Scifer Archiver and Compression

Introduction

Scifer is a data compression and archiving application using a new data compression logic based on Universal Common Logic The algorithms of Scifer are based on UCL^[1]‘ continous entropy , where any number of uniqueness of algorithms can be unified under the UCL^[1] Laws.The BETA version constitutes a few fundamental algorithms.

Features of Compression Algorithm

Theoretically data can be continously compressed, however advanced research and contributions have to be made to UCL^[1], a relatively new theory.
The BETA version focuses more on complex, pattern-less and randomized data. Textual and more literal data which are more compressible using larger dictionary based schemes and advanced frequency analyzed outputs are not implemented in this version.
Consumes the least memory resources compared to any other powerful compression algorithms.
Data can be processed in a buffer-less stream also, with zero information of the previously processed data (no sliding window, no frequency counts, no Huffman or any other frequency based analysis). This can be implemented in hardware or software for real-time compression and decompression of data streams of any type.
Unique archive structure which allows files and data to be directly accessed without extracting.
File addition or removal is not allowed. Ideal for archived file distributions.
The external header makes it possible to view the information of files in the archive in any application like browsers, office automation documents or XML processing viewers without mobilising the large binary data of the archive. Example: show the archive contents in any number of websites while the actual binary archive is present in one location only

^[1] Abbreviation UCL – Universal Common Logic

Archive structure

The archive consists of archive header information in the beginning followed by each file information in sequential order.

Two distinct type of archives can be created in Scifer

Internal header

The archive and file information is stored within the same archive in the beginning followed by binary compressed data.Creates a archive with file extension [ .sen ]

External header

The archive and file information is stored in a separate XML file. The contents of XML can be universally viewed and a more comprehensible view can be obtained using XML viewers like Internet Explorer. The files from the binary archive (.ba) created along with this can be extracted only along with the XML header file.

Archive headers

Internal

Archive information and file information repeated for each file sequentially

Archive Information	Type	Bytes
Header [ HEX = ACEAFACE]	INT	4
Author	STRING	16
Mode Store\|Compress	BOOL	1
Passphrase protected	BOOL	1
Number of Files	INT	4
Number of Folders	INT	4
Extradds	INT	1
Total file size	_INT64	8
Total compressed size	_INT64	8
End of archive pointer	_INT64	8

File Information - Repeated for number of files

Passphrase protection	BOOL	1
If [ Protected = true ]	STRING	12
Filename size	INT	2
Filename	STRING	Filenamesize
File type and information	STRING	24
File size	INT	4
Relative path size	INT	2
Relative path	STRING	Relativepathsize
Attributes	INT	4
Creation time/date	_INT64	8
Modified time/date	_INT64	8
Accessed time/date	_INT64	8
Mode Store\|Compress	BOOL	1
File Identity/Compare	INT	4
File pointer	_INT64	8
If [ mode = compress] Compress Options	INT	1
If[compress option (bit 3)=true] windowsize	INT	2
Compressed file size	INT	4

External similar to Internal

The DTD file can be downloaded from http://www.senthilkumarr.com/scifer/archive.dtd

DTD structure with element definition:-

ROOT

ARMA (TF,Au,Se,Fc,Foc,Os,Cs,Ea,Pe?,At,files)

TF -- original target BA filename --

Au -- author --

Se -- protected --

Fc -- number of files --

Foc -- number of folders --

Os -- total file size --

Cs -- compressed file size --

Ea -- extradds --

Pe -- end of archive pointer --

At -- archive type store | compress --

Files(FM+) -- files information --

FM (Fn,Ty,Fs,Rp,Ar,Cdt,Adt,Mdt,FP,Ct,Md,Co?,Cw?,Cm,Pr,Pw?)

Fn -- filename --

Ty -- file type and information --

Fs -- filesize --

Rp -- relative path --

Ar -- attribute --

Cdt -- creation time/date --

Adt -- accessed time/date --

Mdt -- modified time/date --

FP -- file pointer --

Cf -- file identity/compare --

Md -- mode store | compress --

Co -- compress option --

Cw -- window size --

Cm -- compressed size --

Pr -- passphrase protection --

Pw -- passphrase --

Interface

Main Scifer Dialog

main windows

a) Files/folders to create archive or archive to be extracted .Multiple files/folders are separated by ::

b) Display folder selection dialog

c) Display file selection dialog

d) Path to create archive or extract files.

e) Display path selection dialog.

f) Author for archive creation (max 16 chars).

g) See help for header types

h) See help for types of compression

i) Create an archive based on selection

j) Select type files only for archive creation and filtered substring filename match for extraction.

k) Enter passphrase if archive needs to be protected.

l) Type of files extraction.

m) See view archive information interface.

n) Results of created archives and extracted files.

o) Save results in scifer.log

p) Extract all files based on selection

q) Encrypt files names and header using the same passphrase.

Progress Control Dialog

a) Text to identify archivers current process

b) Percentage of already processed files.

c) File in process along with it size in bytes.

d) Progress of processed file.

e) Compressing size of processing file.

f) Pause process.

g) Close application

h) Commandline arguments for creating archive or extracting from archive.

Passphrase Dialog

This dialog pops up if the file processed is protected and passphrase not supplied in advance.

passphrase

a) Enter alphanumeric passphrase max 12 chars.

Note:if passphrase is incorrect the processing file isNote:if passphrase is incorrect the processing file is created with 0 bytes.

View archive information

A new file exploring technique called ONEVIEW is used in which you can see all the files present in a single

view information
View archive information as per header structure.

b) Complete ONEVIEW all files and details all in one window.

c) Extract the selected file only

d) Extract all the files in the selected file folder.

e) Extract all files in the selected file folder and subfolders.

f) Extract the selected file into c:\temp and execute.

g) Type of extraction

h) Path to extract files.

i) Extract all files.

Creating an Archive

Three methods of creating an archive.

I. From main dialog

Select files/folders to be included in the archive, separate each file/folder with ::
Select the archive path if required, otherwise the default is in the same path as the first file or folder.
Enter the authors name, default is Scifer BETA
Choose header type internal or external, if File structure only is selected then only the XML file information file is created without the binary archive(.ba).
Choose type of compression (refer types of compression)
Click create button or press alt+c to create the archive.

II. From console options for creating an archive.

usage : scifer [flags] <archive> <file1/folder><file2/folder> <...>

Flags

-c: Compression type : 0 Normal(default) | :1 Without Probability

-m: Mode type : 0 File(default) | :1 Stream

-d: Data type : 0 Normal(default) | :1 Text

-h1 External Header File as XML

-xmlon File structure only as xml(no file processing)

-w: window size in bytes maximum 65536(default 32768) (if type = 0)

-SEL selective filetype to compression type processing

-nc no compression(store only)

-ft:*** file specific processing(max 9 chars)

-A: Author (16 bytes)

III From explorer contextmenu

Right click on any file or folder and select the type of archive required.

Extract files from archive

Three methods of extracting files from archive.

I. From main dialog.

From main dialog.

Select the archive file(.sen) or in case of external header then(.xml) file.
If the archive is of type external header and the xml filename does not match the binary archive filename then select xml file followed by binary archive filename.
Select output file directory,if needed.
If selective file have to be extracted then enter filtertype eg.jpg so any filename which has a substring .jpg in its filename only will be extracted (note filter is case sensitive)
Enter passphrase in case archive is protected.
Select type of extract /here default or /scifer (all files will be extracted to /scifer without any folder structure).
Click view information if you want to see the file information of the archive.
Click extract or press alt+e to extract files.

II. From console options for extracting from archive

Extraction :

usage : scifer [flags] <XML header if specified> <..ba>

Flags

-de: (decompression mode)if header in external XML

-d: (decompress)

-v View Files only

-ls Saves to logfile (Archive.log) after viewing only option

-ft: type- (max 10 chars)type refers to filetype of files to view or extract

-es Extract files sequentially without folder structure into scifer folder

III. From explorer context menu

Contextmenu options are same as main dialog options.

Types of compression

Normal

All features included file compression. Not recommended for very large files of precompressed format like multimedia files.

Buffer less (File)

Compression with UCL^[1] only (without any sliding window, dictionary, or frequency trees like Huffman).Not recommended for high literal files like text files or html which are better compressed with dictionary based schemes.

Buffer less(Stream)

Same as buffer less – file but data is compressed with real-time sequencing.

Text

The BETA version does not use any Scifer or UCL^[1]based algorithm but a direct LZA method.

Selective

The archiver uses a pre-defined filetype/filesize ratio based compression method selection.

Store

No compression for files. Suitable for large files like multimedia which are already precompressed where the time and resource utilized to ratio of compression is negligible.

--| Return to Mainpage |