infojunkie · infojunkie · Nov 14, 2023 · Nov 9, 2023 · Nov 9, 2023 · Nov 10, 2023
diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "musicxml-midi",
-  "version": "2.1.2",
+  "version": "2.2.0",
   "description": "MusicXML to MIDI converter",
   "type": "module",
   "directories": {
@@ -52,9 +52,10 @@
     "express-fileupload": "^1.3.1",
     "morgan": "^1.10.0",
     "node-fetch": "^3.3.1",
-    "saxon-js": "^2.3.0",
+    "saxon-js": "2.5.0",
     "unzipit": "^1.4.0",
     "validate-with-xmllint": "^1.2.0",
+    "xml-formatter": "^3.6.0",
     "xslt3": "^2.3.0"
   },
   "devDependencies": {

diff --git a/src/js/musicxml-examples.js b/src/js/musicxml-examples.js
@@ -6,32 +6,220 @@
  */
 
 const URL_EXAMPLES_ROOT = 'https://www.w3.org/2021/06/musicxml40/musicxml-reference/examples/'
+const MUSICXML_VERSION = '4.0'
 
 import fetch from 'node-fetch'
 import * as cheerio from 'cheerio'
+import xmlFormat from 'xml-formatter';
 import fs from 'fs'
 import process from 'process'
 import path from 'path'
+import { parseArgs } from 'node:util'
+import { createRequire } from 'node:module'
+import { validateXMLWithXSD } from 'validate-with-xmllint'
 
-const output = process.argv[2] || ''
-if (output != '' && !fs.existsSync(output)) {
-  console.error(`Missing output dir ${output}`)
-  process.exit(1)
+const require = createRequire(import.meta.url)
+const { version } = require('../../package.json')
+
+const options = {
+  'xml': {
+    type: 'boolean',
+    short: 'x',
+  },
+  'output': {
+    type: 'string',
+    short: 'o'
+  },
+  'help': {
+    type: 'boolean',
+    short: 'h'
+  },
+  'version': {
+    type: 'boolean',
+    short: 'v'
+  },
+  'example': {
+    type: 'string',
+    short: 'e'
+  },
+  'validate': {
+    type: 'boolean'
+  },
+  'source': {
+    type: 'string',
+    short: 's'
+  }
+}
+const { values: args } = (function() {
+  try {
+    return parseArgs({ options, allowPositionals: false })
+  }
+  catch (e) {
+    console.error(e.message)
+    process.exit(1)
+  }
+})()
+if (!('source' in args)) {
+  args['source'] = URL_EXAMPLES_ROOT
 }
 
-async function extractMusicXML(page, title) {
-  const response = await fetch(page)
-  const body = await response.text()
-  const $ = cheerio.load(body)
-  const musicxml = "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n" + $('.xmlmarkup').text()
-  fs.writeFileSync(path.join(output, `${title}.musicxml`), musicxml)
+if ('help' in args) {
+  console.log(`
+Usage: musicxml-examples v${version} [--output|-o /path/to/output] [--example|-e example-slug] [--source|-s url/or/path/to/examples/] [--xml|-x] [--validate] [--version|-v] [--help|-h]
+
+Extracts MusicXML examples from ${args['source']}.
+Use --xml to recreate a valid MusicXML structure around examples that lack it.
+`.trim())
+  process.exit(0)
+}
+
+if ('version' in args) {
+  console.log(`musicxml-examples v${version}`)
+  process.exit(0)
+}
+
+const output = args['output'] || ''
+if (output !== '' && !fs.existsSync(output)) {
+  console.error(`Missing output dir ${output}`)
+  process.exit(1)
 }
 
-const response = await fetch(URL_EXAMPLES_ROOT)
-const main = await response.text()
+const main = await fetchPage(args['source'])
 const $ = cheerio.load(main)
 for (const example of $('body').find('a:has(img)')) {
   const href = $(example).prop('href')
-  console.log(`Extracting ${href}...`)
-  await extractMusicXML(URL_EXAMPLES_ROOT + href, href.replace('/', ''))
+  if ('example' in args && args['example'] !== href.replace('/', '')) continue
+  console.error(`Extracting ${href}...`)
+  await extractMusicXml(args['source'] + href, href.replace('/', ''))
+}
+
+async function extractMusicXml(page, slug) {
+  const body = await fetchPage(page)
+  const $ = cheerio.load(body)
+  const musicxml = scaffoldMusicXml($('.xmlmarkup').text())
+
+  if ('validate' in args) {
+    await validateXMLWithXSD(musicxml, 'src/xsd/musicxml.xsd')
+    .catch(error => {
+      console.error(`Failed to validate MusicXML: ${error.message}`)
+    })
+  }
+
+  if (output !== '') {
+    fs.writeFileSync(path.join(output, `${slug}.musicxml`), musicxml)
+  }
+  else {
+    process.stdout.write(musicxml + '\n')
+  }
+}
+
+async function fetchPage(url) {
+  if (/^http/.test(url)) {
+    const response = await fetch(url)
+    return await response.text()
+  }
+  return fs.readFileSync(path.join(url, 'index.html'), { encoding: 'utf-8' })
+}
+
+function scaffoldMusicXml(xml) {
+  if (!('xml' in args)) {
+    return `<?xml version="1.0" encoding="utf-8"?>\n${xml}`
+  }
+
+  const template = `
+  <?xml version="1.0" encoding="utf-8" standalone="no"?>
+  <!DOCTYPE score-partwise PUBLIC
+      "-//Recordare//DTD MusicXML ${MUSICXML_VERSION} Partwise//EN"
+      "http://www.musicxml.org/dtds/partwise.dtd">
+  <score-partwise version="${MUSICXML_VERSION}">
+    <defaults>
+      <system-layout optional-example="yes"/>
+      <staff-layout optional-example="yes"/>
+      <appearance optional-example="yes"/>
+    </defaults>
+    <credit optional-example="yes"/>
+    <part-list>
+      <part-group optional-example="yes"/>
+      <score-part id="P1">
+        <part-name>placeholder</part-name>
+      </score-part>
+    </part-list>
+    <part id="P1">
+      <measure number="1">
+        <direction optional-example="yes">
+          <direction-type optional-example="yes">
+            <dynamics optional-example="yes"/>
+            <metronome optional-example="yes"/>
+            <scordatura optional-example="yes"/>
+            <accordion-registration optional-example="yes"/>
+          </direction-type>
+        </direction>
+        <attributes>
+          <divisions>1</divisions>
+          <key>
+            <fifths>0</fifths>
+          </key>
+          <time>
+            <beats>4</beats>
+            <beat-type>4</beat-type>
+            <interchangeable optional-example="yes">
+              <time-relation optional-example="yes"/>
+            </interchangeable>
+          </time>
+          <clef>
+            <sign>G</sign>
+            <line>2</line>
+          </clef>
+          <staff-details optional-example="yes"/>
+          <measure-style optional-example="yes"/>
+        </attributes>
+        <harmony>
+          <root>
+            <root-step>C</root-step>
+          </root>
+          <kind use-symbols="yes">major-seventh</kind>
+          <inversion optional-example="yes"/>
+          <degree optional-example="yes"/>
+          <frame optional-example="yes"/>
+        </harmony>
+        <figured-bass optional-example="yes"/>
+        <note>
+          <pitch>
+            <step>C</step>
+            <octave>4</octave>
+          </pitch>
+          <duration>4</duration>
+          <type>whole</type>
+          <notehead-text optional-example="yes"/>
+          <notations>
+            <technical optional-example="yes"/>
+          </notations>
+          <lyric optional-example="yes"/>
+        </note>
+        <barline optional-example="yes"/>
+        <print>
+          <part-name-display optional-example="yes"/>
+          <part-abbreviation-display optional-example="yes"/>
+        </print>
+      </measure>
+    </part>
+  </score-partwise>
+  `.trim()
+
+  // Insert the example fragment into the fully-formed template.
+  // - Find the example's root element in the template
+  // - Replace it with the full example fragment
+  // - Remove optional-example attribute from parents of the example fragment
+  // - Remove all elements that still include attribute optional-example="yes"
+  const src = cheerio.load(xml, { xml: true })
+  const core = src.root().children().first().prop('nodeName')
+  const dst = cheerio.load(template, { xml: { xmlMode: true, lowerCaseTags: true, lowerCaseAttributeNames : true }})
+  if (dst(core).length === 0) {
+    console.error(`${core} element not found in template. Returning verbatim XML.`)
+    return `<?xml version="1.0" encoding="utf-8"?>\n${xml}`
+  }
+  dst(core).replaceWith(src.root())
+  dst(src.root()).parents().removeAttr('optional-example')
+  dst('[optional-example="yes"]').remove()
+  return xmlFormat(dst.html(), { collapseContent: true })
 }
diff --git a/test/data/examples/accent-element.musicxml b/test/data/examples/accent-element.musicxml
@@ -1,16 +1,54 @@
-<?xml version="1.0" encoding="utf-8"?>
-<note default-x="36">
-   <pitch>
-      <step>A</step>
-      <octave>4</octave>
-   </pitch>
-   <duration>4</duration>
-   <voice>1</voice>
-   <type>half</type>
-   <stem default-y="10">up</stem>
-   <notations>
-      <articulations>
-         <accent default-x="-1" default-y="-55" placement="below"/>
-      </articulations>
-   </notations>
-</note>
+<?xml version="1.0" encoding="utf-8" standalone="no"?>
+<!DOCTYPE score-partwise PUBLIC
+      "-//Recordare//DTD MusicXML 4.0 Partwise//EN"
+      "http://www.musicxml.org/dtds/partwise.dtd">
+<score-partwise version="4.0">
+    <defaults>
+    </defaults>
+    <part-list>
+        <score-part id="P1">
+            <part-name>placeholder</part-name>
+        </score-part>
+    </part-list>
+    <part id="P1">
+        <measure number="1">
+            <attributes>
+                <divisions>1</divisions>
+                <key>
+                    <fifths>0</fifths>
+                </key>
+                <time>
+                    <beats>4</beats>
+                    <beat-type>4</beat-type>
+                </time>
+                <clef>
+                    <sign>G</sign>
+                    <line>2</line>
+                </clef>
+            </attributes>
+            <harmony>
+                <root>
+                    <root-step>C</root-step>
+                </root>
+                <kind use-symbols="yes">major-seventh</kind>
+            </harmony>
+            <note default-x="36">
+                <pitch>
+                    <step>A</step>
+                    <octave>4</octave>
+                </pitch>
+                <duration>4</duration>
+                <voice>1</voice>
+                <type>half</type>
+                <stem default-y="10">up</stem>
+                <notations>
+                    <articulations>
+                        <accent default-x="-1" default-y="-55" placement="below"/>
+                    </articulations>
+                </notations>
+            </note>
+            <print>
+            </print>
+        </measure>
+    </part>
+</score-partwise>
diff --git a/test/data/examples/accidental-element.musicxml b/test/data/examples/accidental-element.musicxml
@@ -1,13 +1,51 @@
-<?xml version="1.0" encoding="utf-8"?>
-<note default-x="83">
-   <pitch>
-      <step>A</step>
-      <alter>1</alter>
-      <octave>4</octave>
-   </pitch>
-   <duration>2</duration>
-   <voice>1</voice>
-   <type>quarter</type>
-   <accidental>sharp</accidental>
-   <stem default-y="10">up</stem>
-</note>
+<?xml version="1.0" encoding="utf-8" standalone="no"?>
+<!DOCTYPE score-partwise PUBLIC
+      "-//Recordare//DTD MusicXML 4.0 Partwise//EN"
+      "http://www.musicxml.org/dtds/partwise.dtd">
+<score-partwise version="4.0">
+    <defaults>
+    </defaults>
+    <part-list>
+        <score-part id="P1">
+            <part-name>placeholder</part-name>
+        </score-part>
+    </part-list>
+    <part id="P1">
+        <measure number="1">
+            <attributes>
+                <divisions>1</divisions>
+                <key>
+                    <fifths>0</fifths>
+                </key>
+                <time>
+                    <beats>4</beats>
+                    <beat-type>4</beat-type>
+                </time>
+                <clef>
+                    <sign>G</sign>
+                    <line>2</line>
+                </clef>
+            </attributes>
+            <harmony>
+                <root>
+                    <root-step>C</root-step>
+                </root>
+                <kind use-symbols="yes">major-seventh</kind>
+            </harmony>
+            <note default-x="83">
+                <pitch>
+                    <step>A</step>
+                    <alter>1</alter>
+                    <octave>4</octave>
+                </pitch>
+                <duration>2</duration>
+                <voice>1</voice>
+                <type>quarter</type>
+                <accidental>sharp</accidental>
+                <stem default-y="10">up</stem>
+            </note>
+            <print>
+            </print>
+        </measure>
+    </part>
+</score-partwise>