Re: WOLips HTML validation won't go away

From: Mike Schrag (mschra..dimension.com)
Date: Tue Jul 03 2007 - 00:33:26 EDT

  • Next message: Alex Cone: "Re: WOLips HTML validation won't go away"

    Yeah ... So the deal here is that I actually hacked the crap out of
    this parser to support many wacky things that people do.
    Unfortunately this one was a little tricky to do. I support the
    variant of this where there are no quotes around the name =
    SelectContainerID inside the quoted ID value. However, there is a
    bug at the moment if you use the refactor=>reformat on the HTML file
    where it spits out an extra child node (which is actually how I
    represent that in the model in memory). In 5.4's new parser, it
    wouldn't even try to parse this, but I wanted to attempt to get as
    close as possible to the way WO 5.3's parser works in the validator.
    The fundamental problem is that to be able to do fancy things with
    refactoring, I need an in-memory XML-ish model, but WO components
    really aren't -- they don't resolve into a tree, because WO runtime
    actually is a two-pass parser that doesn't care about a validating
    tree, so I just had to cut some corners to try to shoehorn components
    into an actual dom by whatever means necessary. This case in
    particular was one I was not able to easily resolve. I do have some
    ideas for it, but I just haven't gotten around to implementing them.

    On Jul 3, 2007, at 12:27 AM, Alex Cone wrote:

    > many of the files don't have valid html - at least as I would write
    > it:
    >
    > <div id="<webobject name="SelectContainerID"></webobject>"
    > style="display:none; position: absolute;">
    > <select size="10" id="<webobject name="SearchResultsID"></
    > webobject>"></select>
    > </div>
    >
    > ...webobject tag missing 'name' attribute
    > ...</select> occurred before </webobject>
    > ...</div> occurred before </select>
    >
    > but it is the client's wo components and I just inherited the
    > project. Additionally there indeed seems to be some actual bad
    > HTML which I will have engineers clean up.
    >
    > On the up side, after several rounds of quitting Eclipse and
    > removing .markers and .markers.snap files from the projects in
    > question and restarting I was ultimately able to shut off the HTML
    > validation.
    >
    > Now that I can isolate the compile errors in the java I will have
    > the team remove the actual HTML errors and maybe we can get a fully
    > clean build. Heck, I may even try to get the warnings fixed
    > too :-) Some of the frameworks belong to other teams, but I'll
    > start the good habits at home first.
    >
    > Thanks all!
    >
    > abc
    >
    > On Jul 2, 2007, at 8:25 PM, Mike Schrag wrote:
    >
    >> Can you send me one of these components? My first thought is that
    >> it actually is just invalid HTML, and I don't recall offhand if
    >> it's possible to get amu other validation if it's structurally
    >> invalid. I believe it depends on exactly how invalid it is, but I
    >> can give you a better idea after seeing an example (if that's
    >> possible).
    >>
    >
    >
    > __alex cone
    > ceo c o d e f a b inc
    > ab..odefab.com
    > 212.465.8484 x101
    > http://www.codefab.com
    >
    > If you are not living on the edge, you are taking up too much space.
    >
    >
    >



    This archive was generated by hypermail 2.0.0 : Tue Jul 03 2007 - 00:34:21 EDT