W3C home > Mailing lists > Public > www-dom-ts@w3.org > February 2002

Re: DOM Level 1 HTML TS

From: Dimitris Dimitriadis <dimitris@ontologicon.com>
Date: Wed, 20 Feb 2002 18:36:41 +0100
Cc: Mary Brady <mbrady@nist.gov>, www-dom-ts@w3.org
To: stummala@netscape.com (Sivakiran Tummala)
Message-Id: <66CA6611-2628-11D6-B023-000393556882@ontologicon.com>
Hi Siva

Thanks, sounds good.

If you ned any help in the meanwhile, just send an email to the list. 
Also, check with the list for the most appropriate form to submit the 
tests to the CVS (you could either send them in zipped form to the 
submission list and have one of the moderators submit them for you if 
all are ready at once, or submit them yourselves which would mean we 
have to create an extra account. Logistics..)

/Dimitris


On Tuesday, February 19, 2002, at 01:00 , Sivakiran Tummala wrote:

> Hi,
>
> We are expecting to post around 100 level 1 HTML testcases by 15th of 
> March.
> The coverage in Level2 HTML and Level1 HTML are more or less the same.
>
> Thanks
> Siva
>
>
> Sivakiran Tummala wrote:
>
>> Hi,
>>
>> As Mary said there is no difference between level2 html and level1 html
>> testcases, all the testcases in level2 can be applied to level1 with 
>> out
>> changes
>>
>> Thanks
>> Siva
>>
>> Mary Brady wrote:
>>
>>> With respect to the NIST tests, there are
>>> two issues:
>>>
>>> 1) DOM Level 1 HTML -- these tests are pretty much the
>>> equivalent of the DOM Level 2 HTML tests - 90% of which
>>> just require pointing to the Level 1 DTD -- there are a few
>>> tweaks necessary for level 2, but not too much -- minimal
>>> effort to convert ~ 1 week.
>>>
>>> 2) DOM Level 1 Core (HTML only implementations) -- much
>>> more difficult, as the Level 1 XML tests would have to be
>>> converted -- these tests currently load an xml file and manipulate
>>> the file via dom calls.  In order to convert, the xml file must be
>>> converted to an html file, and each of the tests would have to be
>>> changed.  Most of the tests start with a getElementsByTagName call -
>>> these would have to be changed to be consistent with the HTML
>>> version of this file.  Other tests try to get attributes from various
>>> element
>>> names -- I'm not sure the best way to represent this in HTML.  So
>>> essentially the logic of these tests can be reused, but probably all 
>>> of
>>> them will have to be recoded.
>>>
>>> --Mary
>>>
>>> ----- Original Message -----
>>> From: "Dimitris Dimitriadis" <dimitris@ontologicon.com>
>>> To: <www-dom-ts@w3.org>
>>> Cc: "Sivakiran Tummala" <stummala@netscape.com>; "Mary Brady"
>>> <mbrady@nist.gov>
>>> Sent: Tuesday, February 12, 2002 9:58 AM
>>> Subject: DOM Level 1 HTML TS
>>>
>>>> We've acquired some steam woth keeping up, so once we've released the
>>>> Level 1 TS for core, we should start working with HTML.
>>>>
>>>> There's been two major options brought to the list so far:
>>>>
>>>> 1. Convert as many as possible of the existing NIST level 2 tests to
>>>> level 1 HTML
>>>>
>>>> 2. Convert some of Netscape's HTML tests to our markup
>>>>
>>>> How long would this take?
>>>>
>>>> Also, since both are open options, it seems a good idea to try to
>>>> minimize duplication by stating what tests would be relevant and 
>>>> within
>>>> what time frame.
>>>>
>>>> Best,
>>>>
>>>> /Dimitris
>>>>
>>>>
>
Received on Wednesday, 20 February 2002 12:36:02 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:34:04 UTC