W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2006

Re: automatic extraction of rq23 examples

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Mon, 10 Apr 2006 12:05:28 +0100
Message-ID: <443A3BF8.9050102@hp.com>
To: public-rdf-dawg@w3.org



Jeen Broekstra wrote:
> Dan Connolly wrote:
>> On Thu, 2006-03-23 at 11:33 -0500, Eric Prud'hommeaux wrote:
>>> On Sun, Mar 19, 2006 at 12:08:40PM -0500, Eric Prud'hommeaux wrote:
>>>> getting close. xslt [XS] numbers them and extracts them to html [HT].
>>>> perl script [PL] generates a manifest and does not yet create the
>>>> files.
>>>>
>>>> [XS] http://www.w3.org/2001/sw/DataAccess/rq23/examples-extract.xsl
>>>> [HT] http://www.w3.org/2001/sw/DataAccess/rq23/examples
>>>> [PL] http://www.w3.org/2001/sw/DataAccess/tests/data/extracted-examples/extract
>>> working now. please co and test 
>>>   http://www.w3.org/2001/sw/DataAccess/tests/data/extracted-examples/
>> Has anybody managed to reproduce EricP's results?
>>
>> Jeen, maybe?
> 
> I have not yet ran EricPs scripts to generate this stuff but I have 
> tried to run the generated test cases through my test suite.
> 
> I am getting a couple of parser errors on result set files though, 
> specifically result-5.3.ttl and result-5.5.ttl. Both test cases deal 
> with optionals but the generated result set is malformed (it contains 
> 'empty' value properties). It is easy to fix them by hand (simply remove 
> the malformed binding entries), but I guess it's better if EricP tweaks 
> the Perl script and generates new versions.
> 
> I got plenty of other errors on running these tests by the way but that 
> may be due to our somewhat shaky alpha-stage SPARQL implementation. More 
> to follow.
> 
>> EricP, can you elaborate a bit on the lifecycle? i.e. suppose you
>> edit rq23 now... what do you do to make sure you haven't broken
>> anything?
>>

Progress report on extracted-examples:

The formatting in rq23 has been broken : all results are now
on font size smaller than they shoudl be and as a consequence (in FF)
some examples are unreadable.

==== 18 tests have no dataset:
Failure:  spec-2.1.6: No dataset
Failure:  spec-2.1.6-q1: No dataset
Failure:  spec-2.1.6-q2: No dataset
Failure:  spec-2.1.6-q3: No dataset
Failure:  spec-4.1: No dataset
Failure:  spec-4.1-q1: No dataset
Failure:  spec-10.1.3: No dataset
Failure:  spec-10.1.3-q1: No dataset
Failure:  spec-10.1.4: No dataset
Failure:  spec-10.1.5: No dataset
Failure:  spec-10.3.2: No dataset
Failure:  spec-10.4.1: No dataset
Failure:  spec-10.4.2: No dataset
Failure:  spec-10.4.2-q1: No dataset
Failure:  spec-10.4.2-q2: No dataset
Failure:  spec-10.4.3: No dataset
Failure:  spec-11.6: No dataset
Failure:  spec-11.6-q1: No dataset

==== 3 tests have no query:
spec-2.1.7
spec-7.1-d2
spec-7.1-d5

==== 14 tests have no results (file empty except for comment)
   These are non-SELECT queries.
Test: spec-10.2-r1
Test: spec-10.3
Test: spec-10.3.1
Test: spec-10.3.2
Test: spec-10.3.2-q1
Test: spec-10.3.3
Test: spec-10.4.1
Test: spec-10.4.2
Test: spec-10.4.2-q1
Test: spec-10.4.2-q2
Test: spec-10.4.3
Test: spec-10.5
Test: spec-10.5-r1
Test: spec-10.5-r2

==== Dataset problems with named graphs

Test: spec-8.1

1/ The manifest says the data is for the default graph when
they should be named graphs for the dataset.

2/ Manifest does not express the fact that the data comes from a named graph
but is read from a file.

Test: spec-8.2
Test: spec-8.2
Test: spec-8.3
Test: spec-8.4

As spec-8.1

Test: spec-9.1
Test: spec-9.2
Test: spec-9.3

The manifest identifies the dataset but the queries have FROM/FROM NAMED to
identify the dataset.

==== 5 places where rq23 is wrong.

(what's the process for making changes now we are in CR?)

Test: spec-2.1

2.1.6 are syntax examples and not queries to be executed
on data. Presumably the HTML div structure needs to
reflect this but the appearance and the extraction are
intertwined at the moment.


Test: spec-2.9-r1

=======================================
Failure: Test 13 :: spec-2.9-r1
Got: 0 --------------------------------
----------------
| book | title |
================
----------------
Expected: 1 -----------------------------
-------------------------------------------------------
| title             | book                            |
=======================================================
| "SPARQL Tutorial" | <http://example.org/book/book1> |
-------------------------------------------------------

rq23 is wrong - there is a space in " Bob"


Test: spec-5.3

=======================================
Failure: Test 23 :: spec-5.3
Got: 2 --------------------------------
--------------------------------------------------------------------------
| name    | mbox                      | hpage                            |
==========================================================================
| "Alice" |                           | <http://work.example.org/alice/> |
| "Bob"   | <mailto:bob@work.example> |                                  |
--------------------------------------------------------------------------
Expected: 2 -----------------------------
-------------------------------------------------------------------------
| hpage                            | mbox                     | name    |
=========================================================================
|                                  | <mailto:bob@example.com> | "Bob"   |
| <http://work.example.org/alice/> |                          | "Alice" |
-------------------------------------------------------------------------

rq23 is wrong.
I have corrected this before (in response to a comment) but it seems
to have been overwritten.


Test: spec-6.1-r2

=======================================
Failure: Test 27 :: spec-6.1-r2
Got: 0 --------------------------------
------------------
| title | author |
==================
------------------
Expected: 2 -----------------------------
----------------------------------------------
| title                            | author  |
==============================================
| "SPARQL Query Language Tutorial" | "Bob"   |
| "SPARQL Protocol Tutorial"       | "Alice" |
----------------------------------------------

rq23 is wrong.
(There is no author information in the data)


Test: spec-11.4.1

=======================================
Failure: Test 60 :: spec-11.4.1
Got: 1 --------------------------------
---------
| name  |
=========
| "Bob" |
---------
Expected: 1 -----------------------------
-------------
| givenName |
=============
| "Bob"     |
-------------

** rq23 is wrong


Test: spec-11.4.7

=======================================
Failure: Test 67 :: spec-11.4.7
Got: 1 --------------------------------
----------------
| name  | size |
================
| "Bob" | 42   |
----------------
Expected: 1 -----------------------------
--------------------
| shoeSize | name  |
====================
| 42       | "Bob" |
--------------------

** rq23 is wrong
Received on Monday, 10 April 2006 11:05:41 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:26 GMT