Add a simple parser for content type specifications.

3 files changed, 39 insertions, 0 deletions

diff --git a/libmproxy/console/flowview.py b/libmproxy/console/flowview.py
index eb496732..6038a777 100644
--- a/libmproxy/console/flowview.py
+++ b/libmproxy/console/flowview.py
@@ -209,6 +209,7 @@ class ConnectionView(common.WWrap):
                     break
         else:
             ctype = common.BODY_PRETTY_TYPES[pretty_type]
+
         if ctype and flow.HDR_FORM_URLENCODED in ctype:
             data = utils.urldecode(content)
             if data:
diff --git a/libmproxy/utils.py b/libmproxy/utils.py
index 02bc49cb..0eaada7c 100644
--- a/libmproxy/utils.py
+++ b/libmproxy/utils.py
@@ -265,6 +265,33 @@ def parse_proxy_spec(url):
     return p[:3]
 
 
+def parse_content_type(c):
+    """
+        A simple parser for content-type values. Returns a (type, subtype,
+        parameters) tuple, where type and subtype are strings, and parameters
+        is a dict. If the string could not be parsed, return None.
+
+        E.g. the following string:
+
+            text/html; charset=UTF-8
+
+        Returns:
+
+            ("text", "html", {"charset": "UTF-8"})
+    """
+    parts = c.split(";", 1)
+    ts = parts[0].split("/", 1)
+    if len(ts) != 2:
+        return None
+    d = {}
+    if len(parts) == 2:
+        for i in parts[1].split(";"):
+            clause = i.split("=", 1)
+            if len(clause) == 2:
+                d[clause[0].strip()] = clause[1].strip()
+    return ts[0], ts[1], d
+
+
 def hostport(scheme, host, port):
     """
         Returns the host component, with a port specifcation if needed.
diff --git a/test/test_utils.py b/test/test_utils.py
index 9f07c706..d65fa5b4 100644
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -214,7 +214,18 @@ class u_parse_size(libpry.AutoTree):
         libpry.raises(ValueError, utils.parse_size, "ak")
 
 
+class u_parse_content_type(libpry.AutoTree):
+    def test_simple(self):
+        p = utils.parse_content_type
+        assert p("text/html") == ("text", "html", {})
+        assert p("text") == None
+
+        v = p("text/html; charset=UTF-8")
+        assert v == ('text', 'html', {'charset': 'UTF-8'})
+
+
 tests = [
+    u_parse_content_type(),
     uformat_timestamp(),
     uisBin(),
     uisXML(),