Free provider software for linux

Template::Provider::OpenOffice is a OpenOffice (ODT) Provider for Template Toolkit.

This module extends Template::Provider to automatically extract the content.xml file from an OpenOffice zip file and run it through Template::Toolkit for processing.

We use OpenOffice::OODoc to actually open the document and extract the content.xml file. This gives us the benefit of having the methods available to add/subtract files in addition to parsing and processing the content file if we want to do some custom work in addition to templating.

Direct Access Provider Library in short DAPL is a transport-independent, platform-independent, high-performance API for using the remote direct memory access (RDMA) capabilities of modern interconnect technologies such as InfiniBand, the Virtual Interface Architecture, and iWARP.
This is the reference implementation of that specification by the DAT Collaborative who maintain the DAPL API.
Enhancements:
- Support was added for new IB verbs client register event.
- Some problems supporting create qp without recv cq handle or recv qp resources were fixed.
- Some timeout and long disconnect delay issues exposed during scalability tests were fixed.
- Minor build problems were fixed.
- Support for multiple IB devices was added to dat.conf to support IPoIB HA failover.
- Support for returning local and remote port information was added.
- The RPM spec file was cleaned up.
- Enviroment variables to configure connection management timers were added.

mbridge provides a multicast DNS services bridge.

mbridge provides a bridge between LANs for multicast-announced services. As an example, you can use it to share iTunes libraries outside of your LAN with your peers.

It works by querying for and listening to announcements for instances of the desired service using the multicast DNS (mDNS) service discovery protocol.

Upon discovery of a new instance, it announces itself as the provider for that service on all the other LANs.

It then acts as a proxy between the clients and the real server.

How It Works

mbridge queries for and listens to announcements for instances of the desired service using the multicast DNS (mDNS) service discovery protocol.

Upon discovery of a new instance, mbridge announces itself as the provider for that service on all the other LANs. It then acts as a proxy between the clients and the real server.

phpMyID is a small, fairly lightweight, standalone, single-user Identity Provider for OpenID authentication.
OpenID is an open, decentralized, free framework for user-centric digital identity (I stole that from their website). But what does it mean? Well, basically OpenID is a way to authenticate yourself to various places (websites) by verifying your identity as the owner of a particular URL (say, a website of your own).
Instead of giving a username and password to a login form, you just give it your URL. Youre then directed to your "identity provider" to log in, and when it authenticates you, you go back to the site you were orignally trying to get into. Why this is good or bad, and what the actual implications of it are... well, thats all mostly outta scope as far as phpMyID is concerned. Suffice it to say, phpMyID acts as an "identity provider" so you can log in to OpenID enabled sites.
From a user point of view, OpenID is a neat concept, but it has a few flaws. For example, when I wanted to actually, yknow, get an OpenID thingie of my own so I could log into OpenID sites, I found there was no satisfactory way to get one. My options were:
Download a ready made application to handle serving up identities and create myself an account in it. - There arent many of these, and what few there are seem to be geared towards multi-user setups. Since theres only one of me, this option seemed like overkill.
Download a complex set of libraries and build an application to handle serving up my identity. - Again there are not many of these, and what few there are seem a bit overcomplicated for my needs.
Set up an account with a third party provider. - A bad option for me because I really dont like the idea of coupling a theoretically decentralized framework to one of a handfull of third party providers (and giving them my account credentials). Sure, someone who doesnt have their own site and server technology will probably go this way, but not me.
Since I couldnt find the simple solution that I wanted, I did what any [idiot|geek] would do, and created the missing option for myself: a single user OpenID server. phpMyID.
Usage
phpMyID is a single user (though, if you were so inclined, you could easily turn it into a multi-user setup) IdP, or "Identity Provider" for the OpenID framework. Its a single PHP script with minimal dependancies.
You dont need a database, you dont need to make your filesystem writable, you dont need to download any libraries, and you dont need to recompile PHP. Okay, well, you shouldnt need to do any of that.
Installing phpMyID requires an MD5 hashing utility. Why? Because you have to authenticate to it using a password. phpMyID uses HTTP Digest authentication for security and your password must be encrypted when you enter it during installation. Say it with me: "passwords should never be stored or transmitted in plain text" (one of the advantages of phpMyID and OpenID is that they never are).
For Linux or OSX (or any other Unix-like OS), I suggest using OpenSSL to encrypt your password. For Windows, there are a number of utilities available, but I recommend this one by Colin Plumb. Its public domain code, and it will do exactly what you need (yes, the hash it create is all upper-case - dont worry, phpMyID will convert it for you). You can use PHPs md5 function to generate your hash for you on the fly, but I must discourage doing so. Not only does it take all the fun out, but you have to store your password in plain text to make it go.
Complete installation instructions, including examples of how to encrypt your password, are available in the provided README file (heres the svn version).
Enhancements:
- This release should fix the outstanding (known) "bad signature" errors, adds a number of refinements, includes a testing mode, and provides support for a more versatile set of "big math" functions.

myOpenID is a small, fairly lightweight, standalone, single-user Identity Provider for OpenID authentication. The project comprises a single PHP script that can be used by one individual to run their own personal OpenID "IdP."
The program requires no external libraries, and has very minimal requirements. It should run on almost any server that runs PHP 4 or 5. It requires no database, and does not access the file system beyond session handling.
What matters is that OpenID is a promising new technology, but from a user porint of view, its a pain the butt to manage. When I wanted to actually, yknow, get an OpenID thingie of my own, I found there was no satisfactory way to make it happen. My options were to either download a complex set of libraries and build an application to handle serving up my identity, or to set up an account with a third party provider. Since I dont like bulky libraries, and I really dont like the idea of re-centralizing a decentralized framework through trusting someone else with my account credentials... I did what any good Geek would do. I made myself a third option: myOpenID.
Usage:
myOpenID is a single user (though, if you were so inclined, you could easily turn it into a multi-user setup) IdP, or "Identity Provider" for the OpenID framework. Its a single PHP script with minimal dependancies. You dont need a database, you dont need to make your filesystem writable, you dont need to download any libraries, and you dont need to recompile PHP.
Okay, well, you shouldnt need to do any of that. Installation requires an MD5 hashing utility. Why? Because you have to authenticate to it using a password. myOpenID uses HTTP Digest authentication for security, and you set your password must be encrypted when the script is set up. For Linux or OSX, I suggest OpenSSL. For Windows, there are a number of utilities available. I suggest this one by Colin Plumb will do exactly what you need.
Enhancements:
- Added support for PHP4, including a better mechanism for getting Digest authentication headers which should work with non-Apache servers
- Added a (hidden) logout mode, usable by calling myOpenID.php?openid.mode=logout
- Fixed a bug which occurred when associate mode was called and was unable to produce a secure key. Future queries from that handle would then fail check_authentication mode queries
- The lifetime for a smart-mode client is now same as the default cache life of the session minus ten seconds, which should be just under three hours
- Renamed the sha20 function to sha1_20 (to designate using sha1 to get a 20 byte hash)
- Some code and debug cleanup

Embperl is a framework for building websites with Perl.

For the beginner its an easy to setup and use way of embedding Perl code in HTML pages. Embperl delivers several features that ease the task of creating a websites, including dynamic tables, formfield-processing, session handling, escaping/unescaping, caching and more.

If your demands grow it gives you the power to make your Website object-oriented and build it out of small reusable components. If you dont like the idea of mixing up all your layout and code then Embperl also supports separating it in different objects (e.g. creating an MVC application).

Of course Embperl doesnt tie you to HTML - it allows components to be from different source formats (e.g. HTML, WML, XML, POD, ...) and if necessary transforms them (for example via XSLT) to other output formats. This is achieved by dividing the output generation into small steps, where each is processed by a plugable provider. Advanced users can create their own syntax definitions (for example tag libraries) and extend Embperl by writing their own providers and much more...

Embperl is a server-side tool, which means that its browser-independent. It can run in various ways: under mod_perl, as a CGI script, or offline.

tcpproxy project is a proxy (or tunnel or redirector) for TCP/IP protocols. In standalone mode it waits for incoming connections forwarding them to another machine or starting a local server program.
Several programs with this function or something similiar are around. However, tcpproxys design goal was to let it operate on some kind of firewall.
Main features:
- Extensive logging to syslog,
- Interface based configuration,
- can bind to a particular interface on a multi-homed host,
- sets environment variables before calling a local server program,
- support for external access control programs,
- can be started from inetd or run in standalone mode.
tcpproxy was created with a transparent TCP proxy in mind. When its used to start local server programs (e.g. an FTP server) it can however also work as "port multiplexer" since it requires different configurations for different interfaces (there are no defaults).
Interface based configuration
tcpproxys services are always bound to a certain interface. Suppose you have a multi-homed host (e.g. a firewall) with the IP numbers 192.168.0.1 (part of your LAN) and 10.11.12.13 (connected to the Internet). The configuration
port 119
interface 192.168.0.1
server news.provider.com
forwards then any connection made to your local interface on the NNTP port to the machine news.provider.com. The providers news server appears now to run on your firewall. Furthermore, if you for each port only a single interface where you want to have tcpproxys service, tcpproxy will not even bind to the others. For the example above this means that someone trying to connect to your external interface would only see a closed port.
Now suppose you want to use a second NNTP server from your LAN. You would first configure a second IP number on your internal interface, e.g. 192.168.0.2 and then reconfigure tcpproxy:
port 119
interface 192.168.0.1
server news.provider.com
interface 192.168.0.2
server news.freshmeat.com
Depending on the incoming interface of a client request the connection is forwarded to one of the servers.
In this case the firewalls external interface is opened on port 119 and a port scan would show that theres some kind of service. If however someone connects to the outer interface the connection is immediatly dropped, simply because tcpproxy isnt configured to handle request on the interface 10.11.12.13 and tcpproxy doesnt accept service defaults.
If you like you can extend this configuration to
port 119
interface 192.168.0.1
server news.provider.com
interface 192.168.0.2
server news.freshmeat.com
interface 10.11.12.13
exec /bin/date
for the scanners amusement. But you might also want to write a message to your systems syslog.
Access control
tcpproxy implements access control by calling external, user provided, script, the so called "access control programs" (or in short: acps). I implemented them because I wanted to be able to deny service usage based on anything I like, not just on the clients IP number or its name.