Free band saw software for linux

LibBand is a simple library wrapper written in assembly and C (but I plan to rewrite everything in C) useful if you want to limit the bandwidth used by a program.
It simply overrides some libc functions, using the LD_PRELOAD environment variable, and puts some pauses in the program to clamp the bandwidth at the specified value, for both upload and download operations.
You just have to set UPLOAD_BAND and DOWNLOAD_BAND and run your program, for example:
export LD_PRELOAD=/replace-with-the-path/to/libband.so
export UPLOAD_BAND=1024
export DOWNLOAD_BAND=8192
ftp somewhere.com
The bandwidth must be expressed in bytes/s, and so in the above example, we want the upload band clamped at 1.0 kb/s, the download band at 8.0 kb/s.
Theres a small utility in the scripts/ directory, called lb: you can use it to set everything and start your program from a nice whiptail/dialog environment.
You need NASM to build the binary (but dont worry, a pre-assembled version is included in the package).
Just type "make" to compile, and "make install" to install the library.
Remember that $HOME/lib is the default installation directory, so you usually have to export LD_PRELOAD by using:
export LD_PRELOAD=$HOME/lib/libband.so
The "lb" script is configured to work with this path, but you may change it by editing the script itself (search for the LIBPATH variable).
Please note that at the moment libBand DOESNT WORK with multithreaded and _graphic_ applications.
Enhancements:
- We keep track of sockets by wrapping socket(), accept() and close(). The old fstat() mechanism has been obsoleted.

Vanila is a VERY basic SuperKaramba theme. I made it from an idea I saw on a screen shot I saw somewhere ages ago. Clicking on various bits open new themes - i.e. Click on CPU to see detailed info on your proc. Click on Net to see upload and download daily amounts.

Ink relies on a program called Ink being installed and you having a printer that Ink can talk to.

Oh yeah - the close buttons on the sub-themes dont work. I dont know how to do that but if anyone has any idea or cares to mod it so it works....

Only HTML, text and images are rendered. It has the peculiarity to be entirely written in Python, though based on many C-written pieces.
The idea of it came because I like Operas way of browsing, but it is too buggy on Linux, so when I saw someone had written a HTML rendering widget with rather few dependencies, I decided to write my own web browser.
This has led me to write other software pieces that are needed to run balayer for the most part: python-gdkinput, python-cschtml, gladepyc. The latter was independant from Balayer at start, but finally I decided that Balayer would be a good demonstration for it; At the same time, I had to implement features in gladepyc to implement Balayer, so they both benefitted from each other.
Main features:
- Single window navigation: web pages are displayed in a notebook rather than in separate Windows.
- Clicking on a link displays the new page systematically in a new tab, and does not focus it automatically. This is to ease parallel loading of pages.
- Non-blocking loading: you can continue viewing pages while documents and images are loading.
- Bookmark manager: shows and manage separate bookmark trees for each supported bookmark format. No native bookmark format, it uses others. Planned: Opera, Lynx, Netscape.
- Should be quite fast and light (memory and disk footprint); At least it is on my host.

libDSP project is C++ class library of common digital signal processing functions. There is also a wrapper for the C language.
Current (vector) functions:
- Float to integer conversion with rounding
- Addition (real, complex)
- Subtraction (real, complex)
- Multiply (real, complex)
- Division (real, complex)
- 1/x (real)
- Absolute value (real)
- Square root (real)
- Zero (real, complex)
- Set (real, complex)
- Clip (real)
- Copy (real, complex)
- Convolution (real)
- Correlation (real)
- Autocorrelation (real)
- Dot product (real)
- Min and Max (real)
- Mean (real)
- Median (real)
- Negate (real)
- Normalization (real)
- Product (real)
- Reverse
- Scale to -1..1 and 0..1 (real)
- Sort (real)
- Standard deviation (real)
- Sum (real)
- Square (real)
- Integer-Float and Float-Integer conversion: 8, 12, 16, 24, 32 bits (real)
- Cartesian-Polar and Polar-Cartesian conversion
- Cross correlation (real)
- Energy (real)
- Magnitude
- Power
- Phase
- Decimation
- Interpolation
- RMS (real)
- Variance (real)
- Peak level in dB (real)
- Window functions: Bartlett, Blackman, Blackman-Harris, Cosine tapered, Exponential, Flat-top, Generic cosine, Hamming, Hanning, Kaiser, Kaiser-Bessel, Tukey, Dolph-Chebyshev
- Mix (real)
- Spatialize (real)
- Interleave/Deinterleave (real)
- Rebuffer
- Deg-Rad and Rad-Deg conversion
- FIR filter
- IIR filter
- DFT, DCT, DST
- FFT filter
- Multirate decimation/interpolation with FFT, FIR and IIR filters
- Frequency band flipping
- Hankel

AudioTT is ID3 tag editor, which is FREE to use.
AudioTT supports both ID3 tags (ID3v1 and ID3v2). Editing ID3 tags by this application is very easy and saves your time, beacause it allows you to edit a heap of MP3 files at once.
This function is useful especialy for those who own a big collection of MP3s. Imagine that you have a discography of the QUEEN - it means hundreds of MP3 files. It would be silly to edit ID3 tag of each file separately. Therefore you may load the whole dicography to AudioTT, then write QUEEN to the Band item of ID3 tag and press Save/F12.
If you do so, AudioTT will write QUEEN as a Band to all files of your discography. In exactly the same way you may edit all the items of ID3 tag (Year, Album, Pictures, Official web site of the band/artist...).
AudioTT contains a panel, which shows the detailed information of MP3 files. It contains also another panel, which dynamicaly creates statistic for all your MP3s, which are in that moment loaded into AudioTT (Total time, Average time, Average bitrate, Average sampling rate...). If you are not interested in these information, you may simply hide the panels...
Pictures of jpeg, gif and other types represents a form of binary data similar to audio, video or text data. ID3 tag version 2 (ID3v2) has been proposed so it can contain any type of data. Therefore the new version of AudioTT is now able to handle pictures data. It means that AudioTT is now able to store any count of pictures into ID3 tag. Moreover you can specify a picture by selecting a type (Front cover, Band logotype, ...) and inserting an arbitrary comment.
This all is possible to do with a new tool - Pictures manager. It allows you also to export all images to external files (jpeg, png, gif), remove all images from MP3... This all can be done at one mose click! However, dont forget that adding too many images into ID3 tags might rapidly increase size of your MP3 files!
Main features:
- support of MP3 file format
- support of ID3v1 and ID3v2 tags
- support of 39 items of ID3 tag version 2
- full support of "Pictures", "Comments" ...
- ability to read, edit and save hundreds ID3 tags at once
- comprehensive MP3 details panel
- dynamic creation of MP3s statistic
- multilingual
- multi/cross platform

This is the IRC gateway for Microsoft(tm) Messenger.

You probably wonder what this is and why I wrote this. Some more information can be found in the DESIGN document that is included in this package.

This software is written by Johannes Verelst and is licensed through the GPL license (http://www.gnu.org/licenses/gpl.html).

Note that this is pre-alpha code, it probably does a bit that you want, but dont email me bugreports or feature requests. It is under heavy development.

If this program doesnt run try to install corectly all required perl modules yourself.

The idea for this program came when I was getting annoyed by the fact that I had to use different programs for both IRC and MSN. Since I dont know of any good MSN clients that can do IRC too (on unix, ofcourse), I started thinking about the other way around: an IRC client that can do MSN.

When I saw a piece of software written by Adam Swann that allows easy connectivity to MSN with perl, I decided to write an IRC server in perl that would connect to MSN for you. This is the result.

Sitback is actual just another tar/gzip interface. It evolved from a bash-script I wrote to handle automated backups on Linux fileservers.
The script worked ok, but it soon proved difficult and tiresome to make local modifications to the script to handle various problems or differences.
I began searching the web to find a suitable backup tool, and came up with some candidates. But none of them did exactly what I wanted, or they were just too big and filled with features that I did not need.
One of the things I needed was a way of reporting the result of a backup to office peoples So a printed report, readable by non-nerds was required.
In the end, I desided that a homebrew system was required and began to put this little tool together..
Tar has been, and is, one of the best archiving utilities around, at least on small and medium scale systems (which are what I work on), so I saw no reason to abandon tar.
It is known to be stable and produce error-free archives, why should I then start all over and invent another file-storage protocol ?
Sitback works like a kind of super-intelligent script. Just tell it what you want to backup and where to put it.. Sitback takes care of the rest, including finding the tools needed, which compression to use (if you want that), how to handle the archive device, etc. etc.
Sitback will take care of checking the files, verify the archive and maintain a little database, so that you very quickly can find out on which tape a certain file is located, without using the tape.
Sitback can run a single backup operation, or you can ask it to fork into the background and do automated backups (this is where the printed report is nice to have; no report.. or report says ERROR ..then call for support).
Sitback has a graphic interface based on ncurses, when you run it directly from the shell. Do not worry if you do not have ncurses. Sitback works fine without.
If you run Sitback as a daemon making scheduled backups, no interface is available, but the logfile contains all important messages.
Enhancements:
- Fixed bug with mail-sending.. Message body contained standalone linefeeds, violating the standard.
- Now using stat64 (if available) to allow for large files (larger than 2GB) in the backup.