diff options
| author | Vsevolod Stakhov <vsevolod@highsecure.ru> | 2017-04-17 18:11:37 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2017-04-17 18:11:37 +0100 |
| commit | af76c47e4a31b499fe443e59727e3f909cd763f9 (patch) | |
| tree | 7acf71350bf5bdc87623af83295665732ef760a5 | |
| parent | 43a67aa9d3b68993de487d04d00c328ef051345b (diff) | |
| parent | a69b112bf749cdec95962ea8835bd62e4cfd5db2 (diff) | |
| download | rspamd-af76c47e4a31b499fe443e59727e3f909cd763f9.tar.gz rspamd-af76c47e4a31b499fe443e59727e3f909cd763f9.zip | |
Merge pull request #1596 from cpragadeesh/html_test_improv
HTML content extraction tests improvement
| -rw-r--r-- | test/lua/unit/html.lua | 58 |
1 files changed, 58 insertions, 0 deletions
diff --git a/test/lua/unit/html.lua b/test/lua/unit/html.lua index f1408c25a..489947ffb 100644 --- a/test/lua/unit/html.lua +++ b/test/lua/unit/html.lua @@ -57,6 +57,64 @@ context("HTML processing", function() </body> </html> ]], 'Hello, world!'}, + {[[ +<html lang="en"> + <head> + <meta charset="utf-8"> + <title>title</title> + <link rel="stylesheet" href="style.css"> + <script src="script.js"></script> + </head> + <body> + <!-- page content --> + Hello, world!<br>test</br><br>content</hr>more content<br> + <div> + content inside div + </div> + </body> +</html> + ]], 'Hello, world!\r\ntest\r\ncontent\r\nmore content\r\ncontent inside div\r\n'}, + {[[ +<html lang="en"> + <head> + <meta charset="utf-8"> + <title>title</title> + <link rel="stylesheet" href="style.css"> + <script src="script.js"></script> + </head> + <body> + <!-- tabular content --> + <table> + content + </table> + <table> + <tr> + <th>heada</th> + <th>headb</th> + </tr> + <tr> + <td>data1</td> + <td>data2</td> + </tr> + </table> + + </body> +</html> + ]], 'content heada headb\r\ndata1 data2\r\n'}, + {[[ +<html lang="en"> + <head> + <meta charset="utf-8"> + <title>title</title> + <link rel="stylesheet" href="style.css"> + <script src="script.js"></script> + </head> + <body> + <!-- escape content --> + a b a > b a < b a & b 'a "a" + </body> +</html> + ]], 'a b a > b a < b a & b \'a "a"'}, } for _,c in ipairs(cases) do |