development

create pdf out of rails

a small internal web app used at a company for event administration. event data is input in to a database via a web form, out of this a html form for event registration for users already registered in the companies contact database. Also needed is a simple way to generate a pdf invitation out of the rails model data, without needing to create an extra document in a word processor or similiar...

Using RDF(S) without XML.

RDF can be written in some ways: A RDF triple could be represented through a RDF graph, through FOL expressions like P(X,Y), or - most common used for web development - through XML. There are some good RDF/XML parsers for the most common web development languages like Perl or Python. But is there another way of writing RDF, which is supported by a parser for programming/script languages?

How to setup a Canoo Webtest Environment

I am working as part time softwaretester in a software development company. In the last months the main part of my work was testing the same testcases over and over again. All startet with a few testcases for a relatively small frontend. But as time goes by, the software developed and the functionality grew...and so grew the frontend of the application and the complexety of the testcases. At the beginning i only had about 30 testcases which covered the overall functionality of the application. Now i have to test about 100 testcases in the same time. So some solution had to be found. The great thing of the application i have to test is, that most of the navigation and layout stays the same, even when new functionality is implemented. That makes an automatic regression very easy. My goal was now to find a free, easy to understand, easy to install and easy to use test tool which fits my requirements. I chose "Canoo Webtests". It's an XML based test tool, which is really great. It produces very good testreports, it generates screenshots during the automated testing and it is relatively easy to install. Because its XML based the testcases are very easy to implement, and persons who are not so familiar with programming can understand and implement the testcases very easy.

Checking websites for updates

Lots of websites get updated on an irregular basis. Some may get updated every week, others get updated every hour; some may not get updated for month and users are bored looking up the website regulary and they do not find any new content. In addition, checking websites manually is not very efficient and costs a lot of time. The solution for this problem should provide an easy way to check websites and stay updated. It would be helpful if the tool is able to highlight changes. In addition, the tool should be able to exclude several areas of a website from checking, for example the current time or the "who's online" section of the website. These changes might not be interesting for the user. The tool should of course work on HTML files; it also would be good if changes in images can be detected using a tool. The Solution should also cover non-HTML content, especially flash. Flash is used a lot in the www and this technology should not be left out. However, a solution for HTML at least would work for most websites. CSS changes do not need to be included as CSS should not contain any content but just formatting information.

Managing courses and providing a registration System

The registration system should offer a specified number of courses and a limited number of possible registrations. The system has to ensure that no further registrations are possible if the places are limited. It should be possible to integrate the system into an existing website, including adaptions concerning layout and design. The registration process is limited to a specific time and registered users should be able to change their registrations. The users should be able to log on and view as well as change their current registrations. The backend should provide facilities to add and remove users and should offer registration lists which could be used to check attendance. In addition, collected e-mail addresses could be used to inform audience on time or location change. The technology used should be widespread and easy to use. It should be used for a Website for company fairs (www.discovery.jku.at) and is therefore used only occasionally. It is of major importance that the system is easy to use and to maintain because the organizing team of the fair and also the maintaining team of the website change a lot. The existing infrastructure is an Apache server and a mysql database which ideally should be used for this project as well. Nevertheless, other open source technologies could be integrated.

Using Git DVCS (distributed version control system) and GitHub to effectively collaborate on an open-source software project

While collaboratively working together on an open-source project (which aims to provide a fresh web app framework in this concrete case) and moves at a fast pace it conseqeuntly turned out that using the standard approach of centralized VCS for it soon hit its limits. Therefore, we jointly decided in the community to switch to a decentralized one, namely Git. This DVCS was originally developed by Linus Torvalds for managing the development of the Linux kernel. Its main concept basically is that everybody has his/her own complete "clone" of the code repository. So there isn't a single centralized code repository as in the structure/architecture of Subversion, but a number of decentralized ones. As a consequence Git was also developed to have very powerful and convenient capabilities of merging branches of code repository clones. This feature which is also one of Git's biggest strengths shall be described and explained in more detail. Git's basic usage is mainly command line based and can especially in the beginning pose quite a steep learning curve on new users. So it should be given a short introduction to Git's main concepts and concrete usages as well. Another very nice companion to Git is the Git code repositories hosting site/service GitHub (github.com) which can make using Git simpler and furthermore improve collaborative development processes and tasks of a programming/development project. This shall also be presented and explained.

Setting up a Subversion repository and corresponding Trac app on a (Debian/Apache) server

So the basic problem here is how to best work together productively in a programming/development project and the question which are good tools providing capabilities to potentially support and improve this. One such tool is a version control systems (VCS) which basically helps to keep the code and especially its progress under (version) control in a code repository (transparently). A specific such VCS is Subversion (a.k.a. SVN). This is at the time one of the most popular and up-to-date systems of its kind (i.e., centralized VCS). Its basic slogan is "CVS done right" (which consequently can be seen as Subversion's indirect predecessor system). There are many tools available to work with Subversion from a client/user perspective like stand-alone client apps, integration in editors, e.g., via special plugins or also complete integration in a full-blown IDE (integrated development environment). Now, what's up to do is to actually setup a Subversion system on a server and to create a code repository for a programming/development project. In this case a Debian-based OS with an Apache Web server is chosen as infrastructure. Additionally a Trac app should be installed which among other things offers nice and convenient Web visualization of the code repository (and its progress) to users. When all this is in place users can take the benefits of using Subversion for version control of the code of the project.

Pages

Subscribe to development