[prev in list] [next in list] [prev in thread] [next in thread]
List: kde-commits
Subject: [kdepimlibs] syndication: In proper RSS 1.0, the items in the feed must be listed in the "items" seq
From: Frank Osterfeld <frank.osterfeld () kdab ! com>
Date: 2012-03-27 19:34:33
Message-ID: 20120327193433.7087DA60A9 () git ! kde ! org
[Download RAW message or body]
Git commit 503d2e94d8fdd357bf723441035d4b0c49a20ec4 by Frank Osterfeld.
Committed on 27/03/2012 at 20:11.
Pushed by osterfeld into branch 'master'.
In proper RSS 1.0, the items in the feed must be listed in the "items" sequence, with \
the same URIs used in the item descriptions afterwards. If the URIs don't match, the \
sequence and items are not linked in the RDF, thus no item descriptions are found. \
This is a common error RDF feeds out there and W3C's feed validator doesn't \
complain. Nobody groks RDF anyway, so don't rely on matching URIs. Instead, just \
ignore <items> and get all item resources by type as they occur in the feed.
BUG:295354
M +52 -22 syndication/rdf/document.cpp
M +6 -0 syndication/rdf/parser.cpp
A +102 -0 syndication/tests/rdf/javaworld.xml
A +86 -0 syndication/tests/rdf/javaworld.xml.expected
M +2 -2 syndication/tests/rdf/tagesschau.de.xml
http://commits.kde.org/kdepimlibs/503d2e94d8fdd357bf723441035d4b0c49a20ec4
diff --git a/syndication/rdf/document.cpp b/syndication/rdf/document.cpp
index e336955..997e3c9 100644
--- a/syndication/rdf/document.cpp
+++ b/syndication/rdf/document.cpp
@@ -38,6 +38,8 @@
#include <QtCore/QList>
#include <QtCore/QString>
+#include <QtCore/QStringList>
+#include <QtCore/QVector>
using namespace boost;
@@ -138,37 +140,65 @@ SyndicationInfo Document::syn() const
return SyndicationInfo(resource());
}
+
+struct SortItem {
+ Item item;
+ int index;
+};
+
+struct LessThanByIndex {
+ bool operator()(const SortItem& lhs, const SortItem& rhs) const {
+ return lhs.index < rhs.index;
+ }
+};
+
+static QList<Item> sortListToMatchSequence(QList<Item> items, const QStringList& \
uriSequence) { + QVector<SortItem> toSort;
+ toSort.reserve(items.size());
+ Q_FOREACH(const Item& i, items) {
+ SortItem item;
+ item.item = i;
+ item.index = uriSequence.indexOf(i.resource()->uri());
+ toSort.append(item);
+ }
+ qSort(toSort.begin(), toSort.end(), LessThanByIndex());
+
+ int i = 0;
+ Q_FOREACH(const SortItem& sortItem, toSort) {
+ items[i] = sortItem.item;
+ i++;
+ }
+
+ return items;
+}
+
QList<Item> Document::items() const
{
QList<Item> list;
- if (!resource()->hasProperty(RSSVocab::self()->items()))
- return list;
- NodePtr n = resource()->property(RSSVocab::self()->items())->object();
- if (n->isSequence())
- {
- Sequence* seq = static_cast<Sequence*>(n.get());
+ const QList<ResourcePtr> items = \
resource()->model().resourcesWithType(RSSVocab::self()->item()); + DocumentPtr \
doccpy(new Document(*this)); + Q_FOREACH (const ResourcePtr& i, items)
+ list.append(Item(i, doccpy));
- const QList<NodePtr> items = seq->items();
- QList<NodePtr>::ConstIterator it = items.begin();
- QList<NodePtr>::ConstIterator end = items.end();
+ if (resource()->hasProperty(RSSVocab::self()->items())) {
+ NodePtr n = resource()->property(RSSVocab::self()->items())->object();
+ if (n->isSequence())
+ {
+ Sequence* seq = static_cast<Sequence*>(n.get());
- DocumentPtr doccpy(new Document(*this));
+ const QList<NodePtr> seqItems = seq->items();
- for ( ; it != end; ++it)
- {
- if ((*it)->isResource())
- {
- // well, we need it as ResourcePtr
- // maybe this should go to the node
- // interface ResourcePtr asResource()?
- ResourcePtr ptr = \
resource()->model().createResource((static_cast<Resource*>((*it).get()))->uri());
-
- list.append(Item(ptr, doccpy));
- }
- }
+ QStringList uriSequence;
+ uriSequence.reserve(seqItems.size());
+ Q_FOREACH(const NodePtr& i, seqItems)
+ if (i->isResource())
+ uriSequence.append(static_cast<Resource*>(i.get())->uri());
+ list = sortListToMatchSequence(list, uriSequence);
+ }
}
+
return list;
}
diff --git a/syndication/rdf/parser.cpp b/syndication/rdf/parser.cpp
index 12b4c1c..f675236 100644
--- a/syndication/rdf/parser.cpp
+++ b/syndication/rdf/parser.cpp
@@ -164,6 +164,8 @@ void Parser::ParserPrivate::map09to10(Model model)
void Parser::ParserPrivate::addSequenceFor09(Model model)
{
+ //RDF 0.9 doesn't contain an item sequence, and the items don't have rdf:about, \
so add both +
const QList<ResourcePtr> items = \
model.resourcesWithType(RSS09Vocab::self()->item());
if (items.isEmpty())
@@ -196,6 +198,10 @@ void Parser::ParserPrivate::addSequenceFor09(Model model)
foreach (const ResourcePtr &i, sorted)
{
seq->append(i);
+ // add rdf:about (type)
+ model.addStatement(i, RDFVocab::self()->type(), RSSVocab::self()->item());
+
+ //add to items sequence
model.addStatement(seq, RDFVocab::self()->li(), i);
}
}
diff --git a/syndication/tests/rdf/javaworld.xml \
b/syndication/tests/rdf/javaworld.xml new file mode 100644
index 0000000..dbbc6b0
--- /dev/null
+++ b/syndication/tests/rdf/javaworld.xml
@@ -0,0 +1,102 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--
+RDF 1.0 with the URIs <items> not matching the actual <item>s
+As not matching URIs is a common error, <items> should be ignored and the actual \
item descriptions should be returned (even if that doesn't make sense on RDF level) \
+bug 294354 +-->
+
+
+
+ <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" \
xmlns="http://purl.org/rss/1.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" \
xmlns:rss="http://purl.org/rss/1.0/" \
xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"> + <channel \
rdf:about="http://www.javaworld.com/features/index.html"> + <title>Featured \
Articles</title> + <link>http://www.javaworld.com/features/index.html</link>
+ <description>JW RSS feed for Featured Articles</description>
+ <dc:publisher>Javaworld</dc:publisher>
+ <dc:rights>Copyright(C) 1994 - 2012 Javaworld</dc:rights>
+ <items>
+ <rdf:Seq>
+ <rdf:li \
resource="http://www.javaworld.com/javaworld/jw-02-2012/120228-modernizing-it.html"/> \
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120224-bjarne-stroustrup-interview.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120222-node-js-tools-remake-server-side.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120216-fatal-exception.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120214-jtip-rss-for-android.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120208-mongodb-review.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120206-programmer-personality-types.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120202-environment-aware-maven-build.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-02-2012/120201-opa-programming.html"/>
+ <rdf:li resource="http://www.javaworld.com/javaworld/jw-01-2012/120126-fatal-exception.html"/>
+ </rdf:Seq>
+ </items>
+ </channel>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120228-modernizing-it.html">
+ <title>Electric Cloud CEO: We automate agile development</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120228-modernizing-it.html</link>
+ <description>An explosion of tools and platforms is making app dev much more \
complex. Mike Maciag, CEO of Electric Cloud, offers cross-platform workflow and \
automation to rationalize it all.</description> + <dc:creator/>
+ <dc:date>2012-02-28T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120224-bjarne-stroustrup-interview.html">
+ <title>Stroustrup reveals what's new in C++ 11</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120224-bjarne-stroustrup-interview.html</link>
+ <description>Latest version of language isn't a major overhaul, but \
offers many small upgrades to appeal to different areas of development.</description> \
+ <dc:creator/> + <dc:date>2012-02-24T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120222-node-js-tools-remake-server-side.html">
+ <title>Node.js tools: Server-side JavaScript comes of age</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120222-node-js-tools-remake-server-side.html</link>
+ <description>Node's ecosystem is growing at an exponential rate, which is \
only natural when you start from scratch. Still, the plethora of Node-inspired tools \
amount to a diverse and sustainable infrastructure, which could ultimatey redefine \
the web development stack. </description> + <dc:creator/>
+ <dc:date>2012-02-22T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120216-fatal-exception.html">
+ <title>Google Chrome, HTML5, and the new Web platform</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120216-fatal-exception.html</link>
+ <description>The Chrome dev team is working toward a vision of Web apps that \
offers a clean break from traditional websites.</description> + <dc:creator/>
+ <dc:date>2012-02-16T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120214-jtip-rss-for-android.html">
+ <title>Java tip: Set up an RSS feed for your Android application</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120214-jtip-rss-for-android.html</link>
+ <description>Get a hands-on introduction to using \
<code>javax.xml.parsers.SAXParser</code> to parse an RSS feed in XML \
format. While some familiarity with Java application development is assumed, this \
Java tip is suitable for developers new to Java mobile development.</description> \
+ <dc:creator/> + <dc:date>2012-02-14T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120208-mongodb-review.html">
+ <title>Flexing NoSQL: MongoDB in review</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120208-mongodb-review.html</link>
+ <description>MongoDB shines with broad programming language support, SQL-like \
queries, and out-of-the-box scaling. Check the InfoWorld Review score card to see how \
MongoDB stacks up among NoSQL data stores.</description> + <dc:creator/>
+ <dc:date>2012-02-08T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120206-programmer-personality-types.html">
+ <title>Programmer personality types: 13 profiles in code</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120206-programmer-personality-types.html</link>
+ <description>From Underdocumenters to flat-out Fakers, the world of software \
development is full of characters. Here, InfoWorld's Peter Wayner offers 13 \
developer personality profiles based on coding style.</description> \
+ <dc:creator/> + <dc:date>2012-02-06T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120202-environment-aware-maven-build.html">
+ <title>Develop an environment-aware Maven build process</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120202-environment-aware-maven-build.html</link>
+ <description>Including environment variables in your Maven build process could \
boost your team's efficiency at every stage of the software development \
lifecycle. Java developer Paul Spinelli demonstrates his custom approach to \
environment-aware Maven builds.</description> + <dc:creator/>
+ <dc:date>2012-02-02T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-02-2012/120201-opa-programming.html">
+ <title>Programming Opa: Web development, reimagined</title>
+ <link>http://www.javaworld.com/javaworld/jw-02-2012/120201-opa-programming.html</link>
+ <description>MLstate's Opa streamlines Web app development with a single \
language for client and server, but the bright promise is not without \
pitfalls.</description> + <dc:creator/>
+ <dc:date>2012-02-01T00:00:00Z</dc:date>
+ </item>
+ <item rdf:about="http://www.javaworld.com//javaworld/jw-01-2012/120126-fatal-exception.html">
+ <title>Here's how to solve America's developer shortage</title>
+ <link>http://www.javaworld.com/javaworld/jw-01-2012/120126-fatal-exception.html</link>
+ <description>Employers say they can't find enough workers to staff IT \
jobs. Maybe they've been going about it all wrong.</description> \
+ <dc:creator/> + <dc:date>2012-01-26T00:00:00Z</dc:date>
+ </item>
+ </rdf:RDF>
diff --git a/syndication/tests/rdf/javaworld.xml.expected \
b/syndication/tests/rdf/javaworld.xml.expected new file mode 100644
index 0000000..5d3e0aa
--- /dev/null
+++ b/syndication/tests/rdf/javaworld.xml.expected
@@ -0,0 +1,86 @@
+# Feed begin ######################
+title: #Featured Articles#
+link: #http://www.javaworld.com/features/index.html#
+description: #JW RSS feed for Featured Articles#
+copyright: #Copyright(C) 1994 - 2012 Javaworld#
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120201-opa-programming.html#
+title: #Programming Opa: Web development, reimagined#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120201-opa-programming.html#
+description: #MLstate's Opa streamlines Web app development with a single language \
for client and server, but the bright promise is not without pitfalls.# \
+datePublished: #Wed Feb 1 01:00:00 2012# +dateUpdated: #Wed Feb 1 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120222-node-js-tools-remake-server-side.html#
+title: #Node.js tools: Server-side JavaScript comes of age#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120222-node-js-tools-remake-server-side.html#
+description: #Node's ecosystem is growing at an exponential rate, which is only \
natural when you start from scratch. Still, the plethora of Node-inspired tools \
amount to a diverse and sustainable infrastructure, which could ultimatey redefine \
the web development stack.# +datePublished: #Wed Feb 22 01:00:00 2012#
+dateUpdated: #Wed Feb 22 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120216-fatal-exception.html#
+title: #Google Chrome, HTML5, and the new Web platform#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120216-fatal-exception.html#
+description: #The Chrome dev team is working toward a vision of Web apps that offers \
a clean break from traditional websites.# +datePublished: #Thu Feb 16 01:00:00 2012#
+dateUpdated: #Thu Feb 16 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120208-mongodb-review.html#
+title: #Flexing NoSQL: MongoDB in review#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120208-mongodb-review.html#
+description: #MongoDB shines with broad programming language support, SQL-like \
queries, and out-of-the-box scaling. Check the InfoWorld Review score card to see how \
MongoDB stacks up among NoSQL data stores.# +datePublished: #Wed Feb 8 01:00:00 2012#
+dateUpdated: #Wed Feb 8 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120224-bjarne-stroustrup-interview.html#
+title: #Stroustrup reveals what's new in C++ 11#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120224-bjarne-stroustrup-interview.html#
+description: #Latest version of language isn't a major overhaul, but offers many \
small upgrades to appeal to different areas of development.# +datePublished: #Fri Feb \
24 01:00:00 2012# +dateUpdated: #Fri Feb 24 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120206-programmer-personality-types.html#
+title: #Programmer personality types: 13 profiles in code#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120206-programmer-personality-types.html#
+description: #From Underdocumenters to flat-out Fakers, the world of software \
development is full of characters. Here, InfoWorld's Peter Wayner offers 13 developer \
personality profiles based on coding style.# +datePublished: #Mon Feb 6 01:00:00 \
2012# +dateUpdated: #Mon Feb 6 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120202-environment-aware-maven-build.html#
+title: #Develop an environment-aware Maven build process#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120202-environment-aware-maven-build.html#
+description: #Including environment variables in your Maven build process could \
boost your team's efficiency at every stage of the software development lifecycle. \
Java developer Paul Spinelli demonstrates his custom approach to environment-aware \
Maven builds.# +datePublished: #Thu Feb 2 01:00:00 2012#
+dateUpdated: #Thu Feb 2 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120228-modernizing-it.html#
+title: #Electric Cloud CEO: We automate agile development#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120228-modernizing-it.html#
+description: #An explosion of tools and platforms is making app dev much more \
complex. Mike Maciag, CEO of Electric Cloud, offers cross-platform workflow and \
automation to rationalize it all.# +datePublished: #Tue Feb 28 01:00:00 2012#
+dateUpdated: #Tue Feb 28 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-02-2012/120214-jtip-rss-for-android.html#
+title: #Java tip: Set up an RSS feed for your Android application#
+link: #http://www.javaworld.com/javaworld/jw-02-2012/120214-jtip-rss-for-android.html#
+description: #Get a hands-on introduction to using \
<code>javax.xml.parsers.SAXParser</code> to parse an RSS feed in XML format. While \
some familiarity with Java application development is assumed, this Java tip is \
suitable for developers new to Java mobile development.# +datePublished: #Tue Feb 14 \
01:00:00 2012# +dateUpdated: #Tue Feb 14 01:00:00 2012#
+# Item end ########################
+# Item begin ######################
+id: #http://www.javaworld.com//javaworld/jw-01-2012/120126-fatal-exception.html#
+title: #Here's how to solve America's developer shortage#
+link: #http://www.javaworld.com/javaworld/jw-01-2012/120126-fatal-exception.html#
+description: #Employers say they can't find enough workers to staff IT jobs. Maybe \
they've been going about it all wrong.# +datePublished: #Thu Jan 26 01:00:00 2012#
+dateUpdated: #Thu Jan 26 01:00:00 2012#
+# Item end ########################
+# Feed end ########################
diff --git a/syndication/tests/rdf/tagesschau.de.xml \
b/syndication/tests/rdf/tagesschau.de.xml index 27b9ea1..fbf6906 100644
--- a/syndication/tests/rdf/tagesschau.de.xml
+++ b/syndication/tests/rdf/tagesschau.de.xml
@@ -1,7 +1,7 @@
<?xml version="1.0" encoding="ISO-8859-1"?><rdf:RDF \
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" \
xmlns="http://my.netscape.com/rdf/simple/0.9/"> <!-- Source: www.tagesschau.de
Properties:
- - no-thrills RDF 1.0 feed
+ - no-thrills RDF 0.9 feed
- unescaped quotes in text
-->
<!--
@@ -41,4 +41,4 @@
-</rdf:RDF>
\ No newline at end of file
+</rdf:RDF>
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic