Common extended header

Common extended headers

Common header : 0x00

Introduction
Coomon header has a CRC16 for check of header. So all Level 1 headers with one or more extended headers, all Level 2 headers, and all Level 3 headers need to have common header.

Structure

offset	common header level 1 or 2 no additional info	common header level 3 no additional info	common header level 1 or 2 (UNLHA, Unix) with additional info	common header level 3 (UNLHA, Unix) with additional info	common header level 1 or 2 (OS/2) with additional info	common header level 3 (OS/2) with additional info	offset
00	Extended header identifier	Extended header identifier	Extended header identifier	Extended header identifier	Extended header identifier	Extended header identifier	00
01	CRC of header(L)	CRC of header(L)	CRC of header(L)	CRC of header(L)	CRC of header(L)	CRC of header(L)	01
02	CRC of header(H)	CRC of header(H)	CRC of header(H)	CRC of header(H)	CRC of header(H)	CRC of header(H)	02
03	Size of next extended header(L)	Size of next extended header(LL)	Additional info	Additional info	Additional info(1)	Additional indo(1)	03
04	Size of next extended header(L)	Size of next extended header(LH)	Size of next extended header(L)	Size of next extended header(LL)	...	...	??
05		Size of next extended header(HL)	Size of next extended header(H)	Size of next extended header(LH)	...	Additional info(end)	size - 5
06		Size of next extended header(HH)		Size of next extended header(HL)	...	Size of next extended header(LL)	size - 4
07				Size of next extended header(HH)	Additional info(end)	Size of next extended header(LH)	size - 3
08					Size of next extended header(L)	Size of next extended header(HL)	size - 2
09					Size of next extended header(H)	Size of next extended header(HH)	size - 1

Sample

	+0	+1	+2	+3	+4	+5	+6	+7	+8	+9	+A	+B	+C	+D	+E	+F	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x0000	2E	00	2D	6C	68	30	2D	00	00	00	00	00	00	00	00	0C	.	.	-	l	h	0	-	.	.	.	.	.	.	.	.
0x0010	EE	27	3D	20	02	00	00	4A	05	00	00	CF	4F	07	00	01	î	'	=			.	.	J		.	.	Ï	O		.
0x0020	74	65	73	74	08	00	02	74	65	73	74	FF	00	00	--	--	t	e	s	t		.		t	e	s	t	ÿ	.	.	-	-

Description

Extended header identifier	This is the identifier for specifying an extended header. You must set 0x00 in common header.
CRC of header	This is the CRC16 of the whole header. You must use 0 in this field to calc CRC16. CRC16 polynomial is x¹⁶ + x¹⁵ + x² + 1, and initial value is 0. CRC of header is 0x4FCF in the foregoing sample.
Additional info	This is a Additional information. For details, refer to the after-mentioned About additional info. This field may not exist. According to header.tex which is attached LHA for OS/2, the size of this field is "n" which means variable size, and you must not assume that size of this field is 0. Additional info is not exist in the foregoing sample.
Size of next extended header	This is the byte count of following extended header. Set this field as 0, when there is not following extended header. Size of next extended header is 0x0005 in the foregoing sample.

About additional info

Following is additional info which is written in header.txt that is attached UNLHA32.DLL. It is unknown about that is written in header.tex which is attached LHA for OS/2.

bit NO

caption

description

M
L
B
/\

\/
M
S
B

7 - 3

Time zone(1)

The time zone of the district.
In reference

0	: it is same as environment of the extracting.
other	: time of storeing + 16.

I can not understand it. It is unknown also about the adaptation range.

Encryption flag

Details are not set to encryption.

Time zone(2)

Local(1)/UTC(0)
Time zone(2) is different than Time zone(1). However I can't understand well.
It is unknown also about the adaptation range.

Conversion of new-line code

Text(1)/Binary(0)
New-line code is "\r\n" in Windows, '\r' in Macintosh, '\n' in Unix. This is the flag for conversion from the new-line code of OS which compressed the file to the new-line code of OS which extracted the file. However most editors correspond to these new-line codes now, and conversion may introduce confusion when conversion can not operate correctly, so I think that conversion is unneedful.

Filename header : 0x01

Introduction
Filename header contains the filename. So all Level 1 headers with empty filename field in base header, and All Level 2 headers, and All Level 3 header need to have filename header. Note that you must add an empty filename header in Level 2 Header and Level 3 Header when you store a directory by "-lhd-". However, LHa for UNIX does not have an empty filename header, in case of a stored directory with "-lhd-". Take into consideration the fact that there are such headers, when you write a program that processes a LHA archive.

Structure

offset	Filename header Level 1 and 2	Filename header Level 3
00	Extended header identifier	Extended header identifier
01	Filename(1)	Filename(1)
??	...	...
size - 5	...	Filename(end)
size - 4	...	Size of next extended header(LL)
size - 3	Filename(end)	Size of next extended header(LH)
size - 2	Size of next extended header(L)	Size of next extended header(HL)
size - 1	Size of next extended header(H)	Size of next extended header(HH)

Sample1

	+0	+1	+2	+3	+4	+5	+6	+7	+8	+9	+A	+B	+C	+D	+E	+F	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x0000	2E	00	2D	6C	68	30	2D	00	00	00	00	00	00	00	00	0C	.	.	-	l	h	0	-	.	.	.	.	.	.	.	.
0x0010	EE	27	3D	20	02	00	00	4A	05	00	00	CF	4F	07	00	01	î	'	=			.	.	J		.	.	Ï	O		.
0x0020	74	65	73	74	08	00	02	74	65	73	74	FF	00	00	--	--	t	e	s	t		.		t	e	s	t	ÿ	.	.	-	-

Sample2

Example of empty filename header, when store directory with "-lhd-".

	+0	+1	+2	+3	+4	+5	+6	+7	+8	+9	+A	+B	+C	+D	+E	+F	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x0000	2E	00	2D	6C	68	64	2D	00	00	00	00	00	00	00	00	8F	.	.	-	l	h	d	-	.	.	.	.	.	.	.	.
0x0010	88	2E	3D	20	02	00	00	4A	05	00	00	24	07	03	00	01	ˆ	.	=			.	.	J		.	.	$			.
0x0020	0C	00	02	74	65	73	74	FF	64	69	72	FF	00	00	--	--		.		t	e	s	t	ÿ	d	i	r	ÿ	.	.	-	-

Description

Extended header identifier	This is the identifier for specifying an extended header. You must set 0x01 in filename header.
Filename	This is the filename. Filename is "test" in the foregoing sample1. Filename is not exist foregoing sample2.
Size of next extended header	This is the byte count of following extended header. Set this field as 0, when there is not following extended header. Size of next extended header is 0x0008 in the foregoing sample1, and 0x000C in the foregoing sample2.

Directory name header : 0x02

Introduction
Directory name header contains the path from the base directory to the parent directory of compressed file. When you store a directory by "-lhd-", you must set also including the name of the directory to store. Directory name header may not exist, when it is not needed. Path delimiter is 0xFF which is not used by Shift_JIS that is japanese character code. Moreover, note that directory name is terminated by path delimiter.

Structure

offset	directory name header level 1 or 2	directory name header level 3
00	Extended header identifier	Extended header identifier
01	Directory name(1)	Directory name(1)
??	...	...
size - 5	...	Directory name(end)
size - 4	...	Size of next extended header(LL)
size - 3	Directory name(end)	Size of next extended header(LH)
size - 2	Size of next extended header(L)	Size of next extended header(HL)
size - 1	Size of next extended header(H)	Size of next extended header(HH)

Smaple1

	+0	+1	+2	+3	+4	+5	+6	+7	+8	+9	+A	+B	+C	+D	+E	+F	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x0000	2E	00	2D	6C	68	30	2D	00	00	00	00	00	00	00	00	0C	.	.	-	l	h	0	-	.	.	.	.	.	.	.	.
0x0010	EE	27	3D	20	02	00	00	4A	05	00	00	CF	4F	07	00	01	î	'	=			.	.	J		.	.	Ï	O		.
0x0020	74	65	73	74	08	00	02	74	65	73	74	FF	00	00	--	--	t	e	s	t		.		t	e	s	t	ÿ	.	.	-	-

Sample2

Example of store a directory which is named "test/dir" by "-lhd-".

	+0	+1	+2	+3	+4	+5	+6	+7	+8	+9	+A	+B	+C	+D	+E	+F	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x0000	2E	00	2D	6C	68	64	2D	00	00	00	00	00	00	00	00	8F	.	.	-	l	h	d	-	.	.	.	.	.	.	.	.
0x0010	88	2E	3D	20	02	00	00	4A	05	00	00	24	07	03	00	01	ˆ	.	=			.	.	J		.	.	$			.
0x0020	0C	00	02	74	65	73	74	FF	64	69	72	FF	00	00	--	--		.		t	e	s	t	ÿ	d	i	r	ÿ	.	.	-	-

Description

Extended header identifier	This is the identifier for specifying an extended header. You must set 0x02 in directory name header.
Directory name	This is the path from the base directory to the parent directory of compressed file. Directory name is "test/" in the foregoing sample1. Directory name is "test/dir/" in the foregoing sample2. (It is converted from 0xFF to '/' for easier to understand.)
Size of next extended header	This is the byte count of following extended header. Set this field as 0, when there is not following extended header. Size of next extended header is 0x0000 which means there is no more following extended header, in both of foregoing sample.

Multi-disc header : 0x39

Introduction
Multi-disc header caontains the information for dividing the compressed file. This header is proposed http://www.creative.net/~aeco/jp/lzhspc01.html. I think that there is no implementation which corresponds to multi-disc header, so you should not use multi-disc header.

Construction

offset	Multi-disc header level 1 or 2	Description
00	Extended header identifier	This is the identifier for specifying an extended header. You must set 0x39 in multi-disc header.
01	Span mode	This is the identifier the mode of the data following this header. For details, refer to the after-mentioned About span mode.
02	Beginning offset(LL)	This is the position which shows where the divided data following this header should fit into the whole compression data.
03	Beginning offset(LH)
04	Beginning offset(HL)
05	Beginning offset(HH)
06	Size of data(LL)	This is the byte count of the divided data following this header.
07	Size of data(LH)
08	Size of data(HL)
09	Size of data(HH)
10	Size of next extended header(L)	This is the byte count of following extended header. Set this field as 0, when there is not following extended header.
11	Size of next extended header(H)

About span mode

val	Span mode	description
0	SPAN_COMPLETE	It means that the data following this header is not divided. In this case, multi-disc header is not needed, so you should not use this value.
1	SPAN_MORE	It means that the data following this header is one of divided data and following data is not last data. You should continue to extract until you find a header that has span_mode as SPAN_LAST.
2	SPAN_LAST	It means that the data following this header is last of divided data.

Comment header : 0x3F

Introduction
Comment header contains a comment text of the compressed file. Comment header may not exist, when it is not needed. Comment header is not used so much, because it can not read and write easily and it has not clarified the purpose of use.

Structure

offset	Comment header level 1 and 2	Comment header level 3
00	Extended header identifier	Extended header identifier
01	Comment(1)	Comment(1)
??	...	...
size - 5	...	Comment(end)
size - 4	...	Size of next extended header(LL)
size - 3	Comment(end)	Size of next extended header(LH)
size - 2	Size of next extended header(L)	Size of next extended header(HL)
size - 1	Size of next extended header(H)	Size of next extended header(HH)

Sample

	+0	+1	+2	+3	+4	+5	+6	+7	+8	+9	+A	+B	+C	+D	+E	+F	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x0000	42	00	2D	6C	68	30	2D	00	00	00	00	00	00	00	00	0C	B	.	-	l	h	0	-	.	.	.	.	.	.	.	.
0x0010	EE	27	3D	20	02	00	00	4D	07	00	01	74	65	73	74	08	î	'	=			.	.	M		.		t	e	s	t
0x0020	00	02	74	65	73	74	FF	0F	00	3F	63	6F	6D	6D	65	6E	.		t	e	s	t	ÿ		.	?	c	o	m	m	e	n
0x0030	74	20	74	65	73	74	05	00	40	20	20	05	00	00	B8	84	t		t	e	s	t		.	@				.	.	¸	„
0x0040	00	00	--	--	--	--	--	--	--	--	--	--	--	--	--	--	.	.	-	-	-	-	-	-	-	-	-	-	-	-	-	-

Description

Extended header identifier	This is the identifier for specifying an extended header. You must set 0x3F in comment header.
Comment	This is a comment text of the compressed file. Comment is "comment test" in foregoing sample.
Size of next extended header	This is the byte count of following extended header. Set this field as 0, when there is not following extended header. Size of next extended header is 0x0005 in foregoing sample.

Reference

Joe Jared's LZH format
http://www.osirusoft.com/joejared/lzhformat.html
Eriko Tachibana's document of LHA header(Japanese)
http://kuwa.omosiro.com/KGARC2/ARC/LHAHED15.ZIP
Satoshi Hiramatsu's header.tex which is attached to LHA for OS/2(Japanese)
http://www02.so-net.ne.jp/~hiramatu/lha/lha2-2.08.lzh
Micco's header.txt which is attached to UNLHA32.DLL(Japanese)
http://www2.nsknet.or.jp/~micco/unlha32.htm

Author does not certify about accuracy of this document.
When you found typographical errors or omissions, Please mail to cqw10305@nifty.com.
The company name and product name which are used in this document, it is the trademark or registered trademark of each company generally.

Back