TextExtractor.h

Go to the documentation of this file.

//---------------------------------------------------------------------------------------
// Copyright (c) 2001-2008 by PDFTron Systems Inc. All Rights Reserved.
// Consult legal.txt regarding legal and license information.
//---------------------------------------------------------------------------------------

#ifndef   HPP_CPPTextExtractor
#define   HPP_CPPTextExtractor

#include <PDF/Page.h>
#include <PDF/Rect.h>
#include <Common/UString.h>
#include <C/PDF/TRN_TextExtractor.h>

namespace pdftron { 
        namespace PDF {

class TextExtractor 
{
public:

         TextExtractor();
         ~TextExtractor();

        enum ProcessingFlags
        {
                // Disables expanding of ligatures using a predefined mapping. 
                // Default ligatures are: fi, ff, fl, ffi, ffl, ch, cl, ct, ll, 
                // ss, fs, st, oe, OE. 
                e_no_ligature_exp = 1, 

                // Disables removing duplicated text that is frequently used to 
                // achieve visual effects of drop shadow and fake bold. 
                e_no_dup_remove = 2,

                // Treat punctuation (e.g. full stop, comma, semicolon, etc.) as 
                // word break characters. 
                e_punct_break = 4,

                // Enables removal of text that is obscured by images or 
                // rectangles. Since this option has small performance penalty 
                // on performance of text extraction, by default it is not 
                // enabled.
                e_remove_hidden_text = 8, 

                // Enables removing text that uses rendering mode 3 (i.e. invisible text).
                // Invisible text is usually used in 'PDF Searchable Images' (i.e. scanned 
                // pages with a corresponding OCR text). As a result, invisible text 
                // will be extracted by default.
                e_no_invisible_text = 16
        };

         void Begin(Page page, const Rect* clip_ptr = 0, UInt32 flags = 0);

         int GetWordCount();

         void GetAsText(UString& out_str, bool dehyphen = true);

        enum XMLOutputFlags 
        {
                // Output words as XML elements instead of inline text.
                e_words_as_elements = 1, 

                // Include bounding box information for each XML element. 
                // The bounding box information will be stored as 'bbox' attribute.
                e_output_bbox = 2, 

                // Include font and styling information.
                e_output_style_info = 4
        };

         void GetAsXML(UString& out_xml, UInt32 xml_output_flags = 0);

        class Style 
        {
        public:

                SDF::Obj GetFont();

                UString GetFontName();

                double GetFontSize();

                int GetWeight();

                bool IsItalic();

                bool IsSerif();

                void GetColor(UInt8 rgb[3]);

                bool operator== (const Style& s);
                bool operator!= (const Style& s);

                Style();

                Style(const Style& s);
                Style(TRN_TextExtractorStyle impl);
                TRN_TextExtractorStyle mp_style;
        };

        class Word 
        {
        public:
                 int GetNumGlyphs();

                void GetBBox(double out_bbox[4]);

                void GetQuad(double out_quad[8]);

                void GetGlyphQuad(int glyph_idx, double out_quad[8]);

                Style GetCharStyle(int char_idx);

                Style GetStyle();

                int GetStringLen();

                const Unicode* GetString();

                Word GetNextWord();

                int GetCurrentNum();

                bool IsValid();

                bool operator== (const Word&);
                bool operator!= (const Word&);
                Word();

                Word(TRN_TextExtractorWord impl);
                TRN_TextExtractorWord mp_word;
        };

        class Line {
        public: 

                int GetNumWords();

                bool IsSimpleLine();

                const double* GetBBox();

                void GetQuad(double out_quad[8]);

                Word GetFirstWord();

                Word GetWord(int word_idx);

                Line GetNextLine();

                int GetCurrentNum();

                Style GetStyle();

                int GetParagraphID();

                int GetFlowID();

                bool EndsWithHyphen();

                bool IsValid();

                bool operator== (const Line&);
                bool operator!= (const Line&);
                Line();

                Line(TRN_TextExtractorLine impl);
                TRN_TextExtractorLine mp_line;
        };

         int GetNumLines();


         Line GetFirstLine();


private:
        TRN_TextExtractor mp_extractor;
};



#include <Impl/TextExtractor.inl>

        };      // namespace PDF
};      // namespace pdftron

#endif

---

© 2002-2008 PDFTron Systems Inc.