[prev in list] [next in list] [prev in thread] [next in thread] 

List:       koffice
Subject:    Re: Smaller XML files (Re: using kword and cvs?)
From:       shaheed <srhaque () iee ! org>
Date:       2001-08-26 21:42:04
[Download RAW message or body]

Thomas,

> > I think we are talking at cross-purposes. I meant import filters, I think
> > you are talking about export filters.
>
> Yap, I was. :)

Phew!

> > See my previous reply to David as to my
> > concern with import filters.
>
> Oops, allready replied to that,
> I am continuesly talking about the XML KWord writes. I am not at all
> concernted with the XML which is imported, since that obviously works
> (allthough we may have to look at it if it is not consistent)
>
> Does anyone know if the import filters are actually leaving out data from
> the XML? Just like we are proposing KWord will do?
> I was expecting the import filters to simply create all relevant data.
>
> Again, the interpretation of missing tags will not be different from
> before, the defaults will stay the '0' values, not the styles values. So if
> the current situation works, then there is no problem ;)

I cannot speak for the other import filters. For the MsWord import filter, I 
think there is a problem. As I said, the current code does write out all 
values for all paragraph runs so the problem is not visible today.

However, I was intending to use the default-value-comes-from-the-style 
behaviour to omit the attributes/tags with default values. This will be 
broken if the KWord reading code sets unspecified attributes to "0" rather 
than the value in the style on import!

Now, I can think of one solution for import filters. If they are required to 
specifiy all values, and then KWord itself takes care of checking if a style 
value has been overridden. Then all you have to worry about is handling old 
files when kword adds a new attribute - but we have syntaxVersion for that 
case.

What do you think? This idea is particularly nice for me because it means 
that the current MsWord filter should work as-is!

Thanks, Shaheed

[prev in list] [next in list] [prev in thread] [next in thread]