Re: [Hypercontent-users] NekoHTML

[Alex V. <av...@co...>]

On Mar 11, 2004, at 3:07 PM, Pete Boysen wrote:

> Way ahead of you.  I revised my "TidyHelper" to use Neko and your 
> parseHTMLFragment.  Works great.  I have it already installed for 
> users.
>
> I have had a few other minor glitches:
> 1.   I don't think you popup any dialogs for tables etc. but they 
> always come up too narrow.

I found the HTMLArea table editing to be pretty buggy, which is why I 
turned it off in HyperContent.

> 2.  This may be my own configuration problem but just before the 
> content displays in the editor, I get an Action Cancelled error in IE 
> but the content appears correctly.

Haven't seen that one.  I have integrated some of the HTMLArea cvs 
updates into the HyperContent cvs - I know some other browser oddities 
have been fixed.  You may want to try the updates.

> 3.  The width of the actual editing area is sometimes smaller than the 
> toolbar.

You can always configure line breaks in the toolbar . . .

> 4.  I may have to impose quotas at some point and there is no 
> File.length().  I suppose I could get the byte array and get its 
> length but that would be too costly.

This is a tough one.  I chose not to implement this originally because 
it is difficult to find an inexpensive way of implementing this.  It 
wouldn't be too difficult to track the number of bytes written and 
store it with the file edition properties, but what do we do with File 
objects constructed dynamically from IDataLoaders?  It would become a 
burden on the IDataLoader interface to add a length() method, and in 
many situations that could only be properly computed by cycling through 
all the data in the inputstream and adding up the byte count . . . also 
not too efficient.  You can always call the available() method on the 
inputStream, which for IO files and byte arrays almost always returns 
the total number of bytes.

I'll think about this a bit more.  I'm planning on releasing the 
current CVS code as-is, barring last minute bug fixes, as v1.3 within 
the next week.  After that, perhaps we can evaluate this possible 
extension of the filesystem package more in depth.

>
> At 01:57 PM 3/11/2004, you wrote:
>> Pete,
>>        I just wanted to point out the "parseHTMLFragment" method in 
>> the HyperContent DocumentFactory - you may want to look at that for 
>> an example of how to configure and use neko to parse html.  You may 
>> or may not find my implementation meets your needs; I have it 
>> configured to strip out all style and javascript content.
>>
>> Alex
>>
>>
>>
>> -------------------------------------------------------
>> This SF.Net email is sponsored by: IBM Linux Tutorials
>> Free Linux tutorial presented by Daniel Robbins, President and CEO of
>> GenToo technologies. Learn everything from fundamentals to system
>> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
>> _______________________________________________
>> Hypercontent-users mailing list
>> Hyp...@li...
>> https://lists.sourceforge.net/lists/listinfo/hypercontent-users
>
> Pete Boysen    Iowa State University  pb...@ia...
> (515)294-6663  Ames, Iowa 50011 http://www.public.iastate.edu/~pboysen