strip ANSI esc sequences from logs (#117)

Related #67

Signed-off-by: Tomas Tomecek <ttomecek@redhat.com>

sen/tui/widgets/list/common.py

@@ -1,4 +1,5 @@
 import logging
+import re
 import threading
 import traceback
 
@@ -11,6 +12,62 @@
 logger = logging.getLogger(__name__)
 
 
+# def translate_asci_sequence(s):
+#     # FIXME: not finished
+#     translation_map = {
+#         "34": "dark blue"
+#     }
+#     return translation_map.get(s, "")
+
+
+def strip_from_ansi_esc_sequences(text):
+    """
+    find ANSI escape sequences in text and remove them
+
+    :param text: str
+    :return: list, should be passed to ListBox
+    """
+    # esc[ + values + control character
+    # h, l, p commands are complicated, let's ignore them
+    seq_regex = r"\x1b\[[0-9;]*[mKJusDCBAfH]"
+    regex = re.compile(seq_regex)
+    start = 0
+    response = ""
+    for match in regex.finditer(text):
+        end = match.start()
+        response += text[start:end]
+
+        start = match.end()
+    response += text[start:len(text)]
+    return response
+
+
+# def colorize_text(text):
+#     """
+#     finds ANSI color escapes in text and transforms them to urwid
+#
+#     :param text: str
+#     :return: list, should be passed to ListBox
+#     """
+#     # FIXME: not finished
+#     response = []
+#     # http://ascii-table.com/ansi-escape-sequences.php
+#     regex_pattern = r"(?:\x1b\[(\d+)?(?:;(\d+))*m)([^\x1b]+)"  # [%d;%d;...m
+#     regex = re.compile(regex_pattern, re.UNICODE)
+#     for match in regex.finditer(text):
+#         groups = match.groups()
+#         t = groups[-1]
+#         color_specs = groups[:-1]
+#         urwid_spec = translate_asci_sequence(color_specs)
+#         if urwid_spec:
+#             item = (urwid.AttrSpec(urwid_spec, "main_list_dg"), t)
+#         else:
+#             item = t
+#         item = urwid.AttrMap(urwid.Text(t, align="left", wrap="any"), "main_list_dg", "main_list_white")
+#         response.append(item)
+#     return response
+
+
 class ScrollableListBox(WidgetBase):
     def __init__(self, ui, text, focus_bottom=True):
         self.walker = urwid.SimpleFocusListWalker([])
@@ -25,9 +82,11 @@ def __init__(self, ui, text, focus_bottom=True):
     def set_text(self, text):
         self.walker.clear()
         text = _ensure_unicode(text)
+        # logger.debug(repr(text))
+        text = strip_from_ansi_esc_sequences(text)
         list_of_texts = text.split("\n")
         self.walker[:] = [
-            urwid.AttrMap(urwid.Text(t, align="left", wrap="any"), "main_list_dg", "main_list_white")
+            urwid.AttrMap(urwid.Text(t.strip(), align="left", wrap="any"), "main_list_dg", "main_list_white")
             for t in list_of_texts
         ]
 

tests/test_util.py

@@ -5,6 +5,7 @@
 
 from flexmock import flexmock
 
+from sen.tui.widgets.list.common import strip_from_ansi_esc_sequences
 from sen.util import _ensure_unicode, log_traceback, repeater, humanize_time, \
     OrderedSet
 
@@ -121,3 +122,35 @@ def test_ordered_set():
     assert s == [1, 2]
     s.append(1)
     assert s == [2, 1]
+
+
+# @pytest.mark.parametrize("inp,expected", [
+#     ("aaa", ["aaa"]),
+#     (
+#         "qwe [01;34m asd [01;33mbnm",
+#         ["qwe ", " asd ", "bnm"]
+#     )
+# ])
+# def test_colorize_text(inp, expected):
+#     got = colorize_text(inp)
+#     for idx, c in enumerate(got):
+#         if isinstance(c, str):
+#             assert expected[idx] == c
+#         else:
+#             assert expected[idx] == c[1]
+
+
+@pytest.mark.parametrize("inp,expected", [
+    ("aaa", "aaa"),
+    (
+        "root:x:0:0:root:/root:/bin/b\x1b[01;31m\x1b[Ka\x1b[m\x1b[Ksh",
+        "root:x:0:0:root:/root:/bin/bash"
+    ),
+    (
+        "\x1b[0m\x1b[01;36mbin\x1b[0m\r\n\x1b[01;34mboot\x1b[0m\r\n\x1b[01;34mdev\x1b[0m\r\n",
+        "bin\r\nboot\r\ndev\r\n"
+    )
+])
+def test_strip_from_ansi_seqs(inp, expected):
+    got = strip_from_ansi_esc_sequences(inp)
+    assert got == expected