From: Nicolas G. <nik...@gm...> - 2012-11-07 12:17:25
|
Hi, I have browsed the 1.7.1 code on GitHub, and I've seen quite a couple of interesting things with the new AccessPoint class hierarchy. However the code is nearly undocumented, so I'd appreciate a couple of explanations. First let me explain what I'm trying to do. The infrastructure we use for Wayback at the French National library implies that Wayback acts in proxy mode, but with a single proxy (the proxy is configured once and for all in the browser). We have two additional needs: - also be able to process ArchivalUrl replay requests (as we can use these as permalinks) - be able to choose which WaybackCollection we want to search in, without changing the browser's proxy This worked with Wayback 1.4.1, but the code was quite a hack, since AccessPoint was hardly extensible. I have adapted this code and enhanced it a bit for 1.7.1, but still am not satisfied with the design. Now looking at AccessPointAdapter and ProxyAccessPoint classes, I believe there might be a clean way to implement those functions. Can someone please provide a bit of background about the new AccessPoint class hierarchy? Configuration examples maybe? Thanks in advance, -- Nicolas Giraud --------------------------------------------------------------------------------------------- Développeur Archives du Web - Bibliothèque Nationale de France Web Archiving Developper - National Library of France --------------------------------------------------------------------------------------------- |