Re: [xml] help with xmlTextReaderRead (with DTD validation)

[Daniel Veillard <veillard redhat com>]

On Tue, Jul 29, 2008 at 07:31:27AM -0400, Denis Bourgeois wrote:
> Although I have contributed to a few (free) scientific applications
> (in C and F77) for quite some time, I'm new to xml and libxml2 (this may
> explain the naiveté of my question): I'm writing a C parser for an XML
> input (input only, no file editing or writing out) - so the xmlTextReader
> API (http://xmlsoft.org/xmlreader.html#L1142) seems very appropriate. I
> have gone through the tutorial and am now in the process of integrating
> DTD validation. The DTD validation correctly flags errors, but I can't
> seem to add appropriate code to halt the scanning process once a first
> parsing error occurs (the code below - lightly adapted from the tutorial -
> simply goes on and on, parsing the entire file despite the initial DTD
> parsing errors): --- C code ---

  That's normal. DTD validation error are not fatal errors, so from
a parsing viewpoint it's normal to continue processing.
  Add an xmlTextReaderIsValid() test in your reading loop and break if
it returns a value != 1

> once a first parsing error occurs? Question 2: The screen output indicates
> that the DTD error (lines 7. versus 12.) is immediately flagged after
> the first xmlTextReaderRead call (line 24.); if the xmlTextReader only
> scans one element node at a time, how can it initially flag this error
> (which only occurs later)? Is it building an entire DOM tree to do

  No the reader builds a sliding subset of the tree as it progresses,
unless of course your document is tiny it will never be fully loaded in
memory.

Daniel

-- 
Red Hat Virtualization group http://redhat.com/virtualization/
Daniel Veillard      | virtualization library  http://libvirt.org/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine  http://rpmfind.net/