RE: LocationTemplate-1G test
"Jonathan Marsh" <[email protected]>
| Newsgroups | gmane.comp.web.services.description |
|---|---|
| Message-ID | <01d801c756cc$f38671a0$1501010a@DELLICIOUS> |
Summarizing this thread, this morning's discussion, and the related issues:
- [FIXED] * was improperly encoded in the baseline.
- [QUESTION 1] The spec says what characters MUST be encoded, but there are
also characters that MAY be encoded such as * (and pretty much any other
character except %). Our test suite assumes only the characters that MUST
be are. Should we change this? (I think we should do this
opportunistically, that is, if a testcase is proven to be correct, we simply
add an alternative that matches that implementation's encoding strategy. I
don't think we have any failures because of this at present.)
- [AGREED] Per the last paragraph of 6.8.1, referencing section 3.1 of RFC
3987, some further encoding is performed after the http location templates
are resolved and combined with the {address} property.
- [QUESTION 2] Is this sufficiently clear in the spec? (I think so.)
- [AGREED] Besides the extended characters encoded above, the spec says
implementations SHOULD also encode "<", ">", '"', space, "{", "}", "|", "\",
"^", and "`". Our test suite will currently assume this SHOULD has been
followed.
- [FIXED] There other editorial improvements such as removing the double
negative, reordering bullets, removing query parameter separator from
consideration before the "?".
- [QUESTION 3] Are there additional editorial improvements possible? (I
think so, as reported in
http://lists.w3.org/Archives/Public/www-ws-desc/2007Feb/0193.html).
- [QUESTION 4] Is "&" a harmful character before the "?". If not, we should
add it to the excluded list.
- [QUESTION 5] Are ";" and "=" harmful characters before the "?". If so, we
should remove them from the excluded list.
I'll research proposals for 4 and 5 per my AI, but if there are any other
questions I didn't capture here, let us know!
Jonathan Marsh - http://www.wso2.com - http://auburnmarshes.spaces.live.com
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On
> Behalf Of Youenn Fablet
> Sent: Thursday, February 22, 2007 5:35 AM
> To: keith chapman
> Cc: www-ws-desc
> Subject: Re: LocationTemplate-1G test
>
>
> +1
> If we have a location="?path={value1}&uri={value2}", the intent of the
> specification is that '&' is sent as is in the url.
> Some characters (the non-ascii ones for instance) must of course be
> encoded to form a correct URL.
> Some text may be missing in the specification to describe this precisely.
> Youenn
>
>
> keith chapman wrote:
> > Even option 1 does not speak of encoding the httplocation. It only
> > speaks of encoding instance data. This raises the issue, should the
> > location template itself be encoded?
> >
> > If we say the location template itself will be encoded then the user
> > wont hav an option of sending unencoded stuff in the url (unless
> > through templating). But if we send it raw as it is then the user can
> > do anything he wants.
> >
> > Thanks,
> > Keith.
> >
> > On 2/22/07, *Jean-Jacques Moreau* <[email protected]
> > <mailto:[email protected]>> wrote:
> >
> > Comments inline.
> >
> > keith chapman wrote:
> > > See comments inline
> > >
> > > On 2/22/07, *Jonathan Marsh* < [email protected]
> > <mailto:[email protected]>
> > > <mailto:[email protected] <mailto:[email protected]>>> wrote:
> > >
> > > I thought so too a couple of days ago, but then I found this
> in
> > > 6.8.1.1 <http://6.8.1.1> <http://6.8.1.1>:
> > >
> > > * Percent-encoding MUST be performed using the UTF-8
> > > representation of the character as prescribed by
> > section 6.4
> > > of [/ IETF RFC 3987
> > >
> > <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-
> adjuncts.html?content-type=text/html;%20charset=utf-8#RFC3987%23RFC3987
> > <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-
> adjuncts.html?content-type=text/html;%20charset=utf-
> 8#RFC3987%23RFC3987>>/].
> > >
> > > My understanding of this was that it specifies the encoding
> > mechanism
> > > to be used, when encoding is performed. I think wat it says is
> that
> > > instance data that is encoded should be encoded it in this way. It
> > > does not say anything about encoding the httplocation itself.
> > Yes, this is how the spec was meant to be interpreted (and how I
> > understood the resolution to CR117). However, since the resolution
> is
> > Jonathan's option 1 [1] in the first place, we should probably do
> > as he
> > suggests.
> >
> > Jonathan, shall I go make this more explicit in the spec?
> >
> > JJ.
> >
> > [1]
> > http://lists.w3.org/Archives/Public/www-ws-desc/2007Feb/0039.html
> > <http://lists.w3.org/Archives/Public/www-ws-desc/2007Feb/0039.html>
> > >
> > > Section 6.4 is where I got the list of characters to be
> > encoded in
> > > my earlier mail.
> > >
> > >
> > >
> > > I actually do think we should be encoding characters in the
> > > template before we stick it in a message that takes a URI
> > (if the
> > > message were to take an IRI that would be different, but AIUI
> a
> > > URI goes in the message in SOAP/HTTP).
> > >
> > >
> > >
> > > **Jonathan Marsh** - http://www.wso2.com <http://www.wso2.com>
> -
> > > http://auburnmarshes.spaces.live.com
> > >
> > >
> > >
> > >
> > --------------------------------------------------------------------
> ----
> > >
> > > *From:* keith chapman [mailto:[email protected]
> > <mailto:[email protected]>
> > > <mailto:[email protected]
> > <mailto:[email protected]>>]
> > > *Sent:* Wednesday, February 21, 2007 9:03 PM
> > > *To:* Jonathan Marsh
> > > *Cc:* Youenn Fablet; www-ws-desc
> > >
> > > *Subject:* Re: LocationTemplate-1G test
> > >
> > >
> > >
> > > Hi Jonathan and all,
> > >
> > > Section 6.8.1.1 <http://6.8.1.1> <http://6.8.1.1> [1] in
> > adjuncts does not speak of
> > > encoding characters from the httpLocation it self. As I
> > understood
> > > encoding is performed ONLY on the instance data which replaces
> > > templates. In that case I dont think that {{4(^$@!}}4^@*}}
> [2]
> > > should be encoded. Afterall its the arthor of the web
> > services who
> > > sticks in the httpLocation. May be he wanted to send those
> > > characters as it is in the URL. He should be aware of it. I
> dont
> > > think that we should be encoding the characters in the
> template.
> > >
> > > Should we make this clear in the spec or have
> > I misunderstood it?
> > >
> > > [1]
> > >
> > http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl20/wsdl20-
> adjuncts.html?content-type=text/html;%20charset=utf-
> 8#_http_operation_location_cited_ser
> > <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-
> adjuncts.html?content-type=text/html;%20charset=utf-
> 8#_http_operation_location_cited_ser>
> > >
> > <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-
> adjuncts.html?content-type=text/html;%20charset=utf-
> 8#_http_operation_location_cited_ser
> > <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-
> adjuncts.html?content-type=text/html;%20charset=utf-
> 8#_http_operation_location_cited_ser>>
> > >
> > > [2] datespace/{year}.html?passphrase={time}{{4(^$@!}}4^@*}}
> > >
> > > Thanks,
> > > Keith.
> > >
> > > On 2/21/07, *Jonathan Marsh* < [email protected]
> > <mailto:[email protected]>
> > > <mailto:[email protected] <mailto:[email protected]>>> wrote:
> > >
> > > I looked into message 13 there. The location template in that
> > > testcase is:
> > >
> > > datespace/{year}.html?passphrase={time}{{4(^$@!}}4^@*}}
> > >
> > > This testcase test not only that templates are encoded, but
> that
> > > the IRI is
> > > correctly converted to a URI per RFC3987. Many characters
> over
> > > #xA0 are
> > > %-encoded, as well as "<", ">", '"', space, "{", "}", "|",
> "\",
> > > "^", and
> > > "`".
> > >
> > >
> datespace/2006.html?passphrase=16:40:00%7B4(%5E$@!%7D4%5E@*%7D
> > >
> > > Indeed the "*" should have been omitted from the list of
> > > characters encoded
> > > during this conversion. I've updated the expected results,
> > > putting Canon at
> > > all green on this testcase. Axis2 still fails the testcase
> > because it
> > > doesn't seem to be doing the encoding listed above at all.
> > >
> > > Jonathan Marsh - http://www.wso2.com -
> > > http://auburnmarshes.spaces.live.com
> > >
> > >
> > > > -----Original Message-----
> > > > From: Youenn Fablet [mailto: [email protected]
> > <mailto:[email protected]>
> > > <mailto:[email protected]
> > <mailto:[email protected]>>]
> > > > Sent: Tuesday, February 20, 2007 5:31 AM
> > > > To: keith chapman
> > > > Cc: www-ws-desc; Jonathan Marsh
> > > > Subject: Re: LocationTemplate-1G test
> > > >
> > > > Here is canon location template 1G result.
> > > > There is one red (request 13).
> > > > It seems that our implementation is not %-encoding the '*'
> > character
> > > > while the test framework wants it to be %-encoded.
> > > > As per the current draft, '*' is not in the
> > must-be-encoded set.
> > > > An application may therefore typically choose to %-encode
> > it or not,
> > > > right?
> > > > The test framework should only check that must-be-%-encoded
> > > characters
> > > > are correctly encoded and nothing more.
> > > > I do not know whether this is easily feasible in the
> > current test
> > > > framework though.
> > > > Youenn
> > > >
> > > > keith chapman wrote:
> > > > >
> > > > > Hi Jonathan,
> > > > >
> > > > > Request 11 shows red due to a mistake in the test
> > framework. You
> > > > > forgot to add 000Z to one time element...
> > > > >
> > > > > Thanks,
> > > > > Keith.
> > > > > --
> > > > > Keith Chapman
> > > > > WSO2 Inc.
> > > > > Oxygen for Web Services Developers.
> > > > > http://wso2.org/
> > >
> > >
> > >
> > >
> > > --
> > > Keith Chapman
> > > WSO2 Inc.
> > > Oxygen for Web Services Developers.
> > > http://wso2.org/ <http://wso2.org/>
> > >
> > >
> > >
> > >
> > > --
> > > Keith Chapman
> > > WSO2 Inc.
> > > Oxygen for Web Services Developers.
> > > http://wso2.org/
> >
> >
> >
> >
> > --
> > Keith Chapman
> > WSO2 Inc.
> > Oxygen for Web Services Developers.
> > http://wso2.org/