File archive - Revision history

Halamix2 at 21:24, 23 October 2022

2022-10-23T21:24:15Z

← Older revision		Revision as of 21:24, 23 October 2022
Line 48:		Line 48:

	* each entry value + "address of directory" is a pointer to corresponding ''file descriptor''		* each entry value + "address of directory" is a pointer to corresponding ''file descriptor''
	* each position is calculated, hash function is available at ~~<nowiki>~~https://worms2d.info/Graphics_directory~~</nowiki>~~		* each position is calculated, hash function is available at https://worms2d.info/Graphics_directory

	each '''file descriptor''' have following structure:		each '''file descriptor''' have following structure:

Halamix2 at 12:46, 22 August 2022

2022-08-22T12:46:32Z

← Older revision		Revision as of 12:46, 22 August 2022
Line 22:		Line 22:
	\|}		\|}

	* file data spans up to the "address of directory". These two ~~subpoints~~ are optional, but they provide nice alignment		* file data spans up to the "address of directory". These two points are optional, but they provide a nice alignment:
	** There is <code>1A 00</code> at the end of each file		** There is <code>1A 00</code> at the end of each file
	** There may be <code>00</code> padding after that, so the next file/section starts at offset % 4 ~~== 0~~		** There may be <code>00</code> padding after that, so the next file/section starts at offset aligned to 4 bytes

	Following offsets are counted from "Directory address"		Following offsets are counted from "Directory address"

Halamix2 at 21:18, 1 January 2021

2021-01-01T21:18:19Z

← Older revision		Revision as of 21:18, 1 January 2021
Line 7:		Line 7:
	\|0x0		\|0x0
	\|0x4		\|0x4
	\|Magic, <code>44 49 52 1A</code> (~~"'''~~DIR\x1A~~'''"~~)		\|Magic, <code>44 49 52 1A</code> (<code>DIR\x1A</code>)
	\|-		\|-
	\|0x4		\|0x4

Halamix2 at 16:36, 26 December 2020

2020-12-26T16:36:36Z

← Older revision		Revision as of 16:36, 26 December 2020
Line 73:		Line 73:
	\|}		\|}

	* File descriptors		==== File descriptors ====
	** if two files have the same hash, then hash table points to first file, which have "next hash" to another file descriptor with the same hash. If there is no collision then it's 0.		* if two files have the same hash, then hash table points to first file, which have "next hash" to another file descriptor with the same hash. If there is no collision then it's 0.
	** filename is 4 bytes aligned, terminated by 0x00 (and padded with 0x00 to 4 bytes)		* filename is 4 bytes aligned, terminated by 0x00 (and padded with 0x00 to 4 bytes)

Halamix2: Created page with ".wad file format is exactly the same as [https://worms2d.info/Graphics_directory Worms .dir format], only extension is different. {| class="wikitable" !Offset !Size !Descripti..."

2020-12-26T16:35:34Z

Created page with ".wad file format is exactly the same as [https://worms2d.info/Graphics_directory Worms .dir format], only extension is different. {| class="wikitable" !Offset !Size !Descripti..."

New page

.wad file format is exactly the same as [https://worms2d.info/Graphics_directory Worms .dir format], only extension is different.
{| class="wikitable"
!Offset
!Size
!Description
|-
|0x0
|0x4
|Magic, <code>44 49 52 1A</code> ("'''DIR\x1A'''")
|-
|0x4
|0x4
|File size
|-
|0x8
|0x4
|Directory address
|-
|0xC
|???
|Files data
|}

* file data spans up to the "address of directory". These two subpoints are optional, but they provide nice alignment
** There is <code>1A 00</code> at the end of each file
** There may be <code>00</code> padding after that, so the next file/section starts at offset % 4 == 0

Following offsets are counted from "Directory address"
{| class="wikitable"
!Offset
!Size
!Description
|-
|0x0
|0x4
|Magic, <code>0A 00 00 00</code>
|-
|0x4
|0x1000
|[[File archive#Hash table|Hash table]], 1024 4-byte entries
|-
|0x1004
|???
|File descriptors
|}

== Hash table ==

* each entry value + "address of directory" is a pointer to corresponding ''file descriptor''
* each position is calculated, hash function is available at <nowiki>https://worms2d.info/Graphics_directory</nowiki>

each '''file descriptor''' have following structure:
{| class="wikitable"
!Offset
!Size
!Description
|-
|0x0
|0x4
|Next hash, by default is set to <code>00 00 00 00</code>
|-
|0x4
|0x4
|File data offset in archive
|-
|0x8
|0x4
|File size
|-
|0xC
|0x4
|Filename with path
|}

* File descriptors
** if two files have the same hash, then hash table points to first file, which have "next hash" to another file descriptor with the same hash. If there is no collision then it's 0.
** filename is 4 bytes aligned, terminated by 0x00 (and padded with 0x00 to 4 bytes)