Comments on: About rle_unpack in libevecache http://yannramin.com/2009/12/28/about-rle_unpack-in-libevecache/ Software and life Sat, 10 Jul 2010 10:45:58 +0000 hourly 1 http://wordpress.org/?v=3.0 By: Cadde http://yannramin.com/2009/12/28/about-rle_unpack-in-libevecache/comment-page-1/#comment-1453 Cadde Tue, 16 Mar 2010 15:17:22 +0000 http://yannramin.com/?p=332#comment-1453 Sorry for double comment... But of course this stripped the spacetastic formatting i had used so it might not make any sense at all... Try and copy+pasting it into a monospace font editor and adding the spaces back to it. Sorry for double comment…

But of course this stripped the spacetastic formatting i had used so it might not make any sense at all… Try and copy+pasting it into a monospace font editor and adding the spaces back to it.

]]> By: Cadde http://yannramin.com/2009/12/28/about-rle_unpack-in-libevecache/comment-page-1/#comment-1452 Cadde Tue, 16 Mar 2010 15:15:11 +0000 http://yannramin.com/?p=332#comment-1452 I have been working on a VB.NET (Don't laugh dammit) version of libevecache and because it STILL took me hours to figure this one out i think it's only valid to post my methodology for solving the problem. This is a copy+paste of the commented code in my VB project. ' Observe this methodology for example... ' ' BZero & BLen 0100 1000 0001 0111 0001 0100 0011 ' TZero & TLen 1000 1111 0000 1010 1010 1000 1011 ' Operation Bytes 48 8F 10 7A 1A 48 3B ' Data Bytes CF 1413 04 2440 C0E6 7F39 66B3 CA01 4617 EA58 0A 01 109F 9303 B496 98 9AD3 C901 1579 FF7F 0E ' Result 00CF 1413 0400 0000 0000 0000 0000 2440 C0E6 7F39 66B3 CA01 4617 EA58 0A00 0000 0100 0000 109F 9303 B496 9800 9AD3 C901 0000 0000 1579 FF7F 0E ' ZERO's added 1 9 3 3 1 4 ' Data bytes copied 4 15 1 7 4 5 ' ' TZero, TLen deals with the first 4 bits of the operation byte. ' BZero, BLen deals with the last 4 bits of the operation byte. ' ' TZero is the 4th bit and if it's set it means it should ADD X number of &h00 bytes to the result. ' TLen is the number of Zero's (+1) to add if the TZero flag is set. ' If TZero isn't set then instead of adding null bytes it takes the following (8 - BLen) bytes and adds to the result. ' ' This is repeated with BZero and BLen respectively. ' ' So, for the first byte in the compressed row (&h48) TZero is set and we add TLen(0) + 1 = 1, null bytes to the result. ' BZero is not set so we copy 8 - BLen(4) = 4, bytes from the compressed data row. ' ' We now read the next operand byte and here TZero is set so we insert TLen(7) + 1 = 8, null bytes to the result. ' BZero is also set so we insert another BLen(0) + 1 = 1, null bytes to the result. ' ' Since the last operation only inserted null bytes (9 of them in total) we simply read the next operand. ' TZero isn't set so we copy 8 - TLen(0) = 8, bytes from the compressed row. ' BZero isn't set so we copy 8 - BLen(1) = 7, bytes from the compressed row. ' This means we copied (8 + 7) 15 bytes from the compressed row. ' ' That is half the compressed row done, i leave the rest up to you to grasp. Hope this helps someone. I have been working on a VB.NET (Don’t laugh dammit) version of libevecache and because it STILL took me hours to figure this one out i think it’s only valid to post my methodology for solving the problem.

This is a copy+paste of the commented code in my VB project.

‘ Observe this methodology for example…
‘
‘ BZero & BLen 0100 1000 0001 0111 0001 0100 0011
‘ TZero & TLen 1000 1111 0000 1010 1010 1000 1011
‘ Operation Bytes 48 8F 10 7A 1A 48 3B
‘ Data Bytes CF 1413 04 2440 C0E6 7F39 66B3 CA01 4617 EA58 0A 01 109F 9303 B496 98 9AD3 C901 1579 FF7F 0E
‘ Result 00CF 1413 0400 0000 0000 0000 0000 2440 C0E6 7F39 66B3 CA01 4617 EA58 0A00 0000 0100 0000 109F 9303 B496 9800 9AD3 C901 0000 0000 1579 FF7F 0E
‘ ZERO’s added 1 9 3 3 1 4
‘ Data bytes copied 4 15 1 7 4 5
‘
‘ TZero, TLen deals with the first 4 bits of the operation byte.
‘ BZero, BLen deals with the last 4 bits of the operation byte.
‘
‘ TZero is the 4th bit and if it’s set it means it should ADD X number of &h00 bytes to the result.
‘ TLen is the number of Zero’s (+1) to add if the TZero flag is set.
‘ If TZero isn’t set then instead of adding null bytes it takes the following (8 – BLen) bytes and adds to the result.
‘
‘ This is repeated with BZero and BLen respectively.
‘
‘ So, for the first byte in the compressed row (&h48) TZero is set and we add TLen(0) + 1 = 1, null bytes to the result.
‘ BZero is not set so we copy 8 – BLen(4) = 4, bytes from the compressed data row.
‘
‘ We now read the next operand byte and here TZero is set so we insert TLen(7) + 1 = 8, null bytes to the result.
‘ BZero is also set so we insert another BLen(0) + 1 = 1, null bytes to the result.
‘
‘ Since the last operation only inserted null bytes (9 of them in total) we simply read the next operand.
‘ TZero isn’t set so we copy 8 – TLen(0) = 8, bytes from the compressed row.
‘ BZero isn’t set so we copy 8 – BLen(1) = 7, bytes from the compressed row.
‘ This means we copied (8 + 7) 15 bytes from the compressed row.
‘
‘ That is half the compressed row done, i leave the rest up to you to grasp.

Hope this helps someone.

]]>