@amermathsoc/texml-to-html 15.0.0 → 15.1.0

package/CHANGELOG.md

@@ -2,6 +2,18 @@
 
 All notable changes to this project will be documented in this file. See [standard-version](https://github.com/conventional-changelog/standard-version) for commit guidelines.
 
+## [15.1.0](https://github.com/AmerMathSoc/texml-to-html/compare/v15.0.0...v15.1.0) (2023-11-23)
+
+
+### Features
+
+* **label.js:** preserve title and label ([5dae34f](https://github.com/AmerMathSoc/texml-to-html/commit/5dae34f1dfeaf8fa884303f7de2518b35d52c38b)), closes [#327](https://github.com/AmerMathSoc/texml-to-html/issues/327)
+
+
+### Bug Fixes
+
+* **package:** upgrade linkedom ([34ae8a7](https://github.com/AmerMathSoc/texml-to-html/commit/34ae8a7f48c2d6f2744e22c9f07aed4159a1f896))
+
 ## [15.0.0](https://github.com/AmerMathSoc/texml-to-html/compare/v14.0.1...v15.0.0) (2023-11-02)
 
 

package/README.md

@@ -2,15 +2,22 @@
 
 Converting texml-generated JATS/BITS-like XML to HTML.
 
-The most basic use:
+## Getting started
+
+### Quick example
+
+For a first test run, try [an example](./examples), e.g.,
+
+* Install via npm: `$ npm i @amermathsoc/texml-to-html`
+* Process a test file: `$ node node_modules/@amermathsoc/texml-to-html/examples/cli.js node_modules/@amermathsoc/texml-to-html/text/article.xml > htmlOutput.html`
+
+### Basic usage
 
 ```js
 import fs from 'fs';
 import path from 'path';
-import xml2html from './texml-to-html.js';
+import xml2html from '@amermathsoc/texml-to-html';
 
 const article = xml2html(fs.readFileSync(path.resolve(process.argv[2])).toString()).window.document;
 console.log(article.toString());
 ```
-
-See also [the examples](./examples).

package/lib/elements/label.js

@@ -17,7 +17,7 @@
 import getParentLevel from '../helpers/getParentLevel.js';
 
 /**
- * label and title elements
+ * label and title elements. If both are present, expects label+title (not title+label)
  * @param {HTMLElement} htmlParentNode 
  * @param {Element} xmlnode 
  */
@@ -41,7 +41,8 @@ export default function (htmlParentNode, xmlnode) {
     this.passThrough(htmlParentNode, xmlnode);
     return;
   }
-  // CASE fig, fig-group, table-wrap, table-wrap-group via caption
+  // CASE fig, fig-group, table-wrap, table-wrap-group via caption 
+  // TODO: could we just return?
   if (
     xmlnode.parentNode.tagName === 'table-wrap' ||
     xmlnode.parentNode.tagName === 'table-wrap-group' ||
@@ -54,33 +55,33 @@ 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;
+  if (xmlnode.tagName === 'label' && xmlnode.innerHTML.trim() === '') return; // TODO: I think texml is better about this now (only found in mcl11)
 
   // complex cases
 
-  // Decide output markup (h* or figcaption; wrapping header for subtitles)
+  // 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 heading = (isStatement || isDispFormulaGroup) ? this.createNode('figcaption', '') : this.createNode(`h${level}`, '');
-  htmlParentNode.appendChild(heading);
+  const container = (isStatement || isDispFormulaGroup) ? this.createNode('figcaption', '') : this.createNode(`h${level}`, '');
+  htmlParentNode.appendChild(container);
 
-  // subtitle handling (assumes heading is not figcaption)
+  // 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(heading);
+    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() !== '') {
+  if (previousSibling?.tagName === 'label' && previousSibling.innerHTML.trim() !== '') { //TODO: empty label check unnecessary? cf. earlier TODO re empty label
     const labelSpan = this.createNode('span', '', { 'data-ams-doc': 'label' });
-    heading.appendChild(labelSpan);
+    container.appendChild(labelSpan);
     this.passThrough(labelSpan, previousSibling);
     const labelSeparatorString = isStatement ? ' ' : '. ';
     labelSpan.insertAdjacentText('beforeend', labelSeparatorString);
@@ -90,12 +91,14 @@ export default function (htmlParentNode, xmlnode) {
   const altTitle = xmlnode.parentNode.querySelector(':scope>alt-title');
   const hasAltTitle = altTitle && altTitle.innerHTML !== xmlnode.innerHTML;
   if (this.isBook && hasAltTitle) {
-    heading.setAttribute('data-ams-doc-alttitle', heading.textContent + altTitle.innerHTML); // NOTE assumes previousSibling handled first
+    container.setAttribute('data-ams-doc-alttitle', container.textContent + altTitle.innerHTML); // NOTE assumes previousSibling handled first
   }
 
-  // recurse main node
-  this.passThrough(heading, xmlnode);
+  // add main node and recurse
+  const actualSpan = this.createNode('span', '', { 'data-ams-doc':  xmlnode.tagName });
+  container.appendChild(actualSpan);
+  this.passThrough(actualSpan, xmlnode);
 
-  // faking TeX's punctutation-at-end logic
-  if (isStatement) heading.insertAdjacentText('beforeend', heading.textContent.endsWith('.') ? ' ' : '. ');
+  // faking TeX's punctutation-at-end logic // TODO: should we move punctuation out of the title/label?
+  if (isStatement) container.insertAdjacentText('beforeend', container.textContent.endsWith('.') ? ' ' : '. ');
 };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@amermathsoc/texml-to-html",
-  "version": "15.0.0",
+  "version": "15.1.0",
   "type": "module",
   "description": "A NodeJS library for converting AMS-style JATS XML to HTML",
   "scripts": {
@@ -29,11 +29,11 @@
   },
   "homepage": "https://github.com/AmerMathSoc/texml-to-html#readme",
   "devDependencies": {
-    "c8": "8.0.0",
-    "standard-version": "9.3.2",
-    "tape": "5.5.0"
+    "c8": "8.0.1",
+    "standard-version": "9.5.0",
+    "tape": "5.7.2"
   },
   "dependencies": {
-    "linkedom": "0.14.20"
+    "linkedom": "0.16.4"
   }
 }

package/test/element-book-ref-list.js

@@ -25,7 +25,7 @@ tape('Template: book-back/ref-list ', async function(t) {
   const bibl = document.querySelector('section#reflist[role="doc-bibliography"]');
   t.ok(bibl, 'ref-list becomes section with id, role, doc-level');
   const title = bibl.querySelector('h1');
-  t.equal(title.outerHTML, '<h1>title</h1>', 'ref-list title as heading');
+  t.equal(title.outerHTML, '<h1><span data-ams-doc="title">title</span></h1>', 'ref-list title as heading');
   const list = bibl.querySelector('dl');
   t.ok(list, 'dl created for ref items');
   t.ok(document.querySelector('section#reflistchapter [role="doc-bibliography"] h2'), 'Chapter-level ref-list title gets h2')

package/test/element-disp-formula-group.js

@@ -23,5 +23,5 @@ tape('disp-formula-group', async function(t) {
   const groupChild = document.querySelector('#equations figure[data-ams-doc="statement"]#disp-formula-group');
   t.ok(groupChild, 'disp-formula-group as figure with data-ams-doc="statement" with id')
   t.equal(groupChild.firstElementChild.tagName, 'FIGCAPTION', 'disp-formula-group label');
-  t.equal(groupChild.firstElementChild.innerHTML, 'Formula Group', 'disp-formula-group passes label along');
+  t.equal(groupChild.firstElementChild.innerHTML, '<span data-ams-doc="label">Formula Group</span>', 'disp-formula-group passes label along');
 });

package/test/element-sec-app-front-matter-part-dedication-title-label.js

@@ -29,14 +29,14 @@ tape('sec, app, front-matter-part, dedication, title, label', async function(t)
   t.ok(document.querySelector('#intro2[role="doc-introduction"]'), 'sec with matching title text to role doc-introduction');
   t.ok(document.querySelector('#ded[role="doc-dedication"]'), 'dedication to role doc-dedication');
 
-  t.equal(document.querySelector('#subsec h3').innerHTML, 'Subsection', 'subsection without wrapping section gets correct level');
+  t.equal(document.querySelector('#subsec h3').innerHTML, '<span data-ams-doc="label">Subsection</span>', 'subsection without wrapping section gets correct level');
 
-  t.equal(document.querySelector('#seclabeltitle h2').innerHTML, '<span data-ams-doc="label">Label. </span>Title', 'sec with label+title: heading level and content');
-  t.equal(document.querySelector('#sectitle h2').innerHTML, 'Title', 'sec with title: heading level and content');
-  t.equal(document.querySelector('#seclabel h2').innerHTML, 'Label', 'sec with label: heading level and content');
-  t.equal(document.querySelector('#applabeltitle h2').innerHTML, '<span data-ams-doc="label">Label. </span>Title', 'app with label+title: heading level and content');
-  t.equal(document.querySelector('#apptitle h2').innerHTML, 'Title', 'app with title: heading level and content');
-  t.equal(document.querySelector('#applabel h2').innerHTML, 'Label', 'app with label: heading level and content');
+  t.equal(document.querySelector('#seclabeltitle h2').innerHTML, '<span data-ams-doc="label">Label. </span><span data-ams-doc="title">Title</span>', 'sec with label+title: heading level and content');
+  t.equal(document.querySelector('#sectitle h2').innerHTML, '<span data-ams-doc="title">Title</span>', 'sec with title: heading level and content');
+  t.equal(document.querySelector('#seclabel h2').innerHTML, '<span data-ams-doc="label">Label</span>', 'sec with label: heading level and content');
+  t.equal(document.querySelector('#applabeltitle h2').innerHTML, '<span data-ams-doc="label">Label. </span><span data-ams-doc="title">Title</span>', 'app with label+title: heading level and content');
+  t.equal(document.querySelector('#apptitle h2').innerHTML, '<span data-ams-doc="title">Title</span>', 'app with title: heading level and content');
+  t.equal(document.querySelector('#applabel h2').innerHTML, '<span data-ams-doc="label">Label</span>', 'app with label: heading level and content');
 
   t.ok(document.querySelector('#sectitle header p[data-ams-doc="subtitle"]'), 'sec with title: subtitle to p with data-ams-doc');
   t.ok(document.querySelector('#seclabel header p[data-ams-doc="subtitle"]'), 'sec with title: subtitle to p with data-ams-doc');
@@ -52,15 +52,15 @@ tape('sec, app, front-matter-part, dedication, title, label', async function(t)
   const document2 = book;
   t.ok(document2.querySelector('#chapter[role="doc-chapter"]'), 'sec with specific-use chapter to role doc-chapter'); // NOTE so far, chapters only occur in books but the xslt doesn't check for it
 
-  t.equal(document2.querySelector('#seclabeltitle h2').innerHTML, '<span data-ams-doc="label">Label. </span>Title', 'sec with label+title: heading level and content');
-  t.equal(document2.querySelector('#sectitle h2').innerHTML, 'Title', 'sec with title: heading level and content');
-  t.equal(document2.querySelector('#seclabel h2').innerHTML, 'Label', 'sec with label: heading level and content');
-  t.equal(document2.querySelector('#fmlabeltitle h1').innerHTML, '<span data-ams-doc="label">Label. </span>Title', 'front-matter-part with label+title: heading level and content');
-  t.equal(document2.querySelector('#fmtitle h1').innerHTML, 'Title', 'front-matter-part with title: heading level and content');
-  t.equal(document2.querySelector('#fmlabel h1').innerHTML, 'Label', 'front-matter-part with label: heading level and content');
-  t.equal(document2.querySelector('#applabeltitle h1').innerHTML, '<span data-ams-doc="label">Label. </span>Title', 'app with label+title: heading level and content');
-  t.equal(document2.querySelector('#apptitle h1').innerHTML, 'Title', 'app with title: heading level and content');
-  t.equal(document2.querySelector('#applabel h1').innerHTML, 'Label', 'app with label: heading level and content');
+  t.equal(document2.querySelector('#seclabeltitle h2').innerHTML, '<span data-ams-doc="label">Label. </span><span data-ams-doc="title">Title</span>', 'sec with label+title: heading level and content');
+  t.equal(document2.querySelector('#sectitle h2').innerHTML, '<span data-ams-doc="title">Title</span>', 'sec with title: heading level and content');
+  t.equal(document2.querySelector('#seclabel h2').innerHTML, '<span data-ams-doc="label">Label</span>', 'sec with label: heading level and content');
+  t.equal(document2.querySelector('#fmlabeltitle h1').innerHTML, '<span data-ams-doc="label">Label. </span><span data-ams-doc="title">Title</span>', 'front-matter-part with label+title: heading level and content');
+  t.equal(document2.querySelector('#fmtitle h1').innerHTML, '<span data-ams-doc="title">Title</span>', 'front-matter-part with title: heading level and content');
+  t.equal(document2.querySelector('#fmlabel h1').innerHTML, '<span data-ams-doc="label">Label</span>', 'front-matter-part with label: heading level and content');
+  t.equal(document2.querySelector('#applabeltitle h1').innerHTML, '<span data-ams-doc="label">Label. </span><span data-ams-doc="title">Title</span>', 'app with label+title: heading level and content');
+  t.equal(document2.querySelector('#apptitle h1').innerHTML, '<span data-ams-doc="title">Title</span>', 'app with title: heading level and content');
+  t.equal(document2.querySelector('#applabel h1').innerHTML, '<span data-ams-doc="label">Label</span>', 'app with label: heading level and content');
 
   t.equal(document2.querySelector('#part').getAttribute('role'), 'doc-part', 'part role');
   t.ok(document2.querySelector('#inparttitle h2'), 'sec with title in chapter in part: heading level reduced');

package/test/element-sec-meta.js

@@ -25,7 +25,7 @@ tape('Template: sec-meta => contrib-group, author-comment, abstract/title', asyn
   const secMeta = section.querySelector('section[data-ams-doc="sec-meta"]');
   t.equal(secMeta.getAttribute('data-ams-contributors'), '{"null":[]}', 'book sec-meta');
   t.ok(section.querySelector('h1'), 'title becomes h1');
-  t.equal(section.querySelector('h1').innerHTML, 'Title', 'title content in heading');
+  t.equal(section.querySelector('h1').innerHTML, '<span data-ams-doc="title">Title</span>', 'title content in heading');
   t.ok(section.querySelector('[role="doc-abstract"] h2'), 'abstract title becomes h2');
 
   const document2 = article;

package/test/element-statement-label-title.js

@@ -29,7 +29,7 @@ tape('Template: statement, label, title', async function(t) {
   t.ok(statement1Heading, 'statement heading level in article');
   t.equal(
     statement1Heading.innerHTML,
-    '<span data-ams-doc="label">Label 1 </span>Title 1. ',
+    '<span data-ams-doc="label">Label 1 </span><span data-ams-doc="title">Title 1</span>. ',
     'statement with label+title creates space before and period after title'
   );
   const statement2 = document.querySelector(
@@ -38,7 +38,7 @@ tape('Template: statement, label, title', async function(t) {
   t.ok(statement2, 'statement 2');
   t.equal(
     statement2.querySelector('figcaption').innerHTML,
-    'Label 2. ',
+    '<span data-ams-doc="label">Label 2</span>. ',
     'statement with label creates period after'
   );
   const statement3 = document.querySelector(
@@ -47,7 +47,7 @@ tape('Template: statement, label, title', async function(t) {
   t.ok(statement3, 'statement 3');
   t.equal(
     statement3.querySelector('figcaption').innerHTML,
-    'Title 3. ',
+    '<span data-ams-doc="title">Title 3</span>. ',
     'statement with title creates period after'
   );
   const statementTitlePeriod = document.querySelector(
@@ -55,7 +55,7 @@ tape('Template: statement, label, title', async function(t) {
   );
   t.equal(
     statementTitlePeriod.querySelector('figcaption').innerHTML,
-    'Title 4. ',
+    '<span data-ams-doc="title">Title 4.</span> ',
     'statement with title with period does not have extra period after'
   );
   const statement4 = document.querySelector(
@@ -64,7 +64,7 @@ tape('Template: statement, label, title', async function(t) {
   t.ok(statement4, 'statement 4');
   t.equal(
     statement4.querySelector('figcaption').innerHTML,
-    'Proof. ',
+    '<span data-ams-doc="title">Proof</span>. ',
     'title in proof statement now also gets extra period'
   );
   const statement5 = document.querySelector(
@@ -73,7 +73,7 @@ tape('Template: statement, label, title', async function(t) {
   t.ok(statement5, 'statement 5');
   t.equal(
     statement5.querySelector('figcaption').innerHTML,
-    '<span data-ams-doc="label">Label 5 </span>Proof. ',
+    '<span data-ams-doc="label">Label 5 </span><span data-ams-doc="title">Proof</span>. ',
     'proof statement with label+title'
   );
   t.equal(

package/test/element-stripEmptyLabel.js

@@ -22,5 +22,5 @@ tape('Empty Labels should be stripped', async function(t) {
   t.plan(2);
   const document = article;
   t.equal(document.querySelector('#emptyLabel').innerHTML.trim(), '', 'Statement with no title and empty label');
-  t.equal(document.querySelector('#titleEmptyLabel').innerHTML.trim(), '<figcaption>Title. </figcaption>', 'Statement with no title and empty label');
+  t.equal(document.querySelector('#titleEmptyLabel').innerHTML.trim(), '<figcaption><span data-ams-doc="title">Title</span>. </figcaption>', 'Statement with no title and empty label');
 });