Runtime: Browser + Node.js

Why

DOM parsers like DOMParser or xml2js are heavy dependencies that need a runtime or polyfills. NCBI XML responses have predictable, well-documented structure — a targeted regex reader is faster, lighter, and works everywhere.

@ncbijs/xml extracts tags, blocks, and attributes from NCBI XML without external dependencies or platform assumptions.

• Tag extraction — text content of leaf elements
• Block extraction — full inner content including nested tags, with correct handling of same-name nesting
• Attribute reading — attribute values from opening tags, with entity decoding
• Entity decoding — &, <, >, ", ', and numeric entities (&#x...;, &#...;)
• Tag stripping — remove all XML tags and decode entities in one call

Install

npm install @ncbijs/xml

Quick start

import { readTag, readBlock, readAllTags, stripTags } from '@ncbijs/xml';

const xml = `
<PubmedArticle>
  <MedlineCitation>
    <PMID>12345678</PMID>
    <Article>
      <ArticleTitle>A <i>novel</i> approach</ArticleTitle>
      <Language>eng</Language>
    </Article>
  </MedlineCitation>
</PubmedArticle>`;

readTag(xml, 'PMID');
// => '12345678'

readBlock(xml, 'ArticleTitle');
// => 'A <i>novel</i> approach'

stripTags('A <i>novel</i> approach');
// => 'A novel approach'

readAllTags(xml, 'Language');
// => ['eng']

API

readTag(xml, tagName)

Extract the text content of the first matching tag. Only captures text between the open and close tags — no nested elements.

readTag('<PMID Version="1">12345678</PMID>', 'PMID');
// => '12345678'

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns string | undefined.

readAllTags(xml, tagName)

Extract text content of all matching tags.

readAllTags('<Keyword>cancer</Keyword><Keyword>genomics</Keyword>', 'Keyword');
// => ['cancer', 'genomics']

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns ReadonlyArray<string>.

readBlock(xml, tagName)

Extract the full inner content (including nested tags) between the first matching open/close pair. Handles nested same-name tags correctly.

readBlock('<Abstract><p>First.</p><p>Second.</p></Abstract>', 'Abstract');
// => '<p>First.</p><p>Second.</p>'

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns string | undefined.

readAllBlocks(xml, tagName)

Extract inner content of all matching blocks.

readAllBlocks('<sec><p>A</p></sec><sec><p>B</p></sec>', 'sec');
// => ['<p>A</p>', '<p>B</p>']

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns ReadonlyArray<string>.

readAttribute(xml, tagName, attrName)

Extract the value of an attribute from the first matching tag.

readAttribute('<PMID Version="1">12345678</PMID>', 'PMID', 'Version');
// => '1'

| Parameter | Type | Description | | ---------- | -------- | ----------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. | | attrName | string | Attribute name to read. |

Returns string | undefined.

readTagWithAttributes(xml, tagName)

Extract text content and all attributes from the first matching tag.

readTagWithAttributes('<Keyword MajorTopicYN="Y">cancer</Keyword>', 'Keyword');
// => { text: 'cancer', attributes: { MajorTopicYN: 'Y' } }

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns TagWithAttributes | null.

readAllTagsWithAttributes(xml, tagName)

Extract text content and attributes from all matching tags.

readAllTagsWithAttributes(
  '<DescriptorName UI="D009369" MajorTopicYN="Y">Neoplasms</DescriptorName>',
  'DescriptorName',
);
// => [{ text: 'Neoplasms', attributes: { UI: 'D009369', MajorTopicYN: 'Y' } }]

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns ReadonlyArray<TagWithAttributes>.

readAllBlocksWithAttributes(xml, tagName)

Extract inner content and attributes from all matching blocks.

readAllBlocksWithAttributes(
  '<article-id pub-id-type="doi">10.1234/example</article-id>',
  'article-id',
);
// => [{ content: '10.1234/example', attributes: { 'pub-id-type': 'doi' } }]

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to search. | | tagName | string | Tag name to find. |

Returns ReadonlyArray<BlockWithAttributes>.

stripTags(xml)

Remove all XML tags from a string.

stripTags('<p>A <b>bold</b> statement</p>');
// => 'A bold statement'

| Parameter | Type | Description | | --------- | -------- | -------------------- | | xml | string | XML string to strip. |

Returns string.

removeAllBlocks(xml, tagName)

Remove all occurrences of a block (open tag through close tag, including content). Also removes self-closing elements.

removeAllBlocks('<body><xref>1</xref> text</body>', 'xref');
// => '<body> text</body>'

| Parameter | Type | Description | | --------- | -------- | --------------------- | | xml | string | XML string to modify. | | tagName | string | Tag name to remove. |

Returns string.

decodeEntities(text)

Decode XML entities: &, <, >, ", ', and numeric character references ({, 💡).

decodeEntities('Smith & Jones — 2024');
// => 'Smith & Jones — 2024'

| Parameter | Type | Description | | --------- | -------- | --------------- | | text | string | Text to decode. |

Returns string.

Types

All types are exported for use in your own interfaces:

import type { TagWithAttributes, BlockWithAttributes } from '@ncbijs/xml';