home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
C!T ROM 5
/
ctrom5b.zip
/
ctrom5b
/
DOS
/
DATABASE
/
DATACAN2
/
DATACAN.DOC
< prev
next >
Wrap
Text File
|
1995-04-21
|
25KB
|
544 lines
_______________________________________________________________________
This is a
DEMO/SHAREWARE version
of the DATACAN ver 2.01 program
developed in
ITECH
SLOVENIA, EUROPE
E-Mail: itech.datacan@bbs.abm.si
URL: http://www.abm.si/si21/itech/
_______________________________________________________________________
A short description
DATACAN is a whole new philosophy of the database file storing,
retrieving, extracting... It is not ment to substitute all known
database programs, but to enhance their usability beyond the current
possibilities. DATACAN is a testing, analyzing, storing, reviewing,
selecting, exporting tool which offers a lot what known databases
don't. It may be your choice after see its potentials.
One big problem which always arises, is how to store big files (along
with their searching indexes and with the maintaining programs?) on the
smallest possible amount of disk space (floppy, CD ROM ...). [1]
By solving this problem with ARJ, PKZIP... a new problem arises. How to
maintain compressed files without decompressing them? Stacker or
Dblspace offers only partial answer, because compression factor is
not as big as at classic compressors. [2]
Another big problem is a retrieving software per se. There are some
solutions based on the extended indexing of a several different kinds.
But this methods known also as (total) inverting, need a lot of
additional disk space. This solution is also very limited because
retriever is usually designed only for one kind of file(s). [3]
The third big problem is file sharing among many network users without
overcharging disks space on one and without overcharging CPU and
network traffic on the other side. [4]
A simple yet powerful user interface is a demand from a novice and from
an experienced user. The first kind of users will never use a software
without enough simple interface. Experienced (yesterday's novice) users
have a lot of 'not so simple' needs. [5]
It was always ours deepest believing, that it is the best to handle bus
timetables, 30 or 300 satellite network program schedules, worldwide
air timetables, file directories (my computer or LAN or WAN), a million
cooking recipes, state phone-book, business registers, business
opportunities, chemical substances, abstracts, article lists, book
keeping events, libraries catalogues, C programs, spare parts ... on
one simple, powerful way.
It should be enough to say : TITLE "NEVER OF"
and computer should respond:
_____ __
TITLE : NEVER OUT OF SORTS
SOURCE : SCIENCE
AUTHOR : WHOEVER
.
.
.
.
.
It should be enough to say : FORMULA "?A???O3"
and computer should respond:
_ __
FORMULA : NA2CO3
NAME : SODIUM CARBONATE
PROPERTIES : ... USED ALSO IN SOAP-MAKING ...
.
.
.
.
.
.
It should be also enough to say : PROPERTIES "SOAP MAKING"
and computer should respond:
FORMULA : NA2CO3
NAME : SODIUM CARBONATE ____ ______
PROPERTIES : ... USED ALSO IN SOAP-MAKING ...
.
.
.
.
.
.
It should be enough to say : PROGRAM TYPE "FUTU*", DAY "SATURDAY"
and computer should respond:
____
PROGRAM TYPE : SCIENCE AND FUTURISTIC MAGAZINES
BROADCASTER : CNN INTERNATIONAL
SERIES : FUTURE WATCH
TIME : 22.30
. ________
DAY : SATURDAY
.
.
.
.
and
____
PROGRAM TYPE : SCIENCE AND FUTURISTIC MAGAZINES
BROADCASTER : SKY NEWS
SERIES : BEYOND 2000
TIME : 14:30
. ________
DAY : SATURDAY
.
.
.
One could easily (?) collect together gigabytes of this kind of
information, put them into some kind of database, index them totally,
make appropriate software and faces with the above listed problems.
Converting different types of databases without having the above listed
problems, or better have the listed problems automatically solved, is
the very essence of the DATACAN system.
AD [1],[2]
The converter part of the DATACAN system or CAN.EXE converts many file
types to a special compressed format called CAN. The compression ratio
is up to 10, even more. ZIP size, totally inverted file may seems to
good to be true, but this is the case with the CAN file format.
AD [2]
Without previous decompression of a file not only that searching is
possible, but it is very fast and advanced. For example we can search
very fast through the compressed file for this kind of string "??D??Q!",
where wildcard '?' stands for one or less characters and '!' stands for
exactly one. Instead of previous decompression of a file (or part of a
file), program compress searching string and than search for the
compressed string inside of the compressed data. Then we can export all
matching records without decompressing non matching. All, even more
complex demands are also possible without any previous decompressing.
AD [3]
Retrieving is usually either so called 'free text search' or search by
indexed (beginnings of) fields. Text search as in DOS find or many text
editors' internal function, where any string can be searched for, but
only exact matches can be found. Another disadvantage of this kind of
searching is absence of any more complex tools like operators or field
combination. Searching with indexes is usually limited by beginning of
the field. Total indexing is rare, but when occurs demands a lot of
extra space. Some products offers indexed search by beginning of fields
and full text search which is much slower, but finds inside field
information. DATACAN search is an ideal combination of both searching
techniques with logical and mathematical operators, inter fields
combinations and list of words for all fields. This all does not need
any additional disk space - it is all inside compressed file!
AD [4]
DATACAN.EXE viewer provides network maintaining of CAN files by having
CAN file on the file server without copying duplicates or parts of it
to local disks, but network traffic is not very heavy despite of that.
There is no need for record locks. Any note attached to any record is
stored locally in working directory and is therefore personalized to
user.
AD [5]
DATACAN retriever (at present April 1995 for DOS only) is an intuitive
one. We humans usually know some partial data and want to get a whole
context.
For example we asks:
I know it was SHAKESPEARE or SHACKER or something and I know it was III
Translated question to DATACAN query is:
Author : SHA*
Title : III
And answer is "WILLIAM SHAKESPEARE: RICHARD THE III."
We also asks:
It was MAY or SEPTEMBER, it was 1990 or 1991 and sum ends with '4321'
and it was more than $100000.
Translated question to DATACAN query is:
Date : 199!0!!!
Sum : >100000 *4321
If the number of records exceeds the number to easily browse, we can
always narrow initial conditions by
Date : 199!05!!=199!09!! <19920000
Sum : >100000 *4321
or somehow different. DATACAN QUERY is a simple tool to obtain the
whole context of initial information. Needless to say, with the DATACAN
system everything works very fast inside highly compressed data.
The compression ratio for a CAN file is better for bigger files. But if
CAN file is zipped, it is even smaller. The compression ratio for small
files can be greater then 1, which means that there is no compression
and additional disk space is needed. This is the case only with about
100 Kb files or smaller.
A 136 records DBF file which size is 20,898 bytes
has been zipped to 2,301 bytes or 11%
when canned was 46,160 bytes or 221%
and zipped can has 10,383 bytes or 50%
A 272 records DBF file which size is 41,570 bytes
has been zipped to 4,143 bytes or 10%
when canned was 50,123 bytes or 121%
and zipped can has 12,648 bytes or 30%
A 709 records DBF file which size is 107,994 bytes
has been zipped to 10,151 bytes or 9%
when canned was 61,463 bytes or 56%
and zipped can has 18,701 bytes or 17%
A 148,712 records DBF which size is 136 MEGA bytes
has been zipped to 12 MEGA bytes or 9%
when canned was 16 MEGA bytes or 13%
and zipped can has 8 MEGA bytes or 6%
| + Legend
| transformed +
| file size + + = DBF (untransformed)
| + * = CAN
| + . = ZIP
| + o = zipped CAN
| +
| +
| + .
| + * *
| + * .
| + * .
| * |+ . o o
|* + | o . o
| +o | .
|o + | .
|+ .|
+-----------+---------------------------------------------------------
| DBF file size
records * fields = 30000
This dialog bellow is maybe a little Socratic or Platonic or even
something worse, but it is informative. So please read it if you want
to use DATACAN effectively.
Q. How can we explain the DATACAN system?
A. It is a new way of handling data. It consists of CAN converter which
converts many major formats to DATACAN format and DATACAN program,
which can handle this format. You may convert dBASE and some other
files to EXE as well, so that you don't need the viewer part at all.
Small executable EXE file holds all the data and some powerful
programming features on a small amount of disk. This DBF2EXE or
ASCI2EXE is available only in registered version.
Q. What's DATACAN format?
A. It is a highly compressed and totally indexed file at the same time!
Q. What is the CAN converter?
A. It is a program named CAN.EXE which can convert some files like DBF
or several types of ASCII files into the special CAN format.
Q. How to handle it?
A. CAN C:\DBF\MYFILE.DBF will produce MYFILE.CAN plus DATACAN.EXE file
to maintain MYFILE.CAN plus DATACAN.PIF and DATACAN.ICO. DATACAN.*
are not holding any information about MYFILE.DBF or any other. You
need only one DATACAN.EXE on your computer to handle all CAN files.
Use DATACAN.ICO and DATACAN.PIF for DOS shortcut in Windows 3.x +
environment.
Q. How good it is at compression?
A. Please see the table above.
Q. What do I get beside of compression by converting a file into the
CAN format?
A. A powerful reviewing tool, which displays your data on a very
effective way. You may never see (your) data as good as under
DATACAN.EXE viewer.
Q. What can I do with DATACAN.EXE?
A. Retrieving, extracting, exporting, analyzing data.
Q. What else can be done with data packed into the so called CAN
format?
A. You may search inside by using some innovative techniques which
provides you a very good data error control. You may (re)note your
data. You may not need some special designed programs to work with
your data, just by applying CAN conversion. It can be exported into
several other formats (partly or whole). You may create a whole
library of CAN files and exchange them with others on a far more
effective way than ASCII or DBF or DB files. You may even publish
your data under CAN format along with DATACAN.EXE viewer with no
program royalties other then initial distribution fee. You may even
monitor your big mainframe or server database files and
simultaneously backup them in the CAN format. (Old) COBOL files from
the mainframe tapes just as (today) SQL or ORACLE organized files
can be converted into the CAN format to see them from a different
perspective.
Q. What about LAN?
A. Just put ANYFILE.CAN on the server disk, let say onto the
F:\CANFILES directory and everybody who has access to the
DATACAN.EXE can view ANYFILE.CAN at the same time with others. His,
hers notes and setups will be saved locally on his, hers working
disk:\directory under ANYFILE.DCC and ANYFILE.DCF. It works
perfectly with NOVELL and some other networks. You should run e.g.
F:\DATACAN\DATACAN F:\CANFILES\MYFILE.CAN, or with DC.BAT which
contains the following lines:
@ECHO OFF
IF NOT EXIST F:\CANFILES\%1.CAN GOTO NOEXIST
F:\DATACAN\DATACAN F:\CANFILES\%1.CAN
GOTO ENDING
:NOEXIST
IF NOT EXIST F:\CANFILES\%1 GOTO REALYNO
F:\DATACAN\DATACAN F:\CANFILES\%1
GOTO ENDING
:REALYNO
ECHO NO SUCH FILE %1
:ENDING
Q. What about WINDOWS?
A. For now you may run CAN.EXE (converter) and DATACAN.EXE (viewer)
under DOS window on WINDOWS and OS/2. In fact many of them
simultaneously, what provides you a kind of powerful data crossing.
This limitation of only DOS using will be overcame later this year
by releasing WINDOWS version. Files converted under DOS will be
perfectly suitable for WINDOWS interface reviewing.
Q. What are hardware/software demands?
A. IBM PC compatible with at least 500 kB free conventional memory and
DOS 5.0 or higher. During the conversion a fast computer is
recommended. Viewer (DATACAN.EXE) as converter (CAN.EXE) will use
maximum of available resources.
Q. Why should I convert all my archive DBF files to CAN files and
archive them again?
A. First you will always be able to find and export all records
contains 'Mr. Green's payments greater then 1000 units' or something
like that without decompressing all of them, searching through and
maybe even writing a program for only this occasion. Second you may
need one floppy disk instead of many or one CD-ROM instead of many
for storing data etc.
Q. What are some good examples of searching possibilities?
A. You don't know whether is KIVI or KIWI or even KIWWI. Type KI!?I.
'!' stands for exactly one sign and "?" stands for one sign at most,
so both will cover all combinations.
You may type !!!!!!!!!!!!!!!. The program will find all 15
characters words containing records. If you type !!!!!!!!!!!!!!!??,
the program will find all 15 or 16 or 17 characters words containing
records. You already know what will find !!!!!!!!!!!!!!!* or *.
You may type !YLE-TYLE to get CYLE and KYLE and no TYLE.
This kind of searching is also possible inside of word list. Search
WER* in with F5 invoked list of words and the cursor will stop at
the first word beginning with WER and its frequency and its fourteen
successors with there frequencies will be displayed. F5 invoked list
of words is also recommended to see what to search for at all and to
see perhaps some strange words which are probably near frequently
used word. You may se something like this:
COMPUTER 1236
CONPUTER 7
It is very likely that CONPUTER is misspelling. By pressing ENTER
twice, you will get the whole context of CONPUTER, to see where the
problem might be.
Type -!* to get all the records with THIS field empty. Or !* to get
all the records with THIS field not empty. This is a very good data
error control tool.
Q. Will this CAN converting replace all of my dBASE related software?
A. No, usually only part of it. Sometimes however, when you just want
to make reviewing software for clerks or want to publish CD ROM with
millions of alphanumeric statements, it can be the best way just to
convert the data into one large CAN or EXE file, not designing
special programming tools.
Q. How to install?
A. Make CAN directory on any disk with enough space and copy CAN.EXE
onto it. CAN.EXE file must not be renamed! Everything will take
place inside of this directory, so make sure, that you have at least
twice as much disk space, as the file to convert, occupy. Working
files are deleted automatically after converting and releasing disk
space. If something goes wrong (e.g. power off) and conversion
doesn't complete, please delete the directory $DELETE.ME$
Q. What is the difference between the shareware version and the
registered one?
A. With the shareware version you may convert only the first 10000
records of any file. Format of possible input files is much more
limited. Shareware version don't produce independent EXE files. For
registration fee you are entitled to manual, low cost updates
including Windows when available, E-mail and BBS help and of curse
registered version of DATACAN converter without shareware
limitations.
Q. How to use CAN.EXE file to convert to The CAN format in the first
place?
A. This is the syntax and some examples.
Syntax: CAN [filename]
filename any valid input file (DBF, TXT ...)
Syntax: CAN [/filename.ccf [noask]]
/filename.ccf any valid Can Configuration File
noask batch mode - no questions asked
Example: CAN d:\files\new\FILE.DBF
The FILE.DBF file will be converted to the CAN file.
The filename.ccf is the pilot file for the conversion. Every conversion
makes its own pilot file, so that you can use it again.
A MENU driven conversion will begin:
Input filename
============== the default filename from xxxxxxxx.ccf will be
offered. You may change it.
File format (1 - DBF, 2 - Delimited, 3 - Fix with EOL)
====================================================== press 1, 2 or 3
for desired
file type
If '1' - DBF is selected, no additional information will be necessary,
so the program will only let you change fields names, from those in
DBF format. It's the best to keep them.
If '2' - Delimited is selected, the program will first ask for the
delimiter. You may enter '#9' what means TAB DELIMITED or ',' or '#44'
what both means COMA DELIMITED, or '#254' what means ASCII CHARACTER
254 delimited.
If '3' - Fix with EOL is selected, the program will first ask for
fields' widths. '10-2-32-45-12-2-10-99' is an correct answer example,
which also claims that there are eight fields in every record.
For '2' an '3' default fields' names will be Field01, Field02 ... They
may be changed in the next MENU. Now we recommend change.
Then the following menu appears
1 - Fields to be canned or not
2 - Fields naming
3 - Alphabet setting
4 - Can Configuration File
If '1' is selected, you may toggle any field to be canned or left
behind by entering field number.
If '2' is selected, you may change field name.
If '3' is selected, you may enter alphabet. Any character which is not
in the alphabet is considered as separator. What does it mean you can
see from this example: If the '-' is a member of alphabet, then
DRAGON-FLY will be one word and it will be possible to search for
'*N-FL*' and get it. If '-' is a separator (= not member of the
alphabet) then it will be possible to search for it by 'DRAGON', 'FLY',
'-', 'DRA* *LY' etc., but not by '*N-FL*'
'AltX' is the cancel program choice.
'F10' is the execute choice.
Example: CAN /anyfile.ccf noask
The pilot file anyfile.ccf will be obeyed without possibility to change
anything during conversion.
During every canning a CCF extension (Can Configuration File) will be
produced. You may use it again with some changes.
Output: FILE.CAN or (FILE).EXE
Executable file is into parenthesis to note them immediately among
other EXE files, but parenthesis can be avoided. The shareware version
of converter cannot make EXE, but only CAN files, which needs
DATACAN.EXE to review them. DATACAN.EXE is always produced during
conversion although you may have one from previous canning procedure.
You may delete all but one DATACAN.EXE file.
Executable file (FILE).EXE or FILE.CAN is totally indexed, yet
compressed file which contains all data from FILE.DBF and all possible
index and a powerful programming tools for displaying, retrieving and
exporting data. File for itself already fulfills many needs for
programming.
Demo version converts and displays up to 10,000 records with no more
then 36 fields each. Registered version converts up to more then 2
billions of records.
Q. How to order?
A. Just fill the order form in order_dc.doc file and send it to
FAX No +386 64 714 720
or
E-Mail: itech.datacan@bbs.abm.si
Q. What about the future development?
A. Future improvements of the DATACAN product will be
- more possible input/output formats
- faster conversion
- Windows user interface
- more advanced DQL (DataCan Query Language)
- multi can file support (combining files)
- more compress
- filters in word lists
- field list
- various field types like DATE, NUMERIC, BOOLEAN ...
- a few searching-analyzing innovations like
'towns where Smiths are (not)' or
'the first name is a part of the family name or town name' or
'the sum of Field01 and Field04 is equal to 10000 or Field02'
- and more
Thank you for using DATACAN and contributing with your registration.