string GetPdfPageContent(PDFDoc pdfDoc, int pageNumber)
{
List <PdfString> matchFuncLines = new List <PdfString>();
Page page = pdfDoc.GetPage(pageNumber);
if (page == null)
{
return(null);
}
TextExtractor txt = new TextExtractor();
txt.Begin(page);
TextExtractor.Line line;
TextExtractor.Word word;
string lineString = null;
StringBuilder stringBuilder = new StringBuilder();
for (line = txt.GetFirstLine(); line.IsValid(); line = line.GetNextLine())
{
if (line.GetNumWords() == 0)
{
continue;
}
lineString = null;
for (word = line.GetFirstWord(); word.IsValid(); word = word.GetNextWord())
{
int sz = word.GetStringLen();
if (sz == 0)
{
continue;
}
lineString += word.GetString();
}
if (string.IsNullOrEmpty(lineString))
{
continue;
}
stringBuilder.Append(lineString);
}
txt.Dispose();
return(stringBuilder.ToString());
}
public void ReadAdvanced(string input_path)
{
PDFNet.Initialize();
try
{
PDFDoc doc = new PDFDoc(input_path);
doc.InitSecurityHandler();
Page page = doc.GetPage(1);
if (page == null)
{
ConsoleLog += "Page not found.";
return;
}
TextExtractor txt = new TextExtractor();
txt.Begin(page); // Read the page.
// Other options you may want to consider...
// txt.Begin(page, null, TextExtractor.ProcessingFlags.e_no_dup_remove);
// txt.Begin(page, null, TextExtractor.ProcessingFlags.e_remove_hidden_text);
// ...
// Example 1. Get all text on the page in a single string.
// Words will be separated with space or new line characters.
if (example1_basic)
{
// Get the word count.
ConsoleLog += "Word Count: {0}" + txt.GetWordCount();
ConsoleLog += "\n\n- GetAsText --------------------------\n{0}" + txt.GetAsText();
ConsoleLog += "-----------------------------------------------------------";
}
// Example 2. Get XML logical structure for the page.
if (example2_xml)
{
String text = txt.GetAsXML(TextExtractor.XMLOutputFlags.e_words_as_elements | TextExtractor.XMLOutputFlags.e_output_bbox | TextExtractor.XMLOutputFlags.e_output_style_info);
ConsoleLog += "\n\n- GetAsXML --------------------------\n{0}" + text;
ConsoleLog += "-----------------------------------------------------------";
}
// Example 3. Extract words one by one.
if (example3_wordlist)
{
TextExtractor.Word word;
for (TextExtractor.Line line = txt.GetFirstLine(); line.IsValid(); line = line.GetNextLine())
{
for (word = line.GetFirstWord(); word.IsValid(); word = word.GetNextWord())
{
ConsoleLog += word.GetString();
}
}
ConsoleLog += "-----------------------------------------------------------";
}
// Example 3. A more advanced text extraction example.
// The output is XML structure containing paragraphs, lines, words,
// as well as style and positioning information.
if (example4_advanced)
{
Rect bbox;
int cur_flow_id = -1, cur_para_id = -1;
TextExtractor.Line line;
TextExtractor.Word word;
TextExtractor.Style s, line_style;
// For each line on the page...
for (line = txt.GetFirstLine(); line.IsValid(); line = line.GetNextLine())
{
if (line.GetNumWords() == 0)
{
continue;
}
if (cur_flow_id != line.GetFlowID())
{
if (cur_flow_id != -1)
{
if (cur_para_id != -1)
{
cur_para_id = -1;
ConsoleLog += "</Para>";
}
ConsoleLog += "</Flow>";
}
cur_flow_id = line.GetFlowID();
ConsoleLog += "<Flow id=\"{0}\">" + cur_flow_id;
}
if (cur_para_id != line.GetParagraphID())
{
if (cur_para_id != -1)
{
ConsoleLog += "</Para>";
}
cur_para_id = line.GetParagraphID();
ConsoleLog += "<Para id=\"{0}\">" + cur_para_id;
}
bbox = line.GetBBox();
line_style = line.GetStyle();
Console.Write("<Line box=\"" + bbox.y1 + "," + bbox.y2 + "," + bbox.x1 + "," + bbox.x2 + ">");
PrintStyle(line_style);
ConsoleLog += "";
// For each word in the line...
for (word = line.GetFirstWord(); word.IsValid(); word = word.GetNextWord())
{
// Output the bounding box for the word.
bbox = word.GetBBox();
ConsoleLog += "<Word box=\"{0}, {1}, {2}, {3}\"" + bbox.x1 + bbox.y1 + bbox.x2 + bbox.y2;
int sz = word.GetStringLen();
if (sz == 0)
{
continue;
}
// If the word style is different from the parent style, output the new style.
s = word.GetStyle();
if (s != line_style)
{
PrintStyle(s);
}
ConsoleLog += ">\n" + word.GetString();
ConsoleLog += "</Word>";
}
ConsoleLog += "</Line>";
}
if (cur_flow_id != -1)
{
if (cur_para_id != -1)
{
cur_para_id = -1;
ConsoleLog += "</Para>";
}
ConsoleLog += "</Flow>";
}
}
// Note: Calling Dispose() on TextExtractor when it is not anymore in use can result in increased performance and lower memory consumption.
txt.Dispose();
doc.Close();
ConsoleLog += "Done.";
}
catch (PDFNetException e)
{
ConsoleLog += e.Message;
}
// Sample code showing how to use low-level text extraction APIs.
if (example5_low_level)
{
try
{
LowLevelTextExtractUtils util = new LowLevelTextExtractUtils();
PDFDoc doc = new PDFDoc(input_path);
doc.InitSecurityHandler();
// Example 1. Extract all text content from the document
ElementReader reader = new ElementReader();
PageIterator itr = doc.GetPageIterator();
//for (; itr.HasNext(); itr.Next()) // Read every page
{
reader.Begin(itr.Current());
LowLevelTextExtractUtils u = new LowLevelTextExtractUtils();
u.DumpAllText(reader);
ConsoleLog += u.ConsoleLog;
reader.End();
}
// Example 2. Extract text based on the selection rectangle.
ConsoleLog += "----------------------------------------------------";
ConsoleLog += "Extract text based on the selection rectangle.";
ConsoleLog += "----------------------------------------------------";
Page first_page = doc.GetPage(1);
string field1 = util.ReadTextFromRect(first_page, new Rect(27, 392, 563, 534), reader);
string field2 = util.ReadTextFromRect(first_page, new Rect(28, 551, 106, 623), reader);
string field3 = util.ReadTextFromRect(first_page, new Rect(208, 550, 387, 621), reader);
ConsoleLog += "Field 1: {0}" + field1;
ConsoleLog += "Field 2: {0}" + field2;
ConsoleLog += "Field 3: {0}" + field3;
// ...
reader.Dispose();
doc.Close();
ConsoleLog += "Done.";
}
catch (PDFNetException e)
{
ConsoleLog += e.Message;
}
}
PDFNet.Terminate();
}
static void Main(string[] args)
{
PDFNet.Initialize();
// Relative path to the folder containing test files.
string input_path = "../../TestFiles/";
bool example1_basic = false;
bool example2_xml = false;
bool example3_wordlist = false;
bool example4_advanced = true;
bool example5_low_level = false;
// Sample code showing how to use high-level text extraction APIs.
try
{
using (PDFDoc doc = new PDFDoc(input_path + "newsletter.pdf"))
{
doc.InitSecurityHandler();
Page page = doc.GetPage(1);
if (page == null)
{
Console.WriteLine("Page not found.");
return;
}
using (TextExtractor txt = new TextExtractor())
{
txt.Begin(page); // Read the page.
// Other options you may want to consider...
// txt.Begin(page, null, TextExtractor.ProcessingFlags.e_no_dup_remove);
// txt.Begin(page, null, TextExtractor.ProcessingFlags.e_remove_hidden_text);
// ...
// Example 1. Get all text on the page in a single string.
// Words will be separated with space or new line characters.
if (example1_basic)
{
// Get the word count.
Console.WriteLine("Word Count: {0}", txt.GetWordCount());
Console.WriteLine("\n\n- GetAsText --------------------------\n{0}", txt.GetAsText());
Console.WriteLine("-----------------------------------------------------------");
}
// Example 2. Get XML logical structure for the page.
if (example2_xml)
{
String text = txt.GetAsXML(TextExtractor.XMLOutputFlags.e_words_as_elements | TextExtractor.XMLOutputFlags.e_output_bbox | TextExtractor.XMLOutputFlags.e_output_style_info);
Console.WriteLine("\n\n- GetAsXML --------------------------\n{0}", text);
Console.WriteLine("-----------------------------------------------------------");
}
// Example 3. Extract words one by one.
if (example3_wordlist)
{
TextExtractor.Word word;
for (TextExtractor.Line line = txt.GetFirstLine(); line.IsValid(); line = line.GetNextLine())
{
for (word = line.GetFirstWord(); word.IsValid(); word = word.GetNextWord())
{
Console.WriteLine(word.GetString());
}
}
Console.WriteLine("-----------------------------------------------------------");
}
// Example 3. A more advanced text extraction example.
// The output is XML structure containing paragraphs, lines, words,
// as well as style and positioning information.
if (example4_advanced)
{
Rect bbox;
int cur_flow_id = -1, cur_para_id = -1;
TextExtractor.Line line;
TextExtractor.Word word;
TextExtractor.Style s, line_style;
Console.WriteLine("<PDFText>");
// For each line on the page...
for (line = txt.GetFirstLine(); line.IsValid(); line = line.GetNextLine())
{
if (line.GetNumWords() == 0)
{
continue;
}
if (cur_flow_id != line.GetFlowID())
{
if (cur_flow_id != -1)
{
if (cur_para_id != -1)
{
cur_para_id = -1;
Console.WriteLine("</Para>");
}
Console.WriteLine("</Flow>");
}
cur_flow_id = line.GetFlowID();
Console.WriteLine("<Flow id=\"{0}\">", cur_flow_id);
}
if (cur_para_id != line.GetParagraphID())
{
if (cur_para_id != -1)
{
Console.WriteLine("</Para>");
}
cur_para_id = line.GetParagraphID();
Console.WriteLine("<Para id=\"{0}\">", cur_para_id);
}
bbox = line.GetBBox();
line_style = line.GetStyle();
Console.Write("<Line box=\"{0}, {1}, {2}, {3}\"", bbox.x1.ToString("0.00"), bbox.y1.ToString("0.00"), bbox.x2.ToString("0.00"), bbox.y2.ToString("0.00"));
PrintStyle(line_style);
Console.Write(" cur_num=\"" + line.GetCurrentNum() + "\"" + ">\n");
// For each word in the line...
for (word = line.GetFirstWord(); word.IsValid(); word = word.GetNextWord())
{
// Output the bounding box for the word.
bbox = word.GetBBox();
Console.Write("<Word box=\"{0}, {1}, {2}, {3}\"", bbox.x1.ToString("0.00"), bbox.y1.ToString("0.00"), bbox.x2.ToString("0.00"), bbox.y2.ToString("0.00"));
Console.Write(" cur_num=\"" + word.GetCurrentNum() + "\"");
int sz = word.GetStringLen();
if (sz == 0)
{
continue;
}
// If the word style is different from the parent style, output the new style.
s = word.GetStyle();
if (s != line_style)
{
PrintStyle(s);
}
Console.Write(">{0}", word.GetString());
Console.WriteLine("</Word>");
}
Console.WriteLine("</Line>");
}
if (cur_flow_id != -1)
{
if (cur_para_id != -1)
{
cur_para_id = -1;
Console.WriteLine("</Para>");
}
Console.WriteLine("</Flow>");
}
}
}
Console.WriteLine("</PDFText>");
}
}
catch (PDFNetException e)
{
Console.WriteLine(e.Message);
}
// Sample code showing how to use low-level text extraction APIs.
if (example5_low_level)
{
try
{
LowLevelTextExtractUtils util = new LowLevelTextExtractUtils();
using (PDFDoc doc = new PDFDoc(input_path + "newsletter.pdf"))
{
doc.InitSecurityHandler();
// Example 1. Extract all text content from the document
using (ElementReader reader = new ElementReader())
{
PageIterator itr = doc.GetPageIterator();
//for (; itr.HasNext(); itr.Next()) // Read every page
{
reader.Begin(itr.Current());
LowLevelTextExtractUtils.DumpAllText(reader);
reader.End();
}
// Example 2. Extract text based on the selection rectangle.
Console.WriteLine("----------------------------------------------------");
Console.WriteLine("Extract text based on the selection rectangle.");
Console.WriteLine("----------------------------------------------------");
Page first_page = doc.GetPage(1);
string field1 = util.ReadTextFromRect(first_page, new Rect(27, 392, 563, 534), reader);
string field2 = util.ReadTextFromRect(first_page, new Rect(28, 551, 106, 623), reader);
string field3 = util.ReadTextFromRect(first_page, new Rect(208, 550, 387, 621), reader);
Console.WriteLine("Field 1: {0}", field1);
Console.WriteLine("Field 2: {0}", field2);
Console.WriteLine("Field 3: {0}", field3);
// ...
Console.WriteLine("Done.");
}
}
}
catch (PDFNetException e)
{
Console.WriteLine(e.Message);
}
}
}