remove lxml-dependent code

4 files changed, 19 insertions, 91 deletions

diff --git a/mitmproxy/contentviews/__init__.py b/mitmproxy/contentviews/__init__.py
index b83e7aa6..3857d5e5 100644
--- a/mitmproxy/contentviews/__init__.py
+++ b/mitmproxy/contentviews/__init__.py
@@ -22,7 +22,7 @@ from mitmproxy import exceptions
 from mitmproxy.net import http
 from mitmproxy.utils import strutils
 from . import (
-    auto, raw, hex, json, xml, wbxml, html, javascript, css,
+    auto, raw, hex, json, html_outline, wbxml, javascript, css,
     urlencoded, multipart, image, query, protobuf
 )
 from .base import View, VIEW_CUTOFF, KEY_MAX, format_text, format_dict
@@ -163,10 +163,8 @@ add(auto.ViewAuto())
 add(raw.ViewRaw())
 add(hex.ViewHex())
 add(json.ViewJSON())
-add(xml.ViewXML())
 add(wbxml.ViewWBXML())
-add(html.ViewHTML())
-add(html.ViewHTMLOutline())
+add(html_outline.ViewHTMLOutline())
 add(javascript.ViewJavaScript())
 add(css.ViewCSS())
 add(urlencoded.ViewURLEncoded())
diff --git a/mitmproxy/contentviews/html.py b/mitmproxy/contentviews/html.py
deleted file mode 100644
index c625beef..00000000
--- a/mitmproxy/contentviews/html.py
+++ /dev/null
@@ -1,42 +0,0 @@
-import html2text
-import lxml.etree
-import lxml.html
-
-from mitmproxy.contentviews.base import View, format_text
-from mitmproxy.utils import strutils
-
-
-class ViewHTML(View):
-    name = "HTML"
-    prompt = ("html", "h")
-    content_types = ["text/html"]
-
-    def __call__(self, data, **metadata):
-        if strutils.is_xml(data):
-            parser = lxml.etree.HTMLParser(
-                strip_cdata=True,
-                remove_blank_text=True
-            )
-            d = lxml.html.fromstring(data, parser=parser)
-            docinfo = d.getroottree().docinfo
-            s = lxml.etree.tostring(
-                d,
-                pretty_print=True,
-                doctype=docinfo.doctype,
-                encoding='utf8'
-            )
-            return "HTML", format_text(s)
-
-
-class ViewHTMLOutline(View):
-    name = "HTML Outline"
-    prompt = ("html outline", "o")
-    content_types = ["text/html"]
-
-    def __call__(self, data, **metadata):
-        data = data.decode("utf-8", "replace")
-        h = html2text.HTML2Text(baseurl="")
-        h.ignore_images = True
-        h.body_width = 0
-        outline = h.handle(data)
-        return "HTML Outline", format_text(outline)
diff --git a/mitmproxy/contentviews/html_outline.py b/mitmproxy/contentviews/html_outline.py
new file mode 100644
index 00000000..d6c51b29
--- /dev/null
+++ b/mitmproxy/contentviews/html_outline.py
@@ -0,0 +1,17 @@
+import html2text
+
+from mitmproxy.contentviews import base
+
+
+class ViewHTMLOutline(base.View):
+    name = "HTML Outline"
+    prompt = ("html outline", "o")
+    content_types = ["text/html"]
+
+    def __call__(self, data, **metadata):
+        data = data.decode("utf-8", "replace")
+        h = html2text.HTML2Text(baseurl="")
+        h.ignore_images = True
+        h.body_width = 0
+        outline = h.handle(data)
+        return "HTML Outline", base.format_text(outline)
diff --git a/mitmproxy/contentviews/xml.py b/mitmproxy/contentviews/xml.py
deleted file mode 100644
index a382b09d..00000000
--- a/mitmproxy/contentviews/xml.py
+++ /dev/null
@@ -1,45 +0,0 @@
-import lxml.etree
-
-from . import base
-
-
-class ViewXML(base.View):
-    name = "XML"
-    prompt = ("xml", "x")
-    content_types = ["text/xml"]
-
-    def __call__(self, data, **metadata):
-        parser = lxml.etree.XMLParser(
-            remove_blank_text=True,
-            resolve_entities=False,
-            strip_cdata=False,
-            recover=False
-        )
-        try:
-            document = lxml.etree.fromstring(data, parser)
-        except lxml.etree.XMLSyntaxError:
-            return None
-        docinfo = document.getroottree().docinfo
-
-        prev = []
-        p = document.getroottree().getroot().getprevious()
-        while p is not None:
-            prev.insert(
-                0,
-                lxml.etree.tostring(p)
-            )
-            p = p.getprevious()
-        doctype = docinfo.doctype
-        if prev:
-            doctype += "\n".join(p.decode() for p in prev).strip()
-        doctype = doctype.strip()
-
-        s = lxml.etree.tostring(
-            document,
-            pretty_print=True,
-            xml_declaration=True,
-            doctype=doctype or None,
-            encoding=docinfo.encoding
-        )
-
-        return "XML-like data", base.format_text(s)