Version 5.0.5:
    * Changed attribute type from cont char* to std::string for TAG_LIST
    * fixed OSX build scripts
    * removed unnecessary files
    * added description for console application
    * fixed bad_alloc exception handling
    * fixed header of html output
    * removed TAG_SPACE and TAG_TAB

Version 5.0.4:
    * Building with CMake for full version numbers fixed

Version 5.0.3:
    * Ordered and unordered lists API redesigned.
    * Fixed memory problem for OSX
    * PST parser API redesigned (tags)
    * Page close tag renamed

Version 5.0.2:
    * Fixed pdf parsing if CMap is unavailable
    * Fixed namespaces
    * tableStyleId tag support added to fix crash with some PPTX documents
    * Ignore all XML text nodes that contains only whitespaces to fix crashes with some ODF and OOXML documents
    * Add support for T tag in OOXML formats with whitespace preserve attribute

Version 5.0.1:
    * Empty table row problem fixed in plain text exporter
    * Fixed problem with char width calculation in pdf parser
    * Added exception if file doesn't exist
    * Added debug version for MSVC
    * Added sendTag method in Parser class
    * Added TAG_PAGE tag in pdf parser
    * Added section about tags to documentation
    * Added section about parser parameters to documentation
    * Added Info structure constructor
    * Changed extension from hpp to h for several files
    * Fixed location of temporary pst files to system defined temporary directory
    * Fixed problem with incorrect support for chars in html parser
    * Fixes problem with sending tags in pst parser

Version 5.0.0:
    * Introduced boost signals mechanism
    * Added examples for doctotext api
    * Updated documentation
    * Added next level api (Importer, Exporter, Transformer)
    * Refactoring
    * Added plugins mechanism for parser providers
    * Added support for hyperlinks in new api
    * Added support for msvc
    * 3rdparty directory removed (moved to build environment)
    * Made CMake script platform independent
    * Added support for lists in new api (ODF)
    * Fixed file validation
    * support table for Open Document Formats
    * Added support for hyperlinks in new api
    * Fixed support for odp files
    * Support attachments in eml in new api
    * Fixed xlsx parser (table formatting)
    * Fixed logs for encrypted files
    * Fixed comment support
    * Removed old api
    * Fixed formatting table - multiple line in cell

Version 4.6.1:
	* Fixed problem with unsupported exceptions in c api
	* Switch to DCMTK version 3.6.7 on windows version
	* Build 3rd party libraries with C++17 standard flags

Version 4.6.0:
	*  Updated tesseract library to version 5
	*  Implemented color inversion from white to black text

Version 4.5.0:
	* Added callbacks for style tags in odfooxml parser
	* Added HtmlWriter
	* Added tags with attributes for pst parser (mail-header, folder-header, attachment-header, mail-body, attachment-body)
	* Added filter mechanism for callbacks
	* Added set of standard filters (filterByFolderName, filterByAttachmentType, filterByMailMinCreationTime, filterByMailMaxCreationTime)
	* Fixed bug in pdf parser (missing check if object is valid string or hex string)
	* Added missing api headers: parser_builder.h parser_parameters.h, parser_provider.hpp, parser_wrapper.h, standard_filter.h, writer.h plain_text_writer.h html_writer.h

Version 4.4.2:
	* Added functions in old APIs to choose different "tessdata" directory for
	OCR parsing

Version 4.4.1:
	* Significant binary files size optimization

Version 4.4.0:
	* Significant OCR parser speed improvements
	* Filters support added to PST/OST parser
	* Possibility to cancel OCR parser process added to new API
	* Build environment updated to Debian Bullseye
	* Formatting style support for PST/OST and HTML parser in new API
	* Hardcoded OCR process timeout added to old API
	* CI environment separated from build environment
	* Multithreaded build and automatic tests

Version 4.3.0:
	* DICOM (Digital Imaging and Communications in Medicine) file format parser added
	* Small improvements in new API
	* Missing support for reading data buffer in OCR parser added
	* Unitialized formatting value error in HTML parser in new API fixed
	* Documentation and copyrights headers updated with new file formats

Version 4.2.2:
	* Remove CMake cache from binary packages
	* Add missing ChangeLog file to binary packages
	* Update documentation with new API and new file formats
	* Fix library file name after CMake introduction
	* Fix automatic tests after CTest introduction

Version 4.2.1:
	* Multithreading problems fixed: libmimetic, getpwuid, gmtime
	* CMake and CTest introduced instead of handwritten Makefiles
	* ODF_OOXML parser, image (OCR) parser, TXT parser and RTF parser added to new experimental API
	* PST/OST parser added to old API (with hardcoded limit for number of e-mails processed)
	* OCR parser support in old C API fixed
	* Automatic tests for old C API fixed and improved

Version 4.2.0:
	* Image parser (OCR) added with support for tiff, jpeg, bmp, png, pnm, jfif, jpg and webp file formats
	* Personal Storage Table (.pst) and Off-line Storage Table (.ost) parser added with support for Microsoft Outlook and Microsoft Exchange files
	* New experimental API added with callback support for quick handling of partial results and preserving the structure of documents

Version 4.1.2:
	* PDF parser improved significally (documents that did not open, word spacing, etc.)
	* PDF parser refactoring started.
	* Added support for EML and ODF_OOXML parsers in C API.
	* Fixed problem with duplicated content in EML parser.

Version 4.1.1:
	* Debug symbols separated to addditional files.

Version 4.1.0:
	* PlainTextExtractor can throw exceptions (doctotext::Exception)
	* Improved whitespace support in OpenDocument formats.
	* Debug symbols enabled by default.
	* Building Linux version process cleanups.
	* Building Win32 version process cleanups.
	* Memory error in EML parser fixed.

Version 4.0:
	* fixed bug in utf8 encoding for RT_TextBytesAtom record in PPT parser
	* Initial Open Document Flat XML Parser (fodt, fods, fodg, fodp)
	* Initial EML Parser (with ability to extract attachments)
	* Capabilities of C API has been expanded.
	* Better charset detection in HTML parser.
	* New TXT parser (Can be used to change encoding to UTF8)
	* ListStyle is a class now (not an enum).
	* Initial PDF parser.
	* Whole public interface of DocToText (PlainTextExctractor, Metadata, Link, Exception, FormattingStyle) is available under doctotext namespace.
	* DocToText supports exceptions now (Exception class).
	* Reorganization of url handling. PlainTextExtractor can now return list of links. Supported parsers: HTML/EML/ODF_OOXML/ODFXML.
	* PlainTextExtractor allows now for parsing files from memory buffer.
	* Independence from glib, libgsf, gettext. Pthreads are used instead of gthreads, build-in OLEReader is used instead of libgsf.
	* New iWork parser.
	* New XLSB parser.
	* Extracting number of pages from ODG files fixed.
	* ODG files added to automatic tests.
	* Support for Object Linking and Embedding (OLE) in ODF formats added.
	* Managing libxml2 (initialization and cleanup) can be disabled.
	* Thread-safety fixed in ODF, OOXML and DOC parsers.
	* Better handling of fields in DOC and DOT files.
	* Better handling of headings in ODF documents.
	* Fixes for x86_64 architecture.
	* Improved stability in multithreaded environments.
	* Embedded XLS Workbooks in DOC files supported without creating temporary files.
	* Cleanups in ODF and OOXML parsers.
	* Memory consumption of ODF and OOXML parsers significantly reduced.
	* Better handling of fields in DOCX files.
	* Fixed crash in RTF parser for invalid files.
	* Fixes in XLS parser.
	* Initial port to win64 architecture.
	* Function enter and exit tracking feature useful for debugging.

Version 0.14.0:
	* Initial HTML format support.
	* Initial implementation of metadata extractor (author, creation time, last modified by, last modification time, page count, word count
	* Font commands in XLSX header or footer do not corrupt output any more.
	* Estimate not existing meta values using different techniques.
	* Initial implementation of annotations support (ODT, DOC, DOCX, RTF formats).
	* Download and use precompiled wv2 binaries
	* Initial API documentation (Doxygen)
	* Fallback to other parsers if specified parser fails
	* Fixes and improvements in XLS parser.
	* C++ API extended to allow communicating without STL objects (to mix two STL implementations for example).
	* Line break handling in ODF and OOXML formats fixed.

Version 0.13.0
	* Initial Mac OS X port
	* Initial implementation of shared library
	* C language API added to allow using C++ library from C and other languages
	* Universal binary (i386, x86_64) for Mac OS X
	* Static linking of 3rdparty libraries option for Mac OS X
	* Cleanups
	* First parser selection is according to file extension
	* First parser used can be choosen via command line
	* Embedded XLS Workbooks support in DOC files
	* Table cells encoding problems in DOC files fixed.
	* Case insensitive file extension matching.
	* Do not try other parsers if parser selected by file extension fails.
	* New XLS parser.
	* Small fix in XLS numbers formatting.
	* Small fix in parsing XLS shared string table.
	* Locating files in zip archives (ODF, OOXML) optimized.
	* Initial PPT format support. ODP, PPT, PPTX files added to automatic tests.
	* Fixes in DOC and XLS parsers.
	* Verbose logging can be turned on and off.
	* Logs can be redirected to other C++ stream than standard cerr.
	* Redesigned C++ API.
	* Debug and release versions separated.
	* End of paragraph special symbols support fixed.
	* Download and use more precompiled libraries
	* Small and big compilation fixes
	* Headers and footers support in DOC files added

Version 0.12.0:
	* Formatting tables optimized
	* Build-in libxml2 library upgraded to version 2.7.7
	* Initial xls format support
	* Better default table formatting
	* Support for DOC format moved to DOCParser class
	* RTFParser API uses standard string class only
	* Cleanups
	* Error message is not displayed when checking for ODF or OOXML format fails
	* Table parsing in DOC files improved
	* Link parsing in DOC files improved
	* Cleanups
	* Copyright headers updated
	* Packages in bzip2 format instead of gzip
	* ODS (OpenOffice Calc) files added to automatic tests
	* VERSION file added to binary packages
	* Download and use precompiled libiconv binaries
	* Encoding problems in XLS format support on win32 fixed
	* Use MAKE variable executing make recursively

Version 0.11.0:
	* zip archives (ODF, OOXML) support moved to DocToTextUnzip class
	* --unzip-cmd option implemented to use specified command to unzip files from archives (ODF, OOXML) instead of build-in decompression library
	* --unzip-cmd fixed on win32 - replacing slashes with backslashes in filename
	* --fix-xml option implemented to try to fix corrupted xml files (ODF, OOXML) using custom recursive descent parser before processing
	* some ODF and OOXML formatting issues fixed
	* UTF-8 support in corrupted xml files (ODF, OOXML) fixing routines
	* Makefile dependencies fixed
	* --strip-xml option implemented to strip xml tags instead of parsing them (ODF, OOXML)
	* Entities support in corrupted xml files (ODF, OOXML) fixing routines
	* max tag depth security limit of libxml supported in corrupted xml files (ODF, OOXML) fixing routines
	* a lot of compilation warnings fixed
	* small fix for encoding support in RTF parser
	* --strip-xml option removes duplicated attributes
	* ChangeLog file added to binary releases
	* copyright headers updated

Version 0.10.0:
	* Command line arguments added to change tables, lists and links formatting (for odt files only)
	* glib upgraded to 2.14.5
	* initial OOXML formats support
	* Parsing tables in DOC fixed
	* support for mixed character encodings RTF files created using MS WordPad or MS Word (thanks to John Estrada)
	* ioapi of unzip library changed from winapi to fopen api in win32 version - winapi io in unzip library caused some problems, for example xlsx file could not be open when is open in MS Word
	* allocating memory using expotential grow in UString - big performance boost in rtf parser
	* copyright headers updated


Version 0.9.0:
	* patch for gettext fixing duplicate case value errors
	* glib upgraded to 2.12.11 - duplicate case value errors
	* wstring changed to UString in rtf parser - problems on mingw
	* initial support of charset in rtf parser
	* initial odt format support
	* performing automatic tests after the source code is compiled
	* memory error fixed - unzReadCurrentFile() does not put NULL at the end of the buffer
	* counting function added
	* number of characters read from file was changed to corect value


Version 0.2.0:
	* rtf format support

Version 0.1.0:
	* initial implementation
	* output in utf-8 encoding
	* help, copyright headers, README
	* paragraphs support
