Added a default case to skip processing instructions, doctypes etc

clojure · Feb 9, 2012 · 89fe25e · 89fe25e
1 parent edd0aeb
commit 89fe25e
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 5 deletions.
diff --git a/src/main/clojure/clojure/data/xml.clj b/src/main/clojure/clojure/data/xml.clj
@@ -214,12 +214,10 @@
          (cons (event :characters nil nil text)
                (pull-seq sreader))
          (recur))
-       XMLStreamConstants/COMMENT
-       (recur)
-       XMLStreamConstants/SPACE
-       (recur)
        XMLStreamConstants/END_DOCUMENT
-       nil))))
+       nil
+       (recur);; Consume and ignore comments, spaces, processing instructions etc
+       ))))
 
 (defn source-seq
   "Parses the XML InputSource source using a pull-parser. Returns

diff --git a/src/test/clojure/clojure/data/xml/test_parse.clj b/src/test/clojure/clojure/data/xml/test_parse.clj
@@ -59,3 +59,17 @@
                                                     "there")))]
   (is (= expected (lazy-parse* input)))))
 
+(deftest test-parsing-processing-instructions
+  (let [input "<?xml version=\"1.0\" encoding=\"utf-8\"?>
+                <?xml-stylesheet type='text/xsl' href='someFile.xsl'?>
+                <ATag>With Stuff</ATag>"
+        expected (element :ATag {} "With Stuff")]
+    (is (= expected (parse-str input)))))
+
+(deftest test-parsing-doctypes
+  (let [input "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\"
+               \"foo://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">
+               <html><h1>Heading Stuff</h1></html>"
+        expected (element :html {}
+                          (element :h1 {} "Heading Stuff"))]
+    (is (= expected (parse-str input)))))