This is an automated email from the ASF dual-hosted git repository. mraible pushed a commit to branch feed-authors-and-nofollow-6.1.x in repository https://gitbox.apache.org/repos/asf/roller.git
commit 530f38ed842e812f1d066271f8b12be5286b36b1 Author: Matt Raible <[email protected]> AuthorDate: Tue Nov 17 20:05:15 2026 -0700 Show comment and media file authors once-escaped in feeds and keep existing rel values in addNofollow The comment feed macros and the media file feed now turn author names back into text with unescapeHTML before XML-escaping them, as the entry author macros do, so a name like "A & B" no longer shows as "A & B" in feed readers. FeedAuthorEscapingTest covers the Atom and RSS comment feeds and the media file feed. Utilities.addNofollow adds nofollow to a link's existing rel value instead of adding a second rel attribute, and a ">" inside a quoted attribute value no longer ends the link early. The linear-time test now allows 2 seconds so it doesn't fail on slow build machines. --- .../apache/roller/weblogger/util/Utilities.java | 66 +++++++++-- app/src/main/webapp/WEB-INF/velocity/feeds.vm | 6 +- .../rendering/velocity/FeedAuthorEscapingTest.java | 124 ++++++++++++++++++++- .../roller/weblogger/util/UtilitiesTest.java | 20 +++- 4 files changed, 199 insertions(+), 17 deletions(-) diff --git a/app/src/main/java/org/apache/roller/weblogger/util/Utilities.java b/app/src/main/java/org/apache/roller/weblogger/util/Utilities.java index 401edd55e..4d4188d74 100644 --- a/app/src/main/java/org/apache/roller/weblogger/util/Utilities.java +++ b/app/src/main/java/org/apache/roller/weblogger/util/Utilities.java @@ -43,6 +43,9 @@ public class Utilities { private static final Pattern mLinkPattern = Pattern.compile("<a href=", Pattern.CASE_INSENSITIVE); + private static final Pattern REL_ATTRIBUTE_PATTERN = Pattern.compile( + "\\srel\\s*=\\s*(?:\"([^\"]*)\"|'([^']*)'|([^\\s\"'=<>`]+))", + Pattern.CASE_INSENSITIVE); private static final Pattern OPENING_B_TAG_PATTERN = Pattern.compile( "<b>", Pattern.CASE_INSENSITIVE); private static final Pattern CLOSING_B_TAG_PATTERN = Pattern.compile( @@ -267,29 +270,26 @@ public class Utilities { /** * Code (stolen from Pebble) to add rel="nofollow" string to all links in - * HTML that don't already have it. + * HTML that don't already have it. A link with another rel value gets + * nofollow added to that value. */ public static String addNofollow(String html) { if (html == null || html.length() == 0) { return html; } - // Each link runs to the first ">" after its "href=". Once no ">" is - // left, no later link can close either, so stop there. + // Each link runs to the first ">" after its "href=" that isn't inside + // a quoted attribute value. Once no ">" is left, no later link can + // close either, so stop there. Matcher m = mLinkPattern.matcher(html); StringBuilder buf = new StringBuilder(html.length()); int end = 0; while (m.find()) { - int close = html.indexOf('>', m.end()); + int close = findTagEnd(html, m.end()); if (close < 0) { break; } - String link = html.substring(m.start(), close + 1); buf.append(html, end, m.start()); - if (link.contains("rel=\"nofollow\"")) { - buf.append(link); - } else { - buf.append(link, 0, link.length() - 1).append(" rel=\"nofollow\">"); - } + buf.append(addNofollowToLink(html.substring(m.start(), close + 1))); end = close + 1; m.region(end, html.length()); } @@ -297,6 +297,52 @@ public class Utilities { return buf.toString(); } + /** + * Index of the ">" that ends the tag, skipping any inside quoted attribute + * values, or -1 if the tag doesn't end. + */ + private static int findTagEnd(String html, int from) { + char quote = 0; + char previous = '='; + for (int i = from; i < html.length(); i++) { + char c = html.charAt(i); + if (quote != 0) { + if (c == quote) { + quote = 0; + } + } else if ((c == '"' || c == '\'') && previous == '=') { + quote = c; + } else if (c == '>') { + return i; + } + if (!Character.isWhitespace(c)) { + previous = c; + } + } + return -1; + } + + private static String addNofollowToLink(String link) { + Matcher rel = REL_ATTRIBUTE_PATTERN.matcher(link); + if (!rel.find()) { + return link.substring(0, link.length() - 1) + " rel=\"nofollow\">"; + } + int group = rel.group(1) != null ? 1 : rel.group(2) != null ? 2 : 3; + String value = rel.group(group); + for (String token : StringUtils.split(value)) { + if ("nofollow".equalsIgnoreCase(token)) { + return link; + } + } + String values = value.isBlank() ? "nofollow" : value + " nofollow"; + if (group == 3) { + // unquoted value: quote it so the added token stays part of it + return link.substring(0, rel.start(group)) + '"' + values + '"' + + link.substring(rel.end(group)); + } + return link.substring(0, rel.start(group)) + values + link.substring(rel.end(group)); + } + // ------------------------------------------------------------------------ /** * Replaces occurences of non-alphanumeric characters with an underscore. diff --git a/app/src/main/webapp/WEB-INF/velocity/feeds.vm b/app/src/main/webapp/WEB-INF/velocity/feeds.vm index 62416ad93..51f0a4beb 100644 --- a/app/src/main/webapp/WEB-INF/velocity/feeds.vm +++ b/app/src/main/webapp/WEB-INF/velocity/feeds.vm @@ -92,7 +92,7 @@ <entry> <id>${comment.weblogEntry.permalink}#comment-${comment.timestamp}</id> <title type="html">Re: $utils.escapeXML($comment.weblogEntry.title)</title> - <author><name>$utils.escapeXML($utils.removeHTML($comment.name))</name></author> + <author><name>$utils.escapeXML($utils.unescapeHTML($comment.name))</name></author> <link rel="alternate" type="text/html" href="${comment.weblogEntry.permalink}#comment-${comment.timestamp}"/> <published>$utils.formatIso8601Date($comment.postTime)</published> <updated>$utils.formatIso8601Date($comment.postTime)</updated> @@ -106,7 +106,7 @@ <item> <guid isPermaLink="true">${comment.weblogEntry.permalink}#comment-${comment.timestamp}</guid> <title>Re: $utils.escapeXML($comment.weblogEntry.title)</title> - <dc:creator>$utils.escapeXML($utils.removeHTML($comment.name))</dc:creator> + <dc:creator>$utils.escapeXML($utils.unescapeHTML($comment.name))</dc:creator> <pubDate>$utils.formatRfc822Date($comment.postTime)</pubDate> <description>$utils.escapeXML($comment.content)</description> </item> @@ -118,7 +118,7 @@ <entry> <id>$file.permalink</id> <title type="html">$utils.escapeXML($file.name)</title> - <author><name>$utils.escapeXML($utils.removeHTML($file.creator.screenName))</name></author> + <author><name>$utils.escapeXML($utils.unescapeHTML($file.creator.screenName))</name></author> <link rel="alternate" type="text/html" href="$file.permalink"/> <published>$utils.formatIso8601Date($file.dateUploaded)</published> <updated>$utils.formatIso8601Date($file.lastUpdated)</updated> diff --git a/app/src/test/java/org/apache/roller/weblogger/ui/rendering/velocity/FeedAuthorEscapingTest.java b/app/src/test/java/org/apache/roller/weblogger/ui/rendering/velocity/FeedAuthorEscapingTest.java index 9ab5f5c49..f3afaf5b5 100644 --- a/app/src/test/java/org/apache/roller/weblogger/ui/rendering/velocity/FeedAuthorEscapingTest.java +++ b/app/src/test/java/org/apache/roller/weblogger/ui/rendering/velocity/FeedAuthorEscapingTest.java @@ -29,7 +29,9 @@ import java.util.Properties; import javax.xml.parsers.DocumentBuilderFactory; import org.apache.roller.weblogger.pojos.User; +import org.apache.roller.weblogger.pojos.WeblogEntryComment; import org.apache.roller.weblogger.pojos.wrapper.UserWrapper; +import org.apache.roller.weblogger.pojos.wrapper.WeblogEntryCommentWrapper; import org.apache.roller.weblogger.ui.rendering.model.UtilitiesModel; import org.apache.velocity.VelocityContext; import org.apache.velocity.app.VelocityEngine; @@ -42,9 +44,10 @@ import static org.mockito.Mockito.mock; import static org.mockito.Mockito.when; /** - * Entry authors in the built-in feeds are plain text, so the feed macros must - * turn the HTML-escaped screen name from {@link UserWrapper} back into text - * and XML-escape it once. + * Entry, comment and media file authors in the built-in feeds are plain text, + * so the feed macros must turn the HTML-escaped names from {@link UserWrapper} + * and {@link WeblogEntryCommentWrapper}, and screen names stored already + * encoded, back into text and XML-escape them once. */ public class FeedAuthorEscapingTest { @@ -58,6 +61,13 @@ public class FeedAuthorEscapingTest { {"Tom & "Jerry"", "Tom & \"Jerry\""}, }; + /** Comment author name as submitted, and the text the feed must contain. */ + private static final String[][] COMMENT_NAMES = { + {"A & B", "A & B"}, + {"<b>Bold</b>", "<b>Bold</b>"}, + {"José \"Pepe\" O'Brien", "José \"Pepe\" O'Brien"}, + }; + /** Minimal stand-ins for the model objects the feed macros read. */ public static class StubEntry { private final UserWrapper creator; @@ -107,6 +117,78 @@ public class FeedAuthorEscapingTest { } } + /** Comment stand-in that takes its name from the real comment wrapper. */ + public static class StubComment { + private final WeblogEntryCommentWrapper wrapper; + + StubComment(String name) { + WeblogEntryComment comment = new WeblogEntryComment(); + comment.setName(name); + this.wrapper = WeblogEntryCommentWrapper.wrap(comment, null); + } + + public String getName() { + return wrapper.getName(); + } + + public StubEntry getWeblogEntry() { + return new StubEntry(null); + } + + public String getTimestamp() { + return "0"; + } + + public Date getPostTime() { + return new Date(0); + } + + public String getContent() { + return "Comment"; + } + } + + /** Media file stand-in; media files expose the creator's User directly. */ + public static class StubFile { + private final User creator; + + StubFile(User creator) { + this.creator = creator; + } + + public String getPermalink() { + return "https://example.org/blog/mediaresource/file"; + } + + public String getName() { + return "file.png"; + } + + public User getCreator() { + return creator; + } + + public Date getDateUploaded() { + return new Date(0); + } + + public Date getLastUpdated() { + return new Date(0); + } + + public List<Object> getTags() { + return Collections.emptyList(); + } + + public String getDescription() { + return null; + } + + public String getContentType() { + return "image/png"; + } + } + public static class StubCategory { public String getName() { return "General"; @@ -126,6 +208,8 @@ public class FeedAuthorEscapingTest { when(user.getScreenName()).thenReturn(name); VelocityContext ctx = new VelocityContext(); ctx.put("entries", List.of(new StubEntry(UserWrapper.wrap(user)))); + ctx.put("comments", List.of(new StubComment(name))); + ctx.put("files", List.of(new StubFile(user))); ctx.put("utils", new UtilitiesModel()); StringWriter out = new StringWriter(); engine.evaluate(ctx, out, "test", "#parse('feeds.vm')" + macro); @@ -160,4 +244,38 @@ public class FeedAuthorEscapingTest { .item(0).getTextContent(), xml); } } + + @Test + public void atomCommentAuthorIsText() throws Exception { + for (String[] name : COMMENT_NAMES) { + String xml = render("<feed xmlns=\"http://www.w3.org/2005/Atom\">" + + "#showCommentsAtom10($comments)</feed>", name[0]); + Document feed = parse(xml); + assertEquals(name[1], feed.getElementsByTagNameNS("http://www.w3.org/2005/Atom", "name") + .item(0).getTextContent(), xml); + } + } + + @Test + public void rssCommentCreatorIsText() throws Exception { + for (String[] name : COMMENT_NAMES) { + String xml = render("<rss xmlns:dc=\"http://purl.org/dc/elements/1.1/\"" + + " xmlns:atom=\"http://www.w3.org/2005/Atom\"><channel>" + + "#showCommentsRSS20($comments)</channel></rss>", name[0]); + Document feed = parse(xml); + assertEquals(name[1], feed.getElementsByTagNameNS("http://purl.org/dc/elements/1.1/", "creator") + .item(0).getTextContent(), xml); + } + } + + @Test + public void atomMediaFileAuthorIsText() throws Exception { + for (String[] name : NAMES) { + String xml = render("<feed xmlns=\"http://www.w3.org/2005/Atom\">" + + "#showFilesAtom10($files)</feed>", name[0]); + Document feed = parse(xml); + assertEquals(name[1], feed.getElementsByTagNameNS("http://www.w3.org/2005/Atom", "name") + .item(0).getTextContent(), xml); + } + } } diff --git a/app/src/test/java/org/apache/roller/weblogger/util/UtilitiesTest.java b/app/src/test/java/org/apache/roller/weblogger/util/UtilitiesTest.java index e725e9cae..f56b198df 100644 --- a/app/src/test/java/org/apache/roller/weblogger/util/UtilitiesTest.java +++ b/app/src/test/java/org/apache/roller/weblogger/util/UtilitiesTest.java @@ -170,6 +170,22 @@ public class UtilitiesTest { + " <a href=\"/two\" rel=\"nofollow\">two</a>"; assertEquals(expect3, Utilities.addNofollow(test3)); + // an existing rel gets nofollow added to its value + assertEquals("<a href=\"/x\" rel=\"external nofollow\">x</a>", + Utilities.addNofollow("<a href=\"/x\" rel=\"external\">x</a>")); + assertEquals("<a href=\"/x\" REL='noopener nofollow'>x</a>", + Utilities.addNofollow("<a href=\"/x\" REL='noopener'>x</a>")); + assertEquals("<a href=\"/x\" rel=\"external nofollow\">x</a>", + Utilities.addNofollow("<a href=\"/x\" rel=external>x</a>")); + assertEquals("<a href=\"/x\" rel=\"Nofollow external\">x</a>", + Utilities.addNofollow("<a href=\"/x\" rel=\"Nofollow external\">x</a>")); + + // a ">" inside a quoted attribute value doesn't end the tag + assertEquals("<a href=\"/x\" title=\"a > b\" rel=\"nofollow\">x</a>", + Utilities.addNofollow("<a href=\"/x\" title=\"a > b\">x</a>")); + assertEquals("<a href='/x?a>b' rel=\"nofollow\">x</a>", + Utilities.addNofollow("<a href='/x?a>b'>x</a>")); + assertEquals("no links", Utilities.addNofollow("no links")); assertEquals("<a href=\"x\"", Utilities.addNofollow("<a href=\"x\"")); assertNull(Utilities.addNofollow(null)); @@ -185,7 +201,9 @@ public class UtilitiesTest { assertEquals(escaped, Utilities.transformToHTMLSubset(Utilities.escapeHTML(raw))); assertEquals(raw, Utilities.addNofollow(raw)); long millis = (System.nanoTime() - start) / 1_000_000; - assertTrue(millis < 200, "took " + millis + " ms"); + // a quadratic scan takes far longer than this; leave room for slow + // build machines + assertTrue(millis < 2000, "took " + millis + " ms"); } @Test
