Hello juergen, About http and https, Webdav ucp uses dav:// and davs:// as respective schemes. Should I do it as cmis and cmiss?
On Mon, Jun 24, 2013 at 6:41 PM, Rajath Shashidhara < rajaths.raja...@gmail.com> wrote: > Hello Juergen, > We could also do this by trial and error, > for example try to connect to: > localhost:8080 > if it fails > try > localhost:8080/inmemory > if it fails > try > localhost:8080/inmemory/atom > > now it connects so, > now connect to repo id:A1. > then the path of folder relative to repo can be determined. > > > On Mon, Jun 24, 2013 at 6:38 PM, Rajath Shashidhara < > rajaths.raja...@gmail.com> wrote: > >> Hello Juergen, >> I feel more than the authentication thing right now, its important to >> break the url into parts like: >> hostname, portname, <servlet-path>,repositoryid, path on repository from >> the given url. This is hard because, there seems to be no standard or it. >> https://chemistry.apache.org/java/developing/dev-url.html >> >> this page has a standard for atompub urls. >> But, the since the servletpath can be also seperated by '/' it is >> difficult to break the url into the above mentioned items. >> So, I propose something like that we define that these items must be >> separated by # in the url. >> for example: >> now the url is written as: >> cmis://localhost:8080/inmemory/atom/A1/My_Folder-0-0 >> >> here, the server address is: http://localhost:8080/inmemory/atom >> Repo id: A1 >> Path of folder on repo: /My_Folder-0-0 >> >> So we could define a way to do it this way: >> cmis://localhost:8080/inmemory/atom/#A1/#My_Folder-0-0 >> >> IS that possible? >> >> >> On Mon, Jun 24, 2013 at 4:39 PM, Jürgen Schmidt <jogischm...@gmail.com>wrote: >> >>> On 6/24/13 11:29 AM, Rajath Shashidhara wrote: >>> > Hello, >>> > >>> > Can i follow the uri format mentioned on this page: >>> > https://camel.apache.org/cmis.html >>> >>> in general yes, but I would say the configuration of the CMIS stores >>> should be http/https. Internally you will replace http/https with "cmis" >>> and will use the CMIS Url only. >>> >>> I think it is important to differentiate between http and https >>> connections. If you have better ideas please let us know and discuss. >>> >>> The next step will be to define a configuration scheme to manage 1-n >>> cmis stores. >>> >>> User credentials can be stored in the internal password container that >>> can handle Urls and related user/password combinations. >>> >>> I will search some related info for you ... >>> >>> Juergen >>> >>> >>> >>> >>> >>> > >>> > >>> > >>> > On Mon, Jun 24, 2013 at 2:53 PM, Rajath Shashidhara < >>> > rajaths.raja...@gmail.com> wrote: >>> > >>> >> Hello, >>> >> >>> >> As of now I'm by default connected to my inmemory repository and my >>> path >>> >> supplied to ucp is relative to the root of that repository. >>> >> But, as my task requires I need to develop a way to break any given >>> url >>> >> into the repository url, repository id, port, and the relative path of >>> >> object with respective to the root of this repository. This requires >>> are >>> >> decoding mechanism. >>> >> >>> >> Is there a standard way of encoding a CMIS Url so that I can decode it >>> >> into the respective components? >>> >> I want to do this before continuing on implementing other commands. >>> >> >>> >> -- >>> >> Rajath S, >>> >> M.Sc(Hons.) Physics, >>> >> Birla Institute of Technology and Science - Pilani, >>> >> Pilani >>> >> >>> > >>> > >>> > >>> >>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org >>> For additional commands, e-mail: dev-h...@openoffice.apache.org >>> >>> >> >> >> -- >> Rajath S, >> M.Sc(Hons.) Physics, >> Birla Institute of Technology and Science - Pilani, >> Pilani >> > > > > -- > Rajath S, > M.Sc(Hons.) Physics, > Birla Institute of Technology and Science - Pilani, > Pilani > -- Rajath S, M.Sc(Hons.) Physics, Birla Institute of Technology and Science - Pilani, Pilani
