netlib.strutils -> mitmproxy.utils.strutils

8 files changed, 12 insertions, 149 deletions

diff --git a/netlib/http/headers.py b/netlib/http/headers.py
index 39673f1a..6c30d278 100644
--- a/netlib/http/headers.py
+++ b/netlib/http/headers.py
@@ -2,7 +2,7 @@ import re
 
 import collections
 from netlib import multidict
-from netlib import strutils
+from mitmproxy.utils import strutils
 
 # See also: http://lucumr.pocoo.org/2013/7/2/the-updated-guide-to-unicode/
 
diff --git a/netlib/http/message.py b/netlib/http/message.py
index 1980b0ab..133a53ce 100644
--- a/netlib/http/message.py
+++ b/netlib/http/message.py
@@ -2,7 +2,9 @@ import re
 import warnings
 from typing import Optional
 
-from netlib import encoding, strutils, basetypes
+from mitmproxy.utils import strutils
+from netlib import encoding
+from netlib import basetypes
 from netlib.http import headers
 
 
diff --git a/netlib/http/request.py b/netlib/http/request.py
index dd6f4164..3479fa4c 100644
--- a/netlib/http/request.py
+++ b/netlib/http/request.py
@@ -2,7 +2,7 @@ import re
 import urllib
 
 from netlib import multidict
-from netlib import strutils
+from mitmproxy.utils import strutils
 from netlib.http import multipart
 from netlib.http import cookies
 from netlib.http import headers as nheaders
diff --git a/netlib/strutils.py b/netlib/strutils.py
deleted file mode 100644
index 57cfbc79..00000000
--- a/netlib/strutils.py
+++ /dev/null
@@ -1,142 +0,0 @@
-import re
-import codecs
-
-
-def always_bytes(unicode_or_bytes, *encode_args):
-    if isinstance(unicode_or_bytes, str):
-        return unicode_or_bytes.encode(*encode_args)
-    elif isinstance(unicode_or_bytes, bytes) or unicode_or_bytes is None:
-        return unicode_or_bytes
-    else:
-        raise TypeError("Expected str or bytes, but got {}.".format(type(unicode_or_bytes).__name__))
-
-
-def native(s, *encoding_opts):
-    """
-    Convert :py:class:`bytes` or :py:class:`unicode` to the native
-    :py:class:`str` type, using latin1 encoding if conversion is necessary.
-
-    https://www.python.org/dev/peps/pep-3333/#a-note-on-string-types
-    """
-    if not isinstance(s, (bytes, str)):
-        raise TypeError("%r is neither bytes nor unicode" % s)
-    if isinstance(s, bytes):
-        return s.decode(*encoding_opts)
-    return s
-
-
-# Translate control characters to "safe" characters. This implementation initially
-# replaced them with the matching control pictures (http://unicode.org/charts/PDF/U2400.pdf),
-# but that turned out to render badly with monospace fonts. We are back to "." therefore.
-_control_char_trans = {
-    x: ord(".")  # x + 0x2400 for unicode control group pictures
-    for x in range(32)
-}
-_control_char_trans[127] = ord(".")  # 0x2421
-_control_char_trans_newline = _control_char_trans.copy()
-for x in ("\r", "\n", "\t"):
-    del _control_char_trans_newline[ord(x)]
-
-
-_control_char_trans = str.maketrans(_control_char_trans)
-_control_char_trans_newline = str.maketrans(_control_char_trans_newline)
-
-
-def escape_control_characters(text: str, keep_spacing=True) -> str:
-    """
-    Replace all unicode C1 control characters from the given text with a single "."
-
-    Args:
-        keep_spacing: If True, tabs and newlines will not be replaced.
-    """
-    if not isinstance(text, str):
-        raise ValueError("text type must be unicode but is {}".format(type(text).__name__))
-
-    trans = _control_char_trans_newline if keep_spacing else _control_char_trans
-    return text.translate(trans)
-
-
-def bytes_to_escaped_str(data, keep_spacing=False, escape_single_quotes=False):
-    """
-    Take bytes and return a safe string that can be displayed to the user.
-
-    Single quotes are always escaped, double quotes are never escaped:
-        "'" + bytes_to_escaped_str(...) + "'"
-    gives a valid Python string.
-
-    Args:
-        keep_spacing: If True, tabs and newlines will not be escaped.
-    """
-
-    if not isinstance(data, bytes):
-        raise ValueError("data must be bytes, but is {}".format(data.__class__.__name__))
-    # We always insert a double-quote here so that we get a single-quoted string back
-    # https://stackoverflow.com/questions/29019340/why-does-python-use-different-quotes-for-representing-strings-depending-on-their
-    ret = repr(b'"' + data).lstrip("b")[2:-1]
-    if not escape_single_quotes:
-        ret = re.sub(r"(?<!\\)(\\\\)*\\'", lambda m: (m.group(1) or "") + "'", ret)
-    if keep_spacing:
-        ret = re.sub(
-            r"(?<!\\)(\\\\)*\\([nrt])",
-            lambda m: (m.group(1) or "") + dict(n="\n", r="\r", t="\t")[m.group(2)],
-            ret
-        )
-    return ret
-
-
-def escaped_str_to_bytes(data):
-    """
-    Take an escaped string and return the unescaped bytes equivalent.
-
-    Raises:
-        ValueError, if the escape sequence is invalid.
-    """
-    if not isinstance(data, str):
-        raise ValueError("data must be str, but is {}".format(data.__class__.__name__))
-
-    # This one is difficult - we use an undocumented Python API here
-    # as per http://stackoverflow.com/a/23151714/934719
-    return codecs.escape_decode(data)[0]
-
-
-def is_mostly_bin(s: bytes) -> bool:
-    if not s or len(s) == 0:
-        return False
-
-    return sum(
-        i < 9 or 13 < i < 32 or 126 < i
-        for i in s[:100]
-    ) / len(s[:100]) > 0.3
-
-
-def is_xml(s: bytes) -> bool:
-    return s.strip().startswith(b"<")
-
-
-def clean_hanging_newline(t):
-    """
-        Many editors will silently add a newline to the final line of a
-        document (I'm looking at you, Vim). This function fixes this common
-        problem at the risk of removing a hanging newline in the rare cases
-        where the user actually intends it.
-    """
-    if t and t[-1] == "\n":
-        return t[:-1]
-    return t
-
-
-def hexdump(s):
-    """
-        Returns:
-            A generator of (offset, hex, str) tuples
-    """
-    for i in range(0, len(s), 16):
-        offset = "{:0=10x}".format(i)
-        part = s[i:i + 16]
-        x = " ".join("{:0=2x}".format(i) for i in part)
-        x = x.ljust(47)  # 16*2 + 15
-        part_repr = native(escape_control_characters(
-            part.decode("ascii", "replace").replace(u"\ufffd", u"."),
-            False
-        ))
-        yield (offset, x, part_repr)
diff --git a/netlib/tcp.py b/netlib/tcp.py
index aeb1d447..aed79388 100644
--- a/netlib/tcp.py
+++ b/netlib/tcp.py
@@ -10,7 +10,7 @@ import binascii
 
 from typing import Optional  # noqa
 
-from netlib import strutils
+from mitmproxy.utils import strutils
 
 import certifi
 from backports import ssl_match_hostname
diff --git a/netlib/websockets/frame.py b/netlib/websockets/frame.py
index 02d74112..e022a95c 100644
--- a/netlib/websockets/frame.py
+++ b/netlib/websockets/frame.py
@@ -3,7 +3,7 @@ import struct
 import io
 
 from netlib import tcp
-from netlib import strutils
+from mitmproxy.utils import strutils
 from netlib import utils
 from mitmproxy.utils import human
 from .masker import Masker
diff --git a/netlib/websockets/utils.py b/netlib/websockets/utils.py
index fdec074e..98043662 100644
--- a/netlib/websockets/utils.py
+++ b/netlib/websockets/utils.py
@@ -8,7 +8,8 @@ import base64
 import hashlib
 import os
 
-from netlib import http, strutils
+from netlib import http
+from mitmproxy.utils import strutils
 
 MAGIC = b'258EAFA5-E914-47DA-95CA-C5AB0DC85B11'
 VERSION = "13"
diff --git a/netlib/wsgi.py b/netlib/wsgi.py
index 11e4aba9..5a54cd70 100644
--- a/netlib/wsgi.py
+++ b/netlib/wsgi.py
@@ -3,7 +3,9 @@ import traceback
 import urllib
 import io
 
-from netlib import http, tcp, strutils
+from netlib import http
+from netlib import tcp
+from mitmproxy.utils import strutils
 
 
 class ClientConn: