diff --git a/packages/epub-utils/package.json b/packages/epub-utils/package.json
index 519058ff..d2232bc7 100644
--- a/packages/epub-utils/package.json
+++ b/packages/epub-utils/package.json
@@ -21,7 +21,7 @@
"fs-extra": "^6.0.1",
"tmp": "^0.0.33",
"winston": "^2.4.0",
- "xmldom": "^0.1.27",
+ "xmldom": "https://github.com/fchasen/xmldom.git#a38f7ddb536ab74e9fb549477ba9f9b7ea2d0beb",
"xpath": "^0.0.24"
},
"publishConfig": {
diff --git a/packages/epub-utils/src/epub-parse.js b/packages/epub-utils/src/epub-parse.js
index 520f3ffe..eb16f6a9 100644
--- a/packages/epub-utils/src/epub-parse.js
+++ b/packages/epub-utils/src/epub-parse.js
@@ -17,6 +17,13 @@ const path = require('path');
const xpath = require('xpath');
const winston = require('winston');
+// Error Handler for DOMParser instances
+const errorHandler = {
+ warning: w => winston.warn(w),
+ error: e => winston.warn(e),
+ fatalError: fe => winston.error(fe),
+}
+
function SpineItem() {
this.filepath = "";
this.relpath = "";
@@ -33,7 +40,7 @@ function EpubParser() {
function parseNavDoc(fullpath, epubDir) {
const content = fs.readFileSync(fullpath).toString();
- const doc = new DOMParser().parseFromString(content);
+ const doc = new DOMParser({errorHandler}).parseFromString(content, 'application/xhtml+xml');
// Remove all links
const aElems = doc.getElementsByTagNameNS('http://www.w3.org/1999/xhtml', 'a');
@@ -127,7 +134,7 @@ EpubParser.prototype.parse = function(epubDir) {
EpubParser.prototype.parseData = function(packageDocPath, epubDir) {
const content = fs.readFileSync(packageDocPath).toString();
- const doc = new DOMParser().parseFromString(content);
+ const doc = new DOMParser({errorHandler}).parseFromString(content);
const select = xpath.useNamespaces(
{ opf: 'http://www.idpf.org/2007/opf',
dc: 'http://purl.org/dc/elements/1.1/'});
@@ -168,7 +175,7 @@ EpubParser.prototype.parseData = function(packageDocPath, epubDir) {
EpubParser.prototype.parseContentDocTitle = function(filepath) {
const content = fs.readFileSync(filepath).toString();
- const doc = new DOMParser().parseFromString(content);
+ const doc = new DOMParser({errorHandler}).parseFromString(content, 'application/xhtml+xml');
const select = xpath.useNamespaces({html: "http://www.w3.org/1999/xhtml", epub: "http://www.idpf.org/2007/ops"});
const title = select('//html:title/text()', doc);
if (title.length > 0) {
@@ -182,7 +189,7 @@ EpubParser.prototype.parseContentDocTitle = function(filepath) {
EpubParser.prototype.calculatePackageDocPath = function(epubDir) {
const containerFilePath = `${epubDir}/META-INF/container.xml`;
const content = fs.readFileSync(containerFilePath).toString();
- const doc = new DOMParser().parseFromString(content);
+ const doc = new DOMParser({errorHandler}).parseFromString(content);
const select = xpath.useNamespaces({ ocf: 'urn:oasis:names:tc:opendocument:xmlns:container' });
const rootfiles = select('//ocf:rootfile[@media-type="application/oebps-package+xml"]/@full-path', doc);
// just grab the first one as we're not handling the case of multiple renditions
diff --git a/tests/__tests__/cli.test.js b/tests/__tests__/cli.test.js
index ba62b91b..916569d5 100644
--- a/tests/__tests__/cli.test.js
+++ b/tests/__tests__/cli.test.js
@@ -96,7 +96,17 @@ describe('Running the CLI', () => {
const log = stripAnsi(stdout);
expect(/^warn:\s+The SVG Content Documents in this EPUB will be ignored\./m.test(log)).toBe(true);
});
- });
+ });
+
+ describe('does not raise a warning', () => {
+ test('when a named character reference is used in XHTML', () => {
+ const { stdout, stderr, status } = ace(['issue-182'], {
+ cwd: path.resolve(__dirname, '../data'),
+ });
+ const log = stripAnsi(stdout);
+ expect(/^warn:\s+\[xmldom error\] entity not found/m.test(log)).toBe(false);
+ });
+ });
/*test('with return-2-on-validation-error set to true should exit with return code 2', () => {
// TODO this test won't work until we can specify the CLI option to enable returning 2 on violation(s)
diff --git a/tests/__tests__/regression.test.js b/tests/__tests__/regression.test.js
index d87573cb..1c1751e2 100644
--- a/tests/__tests__/regression.test.js
+++ b/tests/__tests__/regression.test.js
@@ -75,3 +75,15 @@ test('issue #170: heading with `doc-subtitle` role were reported empty', async (
const report = await ace('../data/issue-170');
expect(report['earl:result']['earl:outcome']).toEqual('pass');
});
+
+test('issue #182: named character references are parsed', async () => {
+ const report = await ace('../data/issue-182');
+ expect(report.assertions).toEqual(expect.arrayContaining([
+ expect.objectContaining({
+ "earl:testSubject": {
+ "url": "content_001.xhtml",
+ "dct:title": "Minimal – EPUB"
+ }
+ })
+ ]));
+});
diff --git a/tests/data/issue-182/EPUB/content_001.xhtml b/tests/data/issue-182/EPUB/content_001.xhtml
new file mode 100644
index 00000000..a0415f08
--- /dev/null
+++ b/tests/data/issue-182/EPUB/content_001.xhtml
@@ -0,0 +1,10 @@
+
+
+
+Minimal – EPUB
+
+
+Loomings
+Call me Ishmael.
+
+
diff --git a/tests/data/issue-182/EPUB/nav.xhtml b/tests/data/issue-182/EPUB/nav.xhtml
new file mode 100644
index 00000000..1d538c19
--- /dev/null
+++ b/tests/data/issue-182/EPUB/nav.xhtml
@@ -0,0 +1,12 @@
+
+
+Minimal Nav
+
+
+
+
+
diff --git a/tests/data/issue-182/EPUB/package.opf b/tests/data/issue-182/EPUB/package.opf
new file mode 100644
index 00000000..17cf544e
--- /dev/null
+++ b/tests/data/issue-182/EPUB/package.opf
@@ -0,0 +1,23 @@
+
+
+
+ Minimal EPUB 3.0
+ en
+ NOID
+ 2017-01-01T00:00:01Z
+ structuralNavigation
+ everything OK!
+ noFlashingHazard
+ noSoundHazard
+ noMotionSimulationHazard
+ textual
+ textual
+
+
+