NAME¶
vsdump - VSD/VSS file format reverse engineering
SYNOPSIS¶
vsdump [SUBOPTIONS] FILE
DESCRIPTION¶
This manual page documents briefly the
vsdump command.
vsdump is an ALT Linux (www.altlinux.ru) project of VSD/VSS file format
reverse engineering. VSD/VSS is an OLE file. Main data is stored in
'VisioDocument' file inside this OLE. 'VisioDocument' file has a header with
version number and pointer to the 1st stream. Pointers contain Type, Offset to
the start, Length, and Format of stream. The Format field has a 'compressed'
flag (the 2nd least significant bit).
Format == 0x4* is used for streams that contains some strings.
Format == 0x5* is used for streams that contains some pointers.
Format == 0xd* is used for streams made of chunks.
Type == 15 is used for streams that contains 'Page' data.
Type == 23 is used for streams that contains Icons.
And so on.
To (de)compress some streams Visio uses some version of LZW algorithmwith
options different from other MS-Office applications.
Those streams are made of chunks. Chunks starts with header that contains Type
and Length fields. Most of chunk types has a mandatory fields and some of them
has a discretionary fields. The type of chunk imply an order and format of
mandatory fields and place of start for discretionary fields (if any).
The discrepionary fields made of 'blocks'. Blocks start with length, type fields
and index of variable it is linked to.
Blocks made of 'slices'. Every slice can be single byte operation like
"+" or "*" or 'command' to put one/two bytes data or
IEEE-754 fraction or stringor name or function into formula.
SUBOPTIONS¶
A summary of options is included below.
- help
-
list subcommands
- dump <file> [directory]
- dump pointers and inflated streams
- test <file>
- test command table
AUTHOR¶
vsdump was written by Valek Filippov (frob@df.ru)
This manual page was written by Juan Angulo Moreno <juan@apuntale.com>,
for the Debian project (but may be used by others).