Re: [xml] redicting parts of trees


On Thu, 2005-06-16 at 09:10 -0400, Rob Richards wrote:

I attached the current diffs for tree.c and tree.h.

There would be three public functions:

xmlDOMWrapReconcileNamespaces(xmlFooCtxtPtr ctxt,
  xmlNodePtr elem,
  int options);

xmlDOMWrapAdoptNode(xmlFooCtxtPtr ctxt,
  xmlDocPtr sourceDoc,
  xmlNodePtr node,
  xmlDocPtr destDoc,                
  xmlNodePtr destParent,
  int options);

xmlDOMWrapRemoveNode(xmlFooCtxtPtr ctxt,
  xmlDocPtr doc,
  xmlNodePtr node);

That makes an addition of ~35 KB to tree.c :-/
Maybe a separate file for DOM-wrapper helper functions would be an
option. Since this all is eventually more code than you expected, I
just post the diff here, and would like to wait for Daniel
to return from vacations for discussion, before exposing anything
to the API.

Additionally there are some tests which can be generated
with the attached files:
"test-adopt.xml" - the test definitions
"test-adopt-to-c.xsl" - the transformation file to C

xsltproc -o test-adopt.c test-adopt-to-c.xsd test-adopt.xml

We could transform the test-definitions to python, when/if
the functions will become available to python.

Open issues:
1. What exactly to do with XIncluded pieces?
2. Design the custom context; it should be used for a
   yet-to-write node-copy function as well.
3. More?

If you have a test-scenario in your head, please send it either
by using the structure as in "test-adopt.xml" or just
describe the scenario, so we can add the test.



Attachment: tree.c.diff
Description: Text Data

Attachment: tree.h.diff
Description: Text Data

Attachment: test-adopt-to-c.xsl
Description: Text Data

Attachment: test-adopt.xml
Description: Text Data

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]