAnnotation HighLightTextFragment(Page page, TextFragment textFragment, String commenttext, String author, DateTime datetime)
        {
            if (textFragment.Segments.Count == 1)
                return new HighlightAnnotation(page, textFragment.Segments[1].Rectangle);

            var offset = 0;
            var quadPoints = new Aspose.Pdf.Point[textFragment.Segments.Count * 4];
            foreach (var segment in textFragment.Segments)
            {
                quadPoints[offset + 0] = new Aspose.Pdf.Point(segment.Rectangle.LLX, segment.Rectangle.URY);
                quadPoints[offset + 1] = new Aspose.Pdf.Point(segment.Rectangle.URX, segment.Rectangle.URY);
                quadPoints[offset + 2] = new Aspose.Pdf.Point(segment.Rectangle.LLX, segment.Rectangle.LLY);
                quadPoints[offset + 3] = new Aspose.Pdf.Point(segment.Rectangle.URX, segment.Rectangle.LLY);
                offset += 4;
            }
            var llx = quadPoints.Min(pt => pt.X);
            var lly = quadPoints.Min(pt => pt.Y);
            var urx = quadPoints.Max(pt => pt.X);
            var ury = quadPoints.Max(pt => pt.Y);

            return new HighlightAnnotation(page, new Aspose.Pdf.Rectangle(llx, lly, urx, ury))
            {
                Modified = datetime,
                Title = author,
                QuadPoints = quadPoints,
                Contents = commenttext,
            };
        }

Add replies to annotation:

        private static void addReply(Page page, Aspose.Pdf.Rectangle rect, String contents, String author, DateTime dateTime, int numOfComment)
        {
            HighlightAnnotation highlightAnnotation = new HighlightAnnotation(page, rect);
            Console.WriteLine("reply to " + numOfComment);
            highlightAnnotation.Contents = contents;
            highlightAnnotation.Title = author;
            highlightAnnotation.Modified = dateTime;
            highlightAnnotation.InReplyTo = page.Annotations[numOfComment];
            page.Annotations.Add(highlightAnnotation);
        }

Save the document:

        if (rzlt == numOfComments)
        {
            _document.Save(filename + ".pdf");
            Console.WriteLine(rzlt + " comments copied from word to PDF succesfully");
        }
        else
        {
            Console.WriteLine("something went wrong");
        }

Conclusion

Please have in mind that the presented code converts MS Word document with comments to PDF document where the comments are conveyed to annotations. If you don't need this feature you can simply convert DOCX to PDF using the simple two lines of code:

// Load the document from disk.
Document doc = new Document(dataDir + "Rendering.docx");
// Save the document in PDF format.
doc.Save(dataDir + "SaveDoc2Pdf.pdf");

You can read more in the documentation.

I hope you find this article helpful.
You can find the whole project on this link.

Best regards.