NEW: option: analysisLang

README.md

@@ -114,11 +114,13 @@ If the input is a string containing only whitespace, an empty string is returned
 
 ## Options
 
-| Option      | type          | Default   | Description                                                                                                 |
-| ----------- | ------------- | --------- | ----------------------------------------------------------------------------------------------------------- |
-| `classes`   | Array<String> | `['igl']` | An array of classes to apply to the wrapper element.                                                        |
-| `scription` | Object        | `{}`      | Options to pass to the `scription2dlx` library. See [scription2dlx][scription2dlx] for more details.        |
-| `tag`       | String        | `'div'`   | The HTML tag to wrap each interlinear gloss in. Can also be a custom tag (useful for HTML custom elements). |
+| Option         | type          | Default   | Description                                                                                                                                                                                                                                                                                                                                                  |
+| -------------- | ------------- | --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
+| `analysisLang` | String        | undefined | An [IETF language tag][lang-tags] to use as the default value of the `lang` attribute for any data in the analysis language (metadata, literal translation, free translation, glosses, literal word translation). If `undefined`, no `lang` tag is added, which means that browsers will assume that the analysis language is the same as the HTML document. |
+| `classes`      | Array<String> | `['igl']` | An array of classes to apply to the wrapper element.                                                                                                                                                                                                                                                                                                         |
+| `scription`    | Object        | `{}`      | Options to pass to the `scription2dlx` library. See [scription2dlx][scription2dlx] for more details.                                                                                                                                                                                                                                                         |
+| `tag`          | String        | `'div'`   | The HTML tag to wrap each interlinear gloss in. Can also be a custom tag (useful for HTML custom elements).                                                                                                                                                                                                                                                  |
+| `targetLang`   | String        | undefined | An [IETF language tag][lang-tags] to use as the default value of the `lang` attribute for any data in the target language.                                                                                                                                                                                                                                   |
 
 ## HTML Structure
 
@@ -135,7 +137,7 @@ const options = {
 
 Each line of the interlinear example is typically given a CSS class that matches its line type (with some exceptions, such as `timespan` instead of just `t`). For example, the `\trs` line will result in `<p class=trs>`, and the `\tln-en` line will result in `<p class=tln lang=en>`.
 
-Whenever the language of a gloss or translation is specified, it is passed through to the `lang` attribute of the relevant analysis language elements, as just seen in the `\tln-en` example. Whenever the default language of the text is specified, it is passed through to the `lang` attribute of the relevant target language elements.
+If the language of the text is specified, it is set as the value of the `lang` attribute for data in the target language wherever relevant. Whenever the language of analysis data (metadata, glosses, translations, etc.) is specified, it is passed through to the `lang` attribute of the relevant analysis language elements (`<p class=tln lang=en>`).
 
 When the scription format allows for data in multiple orthographies, the orthography of the data is specified in the `data-ortho` attribute. For example, the following lines of scription are transformed to the HTML that follows:
 
@@ -159,6 +161,7 @@ When the scription format allows for data in multiple orthographies, the orthogr
 **FORTHCOMING**
 
 <!-- Links -->
+[lang-tags]:     https://datatracker.ietf.org/doc/html/rfc5646
 [learn-Node]:    https://nodejs.dev/en/learn/
 [Leipzig]:       https://www.eva.mpg.de/lingua/resources/glossing-rules.php
 [Node]:          https://nodejs.org/

package.json

@@ -27,6 +27,7 @@
   "scripts": {
     "build": "node build/index.js",
     "stop-only": "stop-only --folder src",
+    "start": "mocha test/*.test.js",
     "test": "mocha test/*.test.js"
   },
   "publishConfig": {

src/lines/literal.js

@@ -1,9 +1,10 @@
 import addEmphasis from '../utilities/addEmphasis.js'
 
-export default function createLiteral(data) {
+export default function createLiteral(data, { analysisLang }) {
 
   if (typeof data === `string`) {
-    return `<p class=lit lang=en>${ addEmphasis(data) }</p>`
+    const lang = analysisLang ? `lang='${ analysisLang }'` : ``
+    return `<p class=lit ${ lang }>${ addEmphasis(data) }</p>`
   }
 
   let html = ``

src/lines/metadata.js

@@ -1,4 +1,5 @@
-export default function createHeader(metadata) {
+export default function createHeader(metadata, { analysisLang }) {
   if (!metadata) return ``
-  return `<p class=ex-header>${ metadata }</p>`
+  const lang = analysisLang ? `lang='${ analysisLang }'` : ``
+  return `<p class=ex-header ${ lang }>${ metadata }</p>`
 }

src/lines/translation.js

@@ -1,9 +1,10 @@
 import addEmphasis from '../utilities/addEmphasis.js'
 
-export default function createTranslation(data) {
+export default function createTranslation(data, { analysisLang }) {
 
   if (typeof data === `string`) {
-    return `<p class=tln lang=en>${ addEmphasis(data) }</p>`
+    const lang = analysisLang ? `lang='${ analysisLang }'` : ``
+    return `<p class=tln ${ lang }>${ addEmphasis(data) }</p>`
   }
 
   let html = ``

src/options.js

@@ -1,5 +1,9 @@
 export default function validateOptions(opts) {
 
+  if (`analysisLang` in opts) {
+    if (typeof opts.analysisLang !== `string`) throw new TypeError(`The 'analysisLang' option must be a String.`)
+  }
+
   if (`classes` in opts) {
     if (!Array.isArray(opts.classes)) throw new TypeError(`The 'classes' option must be an array.`)
     if (!opts.classes.every(cls => typeof cls === `string`)) throw new TypeError(`Each of the values of the 'classes' array must be a string.`)

src/utterance.js

@@ -7,17 +7,19 @@ import createTranscript    from './lines/transcript.js'
 import createTranscription from './lines/transcription.js'
 import createTranslation   from './lines/translation.js'
 
-export default function convertUtterance(u, { classes, tag }) {
+export default function convertUtterance(u, options) {
+
+  const { classes, tag } = options
 
   const classString   = classes.join(` `)
-  const header        = createHeader(u.metadata)
-  const literal       = createLiteral(u.literal)
+  const header        = createHeader(u.metadata, options)
+  const literal       = createLiteral(u.literal, options)
   const phonetic      = createPhonetic(u.phonetic)
   const source        = createSource(u.speaker, u.source)
   const timespan      = createTimespan(u.startTime, u.endTime)
   const transcript    = createTranscript(u.transcript)
   const transcription = createTranscription(u.transcription)
-  const translation   = createTranslation(u.translation)
+  const translation   = createTranslation(u.translation, options)
 
   return `<${ tag } class='${ classString }'>
     ${ header }

test/index.test.js

@@ -1,10 +1,12 @@
-import convert          from '../src/index.js'
-import { expect }       from 'chai'
-import { getAttribute } from '@web/parse5-utils'
-import parse            from './utilities/convertAndParse.js'
-import parseClassString from './utilities/parseClassString.js'
+import convert            from '../src/index.js'
+import { expect }         from 'chai'
+import findElementByClass from './utilities/findElementByClass.js'
+import { getAttribute }   from '@web/parse5-utils'
+import { getTextContent } from '../node_modules/@web/parse5-utils/src/index.js'
+import parse              from './utilities/convertAndParse.js'
+import parseClassString   from './utilities/parseClassString.js'
+import { Swahili }        from '../samples/data/data.js'
 
-import { Swahili } from '../samples/data/data.js'
 
 describe(`scription2html`, function() {
 
@@ -62,6 +64,53 @@ describe(`scription2html`, function() {
 
   })
 
+  it(`option: analysisLang = undefined`, function() {
+
+    const scription = `
+    # Swahili
+    \\txn ninakupenda
+    \\m   ni-na-ku-pend-a
+    \\gl  1SG.SUBJ-PRES-2SG.OBJ-love-IND
+    \\lit Te estoy amando.
+    \\tln Te amo.
+    `
+
+    const { dom } = parse(scription)
+
+    const lit  = findElementByClass(dom, `lit`)
+    const tln  = findElementByClass(dom, `tln`)
+    const meta = findElementByClass(dom, `ex-header`)
+
+    expect(getAttribute(lit, `lang`)).to.be.undefined
+    expect(getAttribute(tln, `lang`)).to.be.undefined
+    expect(getAttribute(meta, `lang`)).to.be.undefined
+
+  })
+
+  it(`option: analysisLang`, function() {
+
+    const scription = `
+    # Swahili
+    \\txn ninakupenda
+    \\m   ni-na-ku-pend-a
+    \\gl  1SG.SUBJ-PRES-2SG.OBJ-love-IND
+    \\lit Te estoy amando.
+    \\tln Te amo.
+    `
+
+    const analysisLang = `sp`
+    const { dom }      = parse(scription, { analysisLang })
+
+    const lit  = findElementByClass(dom, `lit`)
+    const tln  = findElementByClass(dom, `tln`)
+    const meta = findElementByClass(dom, `ex-header`)
+
+    expect(getAttribute(lit, `lang`)).to.equal(analysisLang)
+    expect(getAttribute(tln, `lang`)).to.equal(analysisLang)
+    expect(getAttribute(meta, `lang`)).to.equal(analysisLang)
+
+  })
+
   it(`option: classes`, function() {
 
     const classes = [`example`, `interlinear`]