Re: listfield in code source

After further digging, I the cause of the '\n' stripping in code sources is in 
SilvaDocument/transform/ObjectParser.py.

In parse(self, source, fix), if the type is not unicode the string is parsed 
into a dom using xml.dom.minidom.

Apparrently, minidom does not support newlines in attributes.  This can be 
tested using the following simple script:

import xml.dom.minidom
document = """\
<text attr1="link
link2">blah
</text>
"""
dom = xml.dom.minidom.parseString(document)
print dom.documentElement.getAttribute('attr1')

At this point I'm at a loss about where to go from here.  I don't think there 
is anything I can do about this.

Is there any other way I could get a list of lines from a code source in kupu?

Thanks,
Andy


On Friday 03 February 2006 9:24 am, Andy Altepeter wrote:
> Hi,
>
> I have a list field in an external source that requires newlines to be
> present (each line is a separate item).  In the forms-based editor,
> newlines are retained when the source is saved.  However in kupu, the
> newlines are stripped, which breaks this code source.
>
> Is there any way to work around this?  I could probably detect where the
> newlines are "supposed" to be, and insert them on save, but how do I do
> this? Is it possible to use an external validator method in formulator to
> change the value of a field before it is saved?
>
> I could put the newline detection in the rendering script, but then the
> list field in kupu would show one long line instead of multiple items.
>
> I have tried using a multi list field as well, but it behaves in the same
> manner.
>
> Thanks for the assistance,
> Andy
> _______________________________________________
> silva-dev mailing list
> [email protected]
> https://infrae.com/mailman/listinfo/silva-dev