Hey Norman, > I have an existing python based web application and wanted to know if Wt can > be used to build a similar site for a different market, but similarly > structured? > > I have some questions as to how Wt handles certain requirements. > > 1) Storage of data - Currently my 'database' is the actual file system and I > store everything as XML files, for example: >
<snip> > > Here is an example of a .metadata file for a company entry, i.e. > abakuc.metadata: > > <?xml version="1.0" encoding="UTF-8"?> > <metadata xmlns:abakuc="http://xml.abakuc.com/namespaces/abakuc"; > xmlns:dc="http://purl.org/dc/elements/1.1/";> > <abakuc:branch_manager>0</abakuc:branch_manager> > <abakuc:type>other</abakuc:type> > <owner>0</owner> > <abakuc:topic>technology</abakuc:topic> > <dc:description xml:lang="en"></dc:description> > <abakuc:website_is_open>1</abakuc:website_is_open> > <format>company</format> > <dc:title xml:lang="en">Abakuc Limited</dc:title> > <version>20040625</version> > <abakuc:website>http://abakuc.com</abakuc:website> > </metadata> You will need to read these XML files using an XML parser. Wt uses MiniXML for internal purposes, but many alternatives are available. For example, the 'expat' parser gets good critics, although I have never used it myself. The Xerces-C++ parser is another option, although it is a beast to build and use. > 2) Also, now the application allows each user of the company to be able to > fully manage their 'directory' listing and also have their own front end URL > with their own unique 'skin'. > > This is done using Apache ReWrite rules, for example: > > I have three sites: > > 1) A directory site with all companies and users: > > http://uk.expert.travel > > <VirtualHost *:80> > ServerName uk.expert.travel > RewriteEngine On > RewriteRule ^/(.*) http://localhost:8080/expert/$1 [P] > RequestHeader set X-Base-Path expert/ > RewriteLogLevel 9 > RewriteLog logs/uk_expert_rewrite.log > ErrorLog logs/uk_expert_error.log > </VirtualHost> <snip> I think you could keep this Apache setup, and redirect to a wthttpd or another httpd daemon (with FastCGI) which runs the (single) Wt application. It is probably more convenient to redirect to URLs like http://localhost:8080/app.wt/expert..., where app.wt is the deployment path of the application. Everything that extends after the deployment path can be accessed within Wt using WEnvironment::pathInfo() [which I now notice wasn't implemented yet, I've added it just now to CVS]. What is the purpose of X-Base-Path ? Currently, you would not be able to access this header from within Wt -- only selected (standard) headers are saved into the Environment. > I like to use Wt, as there is a reporting module which allows the Manager to > analyze who's registered and how they are performing on the specific > training programme - here is a rough example, which does not work > > http://uk.tp1.expert.travel/;show_users > > But I would like to have an overview with nice charts, pie-charts etc... > built depending on what the Manager selects. I think that should all be relatively straight forward. Regards, koen ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ witty-interest mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/witty-interest
