@amermathsoc/texml-to-html 18.0.0 → 18.2.0

package/CHANGELOG.md

@@ -2,6 +2,36 @@
 
 All notable changes to this project will be documented in this file. See [commit-and-tag-version](https://github.com/absolute-version/commit-and-tag-version) for commit guidelines.
 
+## [18.2.0](https://github.com/AmerMathSoc/texml-to-html/compare/v18.1.0...v18.2.0) (2025-02-05)
+
+
+### Features
+
+* **attrib.js:** add data-ams-doc ([70fa67c](https://github.com/AmerMathSoc/texml-to-html/commit/70fa67ca59b92d1ff18900d430b107ceef7f2d9c)), closes [#462](https://github.com/AmerMathSoc/texml-to-html/issues/462)
+* replace applyHacks() with fixContentModel() ([0cb4ae7](https://github.com/AmerMathSoc/texml-to-html/commit/0cb4ae7fbcdc4006b51d43b8cd0398dd4cbb2b30)), closes [#473](https://github.com/AmerMathSoc/texml-to-html/issues/473)
+* revise tag and target in text equations ([d7244f6](https://github.com/AmerMathSoc/texml-to-html/commit/d7244f6a8212336754a112657b9e14bb9e0e20a9)), closes [#429](https://github.com/AmerMathSoc/texml-to-html/issues/429)
+
+
+### Bug Fixes
+
+* **app.js:** remove unused ternary path ([f459df1](https://github.com/AmerMathSoc/texml-to-html/commit/f459df12b6f4582138e001cb8c7ae0ddf5125ecd)), closes [#470](https://github.com/AmerMathSoc/texml-to-html/issues/470)
+* **disp-quote.js:** remove outdated workaround ([390482a](https://github.com/AmerMathSoc/texml-to-html/commit/390482a82bd84762e3b8d1ec5191b38c50e29d02)), closes [#472](https://github.com/AmerMathSoc/texml-to-html/issues/472)
+* **disp-quote:** footer is last child ([ee46ad7](https://github.com/AmerMathSoc/texml-to-html/commit/ee46ad7a326ac029ca03adbfe63da1d41bbc2798)), closes [#457](https://github.com/AmerMathSoc/texml-to-html/issues/457) [#463](https://github.com/AmerMathSoc/texml-to-html/issues/463)
+* **formula.js:** remove whitespace clean-up ([6f673f4](https://github.com/AmerMathSoc/texml-to-html/commit/6f673f49e146e08b4647a12122f30ddac88f9e1b)), closes [#424](https://github.com/AmerMathSoc/texml-to-html/issues/424)
+* **formula.js:** trim inner whitespace ([2ae7130](https://github.com/AmerMathSoc/texml-to-html/commit/2ae7130ef1d1d27f4e26a46686c7ec211936d0ec)), closes [#471](https://github.com/AmerMathSoc/texml-to-html/issues/471)
+* remove book-title.js ([57610a8](https://github.com/AmerMathSoc/texml-to-html/commit/57610a8a86ba8e3409d78fcc0dfbe813f9d68b4d)), closes [#254](https://github.com/AmerMathSoc/texml-to-html/issues/254) [#448](https://github.com/AmerMathSoc/texml-to-html/issues/448)
+* remove meta-name.js and meta-value.js ([2bbc8e0](https://github.com/AmerMathSoc/texml-to-html/commit/2bbc8e05b75b72d3f60968d9c891050b16935cde)), closes [#254](https://github.com/AmerMathSoc/texml-to-html/issues/254) [#448](https://github.com/AmerMathSoc/texml-to-html/issues/448)
+* remove subject-group.js and subject.js ([9fefa46](https://github.com/AmerMathSoc/texml-to-html/commit/9fefa4616d0ebe2068e1dd0d1db56a83fae3ed8b)), closes [#254](https://github.com/AmerMathSoc/texml-to-html/issues/254) [#448](https://github.com/AmerMathSoc/texml-to-html/issues/448)
+* revise label/title/etc handling ([804904a](https://github.com/AmerMathSoc/texml-to-html/commit/804904a4bf21878e302bc328a4bbdbee0bbe4f90)), closes [#467](https://github.com/AmerMathSoc/texml-to-html/issues/467) [#452](https://github.com/AmerMathSoc/texml-to-html/issues/452)
+* **sec.js:** workaround nested book-app ([1ec8680](https://github.com/AmerMathSoc/texml-to-html/commit/1ec86805829838288453e8a2ff4140c11a79bc02)), closes [#460](https://github.com/AmerMathSoc/texml-to-html/issues/460)
+
+## [18.1.0](https://github.com/AmerMathSoc/texml-to-html/compare/v18.0.0...v18.1.0) (2024-09-09)
+
+
+### Features
+
+* **sec.js:** support style attribute ([84c7699](https://github.com/AmerMathSoc/texml-to-html/commit/84c769970c4841a696e681766a7348a8a6900b9e)), closes [#461](https://github.com/AmerMathSoc/texml-to-html/issues/461)
+
 ## [18.0.0](https://github.com/AmerMathSoc/texml-to-html/compare/v18.0.0-0...v18.0.0) (2024-08-16)
 
 ## [18.0.0-0](https://github.com/AmerMathSoc/texml-to-html/compare/v17.1.0...v18.0.0-0) (2024-08-13)

package/README.md

@@ -118,6 +118,7 @@ Beyond HTML element and attributes, texml-to-html stores data in custom `data-*`
     - container is inside data-ams-doc="math block" elements
     - ams-html uses these to generate the math panel DTs for equations
   - app-group [from book-app-group]
+  - attribution [from attrib]
 - data-ams-doc-contrib
   - {@content-type} [expected: "authors", "editors", "translators", "contributors"]
   - {@contrib-type} [expected: "author", "editor", "translator", "contributor"]
@@ -132,7 +133,7 @@ Beyond HTML element and attributes, texml-to-html stores data in custom `data-*`
   - underline
   - {disp-quote@specific-use}
   - (inline-)graphic{@specific-use}
-  - {@style} [expected: theorem styles]
+  - {@style} [expected: theorem styles, sec styles]
   - boxed (from boxed-text)
 - data-ams-ref
   - {@ref-type} [expected: bibr, fn, disp-formula, sec, fig, table, algorithm, list, statement]
@@ -247,10 +248,11 @@ The following provide a list from the reverse point of view.
   - @notes-type => @data-ams-content-type (and role=dedication for dedications)
   - use cases: dedication (articles), article and section notes (NOTI only), drm notice & epub note (books)
 - toc-entry@specific-use => data-ams-ref
+- attrib => data-ams-doc="attribution"
 - attributes
   - @disp-level => data-ams-doc-level [data-ams-doc-level is also added to some elements that lack disp-level]
   - @content-type => data-ams-content-type
-  - @style => data-ams-style
+  - @style => data-ams-style [e.g., statement, sec]
   - @specific-use => data-ams-specific-use [sometimes mapped to other attributes, e.g., style]
   - @has-qed-box => data-ams-qed-box
   - @position => data-ams-position

package/lib/elements/abstract.js

@@ -30,5 +30,6 @@ export default function (htmlParentNode, xmlnode) {
   });
   mapAttributes(section, xmlnode);
   htmlParentNode.appendChild(section);
+  this.createHeading(section, xmlnode);
   this.passThrough(section, xmlnode);
 };

package/lib/elements/app.js

@@ -24,12 +24,13 @@ import mapAttributes from '../helpers/mapAttributes.js';
 export default function (htmlParentNode, xmlnode) {
   const section = this.createNode('section', '', {
     role: 'doc-appendix',
-    'data-ams-doc-level': this.isBook ? 0 : 1,
+    'data-ams-doc-level': 1,
   });
   const titleChild = xmlnode.querySelector('title');
   if (titleChild && titleChild.textContent.startsWith('Acknowledg'))
     section.setAttribute('role', 'doc-acknowledgments');
   htmlParentNode.appendChild(section);
   mapAttributes(section, xmlnode);
+  this.createHeading(section, xmlnode);
   this.passThrough(section, xmlnode);
 };

package/lib/elements/attrib.js

@@ -20,22 +20,7 @@
  * @param {Element} xmlnode 
  */
 export default function (htmlParentNode, xmlnode) {
-  let actualParent = htmlParentNode;
-  if (xmlnode.parentNode.tagName === 'disp-quote') {
-    const footer = this.createNode('footer');
-    htmlParentNode.appendChild(footer);
-    actualParent = footer;
-  }
-  if (
-    xmlnode.parentNode.tagName === 'fig' ||
-    xmlnode.parentNode.tagName === 'fig-group'
-  ) {
-    // NOTE there should be a figcaption element (cf. caption())
-    // NOTE so far this only occurs in mbk103, clrm067 to attribute graphics in figures
-    actualParent = htmlParentNode.querySelector(':scope>figcaption');
-  }
-  const span = this.createNode('span');
-  actualParent.insertAdjacentText('beforeend', ' '); // NOTE needed inside fig-caption
-  actualParent.appendChild(span);
+  const span = this.createNode('span', '', { 'data-ams-doc': 'attribution' });
+  htmlParentNode.appendChild(span);
   this.passThrough(span, xmlnode);
 };

package/lib/elements/book-meta-json.js

@@ -115,7 +115,7 @@ export function generateBookJson(bookMetaNode) {
     })
 
     // book-title-group
-    const title = bookMetaNode.querySelector('book-title-group > book-title'); // NOTE title might contain math-mode (no use case yet)
+    const title = bookMetaNode.querySelector('book-title-group > book-title'); // NOTE title might contain complex markup (no use case yet; however e.g. surv279 has math-as-unicode). See also head.js
     const subtitle = bookMetaNode.querySelector('book-title-group > subtitle'); // NOTE subtitle might contain math-mode (no use case yet)
     const alttitle = bookMetaNode.querySelector('book-title-group > alt-title') || title;
     bookMetaJson.book.title = this.passthroughIntoHTMLString(title);

package/lib/elements/disp-formula-group.js

@@ -26,5 +26,12 @@ export default function (htmlParentNode, xmlnode) {
     id: xmlnode.id,
   });
   htmlParentNode.appendChild(figure);
+  const label = xmlnode.querySelector(':scope>label');
+  if (label) {
+    const figcaption = this.createNode('figcaption');
+    figure.appendChild(figcaption);
+    this.redirectRecurseTheDom(figcaption, label);
+  }
+
   this.passThrough(figure, xmlnode);
 };

package/lib/elements/disp-quote.js

@@ -26,8 +26,11 @@ export default function (htmlParentNode, xmlnode) {
   if (contentType === 'epigraph')
     blockquote.setAttribute('role', 'doc-epigraph');
   else blockquote.setAttribute('data-ams-style', specificUse);
-  if (htmlParentNode.tagName === 'P')
-    htmlParentNode.insertAdjacentElement('afterend', blockquote);
-  else htmlParentNode.appendChild(blockquote);
+  htmlParentNode.appendChild(blockquote);
+  // handle attrib
+  const attrib = xmlnode.querySelector(':scope>attrib');
+  const footer = this.createNode('footer');
+  if (attrib) this.redirectRecurseTheDom(footer, attrib);
   this.passThrough(blockquote, xmlnode);
+  if (attrib) blockquote.appendChild(footer);
 };

package/lib/elements/fig.js

@@ -31,8 +31,9 @@ export default function (htmlParentNode, xmlnode) {
 
   const label = xmlnode.querySelector(':scope>label');
   const caption = xmlnode.querySelector(':scope>caption');
-
-  if (label || caption) {
+  const attrib = xmlnode.querySelector(':scope>attrib');
+  
+  if (label || caption || attrib) {
     const figcaption = this.createNode('figcaption');
     figure.appendChild(figcaption);
     if (label) {
@@ -44,6 +45,10 @@ export default function (htmlParentNode, xmlnode) {
     if (caption) {
       this.passThrough(figcaption, caption);
     }
+    if (attrib) {
+      figcaption.insertAdjacentText('beforeend', ' '); //NOTE figcaptioninline content
+      this.redirectRecurseTheDom(figcaption, attrib);
+    }
   }
   // process remaining children
   this.passThrough(figure, xmlnode);

package/lib/elements/formula.js

@@ -46,7 +46,6 @@ export default function (htmlParentNode, xmlnode) {
   const tagContainer = `<span hidden data-ams-doc="tags"></span>`; //NOTE we store copies of tags for easier re-use downstream (cf. tags.js) except for text equations
   if (hasLinkedTag) span.insertAdjacentHTML('afterbegin', tagContainer)
   this.passThrough(span, xmlnode);
-  // NOTE we (sometimes?) get extra whitespace from childnodes; needs test
   const text = span.innerHTML;
-  span.innerHTML = text.replace(/[ \n]+/g, ' ');
+  span.innerHTML = text.trim();  // NOTE: texml (sometimes) adds whitespace around the tex-math child; this causes undesired whitespace (e.g., theorem titles like `($x$)` becoming `( $x$ )`) TODO: revisit after AmerMathSoc/texml#234 is closed
 };

package/lib/elements/label.js

@@ -14,58 +14,14 @@
  *  limitations under the License.
  */
 
-import getParentLevel from '../helpers/getParentLevel.js';
-
 /**
- * label and title elements. If both are present, expects label+title (not title+label)
+ * label and title elements, usually redirected
  * @param {HTMLElement} htmlParentNode 
  * @param {Element} xmlnode 
  */
 export default function (htmlParentNode, xmlnode) {
-  // CASE label followed by a title -- we skip (and pull in the label later on when processing title)
-  if (xmlnode.nextElementSibling?.tagName === 'title') return;
-  // CASE empty label
-  if (xmlnode.tagName === 'label' && xmlnode.innerHTML.trim() === '') return;
-
-  // complex cases
-
-  // Decide container (h* or figcaption; wrapping header for subtitles)
-  const isStatement = xmlnode.parentNode.tagName === 'statement';
-  const isDispFormulaGroup = xmlnode.parentNode.tagName === 'disp-formula-group';
-  const level = getParentLevel(htmlParentNode) + 1;
-  const container = (isStatement || isDispFormulaGroup) ? this.createNode('figcaption', '') : this.createNode(`h${level}`, '');
-  htmlParentNode.appendChild(container);
-
-  // subtitle handling (assumes container is not figcaption)
-  const subtitleSibling = xmlnode.parentNode.querySelector(':scope>subtitle');
-  if (subtitleSibling && subtitleSibling.innerHTML.trim() !== '') {
-    // wrap heading in header
-    const header = this.createNode('header');
-    htmlParentNode.appendChild(header);
-    header.appendChild(container);
-    // recurse subtitle
-    this.recurseTheDom(header, subtitleSibling);
-  }
-
-  // Pull in label (if title+label and we're processing title)
-  const previousSibling = xmlnode.previousElementSibling;
-  if (previousSibling?.tagName === 'label' && previousSibling.innerHTML.trim() !== '') {
-    const labelSpan = this.createNode('span', '', { 'data-ams-doc': 'label' });
-    container.appendChild(labelSpan);
-    this.passThrough(labelSpan, previousSibling);
-    labelSpan.insertAdjacentText('afterend', ' '); // NOTE: adding (in HTML meaningful) space after label & before title to avoid them from smashing together.
-  }
-
-  // CASE Book
-  const altTitle = xmlnode.parentNode.querySelector(':scope>alt-title');
-  const hasAltTitle = altTitle && altTitle.innerHTML !== xmlnode.innerHTML;
-  if (this.isBook && hasAltTitle) {
-    container.setAttribute('data-ams-doc-alttitle', container.textContent + altTitle.innerHTML); // NOTE assumes previousSibling handled first
-  }
-
-  // add main node and recurse
-  const actualSpan = this.createNode('span', '', { 'data-ams-doc':  xmlnode.tagName });
-  container.appendChild(actualSpan);
-  // NOTE: no space after title/label if there's only one. Downstream might insert some (e.g., with inlined headings).
-  this.passThrough(actualSpan, xmlnode);
+  if (xmlnode.innerHTML.trim() === '') return;
+  const span = this.createNode('span', '', { 'data-ams-doc': xmlnode.tagName });
+  htmlParentNode.appendChild(span);
+  this.passThrough(span, xmlnode);
 };

package/lib/elements/notes.js

@@ -32,5 +32,6 @@ export default function (htmlParentNode, xmlnode) {
   htmlParentNode.appendChild(section);
   if (this.isBook) section.setAttribute('data-ams-doc-level', '0');
 
+  this.createHeading(section, xmlnode);
   this.passThrough(section, xmlnode);
 };

package/lib/elements/ref-list.js

@@ -36,6 +36,7 @@ export default function (htmlParentNode, xmlnode) {
   });
   mapAttributes(section, xmlnode);
   htmlParentNode.appendChild(section);
+  this.createHeading(section, xmlnode);
   if (!isNestedRefList) {
     section.setAttribute('role', 'doc-bibliography');
   }
@@ -43,7 +44,6 @@ export default function (htmlParentNode, xmlnode) {
     this.passThrough(section, xmlnode);
     return;
   }
-  this.recurseTheDom(section, xmlnode.querySelector('title'));
   const dl = this.createNode('dl', '');
   section.appendChild(dl);
   xmlnode.querySelectorAll('ref').forEach(this.recurseTheDom.bind(null, dl));

package/lib/elements/sec.js

@@ -53,6 +53,8 @@ export default function (htmlParentNode, xmlnode) {
   });
   htmlParentNode.appendChild(section);
 
+  if (xmlnode.hasAttribute('style')) section.setAttribute('data-ams-style', xmlnode.getAttribute('style'));
+
   if (specificUse === 'part') {
     section.setAttribute('role', 'doc-part');
   }
@@ -80,7 +82,8 @@ export default function (htmlParentNode, xmlnode) {
   if (tagName === 'book-app-group') {
     section.setAttribute('data-ams-doc', 'app-group');
   }
-  if (tagName === 'book-app') section.setAttribute('role', 'doc-appendix');
+  if (tagName === 'book-app' && !xmlnode.parentNode.closest('book-app')) section.setAttribute('role', 'doc-appendix'); // TODO: revisit after AmerMathSoc/texml#223
 
+  this.createHeading(section, xmlnode);
   this.passThrough(section, xmlnode);
 };

package/lib/elements/secheading.js

@@ -25,16 +25,17 @@ export default function (htmlParentNode, xmlnode) {
   const parentLevel = htmlParentNode.closest('[data-ams-doc-level]')?.getAttribute('data-ams-doc-level') || 5;
   const span = this.createNode('span', '', {
     'data-ams-doc': 'secheading',
-    'data-ams-doc-level': parseInt(parentLevel)+1,
+    'data-ams-doc-level': parseInt(parentLevel) + 1,
   });
   mapAttributes(span, xmlnode);
 
   htmlParentNode.appendChild(span);
+  // NOTE: secheadings are re-written to headings downstream so we keep them flat here
   const label = xmlnode.querySelector('label');
   const title = xmlnode.querySelector('title');
   if (label) {
     this.passThrough(span, label);
-    if (title) span.insertAdjacentText('beforeend', ' '); //TODO: cf. & unify with label.js 
+    if (title) span.insertAdjacentText('beforeend', ' ');
   }
   if (!title) return;
   this.passThrough(span, title);

package/lib/elements/statement.js

@@ -27,5 +27,16 @@ export default function (htmlParentNode, xmlnode) {
   });
   mapAttributes(figure, xmlnode);
   htmlParentNode.appendChild(figure);
+
+  const label = xmlnode.querySelector(':scope>label:not(:empty)');
+  const title = xmlnode.querySelector(':scope>title:not(:empty)');
+  if (label || title) {
+    const figcaption = this.createNode('figcaption');
+    figure.appendChild(figcaption);
+    this.redirectRecurseTheDom(figcaption, label);
+    if (label && title) figcaption.insertAdjacentText('beforeend', ' ');
+    this.redirectRecurseTheDom(figcaption, title);
+  }
+
   this.passThrough(figure, xmlnode);
 };

package/lib/elements/subtitle.js

@@ -14,15 +14,12 @@
  *  limitations under the License.
  */
 
-import getParentLevel from '../helpers/getParentLevel.js';
-
 /**
- * subtitle element
+ * subtitle element, usually redirected
  * @param {HTMLElement} htmlParentNode 
  * @param {Element} xmlnode 
  */
 export default function (htmlParentNode, xmlnode) {
-  // NOTE the subtitle is also touched in label.js (which creates a header as htmlParent)
   const p = this.createNode('p', '', { 'data-ams-doc': 'subtitle' });
   htmlParentNode.appendChild(p);
   this.passThrough(p, xmlnode);

package/lib/elements/tag.js

@@ -37,11 +37,11 @@ export default function (htmlParentNode, xmlnode) {
     }
     // rarely, it appears in "text equations" (cf. formula.js, target.js)
     else if (xmlnode.closest('disp-formula[content-type="text"]')) {
-        const span = this.createNode('span', '', { 'data-ams-doc': 'label', id: xmlnode.parentNode.id }); // NOTE: tag should always have a <target> parent with an ID.
+        const span = this.createNode('span', '', { 'data-ams-doc': 'label'});
         if (xmlnode.getAttribute('parens') === 'yes') span.insertAdjacentText('afterbegin', '(')
         this.passThrough(span, xmlnode);
         if (xmlnode.getAttribute('parens') === 'yes') span.insertAdjacentText('beforeend', ')')
-        htmlParentNode.prepend(span);
+        htmlParentNode.prepend(span); //NOTE: we prepend to get left-side tags even when there is no target; but cf. AmerMathSoc/texml#232 
     }
     // tag extraction 
     // if some tag in the equation is linked, we store all tags in a special element in the HTML parent

package/lib/elements/target.js

@@ -27,13 +27,10 @@ export default function (htmlParentNode, xmlnode) {
     node2macro.apply(this, [htmlParentNode, xmlnode, `cssId{${xmlnode.id}}`, false]);
     return;
   }
-  // NOTE: inside a "text equation" we pass it through; the tag child will pick up the ID.
-  if (xmlnode.parentNode.tagName === 'disp-formula' && xmlnode.parentNode.getAttribute('content-type') === 'text') {
-    this.passThrough(htmlParentNode, xmlnode);
-    return;
-  }
   const span = this.createNode('span');
-  htmlParentNode.appendChild(span);
+  const isTextEquation = (xmlnode.parentNode.tagName === 'disp-formula' && xmlnode.parentNode.getAttribute('content-type') === 'text');
+  if (isTextEquation) htmlParentNode.prepend(span); // NOTE: inside a "text equation" we prepend to get a left-side label, but cf. also AmerMathSoc/texml#232 
+  else htmlParentNode.appendChild(span);
   mapAttributes(span, xmlnode);
   this.passThrough(span, xmlnode);
 };

package/lib/elements/{subject.js → title-group.js}

@@ -1,5 +1,5 @@
 /*!
- *  Copyright (c) 2023 American Mathematical Society
+ *  Copyright (c) 2025 American Mathematical Society
  *
  *  Licensed under the Apache License, Version 2.0 (the "License");
  *  you may not use this file except in compliance with the License.
@@ -15,12 +15,10 @@
  */
 
 /**
- * subject element
+ * title-group element mapped to h1
  * @param {HTMLElement} htmlParentNode 
  * @param {Element} xmlnode 
  */
 export default function (htmlParentNode, xmlnode) {
-  const dd = this.createNode('dd');
-  htmlParentNode.appendChild(dd);
-  this.passThrough(dd, xmlnode);
+    this.createHeading(htmlParentNode, xmlnode);
 };

package/lib/head.js

@@ -27,7 +27,7 @@ export const setHead = (xmldoc, htmldoc) => {
       xmldoc.querySelector('front>article-meta>title-group>alt-title') ||
       xmldoc.querySelector(
         'book-meta>book-title-group>book-title, front>article-meta>title-group>article-title'
-      );
+      );// NOTE title might contain complex markup (no use case yet; however e.g. surv279 has math-as-unicode). See also book-meta-json.js
     htmldoc.title = xmlTitle ? xmlTitle.textContent : 'AMS Publication';
   
     const root = xmldoc.querySelector('article, book');

package/lib/postProcessing.js

@@ -0,0 +1,81 @@
+/*!
+ *  Copyright (c) 2025 American Mathematical Society
+ *
+ *  Licensed under the Apache License, Version 2.0 (the "License");
+ *  you may not use this file except in compliance with the License.
+ *  You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ *  Unless required by applicable law or agreed to in writing, software
+ *  distributed under the License is distributed on an "AS IS" BASIS,
+ *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ *  See the License for the specific language governing permissions and
+ *  limitations under the License.
+ */
+
+// From https://developer.mozilla.org/en-US/docs/Web/Guide/HTML/Content_categories#phrasing_content
+// Note: 'a', 'del', 'ins', 'link', 'map', 'meta' have additional constraints that we ignore since they shouldn't occur in our content
+const phrasingContentTags = ['abbr', 'audio', 'b', 'bdo', 'br', 'button', 'canvas', 'cite', 'code', 'data', 'datalist', 'dfn', 'em', 'embed', 'i', 'iframe', 'img', 'input', 'kbd', 'label', 'mark', 'math', 'meter', 'noscript', 'object', 'output', 'picture', 'progress', 'q', 'ruby', 'samp', 'script', 'select', 'small', 'span', 'strong', 'sub', 'sup', 'svg', 'textarea', 'time', 'u', 'var', 'video', 'wbr', 'a', 'del', 'ins', 'link', 'map', 'meta']
+
+/**
+ * Checks if node is an element that is not phrasing content.
+ * @param {Node} node 
+ * @returns Boolean
+ */
+const isProperFlow = node =>
+    node.nodeType === 1
+    && !phrasingContentTags.includes(node.tagName.toLowerCase())
+    && !node.tagName.includes('-');
+
+/**
+ * Helper to remove whitespace-only elements
+ * @param {HTMLElement} node 
+ */
+const removeIfEmpty = node => {
+    if (node.innerHTML.trim() === '') node.remove();
+}
+
+/**
+ * Split paragraph node into multiple nodes if it contains non-phrasing content.
+ * @param {HTMLElement} node 
+ */
+const splitAtNonPhrasing = node => {
+    const document = node.ownerDocument;
+    const tagName = node.tagName.toLowerCase();
+    const childNodes = [...node.childNodes];
+    const properFlowChild = childNodes.find(isProperFlow);
+    if (!properFlowChild) return;
+    let currentNode = node;
+    let index = 0
+    while (index < childNodes.length) {
+        const child = childNodes[index];
+        if (isProperFlow(child)) {
+            currentNode.insertAdjacentElement('afterend', child);
+        }
+        else {
+            currentNode.appendChild(child);
+        }
+        // continuation
+        index++;
+        if (!childNodes[index]) break;
+        if (isProperFlow(child) && !isProperFlow(childNodes[index])) {
+            const newNode = document.createElement(tagName);
+            child.insertAdjacentElement('afterend', newNode);
+            removeIfEmpty(currentNode);
+            currentNode = newNode;
+        }
+        else if (isProperFlow(child)) {
+            currentNode = child;
+        }
+    }
+    removeIfEmpty(currentNode);
+}
+
+/**
+ * Wrapper around fixes for discrepancies between XML and HTML content models
+ * @param {Document} document 
+ */
+export const fixContentModel = document => {
+    document.querySelectorAll('p').forEach(splitAtNonPhrasing);
+}