On Mon, Oct 21, 2013 at 11:50 AM, Bogdan Cristea <crist...@gmail.com> wrote:

> On 10/20/2013 06:53 PM, Nikolay Sivov wrote:
>
>> On 10/20/2013 19:13, Bogdan Cristea wrote:
>>
>>> First, I am puzzled by the way I can obtain the text contained by h1
>>> element. I am using the node from a previous xmlTextReaderRead() call. If I
>>> try to obtain the current node in XML_READER_TYPE_TEXT case the node
>>> pointer is NULL.
>>>
>> Well, that's not surprising. 'h1' in your example contains 3 children -
>> text, element, text - in that order. So when you request node->children you
>> get first child which is a text node, its content is in 'content' field. I
>> don't see a problem here.
>>
>>  Second, I don't know how to obtain the text after <img> element which
>>> still belongs to <h1> element. s there a way to do so ?
>>>
>> It looks wrong to use a node pointer returned from previous reader
>> iteration. As I understand it could be reused, and previous node is freed
>> when you're done with it. So in your example you store previous element
>> pointer which is 'img' and later try to use it to get some text (which
>> can't be outside of 'img' scope by the way)? Are you sure
>> xmlTextReaderCurrentNode() doesn't return anything for text node?
>>
>
> Hi
>
> Indeed, I get <h1> element text when I look for the content its children
> (<img> is between the text); However, I still need to wait for node type
> XML_READER_TYPE_TEXT in order to obtain the text. Currently first I detect
> a XML_READER_TYPE_ELEMENT, which is <h1>, after this I have
> XML_READER_TYPE_TEXT when I get the text before and after <img> element,
> then I get another XML_READER_TYPE_ELEMENT which is <img> and its
> attributes. Inside XML_READER_TYPE_TEXT a call to
> xmlTextReaderCurrentNode() gives me a NULL pointer, so I find a little
> strange the way libxml2 API works, althow not completely forbidden.
>
> Hah, so you're saying that first XML_READER_TYPE_TEXT gives you not-null
pointer with xmlTextReaderCurrentNode()? And this returned node contains
concatenated text from first and last children of 'h1'? That strange.


> regards
> Bogdan
>
>
>
_______________________________________________
xml mailing list, project page  http://xmlsoft.org/
xml@gnome.org
https://mail.gnome.org/mailman/listinfo/xml

Reply via email to