Full-Text Search

What is Full-Text Search?

PDF documents consist of many parts with the textual content often being the most important. Text search, and text extraction, are two common tasks required by both PDF developers as well the end users of PDF software. In order to search for text in a document, the text content must first be extracted from the PDF which can be difficult without our SDK!

Using an index allows a document to be searched quickly as the text extraction phase only needs to be completed once. This allows the search operation to be scaled up to allow searching of large sets of documents.

Simplifying the search process with PDF SDK

PDF SDK offers the fastest text search technology in the market. The biggest challenge with text extraction, or text searching, is that the PDF format allows any character to be displayed at any position on the page, in any size and with any rotation. Characters on the page are not required to appear in any certain order, either within a page, on a line, or even within a word!

Words can be split (for example, with a hyphen at the end of a line) and certain characters can be combined into one (for example, the fi character instead of separate f and i). When searching for a phrase, the word of the phrase could be on different lines.

The visual representation of the characters (known as glyphs) can be sourced from any number of fonts stored within the PDF. In addition, many different font formats are allowed, eg. TrueType or PostScript. Foxit PDF SDK has already accounted for all of this, and, out of the box, helps you set up foolproof full text search capabilities for your application.

Trusted by some of the biggest companies in the world

Sample Code

Full-Text Search

public void doPdfToText() {
  int indexPdf = mFilePath.lastIndexOf(".");
  int indexSep = mFilePath.lastIndexOf("/");

  String outputFilePath = "output.txt";
  String strText = "";

  PDFDoc doc = Common.loadPDFDoc(mContext, mFilePath, null);
  PDFPage page = null;
  try {
      int pageCount = doc.getPageCount();

      // Traverse pages and get the text string.
      for (int i = 0; i < pageCount; i++) {
          page = Common.loadPage(mContext, doc, i, PDFPage.e_ParsePageNormal);
          TextPage textSelect = new TextPage(page, TextPage.e_ParseTextNormal);
          strText += textSelect.getChars(0, textSelect.getCharCount()) + "\r\n";
          page.delete();
      }
  } catch (PDFException e) {
      // PDF to text error
      return;
  } finally {
      Common.releaseDoc(mContext, doc);
  }

  // Output the text string to the text file
  FileWriter fileWriter = null;
  try {
      File fileTxt = new File(outputFilePath);
      fileWriter = new FileWriter(fileTxt);
      fileWriter.write(strText);
      fileWriter.flush();
      fileWriter.close();
  } catch (IOException e) {
      // Error
      return;
  }
}