User Manual For htetris Version 1.0 Henrik Sandin 1999

Implementation details

The bricks are represented with matrix structures in the game as well as the editor. An element in such a matrix represents one square in a brick. A matrix is never larger than the actual rectangle which constitutes the size of a brick which is measured in number of squares wide and high respectively.
In the editor, a brick-matrix consists of ones and zeros where a one represents a colored square and a zero represents an uncolored (black, since the backgroud of the edit pane is black) square. A string representation of such matrices is used when they are saved to file.
When a brick is used in the game, the brick-matrix elements plays a different role. The area where the bricks fall also has a matrix representation where every element, just like the brick matrices in the editor context contains one or zero, one representing a position where a brick-square is permanently stuck and zero representing a position that is "free".

When a brick is shown and falling, its matrix conceptually resides on top of the background matrix. At all times a brick-matrix keeps updated information on where a particular square is as well as if that square is colored or not. A brick-matrix element contains a record which in turn contains information about the current row and column coordinates of the background matrix and whether that square is colored or should be drawn transparent (not drawn). When a brick changes position (falls one step or is moved to the left or to the right), its matrix is updated accordingly.
When a brick is considered to be stuck somewhere, the background matrix is updated by looking at the current information in the current brick-matrix. The determining of whether a brick is stuck or can/can not be moved, is done by looking at the surrounding elements relative to a brick's current position in the background matrix. An element in a brick matrix which is market as "colored" can never be located "on top" of an element in the background matrix which contains a one.

The actual drawing and erasing of the bricks is based on the background matrix indeces where a brick currently resides. A brick square has a constant width and height, so it is only a matter of multiplying that constant number of pixels by the matrix row or column number to determining where the brick image should be drawn.

Graphically, a brick is a rectangular image(string) which is drawn using the procedure DrawImage() which support transparency in drawing. This is useful since bricks are shaped as they are.
Erasing of a brick is done by a series of EraseArea() calls each of which is erasing one square of the brick. This is a little bit slow but is necessary to prevent already stuck bricks from being overwritten. This might happen if a falling brick is erased by clearing one single rectangle covering the whole brick when it is close enough to already stuck ones.