Hi Raif, The main reason for using manual DOM-based parsing of XML files is mainly to prevent any dependency beside JDK/JRE API.
Also, the restlet.xsd schema is trying to match exactly the structure and property of Restlet API classes. In some way, the API classes are the DAO beans... definitely a special case in my mind. Another major concern is keeping the core Restlet as compact as possible. It would be possible to add those DAO beans in the Restlet engine package though without cluttering the Restlet API itself. In a normal project, I would definitely not write the XML code by hand and prefer a generated approach like JAXB or EMF. Anyway, you make a good point regarding the support of other serialization formats (JSON would be nice). I suggest that you open a RFE to cover those requirements and remarks so we keep thinking about them in the future. Regarding the next steps for security, I'll reply to the other thread "securing Restlet" :) Best regards, Jerome Louvel -- Restlet ~ Founder and Lead developer ~ http://www.restlet.org Noelios Technologies ~ Co-founder ~ http://www.noelios.com -----Message d'origine----- De : Raif S. Naffah [mailto:[email protected]] Envoye : jeudi 25 decembre 2008 08:03 A : [email protected] Cc : Jerome Louvel Objet : Re: contributing - areas of interest hello Jerome, On Wednesday 24 December 2008 06:56:47 Jerome Louvel wrote: > Hi Raif, > > Cool! > > * Java Bindings to XML Schemas: we already support JAXB and JiBX > technologies. What would adding an extension for XMLBeans add > (pros/cons)? There are many serializations technologies to consider > (Castor comes to mind) so we need to be very selective in what we add to > the main Restlet project. i thought of this while working on the patch for supporting "init-param" (Issue #670) in the org.restlet package and not as another Extension. the current code relies only on the DOM structure of a restlet.xml file. my suggestion is to replace those bits by (a) pre-compiling schema-to-Java bindings for whatever is defined in the Components.xsd --which today is not much but could be a subject of change depending on how authentication and authorization will be configured in the future-- and (b) replacing the Component and other classes configuration and initialization based on the Data Access Objects representing a successful parsing of a restlet.xml. the current code has the advantage of not relying on other than a JDK but also has the disadvantage IMO of not separating the configuration from the other life-cycle states of software components. in addition having those Java Binding classes would allow configuring a Restlet application through other than a restlet.xml file. > * authorization toolkit: thanks for the pointer to Tomcat Realms, I've > added a comment about it in issue #505. This is definitely an area that > would benefit from contributions during Restlet 1.2 development cycle! great. > * KML/KMZ support: this might be a too application specific to fit into > the main Restlet project, but you could start a dedicated open source > project (for example on Google Code), that would leverage Restlet. We > would be happy to point users to it and support you. makes sense. what's the next step? > Best regards, > Jerome Louvel > -- > Restlet ~ Founder and Lead developer ~ http://www.restlet.org > Noelios Technologies ~ Co-founder ~ http://www.noelios.com > > > -----Message d'origine----- > De : Raif S. Naffah [mailto:[email protected]] > Envoye : lundi 8 decembre 2008 09:05 > A : [email protected] > Objet : contributing - areas of interest > > hello there, > > (i'm sending this again since it does not look like it made it to the > list the first time around). > > i'd like to contribute to this project in my free time. > > areas of interests which i've identified so far --and am personally > interested in-- are: > > * Java Bindings to XML Schemas: replace XML parsing with pre-generated > classes (XMLBeans xmlbeans.apache.org) from published/known schemas; e.g. > the parsing of the restlet.xml. > > * authorization toolkit: something equivalent to what is available with > servlets in Apache Tomcat; see http://tomcat.apache.org/tomcat-6.0- > doc/realm-howto.html. > > * KML/KMZ support: something beyond issue #677; more like a SPI toolkit > to allow building KML files (e.g. from a CSV document). > > > if this is of interest to the project maintainers, then pls. let me know > how to proceed next. > > ------------------------------------------------------ > http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=98 >1057 > > ------------------------------------------------------ > http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=99 >0910 -- cheers; rsn ------------------------------------------------------ http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=992116 ------------------------------------------------------ http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=992808
