Dims
Right now the latest Synapse Configuration Language docs are on the wiki
(http://wiki.apache.org/incubator/Synapse/InProgress/SynapseConfigurationLanguage)
as we are not changing the docs for M2..and currently some changes are
taking place. Shall I place the Wiki src for the above page under Xdocs
instead of its XML/Html?
thanks
asankha
Davanum Srinivas wrote:
Asankha,
Could you please check in a version (html or xml) to svn under xdocs?
So that we can track the changes. I'd like to help rewrite some of the
sentences to better describe what the functionalities are. Let's nail
it down then see if we want to do the xsd thing or not.
thanks,
dims
On 7/15/06, Sanjiva Weerawarana <[EMAIL PROTECTED]> wrote:
We'll have to ask Asankha what he had in mind .. in my mind it was
always that @source was used to select what to apply the filter to and
xpath or regex were ways of filtering.
Sanjiva.
On Fri, 2006-07-14 at 21:19 -0400, Davanum Srinivas wrote:
> Hi,
>
> Forget about XSD and validation....Am just trying to decipher the
> intent from the verbiage / examples on the wiki page...Please bear
> with me. I think i am getting the point. But just for clarification.
>
> #1: The example on the page says:
> <filter xpath="//*[wsx:symbol='MSFT']"
xmlns:wsx="http://www.webserviceX.NET/";>
>
> So you are saying that it is implicitly
>
> <filter source="/soap:Envelope" xpath="//*[wsx:symbol='MSFT']"
> xmlns:wsx="http://www.webserviceX.NET/";>
>
> Because you need to run the xpath on something and that something is
> the soap envelope. Right?
>
> #2: If #1 is true, then the syntax on the wiki has an implicit source,
> but one cannot specify a source explicitly when xpath is specified.
> Right?
>
> Thanks,
> dims
>
> On 7/14/06, Sanjiva Weerawarana <[EMAIL PROTECTED]> wrote:
> > On Wed, 2006-07-12 at 13:00 -0400, Davanum Srinivas wrote:
> > > Let me try another variation to see if folks like it:
> > >
> > > Syntax:
> > > <filter source="xpath" [regex="string"]>
> > > mediator+
> > > </filter>
> > >
> > > Explanation:
> > > If the regular expression is present then the <filter> mediator
> > > matches the evaluation result of a source xpath expression
against the
> > > given regular expression. If the regex is absent, then the
mediator
> > > tests the given source xpath expression as a boolean
expression. In
> > > either case, If the test succeeds, the filter mediator will
execute
> > > the enclosed mediators in sequence.
> >
> > This is not what @source means .. it identifies the data to
filter; the
> > "source" for filtration. If @source is missing, the data is
> > the /soap:Envelope element. If its there it can point to any
place in
> > the message and then the regex or xpath matches against that node.
> >
> > So your proposal has lost function.
> >
> > Again, I'm -1 to the *principle*: If the language syntax is bust
let's
> > fix it. But let's *not* convolute the syntax to make a nice(r)
schema.
> >
> > Sanjiva.
> >
> >
> >
---------------------------------------------------------------------
> > 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]
