xmlfuse-js

XML representations as a JSON stream. Convenient for content-oriented XML tasks.

This is a JS port of Python package xmlfuse.

Installation

npm i @innodatalabs/xmlfuse-js --save

Building and testing:

make

API

import { fromString, toString } from '@innodatalabs/lxmlx-js';
import { fuse } from '@innodatalabs/xmlfuse-js';

const xml1 = fromString('<span>Hello, <i>world!</i></span>');
const xml2 = fromString('<span><b>Hello</b>, world!</span>');

const xml = fuze(xml1, xml2)
toString(xml) === '<span><b>Hello</b>, <i>world!</i></span>'
// true

Input documents must have exactly the same text

Error is raised if text differs. Whitespace does matter!

Example:

const xml1 = fromString('<span>Hello</span>');
const xml2 = fromString('<span>Good bye</span>');

const xml = fuze(xml1, xml2);
// throws Error('Text is different')

Conflicting markup

Sometimes it is not possible to merge two markups, because tags intersect. In such a case one has a choice:

a. Raise an exception and let caller handle the problem b. Resolve by segmenting one of the markups

We treat first document as master, and second as slave. Master markup is never segmented. If there is a conflict between master and slave markups (and if autoSegment option is true), fuse() will segment slave to make markup consistent.

Example:

const xml1 = fromString('<span>Hel<i>lo, world!</i></span>');
const xml2 = fromString('<span><b>Hello</b>, world!</span>');

const xml = fuze(xml1, xml2);
toString(xml) === '<span><b>Hel<i>lo</i></b></i>, <i>world!</i></span>';
// true

Set autoSegment flag to false to prevent segmentation. Error will be raised instead, if conflict detected.

Ambiguities

When master ans slave markups wrap the same text, there is a nesting ambuguity - which tag should be inner?

We resolve this by consistently trying to put slave markup inside the master. This behavior can be changed by setting the flag preferSlaveInner to false.

Example:

const xml1 = fromString('<span><i>Hello</i>, world!</span>');
const xml2 = fromString('<span><b>Hello</b>, world!</span>');

const xml = fuze(xml1, xml2, {preferSlaveInner: true});
toString(xml) === '<span><b><i>Hello</i></b>, world!</span>';
// true

const xml = fuze(xml1, xml2, {preferSlaveInner: false});
toString(xml) == b'<span><i><b>Hello</b></i>, world!</span>';
// true

Slave top-level tag is dropped

Note that top-level tag from slave is not merged. It is just dropped. If you want it to be merged into the output, set stripSlaveTopTag: false.

fuse() signature

function fuse(xml1, xml2, options) { ... }

Where:

• xml1 is the master XML document (LXML Element object, see http://lxml.de)
• xml2 is the slave XML document

Returns fused XML document

Recognized options:

• preferSlaveInner controls ambigiuty resolution
• autoSegment allows slave markup segmentation in case of conflicting markup
• stripSlaveTopTag allows fuse to ignore top-level tag from the slave XML
• nsmap provides namespace mapping for building the output document (see lxmlx-js doc for more details on namespaces)