ANSIString in progress. Checking in what I have so far.
This commit is contained in:
Kelketek
parent
81c57da56b
commit
63761f66ec
1 changed files with 195 additions and 82 deletions
|
|
@ -70,87 +70,6 @@ class ANSIParser(object):
|
||||||
an extra { for Merc-style codes
|
an extra { for Merc-style codes
|
||||||
"""
|
"""
|
||||||
|
|
||||||
def __init__(self):
|
|
||||||
"Sets the mappings"
|
|
||||||
|
|
||||||
# MUX-style mappings %cr %cn etc
|
|
||||||
|
|
||||||
self.mux_ansi_map = [
|
|
||||||
# commented out by default; they (especially blink) are
|
|
||||||
# potentially annoying
|
|
||||||
(r'%r', ANSI_RETURN),
|
|
||||||
(r'%t', ANSI_TAB),
|
|
||||||
(r'%b', ANSI_SPACE),
|
|
||||||
#(r'%cf', ANSI_BLINK),
|
|
||||||
#(r'%ci', ANSI_INVERSE),
|
|
||||||
(r'%cr', ANSI_RED),
|
|
||||||
(r'%cR', ANSI_BACK_RED),
|
|
||||||
(r'%cg', ANSI_GREEN),
|
|
||||||
(r'%cG', ANSI_BACK_GREEN),
|
|
||||||
(r'%cy', ANSI_YELLOW),
|
|
||||||
(r'%cY', ANSI_BACK_YELLOW),
|
|
||||||
(r'%cb', ANSI_BLUE),
|
|
||||||
(r'%cB', ANSI_BACK_BLUE),
|
|
||||||
(r'%cm', ANSI_MAGENTA),
|
|
||||||
(r'%cM', ANSI_BACK_MAGENTA),
|
|
||||||
(r'%cc', ANSI_CYAN),
|
|
||||||
(r'%cC', ANSI_BACK_CYAN),
|
|
||||||
(r'%cw', ANSI_WHITE),
|
|
||||||
(r'%cW', ANSI_BACK_WHITE),
|
|
||||||
(r'%cx', ANSI_BLACK),
|
|
||||||
(r'%cX', ANSI_BACK_BLACK),
|
|
||||||
(r'%ch', ANSI_HILITE),
|
|
||||||
(r'%cn', ANSI_NORMAL),
|
|
||||||
]
|
|
||||||
|
|
||||||
# Expanded mapping {r {n etc
|
|
||||||
|
|
||||||
hilite = ANSI_HILITE
|
|
||||||
normal = ANSI_NORMAL
|
|
||||||
self.ext_ansi_map = [
|
|
||||||
(r'{r', hilite + ANSI_RED),
|
|
||||||
(r'{R', normal + ANSI_RED),
|
|
||||||
(r'{g', hilite + ANSI_GREEN),
|
|
||||||
(r'{G', normal + ANSI_GREEN),
|
|
||||||
(r'{y', hilite + ANSI_YELLOW),
|
|
||||||
(r'{Y', normal + ANSI_YELLOW),
|
|
||||||
(r'{b', hilite + ANSI_BLUE),
|
|
||||||
(r'{B', normal + ANSI_BLUE),
|
|
||||||
(r'{m', hilite + ANSI_MAGENTA),
|
|
||||||
(r'{M', normal + ANSI_MAGENTA),
|
|
||||||
(r'{c', hilite + ANSI_CYAN),
|
|
||||||
(r'{C', normal + ANSI_CYAN),
|
|
||||||
(r'{w', hilite + ANSI_WHITE), # pure white
|
|
||||||
(r'{W', normal + ANSI_WHITE), # light grey
|
|
||||||
(r'{x', hilite + ANSI_BLACK), # dark grey
|
|
||||||
(r'{X', normal + ANSI_BLACK), # pure black
|
|
||||||
(r'{n', normal) # reset
|
|
||||||
]
|
|
||||||
|
|
||||||
# xterm256 {123, %c134,
|
|
||||||
|
|
||||||
self.xterm256_map = [
|
|
||||||
(r'%c([0-5]{3})', self.parse_rgb), # %c123 - foreground colour
|
|
||||||
(r'%c(b[0-5]{3})', self.parse_rgb), # %cb123 - background colour
|
|
||||||
(r'{([0-5]{3})', self.parse_rgb), # {123 - foreground colour
|
|
||||||
(r'{(b[0-5]{3})', self.parse_rgb) # {b123 - background colour
|
|
||||||
]
|
|
||||||
|
|
||||||
# obs - order matters here, we want to do the xterms first since
|
|
||||||
# they collide with some of the other mappings otherwise.
|
|
||||||
self.ansi_map = self.xterm256_map + self.mux_ansi_map + self.ext_ansi_map
|
|
||||||
|
|
||||||
# prepare regex matching
|
|
||||||
self.ansi_sub = [(re.compile(sub[0], re.DOTALL), sub[1])
|
|
||||||
for sub in self.ansi_map]
|
|
||||||
|
|
||||||
# prepare matching ansi codes overall
|
|
||||||
self.ansi_regex = re.compile("\033\[[0-9;]+m")
|
|
||||||
|
|
||||||
# escapes - these double-chars will be replaced with a single
|
|
||||||
# instance of each
|
|
||||||
self.ansi_escapes = re.compile(r"(%s)" % "|".join(ANSI_ESCAPES), re.DOTALL)
|
|
||||||
|
|
||||||
def parse_rgb(self, rgbmatch):
|
def parse_rgb(self, rgbmatch):
|
||||||
"""
|
"""
|
||||||
This is a replacer method called by re.sub with the matched
|
This is a replacer method called by re.sub with the matched
|
||||||
|
|
@ -259,6 +178,84 @@ class ANSIParser(object):
|
||||||
string = self.ansi_regex.sub("", string)
|
string = self.ansi_regex.sub("", string)
|
||||||
return string
|
return string
|
||||||
|
|
||||||
|
# MUX-style mappings %cr %cn etc
|
||||||
|
|
||||||
|
mux_ansi_map = [
|
||||||
|
# commented out by default; they (especially blink) are
|
||||||
|
# potentially annoying
|
||||||
|
(r'%r', ANSI_RETURN),
|
||||||
|
(r'%t', ANSI_TAB),
|
||||||
|
(r'%b', ANSI_SPACE),
|
||||||
|
#(r'%cf', ANSI_BLINK),
|
||||||
|
#(r'%ci', ANSI_INVERSE),
|
||||||
|
(r'%cr', ANSI_RED),
|
||||||
|
(r'%cR', ANSI_BACK_RED),
|
||||||
|
(r'%cg', ANSI_GREEN),
|
||||||
|
(r'%cG', ANSI_BACK_GREEN),
|
||||||
|
(r'%cy', ANSI_YELLOW),
|
||||||
|
(r'%cY', ANSI_BACK_YELLOW),
|
||||||
|
(r'%cb', ANSI_BLUE),
|
||||||
|
(r'%cB', ANSI_BACK_BLUE),
|
||||||
|
(r'%cm', ANSI_MAGENTA),
|
||||||
|
(r'%cM', ANSI_BACK_MAGENTA),
|
||||||
|
(r'%cc', ANSI_CYAN),
|
||||||
|
(r'%cC', ANSI_BACK_CYAN),
|
||||||
|
(r'%cw', ANSI_WHITE),
|
||||||
|
(r'%cW', ANSI_BACK_WHITE),
|
||||||
|
(r'%cx', ANSI_BLACK),
|
||||||
|
(r'%cX', ANSI_BACK_BLACK),
|
||||||
|
(r'%ch', ANSI_HILITE),
|
||||||
|
(r'%cn', ANSI_NORMAL),
|
||||||
|
]
|
||||||
|
|
||||||
|
# Expanded mapping {r {n etc
|
||||||
|
|
||||||
|
hilite = ANSI_HILITE
|
||||||
|
normal = ANSI_NORMAL
|
||||||
|
ext_ansi_map = [
|
||||||
|
(r'{r', hilite + ANSI_RED),
|
||||||
|
(r'{R', normal + ANSI_RED),
|
||||||
|
(r'{g', hilite + ANSI_GREEN),
|
||||||
|
(r'{G', normal + ANSI_GREEN),
|
||||||
|
(r'{y', hilite + ANSI_YELLOW),
|
||||||
|
(r'{Y', normal + ANSI_YELLOW),
|
||||||
|
(r'{b', hilite + ANSI_BLUE),
|
||||||
|
(r'{B', normal + ANSI_BLUE),
|
||||||
|
(r'{m', hilite + ANSI_MAGENTA),
|
||||||
|
(r'{M', normal + ANSI_MAGENTA),
|
||||||
|
(r'{c', hilite + ANSI_CYAN),
|
||||||
|
(r'{C', normal + ANSI_CYAN),
|
||||||
|
(r'{w', hilite + ANSI_WHITE), # pure white
|
||||||
|
(r'{W', normal + ANSI_WHITE), # light grey
|
||||||
|
(r'{x', hilite + ANSI_BLACK), # dark grey
|
||||||
|
(r'{X', normal + ANSI_BLACK), # pure black
|
||||||
|
(r'{n', normal) # reset
|
||||||
|
]
|
||||||
|
|
||||||
|
# xterm256 {123, %c134,
|
||||||
|
|
||||||
|
xterm256_map = [
|
||||||
|
(r'%([0-5]{3})', parse_rgb), # %123 - foreground colour
|
||||||
|
(r'%(-[0-5]{3})', parse_rgb), # %-123 - background colour
|
||||||
|
(r'{([0-5]{3})', parse_rgb), # {123 - foreground colour
|
||||||
|
(r'{(-[0-5]{3})', parse_rgb) # {-123 - background colour
|
||||||
|
]
|
||||||
|
|
||||||
|
# obs - order matters here, we want to do the xterms first since
|
||||||
|
# they collide with some of the other mappings otherwise.
|
||||||
|
ansi_map = xterm256_map + mux_ansi_map + ext_ansi_map
|
||||||
|
|
||||||
|
# prepare regex matching
|
||||||
|
ansi_sub = [(re.compile(sub[0], re.DOTALL), sub[1])
|
||||||
|
for sub in ansi_map]
|
||||||
|
|
||||||
|
# prepare matching ansi codes overall
|
||||||
|
ansi_regex = re.compile("\033\[[0-9;]+m")
|
||||||
|
|
||||||
|
# escapes - these double-chars will be replaced with a single
|
||||||
|
# instance of each
|
||||||
|
ansi_escapes = re.compile(r"(%s)" % "|".join(ANSI_ESCAPES), re.DOTALL)
|
||||||
|
|
||||||
ANSI_PARSER = ANSIParser()
|
ANSI_PARSER = ANSIParser()
|
||||||
|
|
||||||
|
|
||||||
|
|
@ -279,3 +276,119 @@ def raw(string):
|
||||||
Escapes a string into a form which won't be colorized by the ansi parser.
|
Escapes a string into a form which won't be colorized by the ansi parser.
|
||||||
"""
|
"""
|
||||||
return string.replace('{', '{{').replace('%', '%%')
|
return string.replace('{', '{{').replace('%', '%%')
|
||||||
|
|
||||||
|
|
||||||
|
def group(lst, n):
|
||||||
|
for i in range(0, len(lst), n):
|
||||||
|
val = lst[i:i+n]
|
||||||
|
if len(val) == n:
|
||||||
|
yield tuple(val)
|
||||||
|
|
||||||
|
|
||||||
|
class ANSIString(str):
|
||||||
|
"""
|
||||||
|
String-like object that is aware of ANSI codes.
|
||||||
|
|
||||||
|
This isn't especially efficient, as it doesn't really have an
|
||||||
|
understanding of what the codes mean in order to eliminate
|
||||||
|
redundant characters, but a proper parser would have to be written for
|
||||||
|
that.
|
||||||
|
"""
|
||||||
|
|
||||||
|
def __new__(cls, *args, **kwargs):
|
||||||
|
string = args[0]
|
||||||
|
args = args[1:]
|
||||||
|
parser = kwargs.get('parser', ANSI_PARSER)
|
||||||
|
string = parser.parse_ansi(string)
|
||||||
|
return super(ANSIString, cls).__new__(ANSIString, string, *args)
|
||||||
|
|
||||||
|
def __init__(self, *args, **kwargs):
|
||||||
|
self.parser = kwargs.pop('parser', ANSI_PARSER)
|
||||||
|
super(ANSIString, self).__init__(*args, **kwargs)
|
||||||
|
self.raw_string = super(ANSIString, self).__str__()
|
||||||
|
self.clean_string = self.parser.parse_ansi(
|
||||||
|
self.raw_string, strip_ansi=True)
|
||||||
|
for func_name in [
|
||||||
|
'count', 'startswith', 'endswith', 'find', 'index', 'isalnum',
|
||||||
|
'isalpha', 'isdigit', 'islower', 'isspace', 'istitle', 'isupper',
|
||||||
|
'rfind', 'rindex']:
|
||||||
|
setattr(self, func_name, _query_super(func_name))
|
||||||
|
self._code_indexes, self._char_indexes = self._get_indexes()
|
||||||
|
|
||||||
|
def __len__(self):
|
||||||
|
return len(self.clean_string)
|
||||||
|
|
||||||
|
def __getslice__(self, i, j):
|
||||||
|
return self.__getitem__(slice(i, j))
|
||||||
|
|
||||||
|
def _slice(self, item):
|
||||||
|
slice_indexes = self._char_indexes[item]
|
||||||
|
if not slice_indexes:
|
||||||
|
return ANSIString('')
|
||||||
|
try:
|
||||||
|
string = self[item.start].raw_string
|
||||||
|
except IndexError:
|
||||||
|
return ANSIString('')
|
||||||
|
last_mark = slice_indexes[0]
|
||||||
|
for i in slice_indexes[1:]:
|
||||||
|
for index in range(last_mark, i):
|
||||||
|
if index in self._code_indexes:
|
||||||
|
string += self.raw_string[index]
|
||||||
|
last_mark = i
|
||||||
|
try:
|
||||||
|
string += self.raw_string[i]
|
||||||
|
except IndexError:
|
||||||
|
pass
|
||||||
|
return ANSIString(string)
|
||||||
|
|
||||||
|
def __getitem__(self, item):
|
||||||
|
if isinstance(item, slice):
|
||||||
|
return self._slice(item)
|
||||||
|
item = self._char_indexes[item]
|
||||||
|
clean = self.raw_string[item]
|
||||||
|
|
||||||
|
result = ''
|
||||||
|
for index in range(0, item + 1):
|
||||||
|
if index in self._code_indexes:
|
||||||
|
result += self.raw_string[index]
|
||||||
|
return ANSIString(result + clean)
|
||||||
|
|
||||||
|
def _get_indexes(self):
|
||||||
|
matches = [
|
||||||
|
(match.start(), match.end())
|
||||||
|
for match in self.parser.ansi_regex.finditer(self.raw_string)]
|
||||||
|
code_indexes = []
|
||||||
|
# These are all the indexes which hold code characters.
|
||||||
|
for start, end in matches:
|
||||||
|
code_indexes.extend(range(start, end))
|
||||||
|
|
||||||
|
flat_ranges = []
|
||||||
|
# We need to get the ones between them, but the code might start at
|
||||||
|
# the beginning, and there might be codes at the end.
|
||||||
|
for tup in matches:
|
||||||
|
flat_ranges.extend(tup)
|
||||||
|
# Is the beginning of the string a code character?
|
||||||
|
if flat_ranges[0] == 0:
|
||||||
|
flat_ranges.pop(0)
|
||||||
|
else:
|
||||||
|
flat_ranges.insert(0, 0)
|
||||||
|
# How about the end?
|
||||||
|
end_index = (len(self.raw_string) - 1)
|
||||||
|
if flat_ranges[-1] == end_index:
|
||||||
|
flat_ranges.pop()
|
||||||
|
else:
|
||||||
|
flat_ranges.append(end_index)
|
||||||
|
char_indexes = []
|
||||||
|
for start, end in list(group(flat_ranges, 2)):
|
||||||
|
char_indexes.extend(range(start, end))
|
||||||
|
# The end character will be left off if it's a normal character. Fix
|
||||||
|
# that here.
|
||||||
|
if end_index in flat_ranges:
|
||||||
|
char_indexes.append(end_index)
|
||||||
|
return code_indexes, char_indexes
|
||||||
|
|
||||||
|
|
||||||
|
def _query_super(func_name):
|
||||||
|
def query_func(self, *args, **kwargs):
|
||||||
|
getattr(self.raw_string, func_name)(self, *args, **kwargs)
|
||||||
|
return query_func
|
||||||
Loading…
Add table
Add a link
Reference in a new issue