Filtering XML nodes in memory without DOM?

Question

Given a big XML I want to have PART of the XML structure in a tree in memory (THIS IS NOT DOM!!) for example given

<MAIN>
    <A>
        <B>1</B>
        <C>2</C>
        <D>3</D>
            <H>7</H>
            <I>9</I>
        <E>2</D>
    </A>
    <A>
        <B>4</B>
        <C>5</C>
        <D>6</D>
            <H>3</H>
            <I>0</I>
        <E>3</D>
    </A>
</MAIN>

after the user selects nodes B and H, I want to have in memory

<MAIN>
    <A>
        <B>1</B>
        <C>2</C>
        <D>3</D>
            <H>7</H>
    </A>
    <A>
        <B>4</B>
        <C>5</C>
        <D>6</D>
            <H>3</H>
    </A>
</MAIN>

sometimes the user wants to filter by node contents, like all B nodes greater than 3, etc.

is there a (SAX?) library to do such thing? preferably in Smalltalk but any other language will be fine.

score 1 · Answer 1 · answered Nov 12 '11 at 12:56

1

There is a SAX library included in VisualWorks.

answered Nov 12 '11 at 12:56

Alan Knight

2,759
1
15
13

Does that SAX library enables building a DOM-subtree and edit it? – Juan Aguerre Nov 12 '11 at 18:08

score 0 · Answer 2 · answered Nov 12 '11 at 17:35

0

In java land, using either JDOM or DOM4J (and possibly also XOM), you can write a customized builder to achieve what you want.

An example for DOM4J

answered Nov 12 '11 at 17:35

forty-two

12,204
2
26
36

score 0 · Answer 3 · answered Nov 15 '11 at 05:27

0

An STX processor allows one to process a large document in streaming fashion, but to retain (possibly transformed) portions of that document in variables or buffers.

answered Nov 15 '11 at 05:27

WReach

18,098
3
49
93

Filtering XML nodes in memory without DOM?

3 Answers3