XML Normalize

This program allows normalizing arbitrary xml files. Normalization can be configured:

• sort sibling elements based on some attribute value
• remove unwanted nodes
• trim texts
• normalize whitespaces/line breaks

This can be used as a post-/pre-processing step to keep diffs small for generated xml files.

Usage

Either install via npm i -g xml_normalize or run directly with npx xml_normalize.

Usage: npx xml_normalize [options]

Options:
  -i, --input-file <inputFile>       input file
  -o, --output-file <outputFile>     output file - if not provided result is printed to stdout
  -r, --remove-path <removePath...>  simple XPath(s) to remove elements - e.g. "/html/head[1]/script"
  -s, --sort-path <sortPath>         simple XPath that references an attribute to sort - e.g. "/html/head[1]/script/@src"
  --no-pretty                        Disable pretty format output
  --no-trim                          Disable trimming of whitespace at the beginning and end of text nodes (trims only pure text nodes)
  --no-attribute-trim                Disable trimming whitespace at the beginning and end of attribute values
  -tf, --trim-force                  Trim the whitespace at the beginning and end of text nodes (trims as well text adjacent to nested nodes)
  -n, --normalize-whitespace         Normalize whitespaces inside text nodes and attribute values
  -d, --debug                        enable debug output
  -h, --help                         display help for command

Options and Examples

Sorting

Allows to sort siblings at a specific path with the same tag name lexicographically based on a specific attribute value.

Example:

<root>
  <node>
    <child id="z">should be last</child>
    <child id="a">should be first</child>
  </node>
  <node>
    <child id="y">should be last</child>
    <child id="b">should be first</child>
  </node>
</root>

npx xml_normalize -s /root/node/child/@id will create:

<root>
  <node>
    <child id="a">should be first</child>
    <child id="z">should be last</child>
  </node>
  <node>
    <child id="b">should be first</child>
    <child id="y">should be last</child>
  </node>
</root>

Removing

Allows to remove nodes in a specific path.

Example:

<root>
  <node>
    <child id="z">should be removed</child>
    <child id="a">should be removed</child>
  </node>
  <node>
    <child id="y">should stay</child>
    <child id="b">should stay</child>
  </node>
</root>

npx xml_normalize -r /root/node[1]/child will create:

<root>
  <node/>
  <node>
    <child id="b">should stay</child>
    <child id="y">should stay</child>
  </node>
</root>

npx xml_normalize -r /root/node/child instead, will create:

<root>
  <node/>
  <node/>
</root>

Normalize whitespace

This option replaces any number of consecutive whitespace, tab, new line characters with a single whitespace (in text nodes).

Example:

<root>
  <node>
    <child id="z">some    xml
    has messed up 
    formatting
    </child>
      
      
    <child id="sometimes      even attributes are messed 
    up">some more     mess</child>
  </node>
</root>

npx xml_normalize --normalize-whitespace will create:

<root>
  <node>
    <child id="z">some xml has messed up formatting</child>
    <child id="sometimes even attributes are messed up">some more mess</child>
  </node>
</root>

Paths for sorting and removing

Paths are a simple subset of XPaths.

/ROOT/NODE_NAME[INDEX]/ANOTHER_NODE

Supported:

• Only absolute paths
• Index access (note in XPath indices are 1-based!)
• Simple predicates using the following functions (parameters can be string (double quotes) or XPaths):
  • starts-with(str,prefix)
  • contains(str,contained)
• Node wildcard - e.g /root/* to select all nodes in root of any type.
• Attribute reference in last node - e.g. /root/node/@id.

What is this good for?

This helps to bring xml in a standardized form, so that changes can easily be spotted in diff tool or git pull request.

For example, you could run it as a post processing/pre commit script when re-generating XLIFF translation files (or getting them back from your beloved translator in a messed up form).

Contribute

PRs always welcome :-)