Mini-XML Home Page

Current Release: v1.3, December 21, 2003
[ Download Source (.tar.gz 77k) | Download Linux RPM (.i386.rpm 67k) | Change Log | Documentation | Rate/Make Comments ]

Introduction

Mini-XML is a small XML parsing library that you can use to read XML and XML-like data files in your application without requiring large non-standard libraries. Mini-XML only requires an ANSI C compatible compiler (GCC works, as do most vendors' ANSI C compilers) and a "make" program.

Mini-XML provides the following functionality:

Reading and writing of UTF-8 encoded XML files and strings.
Data is stored in a linked-list tree structure, preserving the XML data hierarchy.
Supports arbitrary element names, attributes, and attribute values with no preset limits, just available memory.
Supports integer, real, opaque ("cdata"), and text data types in "leaf" nodes.
Functions for creating and managing trees of data.
"Find" and "walk" functions for easily locating and navigating trees of data.

Mini-XML doesn't do validation or other types of processing on the data based upon schema files or other sources of definition information, nor does it support character entities other than those required by the XML specification. Also, since Mini-XML does not support the UTF-16 encoding, it is technically not a conforming XML consumer/client.

Building Mini-XML

Mini-XML comes with an autoconf-based configure script; just type the following command to get things going:

./configure

The default install prefix is /usr/local, which can be overridden using the --prefix option:

./configure --prefix=/foo

Other configure options can be found using the --help option:

./configure --help

Once you have configured the software, type "make" to do the build and run the test program to verify that things are working, as follows:

make

Installing Mini-XML

The "install" target will install Mini-XML in the lib and include directories:

make install

Once you have installed it, use the "-lmxml" option to link your application against it.

Documentation

The documentation is currently a work in progress. Aside from the information that follows, the documentation page provides a handy reference and is automatically generated using Mini-XML. You can also look at the testmxml.c and mxmldoc.c source files for examples of using Mini-XML.

The Basics

Mini-XML provides a single header file which you include:

#include <mxml.h>

Nodes are defined by the mxml_node_t structure; the type member defines the node type (element, integer, opaque, real, or text) which determines which value you want to look at in the value union. New nodes can be created using the mxmlNewElement(), mxmlNewInteger(), mxmlNewOpaque(), mxmlNewReal(), and mxmlNewText() functions. Only elements can have child nodes, and the top node must be an element, usually "?xml".

Each node has pointers for the node above (parent), below (child), to the left (prev), and to the right (next) of the current node. If you have an XML file like the following:

    <?xml version="1.0"?>
    <data>
        <node>val1</node>
        <node>val2</node>
        <node>val3</node>
        <group>
            <node>val4</node>
            <node>val5</node>
            <node>val6</node>
        </group>
        <node>val7</node>
        <node>val8</node>
        <node>val9</node>
    </data>

the node tree returned by mxmlLoadFile() would look like the following in memory:

    ?xml
      |
    data
      |
    node - node - node - group - node - node - node
      |      |      |      |       |      |      |
    val1   val2   val3     |     val7   val8   val9
                           |
                         node - node - node
                           |      |      |
                         val4   val5   val6

where "-" is a pointer to the next node and "|" is a pointer to the first child node.

Once you are done with the XML data, use the mxmlDelete() function to recursively free the memory that is used for a particular node or the entire tree:

mxmlDelete(tree);

Loading and Saving XML Files

You load an XML file using the mxmlLoadFile() function:

FILE *fp;
mxml_node_t *tree;

fp = fopen("filename.xml", "r");
tree = mxmlLoadFile(NULL, fp, MXML_NO_CALLBACK);
fclose(fp);

The third argument specifies a callback function which returns the value type of the immediate children for a new element node: MXML_INTEGER, MXML_OPAQUE, MXML_REAL, or MXML_TEXT. This function is called after the element and its attributes have been read, so you can look at the element name, attributes, and attribute values to determine the proper value type to return. The default value type is MXML_TEXT if no callback is used.

Similarly, you save an XML file using the mxmlSaveFile() function:

FILE *fp;
mxml_node_t *tree;

fp = fopen("filename.xml", "w");
mxmlSaveFile(tree, fp, MXML_NO_CALLBACK);
fclose(fp);

Callback functions for saving are used to optionally insert whitespace before and after elements in the node tree. Your function will be called up to four times for each element node with a pointer to the node and a "where" value of MXML_WS_BEFORE_OPEN, MXML_WS_AFTER_OPEN, MXML_WS_BEFORE_CLOSE, or MXML_WS_AFTER_CLOSE. The callback function should return 0 if no whitespace should be added and the character to insert (space, tab, newline) otherwise.

The mxmlLoadString(), mxmlSaveAllocString(), and mxmlSaveString() functions load XML node trees from and save XML node trees to strings:

char buffer[8192];
char *ptr;
mxml_node_t *tree;

...
tree = mxmlLoadString(NULL, buffer, MXML_NO_CALLBACK);

...
mxmlSaveString(tree, buffer, sizeof(buffer), MXML_NO_CALLBACK);

...
ptr = mxmlSaveAllocString(tree, MXML_NO_CALLBACK);

Finding and Iterating Nodes

The mxmlWalkPrev() and mxmlWalkNext()functions can be used to iterate through the XML node tree:

mxml_node_t *node = mxmlWalkPrev(current, tree, MXML_DESCEND);

mxml_node_t *node = mxmlWalkNext(current, tree, MXML_DESCEND);

In addition, you can find a named element/node using the mxmlFindElement() function:

mxml_node_t *node = mxmlFindElement(tree, tree, "name", "attr",
                                    "value", MXML_DESCEND);

The name, attr, and value arguments can be passed as NULL to act as wildcards, e.g.:

/* Find the first "a" element */
node = mxmlFindElement(tree, tree, "a", NULL, NULL, MXML_DESCEND);

/* Find the first "a" element with "href" attribute */
node = mxmlFindElement(tree, tree, "a", "href", NULL, MXML_DESCEND);

/* Find the first "a" element with "href" to a URL */
node = mxmlFindElement(tree, tree, "a", "href",
                       "http://www.easysw.com/~mike/mxml/", MXML_DESCEND);

/* Find the first element with a "src" attribute*/
node = mxmlFindElement(tree, tree, NULL, "src", NULL, MXML_DESCEND);

/* Find the first element with a "src" = "foo.jpg" */
node = mxmlFindElement(tree, tree, NULL, "src", "foo.jpg", MXML_DESCEND);

You can also iterate with the same function:

mxml_node_t *node;

for (node = mxmlFindElement(tree, tree, "name", NULL, NULL, MXML_DESCEND);
     node != NULL;
     node = mxmlFindElement(node, tree, "name", NULL, NULL, MXML_DESCEND))
{
  ... do something ...
}

The MXML_DESCEND argument can actually be one of three constants:

MXML_NO_DESCEND means to not to look at any child nodes in the element hierarchy, just look at siblings at the same level or parent nodes until the top node or top-of-tree is reached. The previous node from "group" would be the "node" element to the left, while the next node from "group" would be the "node" element to the right.
MXML_DESCEND_FIRST means that it is OK to descend to the first child of a node, but not to descend further when searching. You'll normally use this when iterating through direct children of a parent node, e.g. all of the "node" elements under the "?xml" parent node in the example above. This mode is only applicable to the search function; the walk functions treat this as MXML_DESCEND since every call is a first time.
MXML_DESCEND means to keep descending until you hit the bottom of the tree. The previous node from "group" would be the "val3" node and the next node would be the first node element under "group". If you were to walk from the root node "?xml" to the end of the tree with mxmlWalkNext(), the order would be:
```
    ?xml
    data
    node
    val1
    node
    val2
    node
    val3
    group
    node
    val4
    node
    val5
    node
    val6
    node
    val7
    node
    val8
    node
    val9
```
If you started at "val9" and walked using mxmlWalkPrev(), the order would be reversed, ending at "?xml".

Getting Help and Reporting Problems

You can email me at "mxml at easysw dot com" to report problems and/or ask for help. Just don't expect an instant response, as I get a lot of email...

Legal Stuff

This library is free software; you can redistribute it and/or modify it under the terms of the GNU Library General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Library General Public License for more details.

You should have received a copy of the GNU Library General Public License along with this library; if not, write to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.