'''' This code is part of Document Solutions for PDF .NET demos.'' Copyright (c) MESCIUS inc. All rights reserved.''Imports System.IOImports System.DrawingImports System.LinqImports System.Collections.GenericImports GrapeCity.Documents.PdfImports GrapeCity.Documents.TextImports GrapeCity.Documents.Pdf.TextMapImports GrapeCity.Documents.Pdf.StructureImports GrapeCity.Documents.Pdf.Recognition.StructureImports GCTEXT = GrapeCity.Documents.Text
'' Find tables and read their data using structure tags.Public Class ReadTagsTableData    Private _tf As TextFormat    Private _tfHdr As TextFormat    Private _tfPgHdr As TextFormat    Private _margin As Single = 72
    Public Function CreatePDF(ByVal stream As Stream) As Integer        '' Set up some text formats:        _tf = New TextFormat() With {            .Font = GCTEXT.Font.FromFile(Path.Combine("Resources", "Fonts", "NotoSans-Regular.ttf")),            .FontSize = 9,            .ForeColor = Color.Black        }        _tfHdr = New TextFormat(_tf) With {            .Font = GCTEXT.Font.FromFile(Path.Combine("Resources", "Fonts", "NotoSans-Bold.ttf")),            .FontSize = 11,            .ForeColor = Color.DarkBlue        }        _tfPgHdr = New TextFormat(_tf) With {            .FontSize = 11,            .ForeColor = Color.Gray        }
        '' The resulting PDF:        Dim doc = New GcPdfDocument()        Using s = File.OpenRead(Path.Combine("Resources", "PDFs", "C1Olap-QuickStart.pdf"))            Dim source = New GcPdfDocument()            source.Load(s)            PrintAllTables(doc, source)        End Using        ' Save the PDF:        doc.Save(stream)        '' Done:        Return doc.Pages.Count    End Function
    Private Sub PrintAllTables(doc As GcPdfDocument, source As GcPdfDocument)        '' Get the LogicalStructure and top parent element:        Dim ls As LogicalStructure = source.GetLogicalStructure()        If ls Is Nothing OrElse ls.Elements Is Nothing OrElse ls.Elements.Count = 0 Then            ' No structure tags found:            Util.AddNote("No structure tags were found in the source document.", doc.Pages.Add())            Return        End If        ' The root element:        Dim root As Element = ls.Elements(0)
        '' Find and print all tables:        Dim tables = New List(Of (TextLayout, Page))()        root.Children.ToList().FindAll(Function(e_) e_.StructElement.Type = "Table").ForEach(Sub(t_) tables.Add(PrintTable(t_)))        ' Group tables by the page they were found on:        Dim tablesByPage = tables.GroupBy(Function(t_) t_.Item2.Index)        ' For each page, print all tables found on that page,        ' followed by the original page for reference:        For Each tbp In tablesByPage            ' The page that will contain the extracted table data:            Dim pgTables = doc.NewPage()            ' The page that will contain the source page for reference:            Dim pgSrc = doc.NewPage()            ' Print the original page:            tbp.First().Item2.Draw(pgSrc.Graphics, pgSrc.Bounds)            ' Add a page header:            pgSrc.Graphics.DrawString($"Page {tbp.First().Item2.Index + 1} of the source PDF",                _tfPgHdr, New RectangleF(0, 0, pgSrc.Size.Width, _margin), TextAlignment.Center, ParagraphAlignment.Center, False)            Dim maxHeight As Single = pgTables.Size.Height - _margin * 2            Dim y As Single = _margin            ' Print all table data. For simplicity sake we assume that all table data will fit on a single page:            For Each t In tbp                t.Item1.MaxHeight = maxHeight                t.Item1.MaxWidth = pgTables.Size.Width - _margin * 2                pgTables.Graphics.DrawTextLayout(t.Item1, New PointF(_margin, y))                maxHeight -= t.Item1.ContentHeight + _margin                y += t.Item1.ContentHeight + _margin            Next        Next    End Sub
    Private Function PrintTable(e As Element) As (TextLayout, Page)        If e.Type <> "Table" Then            Throw New Exception($"Unexpected: element type must be 'Table' but it is '{e.Type}'.")        End If
        Dim table = New List(Of List(Of IList(Of ITextParagraph)))()        Dim maxCols = 0        ' Select all child elements with type TR - table rows:        Dim selectRows As Action(Of IReadOnlyList(Of Element)) = Nothing        selectRows = Sub(elements)            For Each ec As Element In elements                If ec.HasChildren Then                    If ec.StructElement.Type = "TR" Then                        Dim cells = ec.Children.ToList().FindAll(Function(e_) e_.StructElement.Type = "TD").ToArray()                        maxCols = Math.Max(maxCols, cells.Length)                        Dim tableCells = New List(Of IList(Of ITextParagraph))()                        For Each cell In cells                            tableCells.Add(cell.GetParagraphs())                        Next                        table.Add(tableCells)                    Else                        selectRows(ec.Children)                    End If                End If            Next        End Sub        selectRows(e.Children)
        '' Show table:        Dim sourcePage = FindPage(e.StructElement)        If sourcePage Is Nothing Then            Throw New Exception("Unexpected: could not find the default page for the table.")        End If
        Dim tl = New TextLayout(72)
        ' Add table data to the text layout:        tl.Append($"{vbLf}Table on page {sourcePage.Index + 1} of the source document has {maxCols} column(s) and {table.Count} row(s)." & vbLf & "Data by row:", _tfHdr)        tl.AppendParagraphBreak()        Dim irow = 0        For Each row In table            Dim icol = 0            For Each cell In row                For Each para In cell                    tl.Append(para.GetText())                Next                If row.IndexOf(cell) <= row.Count Then                    tl.Append(vbTab)                Else                    tl.AppendLine()                End If                icol += 1            Next            irow += 1            tl.AppendLine()        Next        Return (tl, sourcePage)    End Function
    Private Function FindPage(se As StructElement) As Page        If se.DefaultPage IsNot Nothing Then            Return se.DefaultPage        End If        If se.HasChildren Then            For Each child In se.Children                Dim p = FindPage(child)                If p IsNot Nothing Then                    Return p                End If            Next        End If        Return Nothing    End FunctionEnd Class