Re: Retrieve url path (host name plus path)
florent andré <[email protected]>
| Newsgroups | gmane.comp.cms.lenya.user |
|---|---|
| Message-ID | <[email protected]> |
Ok,
Choose another design process [1] :
- get siteFragment
- transform uri in site fragment with parametrized uuid2url and proxy
transformer
- create sitemap.xml from modified siteFragment
This process work well, and is more "stable" I think but the problem is :
When you *don't have* a lenya proxy definition, url rendering is *only
relative* and not absolute.
Considering like a proxy tranformer bug or this is normal ?
Thanks
Have a good day
[1]
* can be see here :
http://svn.4sengines.eu/modcontrib/seo/branches/0.1/sitemap.xmap
* description, installation, etc, is here :
http://wiki.apache.org/lenya/ModulesIdeas
any feedback is welcome.
florent andré wrote:
> hi,
>
> Here comes my experiences (can be usefull) and my problem
>
> on localhost, no *lenya* proxy stuff here :
> - {request:requestURI} give me /lenya/default/live/sitemap.xml
> - {request:serverName} :: localhost
> - {request:scheme} :: http
> - {request:serverPort} :: 8080
> - {request:servletPath} :: /default/live/sitemap.xml
> - {request-header:host} :: localhost:8080
>
> - so in that case parameter definition is :
> {request:scheme}://{request-header:host}{request:requestURI}
>
> * BUT, *with this lenya proxy* definition :
> <proxy area="live" ssl="false" url="http://toto.com"/>
>
> * answer for {request:scheme}://{request-header:host}{request:requestURI}
>
> * is : http://toto.com/lenya/killbookmakers/live/sitemap.xml
>
> * Arrggg, not ok ! :)
>
> * so, new test :
>
> * {proxy:{request:scheme}://{request-header:host}{request:requestURI}} :
> throw an error
> * {proxy:{request:requestURI}} :: /lenya/lenya/toto/live/sitemap.xml
> * {proxy:{request:servletPath}} : http://toto.com/sitemap.xml
>
> COOOOLL ! I find it : {proxy:{request:servletPath}}.
>
> So, no generic solution when proxy definition or not...
> * no proxy : {request:scheme}://{request-header:host}{request:requestURI}
> * proxy : {proxy:{request:servletPath}}
>
> Someone know or can point me a solution for generic definition if i'm
> behind proxy or not ??
>
> cheers
>
>
>
>
>
>
> [email protected] wrote:
>> Hi --
>>
>> Have you checked out the Cocoon wiki on input modules? The request
>> input module should do what you want, e.g.
>> {request:requestURI}?{request:queryString} would return the entire
>> URL, including request params. If the Apache proxy messes with the
>> headers, you can always set ProxyPreserveHost On.
>>
>> Ref: [ http://wiki.apache.org/cocoon/InputModules ] (and no, it's not
>> obsolete, although it says so at the top of the page).
>>
>> Hope this helped,
>>
>> /Petteri
>>
>> On Apr 15, 2010 3:32pm, Florent André
>> <[email protected]> wrote:
>> > Hi all,
>> >
>> >
>> >
>> > As I currently work on google's sitemap generation, I have to know the
>> >
>> > host name and path of the request.
>> >
>> >
>> >
>> > I have a look into inputs modules but I only see one that gave
>> information
>> >
>> > about "lenya document" (required to have an uuid).
>> >
>> >
>> >
>> > In my use case, I don't have a lenya doc... Here comes the design main
>> >
>> > lines :
>> >
>> >
>> >
>> > - url finish with sitemap.xml (for example
>> http://example.com/sitemap.xml
>> >
>> > , http://www.example.com/toto/foo/sitemap.xml)
>> >
>> >
>> >
>> > - publication's sitemap.XMAP intercept sitemap.xml ending and mount a
>> >
>> > sitemap.xmap in seo module
>> >
>> >
>> >
>> > - in seo module, I have
>> >
>> > -- a match on sitemap.xml
>> >
>> > -- a navigation module generator
>> >
>> > -- a navigationToSitemap transformer
>> >
>> >
>> >
>> > - into this transformer I have to know the entire url for generate
>> a good
>> >
>> > sitemap.xml, I think to do this throw a
>> > value="{anInputModule:key}">
>> >
>> >
>> >
>> > The base url retrive have to be "proxyied", have to be the same as the
>> >
>> > client see...
>> >
>> >
>> >
>> > Any help would be great appreciated,
>> >
>> >
>> >
>> > Have a good day
>> >
>> > ++
>> >
>> >
>> >
>> > ---------------------------------------------------------------------
>> >
>> > To unsubscribe, e-mail: [email protected]
>> >
>> > For additional commands, e-mail: [email protected]
>> >
>> >
>> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>