Skip to content

FileEncoding.md

Latest commit

 

History

History
85 lines (63 loc) · 2.4 KB

FileEncoding.md

File metadata and controls

85 lines (63 loc) · 2.4 KB
Raw

File Encoding

The image files (.ICN, .CMP) are compressed using Run-Length Encoding. There are two different formats of run-length encoding used in Crescent Hawks Inception.

When open an image file the first two bytes are the file size (minus those two bytes). The 3rd byte informs the code which of the follow run-length encoding algorithms to use.

3rd Byte Format Notes
01 Format01 Linear read
02 Format02 160 Byte Object loop read

Format01

Run length encoding algorithm:

Start:

Read a Byte, this is now the Index Byte.

It a value is negative it is flipped to the positive (-2 to 2, -8 to 8)

If Index Byte is non-Zero

  • Index Byte becomes Run-Length Counter
  • Next byte becomes output value.
  • Write to output.
  • Repeat until Run-Length Counter is Zero.
  • Goto Start;

If Index Byte is Zero

  • Next byte becomes Run-Length Counter
  • Next byte becomes output value.
  • Write to output same value for Run-Length Counter times.
  • Goto Start;

Format02

The run-length encoding is identical to Format01. However the file is read as a sequence of 160 byte object s. First byte is extracted per object.

160 Byte Object 160 Byte Object
01 .. .. .. D2 .. ..

Two objects per x line. (X line is 320 pixels)

160 Byte Object 160 Byte Object
01 .. .. .. D2 .. ..
03 .. .. .. DE .. ..
E3 .. .. .. F2 .. ..
D1 .. .. .. A3 .. ..
41 .. .. .. DD .. ..
200 lines total 200 lines total

200 Y lines to the end of the file. (320 x 200 file size)

160 Byte Object 160 Byte Object
01 D5 .. .. D2 4B ..

Then the write is reset to the second byte per object.

const short YAxisSize = 200;
const short XAxisObjectNextByteOffset = 31999;

...
...
...

if (YAxisRemaining == 0)
    {
      YAxisRemaining = YAxisSize;
      log.Debug($"Buffer_Index: {DecodedBuffer_Index}");
      DecodedBuffer_Index -= XAxisObjectNextByteOffset; //Move back to beginning but second byte
      log.Debug($"After RunLength Reset Buffer_Index: {DecodedBuffer_Index}");
    }

This behaviour is controller by the -31999 (XAxisObjectNextByteOffset) and 200 (YAxisSize) offsets seen in the Format02 algorithm.

This results in the read slowly progressing through the 160 byte object reading byte by byte until the 160 byte and then once the last object is read the output is complete.