WPTools - the future

    • Offizieller Beitrag

    Although V4.x will be supported with at least 1 more releases the work on WPTools V5 has begun.

    WPTools 5 is designed to work with the latest compilers. To make this possible some changes to the RTF Engine will be necessary, most of the pointer arithmetic has to be replaced. The modified concept introduces new options which are unthinkable with the current concept so WPTools 5 should be the base for another 7 successful word processing years :-)

    We try to make the changes in a way that your code should run as it used to. Code which access pointers (WPRichText.FirstPar, CPAttr ...) will require changes. Sometimes only the ^sign has to be removed, in case the code lists the lines of a paragraph (lin := par.line; while lin<>nil do ... and of course pc and pa arithmetic) this code must be changed. But also here, the changes shall be marginal and be rewared by an easier to understand and extremely stable code. Undocumented procedures and variables of the 'RTFEngine' (WPRichText.Memo) will be changed or replaced to keep WPTools up to date and improve the performace. The way paragraph border attributes are stored will be changed, too.

    Some components will not be included in future. This are the almost never used components TWPWallPaper, WPTAltStatusBar and TWPContainer + TWPResLabel.
    The new version will alway use the Action classes so the components TWPComboBox, TWPToolButton and TWPToolControl which were mainly created to create GUI without the availability of action classes will also have to go.

    WPTools 5 will support Delphi 5 or higher and C++Builder 5 & 6.
    It should be ready in the first half of 2004. (Delphi 4 does not provide the necessary compiler features.)

    ...comments are welcome...

    Julian Ziersch
    wpCubed GmbH

  • Please take this unique opportunity to make the properties of WPTools really object-oriented.

    [longer post edited]

  • My opinion is that a mix of various programming techniques have to be considered in order to develop a robust-and-effective system.
    Object-oriented code is not the bible: it has many plus but slows down the overall performance and usually it is not the best way to build kernel software with speed requirements. I hope the new version will still have the focus on efficiency.
    Have a good job, Julian!