Import hubbub -- an HTML parsing library.

Plenty of work still to do (like tree generation ;)

svn path=/trunk/hubbub/; revision=3359

1 files changed, 82 insertions, 0 deletions

diff --git a/test/data/csdetect/tests2.dat b/test/data/csdetect/tests2.dat
new file mode 100644
index 0000000..dd43f85
--- /dev/null
+++ b/test/data/csdetect/tests2.dat
@@ -0,0 +1,82 @@
+#data
+<meta
+#encoding
+windows-1252
+
+#data
+<
+#encoding
+windows-1252
+
+#data
+<!
+#encoding
+windows-1252
+
+#data
+<meta charset = "
+#encoding
+windows-1252
+
+#data
+<meta charset=EUC-jp
+#encoding
+windows-1252
+
+#data
+<meta <meta charset='EUC-jp'>
+#encoding
+EUC-jp
+
+#data
+<meta       charset    =     'EUC-jp'>
+#encoding
+EUC-jp
+
+
+#data
+<!-- -->
+<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
+#encoding
+utf-8
+
+#data
+<!-- -->
+<meta http-equiv="Content-Type" content="text/html; charset=utf
+#encoding
+windows-1252
+
+#data
+<meta http-equiv="Content-Type<meta charset="utf-8">
+#encoding
+windows-1252
+
+#data
+<meta http-equiv="Content-Type" content="text/html; charset='utf-8'">
+#encoding
+utf-8
+
+#data
+<meta http-equiv="Content-Type" content="text/html; charset='utf-8">
+#encoding
+windows-1252
+
+#data
+<meta                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 
+#encoding
+windows-1252
+
+#data
+<meta charset                    =                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            
+#encoding
+windows-1252
+
+#data
+<meta charset=                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            utf-8
+#encoding
+windows-1252
+
+#data
+<meta content = "text/html;
+#encoding
+windows-1252