TinyScan - Help

6.2.3

Home
Screenshots
Features
Awards & Translations
Download
Help
Forums
Author
Donate

1. Purpose of TinyScan

Un autre regard sur vos données.
A disk usage utility. Output a table and a tree. Export the result to Excel, Access, MySQL, etc. Report disk usage of remote computers.

A. Functional Help

2. Application Menus

File

Open...		Open a previously saved scan.
		Skip Files	Only load the folders details.
Open MySQL...		Open a scan saved in MySQL.

Save...		Save a scan.
		MySQL	Generate an additional file for MySQL.
Save Tree...		Save the tree. Text only, no data.

Exit

Scan

Scan...
		Skip Files	Scan the folders only. Does not display the files.
		MySQL	Generate an additional file for MySQL.
		Checksum	Compute the checksum. This will take some time. Currently, it is used to find duplicate files with SQL after importing in a database.
	*	Folder To Scan:
		Scan To File:

3. Output Table Structure

ID	The line number (an index).
Path
Name
Extension	For a file.
Folder	True or false. For sorting.
Type	Archive, audio, binary, etc.
Size (MB)	The size in MB.
Size (bytes)
Items	The number of items in a folder.
Date	Like 1999-12-31.
Time
Depth	How far in the tree?
Full Path Length	Length of the full path (path + name).
Checksum	Adler-32. Like a fingerprint, it uniquely identifies a file. Not really useful at this time.
Parent	ID of the parent folder. Used to build the tree.
ScanDate	The date of the scan. For Access and MySQL.
ScanTime	The time of the scan. For Access and MySQL.
Node	A one word description of your data. For Access and MySQL.

You can move the columns.

4. Tree

The same data in a tree. You can move around with the mouse or the arrow keys.

5. Sorting

Click on a column header to sort: not sorted, ascending and descending (1st, 2nd, 3rd mouse click).
Ctrl + mouse click for a compound sort.
For example, sort on folder descending (2 mouse clicks) then press the Ctrl key (keep it pressed) then sort on path ascending and on name ascending.
You can sort by folders, then press the Ctrl key, then sort by files then size, etc.
- Sorting the table will also sort the tree. Click on "Refresh".

6. Status Bar

Click on the status bar to clean it.

7. Network drive

You can scan a network drive. You can also a UNC path directly. Java does not(!) need to be installed on the remote computer.

8. Command Line

TinyScan can also be run from the command line (= text mode).
On Windows, if you have it on your desktop, open a Command Prompt:
cd Desktop
java -jar tinyscan.jar --help

8.1 Scheduling

You can use TinyScan from the command line to schedule a scan. On Windows, create a batch file:
@echo off
java -jar c:\tinyscan\tinyscan.jar c:\ -f c:\tinyscan\panda.txt
Then go to Control Panel -> Scheduled Tasks to schedule it.

On Linux, you can put a script in /etc/cron.weekly for example. You can also edit the file /etc/crontab or use the "crontab" command. Or use "at" for a one shot.

8.2 Network Scan

You can also scan several computers on your (Windows) network.
java -jar c:\tinyscan\tinyscan.jar -s 192.168.0.5 -e 192.168.0.10 c$ -F c:\scan
java -jar c:\tinyscan\tinyscan.jar -s 192.168.0.5 -e 192.168.0.10 "c$\Documents and Settings" -F c:\scan

The network scan is slow.
- You can count the number of licences used for a particular program by scanning for a specific folder on a range of computers.
- Check a "dll" version.
- It is not intented to spy. You can easily prevent scans by changing the permissions on a folder. You can also audit your folders for intrusion. Also each member of your staff should have his own account for administration.

9. System Account

You can stay in your session and scan with a system account. Open a command prompt:
runas /user:dom1\staff5 tinyscan_high_memory.bat

10. Importing the data into another program

You can save the scan in a file and reopen it later. With TinyScan or another program.

- TinyScan: File-> Open. If the result is too big (too much rows), check "Skip Files" in the Open dialog box.

- Excel (2003):
File-> Open. In Files of type: chooses All Files (*.*).
Or right-click the text file then chooses Open With-> Excel.
Once the file is opened in Excel, you can decide to save it to Excel's format: File -> Save As -> Save as type: ...(*.xls).

- Use Access (2003) if you have more than 65,536 rows.
File-> Open. In Files of type: chooses All Files (*.*).
Or right-click the text file then chooses Open With-> Access.
For date and scandate, choose the Data Type: "Text". For sizeb and checksum, choose Double.

Better:
Open tinyscan.mdb. If asked to convert the database, answer "Convert Database".
Then File-> Get External Data -> Import...
-> File of type: Text File (*.txt; *.csv; *.tab; *.asc).
Delimited -> Tab. Check "First Row contains Field Names". And select the Text Qualifier: "
Then choose "In an Existing Table" -> Import to table: scan.

- OpenOffice Calc:
Use the file type Text CSV (*.csv,*.txt) to import the data.

- OpenOffice Base:
File -> New -> Database
Connect to an existing database
Text
Next>>
For the "Path to text files", choose the folder containing the scan results.
Type of files: Plain text files (*.txt)
Field Separator: {Tab}
Text Separator: "

- Gnumeric won't ask any question.

- MySQL / EasyPHP / phpMyAdmin:
phpMyAdmin won't import files above 2MB. Follow the instructions from the file filename-mysql.txt.
Then, you can do your queries through phpMyAdmin's web interface.
Or directly from MySQL (there is a pager and a history between sessions on the Linux version). Or with MySQL Query Browser.

11. SQL Queries

mysql>
# Displaying the scan history:
USE tinyscan;
DROP VIEW v1;
CREATE VIEW v1 AS
SELECT DISTINCT node, path, name, scandate, scantime
FROM scan
WHERE parent = 0
ORDER BY node, scandate, scantime;
SELECT * FROM v1;

node	path	name	scandate	scantime
pc1	C:\		2006-09-30	14:15:31
pc2	C:\		2006-09-30	15:20:37
usbdisk	/media/	disk	2006-09-15	19:18:55

# Sample query:
SELECT path, name FROM scan
WHERE type = 'audio'
AND sizemb >1
AND node = 'usbdisk'
AND scandate = '2006-09-30';   # MySQL, Access

# To keep the queries simple, we remove the history:
DELETE FROM scan WHERE node='usbdisk' AND scandate < '2006-09-30';

SELECT path, name, date, time FROM scan
WHERE date BETWEEN '2006-07-01' AND '2006-08-01'
AND path LIKE '%documents%'
AND path NOT LIKE '%programmation%'
AND TYPE = 'file'
AND node = 'usbdisk';   # MySQL (with Access, use * instead of %)

mysql>
# We recompute sizemb to avoid rounding errors
SELECT ext, ROUND( SUM( sizeb ) /1024 /1024, 1 ) AS sizemb, COUNT( * )
FROM scan
WHERE node = 'usbdisk'
GROUP BY ext
ORDER BY sizemb DESC;   # MySQL

ext	sizemb	COUNT( * )
dll	3231452.6	81152
vmdk	7462.9	27183
exe	3748.4	8212
cab	3128.7	3956
iso	2979.5	7
sys	2599.4	3008
msp	1157.7	111
ZIP	1074.3	355
pdf	890.9	1066
pst	733.2	17
jpg	711.7	5647

...

# Looking for duplicate files:
##############################
# This is not efficient and can hang MySQL until the request ends.

DROP TABLE duplicates;
CREATE TABLE duplicates (
csum BIGINT
);

INSERT INTO duplicates(csum) (
SELECT csum
FROM (
SELECT DISTINCT COUNT(*), csum
FROM scan
WHERE csum !=-1
AND sizeb > 0
AND node = 'usbdisk'
GROUP BY csum
HAVING COUNT(*) >1
)qdupl
);

# This query may take some time to proceed:
SELECT path, name, sizemb, scan.csum
FROM scan, duplicates
WHERE scan.csum = duplicates.csum
AND sizemb > 1
order by 4;
############################## MySQL

B. Technical Help

12. Out of memory

TinyScan can run out of memory when scanning a disk or opening a saved scan. In this case:
- Use tinyscan_high_memory.bat.
- Scan a folder instead of a whole hard drive.
- Or choose the option "Skip files".
- You can also choose "Scan To File" if you plan to use the data with another application.

13. Caution

When scanning, TinyScan is processor intensive and does a lot of disk access.
So if you plan to scan a server:
- You should not do it at peak time.
- Or you should give the process a low priority.
- Or scan a drive remotly, which will at least put the processor load on the client.

Computing the checksum is processor intensive and makes the scan much slower. The Adler-32 checksum can be forged easily and is therefore unsafe for protecting against intentional modification.

14. Bugs

When the checksum option is activated, TinyScan can stop processing folders. So some folders won't be listed. In this case, scan a folder instead of a drive.
On Linux, the "dev" and "proc" folder give improper results. Actually all the special files, like character special, socket, etc.

15. Java

TinyScan is written in Java with the NetBeans IDE. It uses the Swing API.

C. And more

16. To do

In any order:
- Opening a file/folder when double-clicking.
- Saving in Unicode. This would allow for Chinese or Russian characters for example.
- Visual track changes in the tree.
- Scanning a web site to display its structure.

17. License

Open source, BSD license.

18. Credits

Author: Thierry Salmon.
Design contributions and many suggestions: Yves Vostier.
Some suggestions: Eric Salmon, Jean-Philippe Noël, Frédéric de Loneux and Oleg Rezabek.

19. Home Page

http://tinyscan.sourceforge.net

Voilà.