def digit2word(key):
key=key.replace(' U.S.','').replace(',','').replace('.','').strip()
if key.replace('0','')=='':
return 'zero'
if key=='-0':
return 'minus zero'
try:
if key.endswith("'s"):
return digit2word(key[:-2])+"'s"
if key.endswith("s"):
return digit2word(key[:-1])+"'s"
key=str(roman.fromRoman(key.replace('IIII','IV')))
except:
pass
try:
key=str(roman.fromRoman(key.split()[0].replace('IIII','IV')))
except:
pass
try:
text = p.number_to_words(key,decimal='point',andword='', zero='o')
if re.match(r'^0\.',key):
text = 'zero '+text[2:]
if re.match(r'.*\.0$',key): text = text[:-2]+' zero'
text = text.replace('-',' ').replace(',','')
return text.lower()
except: return key
def testFromRomanCase():
"""fromRoman should only accept uppercase input"""
for integer in range(1, 5000):
numeral = roman.toRoman(integer)
roman.fromRoman(numeral.upper())
with pytest.raises(roman.InvalidRomanNumeralError):
roman.fromRoman(numeral.lower())
def roman2dec() -> int:
"""
Entry point for `se roman2dec`
"""
import roman
parser = argparse.ArgumentParser(description="Convert a Roman numeral to a decimal number.")
parser.add_argument("-n", "--no-newline", dest="newline", action="store_false", help="don’t end output with a newline")
parser.add_argument("numbers", metavar="NUMERAL", nargs="+", help="a Roman numeral")
args = parser.parse_args()
lines = []
if not sys.stdin.isatty():
for line in sys.stdin:
lines.append(line.rstrip("\n"))
for line in args.numbers:
lines.append(line)
for line in lines:
try:
if args.newline:
print(roman.fromRoman(line.upper()))
else:
print(roman.fromRoman(line.upper()), end="")
except roman.InvalidRomanNumeralError:
se.print_error("Not a Roman numeral: {}".format(line))
return se.InvalidInputException.code
return 0
def testFromRomanCase(self):
"""fromRoman should only accept uppercase input"""
for integer in range(1, 4000):
numeral = roman.toRoman(integer)
roman.fromRoman(numeral.upper())
self.assertRaises(roman.InvalidRomanNumeralError,
roman.fromRoman, numeral.lower())
def infer_production_century(br, de):
"""
Estimate the production century for an author.
If born on or after the 80th year of a century,
say it's the following century.
If no dates (only centuries), and centuries differ,
output a .5 date (e.g. birth XVI, death XVII => 16.5).
Output False when birth or death date missing.
"""
if str(br) == '0' or str(de) == '0':
return False
try:
byear = int(br)
dyear = int(de)
except ValueError:
try:
byear = roman.fromRoman(br)
dyear = roman.fromRoman(de)
if byear == dyear:
pcen = byear
else:
assert dyear == byear + 1
pcen = byear + 0.5
return pcen
except roman.InvalidRomanNumeralError:
return False
if byear % 100 >= 80:
pcen = floor(byear / 100) + 2
assert floor(dyear / 100) + 1 == pcen
else:
pcen = floor(byear / 100) + 1
return pcen
def testFromRomanCase(self):
"""fromRoman should only accept uppercase input"""
for integer in range(1, 5000):
numeral = roman.toRoman(integer)
roman.fromRoman(numeral.upper())
self.assertRaises(roman.InvalidRomanNumeralError, roman.fromRoman,
numeral.lower())
def is_roman_number(roman_as_str):
""" returns True in case provided value is valid roman number, False otherwise """
try:
fromRoman(roman_as_str)
except InvalidRomanNumeralError:
return False
else:
return True
def roman_magic(self, v1, op, v2):
rom_v1 = str(fromRoman(v1))
rom_v2 = str(fromRoman(v2))
operation = rom_v1+op+rom_v2
return toRoman(eval(operation))
def testFromRomanCase(self):
"""fromRoman should only accept uppercase input"""
for integer in range(1, 5000):
numeral = roman.toRoman(integer)
roman.fromRoman(numeral.upper()) # assume nothing about toRoman() in the fromRoman() test!
# also, this is an implicit test that this line does NOT raise
self.assertRaises(roman.InvalidRomanNumeralError,
roman.fromRoman, numeral.lower())
def is_roman(goo):
try:
roman.fromRoman(goo)
result = True
except roman.InvalidRomanNumeralError:
result = False
return result
def isRoman(txt):
# helper function, checks if string contains valid roman number
# uses pip module roman: https://pypi.python.org/pypi/roman
try:
roman.fromRoman(txt)
return True
except roman.InvalidRomanNumeralError:
return False
def replace_chapter_number(match):
string = match.group(1)
try:
string = str(roman.fromRoman(string))
except roman.InvalidRomanNumeralError:
if string.startswith('L'):
string = str(roman.fromRoman(string[1:]))
return f'Chapitre {string}.'
def start_end(self):
match = PAGE_ORDER_RX.match(self.name)
if match is not None:
start = match.group("start")
end = match.group("end")
if start.isdigit():
return int(start), int(end)
start = roman.fromRoman(start)
end = roman.fromRoman(end)
return start, end
def is_roman(text: str) -> bool:
"""Check if the given string is a Roman number. Return True if it is, False if not.
"""
text = text.strip().upper()
text = re.sub(r'[^A-Z]', "", text)
try:
fromRoman(text)
except NoRoman:
return False
else:
return True
def getMinimalFormOfNumeral(numeral):
value = 0
while (len(numeral) > 1):
if (roman.fromRoman(numeral[0]) < roman.fromRoman(numeral[1])):
value -= roman.fromRoman(numeral[0])
else:
value += roman.fromRoman(numeral[0])
numeral = numeral[1:]
value += roman.fromRoman(numeral[0])
return roman.toRoman(value)
def convertToNum(input_type, s):
if input_type == "R":
num = roman.fromRoman(s)
elif input_type == "r":
num = roman.fromRoman(s.upper())
elif input_type == "a" or input_type == "A":
num = fromCharString(s.lower())
elif input_type == "i" or input_type == "f":
num = float(s)
else:
num = 0
return num
def replace_roman_numerals_in_match(m):
s = m.group(3)
s = s.upper()
try:
if s:
if m.group(8):
return u"{}{}:{}".format(m.group(1), roman.fromRoman(s), m.group(8))
else:
return u"{}{}{}".format(m.group(1), roman.fromRoman(s), m.group(7))
else:
return m.group(0)
except:
return m.group(0)
def romanDigitRangeCorrector(cite_in):
dash_expression = re.compile('[-–—]')
if dash_expression.search(cite_in):
first_roman_num = re.split(dash_expression, cite_in)[0]
second_roman_num = second_num = re.split(dash_expression, cite_in)[1]
first_num = str(roman.fromRoman(first_roman_num.upper()))
second_num = str(roman.fromRoman(second_roman_num.upper()))
arabic_range = first_num + '-' + second_num
return arabic_range
else:
return str(roman.fromRoman(cite_in.upper()))
def normalize_string(self, input_string):
"""
As we have several variations of the Canon Mark models we have to clean it up
1) break down the input string into tokens
2) check if one of those tokens is a variation of "Mark" (i.e. mark, mk, MarkIII, 5dmark, etc)
3) check if the mark model number is in roman and covert it to INT
4) eliminating spaces and lowering case
"""
tokens = [t for t in input_string.split()]
normalized_tokens = []
lookup_pattern = re.compile("(m(?:ar)?k\s*)([ivx\d]*)",
flags=re.IGNORECASE)
roman_nums = roman.romanNumeralPattern
for i, token in enumerate(tokens, 0):
# check if the token is a variation of mark
lookup = lookup_pattern.search(token)
if lookup:
# some strings have the 5d concatenated to the model, let's separate it
if re.search('5d', token, re.IGNORECASE):
token = '5D Mark'
else:
token = 'Mark'
model_nbr = lookup.group(2)
# check if the model number is part of the same token and covert it to ,INT
if model_nbr != '':
if model_nbr.isdigit():
pass
else:
model_nbr = roman.fromRoman(model_nbr.upper())
token = token + ' ' + str(model_nbr)
# check if the token is a roman numeral and covert it to INT
is_roman = roman_nums.search(token)
if is_roman:
token = str(roman.fromRoman(token))
normalized_tokens.append(token)
# convert the list of normalized tokens into a single string and normalize it
output_string = ' '.join(normalized_tokens)
output_string = output_string.strip()
output_string = output_string.lower()
output_string = output_string.replace(" ", "")
return output_string
def onLine(l):
global PLAY
global PLAY_LINES
global line
global inAct
global inScene
global inCharacter
global currCharacter
global firstBlock
l = l.rstrip()
if isSection(l, START_ACT):
inAct = roman.fromRoman(l.split()[1])
if DEBUG:
print('Act', inAct)
createIfEmpty(PLAY, inAct)
inScene = False
inCharacter = False
return
if isSection(l, START_SCENE):
inScene = roman.fromRoman(l.split()[1][:-1])
if DEBUG:
print('Scene', inScene)
createIfEmpty(PLAY[inAct], inScene)
PLAY[inAct][inScene] = []
inCharacter = False
line = 0
return
if isSection(l, START_CHARACTER) and \
l[len(START_CHARACTER) + 1] != ' ' and \
len([word for word in l.split() if word[0].isupper()]) == len(l.split()):
flushCharacter()
inCharacter = l.strip().upper()
currCharacter = ''
firstBlock = True
if DEBUG:
print('Character', inCharacter)
return
if inAct and inScene and inCharacter and firstBlock:
if len(l.strip()) == 0 and firstBlock:
if len(currCharacter) > 0:
firstBlock = False
return
line += 1
PLAY_LINES[''.join(
[str(inAct),
str(inScene), inCharacter,
l.strip().upper()])] = line
currCharacter += l + '\n'
def _ordinal2word(wordsList, indice):
"""Convert an ordinal number to a written
word.
i.e. 1er --> premier
param strNumber: an utf-8 ordinal number
return a 'written' ordinal number
"""
strNumber = NumberFormula._normalizeNumber(wordsList[indice])
if strNumber.encode('utf-8') == u"1ère".encode('utf-8'):
return u"première"
strNewNumber = re.sub(u"[erèm]", "", strNumber)
if NumberFormula._isCardinalNumber(strNewNumber):
strNewNumber = num2words(int(strNewNumber), ordinal=True, lang='fr')
elif NumberFormula._isRomanNumber(strNewNumber):
#Roman to cardinal
strNewNumber = strNewNumber.encode('utf-8')
cardinalNumber = fromRoman(strNewNumber)
#Digits to ordinal
strNewNumber = num2words(cardinalNumber, ordinal=True, lang='fr')
else:
strNewNumber = strNumber
strNewNumber = re.sub(r'vingtsi','vingti',strNewNumber)
strNewNumber = re.sub(r'centsi','centi',strNewNumber)
strNewNumber = re.sub(r'millionsi','millioni',strNewNumber)
strNewNumber = re.sub(r'milliardsi','milliardi',strNewNumber)
return strNewNumber
def get_orf_data(chromo, start, end, orf_annotation = None):
if orf_annotation is None:
sys.exit("For ORF annotation, a GFF file needs to be provided. Such as this one: https://github.com/jianlingzhong/COMPETE_examples/blob/master/saccharomyces_cerevisiae.20080621.gff")
genes = pd.read_csv(orf_annotation, sep = '\t', comment='#', header = None)
# only need the gene features in the gff
genes = genes.loc[genes.iloc[:, 2] == 'gene', ]
# don't need Mito and 2-micron chromosome features
genes = genes.loc[(genes.iloc[:, 0] != 'chrMito') & (genes.iloc[:, 0] != '2-micron')]
# just need the following columns
genes = genes.iloc[:, [0, 3,4,6,8]]
genes.rename(columns={0:'chromo', 3:'start', 4:'end', 6:'strand', 8:'name'}, inplace = True)
# convert 'chrI' to 1
genes.chromo = genes.chromo.apply(lambda x: roman.fromRoman(x[3:]))
genes = genes.loc[(genes.chromo == chromo)]
selection = ~((genes.end < start) | (genes.start > end))
genes = genes.loc[selection]
genes.name = genes.name.apply(find_name)
return genes
def _parse_stenogram_meta(self, response, meta_xs):
meta = {}
source = self._get_source(response.url, 'p_id')
meta['source'] = source
meta['_id'] = source['id']
date_match = meta_xs.re(date_re)
if date_match:
year = int(date_match[0])
month = month_names_map[date_match[1]]
day = int(date_match[2])
meta['date'] = date(year, month, day)
sitting_match = meta_xs.re(sitting_re)
if sitting_match:
meta['sitting_time'] = sitting_match[0].title()
meta['sitting_no'] = sitting_match[1]
else:
sitting_no_match = meta_xs.re(sitting_no_re)
if sitting_no_match:
meta['sitting_no'] = sitting_no_match[0]
session_match = meta_xs.re(session_re)
if session_match:
meta['session_no'] = roman.fromRoman(session_match[0])
meta['session_season'] = session_match[1].title()
if 'date' in meta:
meta['session'] = '{} {}'.format(meta['date'].year,
meta['session_season'])
return meta
def sort_collections(t):
sort_key = t[0]
if ": Chartae Latinae" in t[0]:
key_parts = sort_key.split()
key_parts[-1] = '{:04}'.format(roman.fromRoman(key_parts[-1]))
sort_key = ' '.join(key_parts)
return sort_key
def nameclean(orig):
name = orig.encode('ascii', errors='ignore').decode('ascii')
name = name.upper().strip()
name = name.replace('GOTY', 'GAME OF THE YEAR')
ignoredphrases = [
'\(', '\)', 'the', 'with', 'Early Access', 'bundle', 'and', '&', 'in',
'vr', 'beta', 'Double Pack', 'Pack', 'Free to Play', 'Edition'
]
for phrase in ignoredphrases:
name = re.sub(r'\b{0}\b'.format(phrase), '', name, flags=re.IGNORECASE)
name = re.sub(r'\bone\b', '1', name, flags=re.IGNORECASE)
name = re.sub(r'\btwo\b', '2', name, flags=re.IGNORECASE)
name = re.sub(r'\b40k\b', '40,000', name, flags=re.IGNORECASE)
name = re.sub(r'\+', 'PLUS', name, flags=re.IGNORECASE)
name = re.sub(r'[^\w\s]', ' ', name, flags=re.IGNORECASE)
name = re.sub(
r'(\A|\s)(\b(?=[MDCLXVI]+\b)M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})\b)(\s|\Z)',
lambda x: x.group(1) + str(roman.fromRoman(x.group(2).strip(
))) + x.group(6) if (x.group(0).strip()) else '',
name,
flags=re.IGNORECASE)
name = re.sub(r'(\w*)s\b', r'\1', name, flags=re.IGNORECASE)
words = name.split()
name = ' '.join(sorted(set(words), key=words.index))
name = re.sub(r'\s+', '', name, flags=re.IGNORECASE).strip()
return name
def setValues(siglum):
''' Take Roman numerals in the text (already marked up with <num>)
and insert a @value attribute with its arabic value.
The 'siglum' is the XML filename to parse ('g' for 'g.xml'; 'bonetti' for 'bonetti.xml' etc.).
Then write the modified tree to another XML file ('numerals-g.xml'; 'numerals-bonetti.xml' etc.).
'''
n = myconst.ns
tree = etree.parse('../xml/%s.xml' % siglum)
numbers = tree.findall('.//t:num', n)
for number in numbers:
if number.get('value') is None:
content = getRomanContent(number, checkallnumbers=False)
if content == '':
print('foo', number.text)
# print(content.upper(), end='\t') # debug
try:
myvalue = roman.fromRoman(content.upper())
#print(roman.fromRoman(content.upper())) # debug
number.set('value', str(myvalue))
number.set('type', 'guessedvalue')
except roman.InvalidRomanNumeralError:
print('Numero romano non parsabile:', content.upper())
number.set('type', 'foo')
else:
#print('Il valore di @value era già settato a', number.get('value'))
pass
tree.write('../xml/numerals-%s.xml' % (siglum),
encoding='UTF-8',
method='xml',
pretty_print=True,
xml_declaration=True)
def token_features(y):
x=y.strip()
parts=x.split('.',2)
pattern = re.compile('[1-9][0-9]*(\.)?([1-9][0-9]*(\.([1-9][0-9]*)?)?)?')
m = pattern.match(x)
if x=="Table" or x=="TABLE" or x=="Figure" or x=="FIGURE" or x=="Fig." or x=="FIG.":
return "0"
if m and m.span()==(0, len(x)):
return "1"
try:
roman.fromRoman(parts[0].upper())
return "1"
except:
if x[0].isupper():
return "2"
return "3"
def sgd_gff2dataframe(input_annot_gff_filename, ok_features_list, include_dubious=False):
"""
loads sgd annotation gff file to dataframe\
leave only features of the types in the input list
filters mitochondrial and 2-micron genes
"""
gff_df = gff2dataframe(input_annot_gff_filename)
gff_df['Name'] = gff_df['Name'].str.strip()
#gff_df = gff_df[gff_df['feature'].isin(['gene','CDS']) ]
# just coding sequence
gff_df = gff_df[gff_df['feature'].isin(ok_features_list) ] # ['CDS']
# filtering dubious and not in standard chromosomes
if include_dubious:
idx = ~gff_df['chrom'].isin(['chrMito', '2-micron'])
else:
idx = gff_df['Note'].apply(lambda x: not str(x).startswith('Dubious')) & ~gff_df['chrom'].isin(['chrMito', '2-micron'])
gff_df = gff_df[idx]
# fixing chromosome to be chrXX where XX is integer
gff_df['chrom'] = gff_df['chrom'].apply(lambda x: 'chr' + str(fromRoman(str(x).replace('chr',''))).zfill(2) )
# removing "SGD:" from the id column
gff_df['dbxref'] = gff_df['dbxref'].apply(lambda x: str(x).replace('SGD:',''))
gff_df['start'] = gff_df['start'].apply(int)
return(gff_df)
def arabicOrRomanToInt(s):
m = re.match(r'\d+$', s)
if m:
i = int(s)
else:
i = roman.fromRoman(s)
return i
def replace_ordinal(token):
"""10.序数词"""
res = token.strip()
replace_list = [
"th", "TH", "Th", 'st', 'ST', 'nd', 'ND', 'rd', 'RD', ".", "º", ",",
"ª"
]
ends = ""
if res.endswith("'s"):
ends = "'s"
res = res[:-2]
if res.endswith("s"):
ends = "s"
res = res[:-1]
for one in replace_list:
res = res.replace(one, "").strip()
if res.isdigit():
res = numstr2word(float(res), ordinal=True)
else:
# 可能是罗马数字
try:
num = roman.fromRoman(res)
res = numstr2word(num, ordinal=True)
res = "the " + res
except roman.InvalidRomanNumeralError:
pass
# print("roman.InvalidRomanNumeralError", res)
return res + ends
def load():
a=pd.read_csv('/home/arya/storage/Data/Yeast/BurkeYeast/freq_tables/snps.txt',sep='\t').rename(columns={'chr':'CHROM','pos':'POS'}).set_index(['CHROM','POS']).iloc[:,2:]
c=pd.Series(a.columns)
founders=a[c[c.apply(lambda x: x[0]=='F')]]
founders.loc[:,map(lambda x:'cov' in x,founders.columns)].sum(1)
def toint(x):
try:return int(x)
except: return x
cd=a[c[c.apply(lambda x: 'maf' ==x[:3]) |c.apply(lambda x: 'cov' ==x[:3])]]
cd.columns=map(lambda x:tuple(map(lambda y: toint(y.replace('cov','D').replace('maf','C')) ,x.split('_'))), cd.columns)
cd.columns=pd.MultiIndex.from_tuples(cd.columns)
cd=pd.concat([(cd['C']* cd['D']).round().astype(int),cd['D']],1,keys=['C','D']).reorder_levels([1,2,0],axis=1)
cd.columns.names=['REP','GEN','READ']
ancestors=a[c[c.apply(lambda x: 'anc' in x)]].rename(columns={'anc_maf':'C','anc_cov':'D'})
ancestors['C']=(ancestors['C']*ancestors['D']).round().astype(int)
ancestors=pd.concat([ancestors],1,keys=[0])
R=cd.columns.levels[0]
ancestors=pd.concat([ancestors for _ in R],1,keys=R)
ancestors.columns.names=['REP','GEN','READ']
cd=pd.concat([cd,ancestors],1).sort_index(1).sort_index()
weekToGen={0:0, 6:180, 12:360, 18:540}
cd=cd.T.reset_index()
cd.GEN=cd.GEN.replace(weekToGen)
cd=cd.set_index(['REP','GEN','READ']).T.sort_index(1).sort_index()
try:
import roman
cd=cd.reset_index()
cd.CHROM=cd.CHROM.apply(lambda x:roman.fromRoman(x[3:]))
cd=cd.set_index(['CHROM','POS'])
except:
pass
return
def replace_cardinal(token):
"""5.普通数字"""
token = token.replace('U.S.', "")
token = token.replace(" ", "")
token = token.replace('"', "")
token = token.replace(',', "")
token = token.replace('.', "")
token = token.replace(':', "")
minus = False
if len(token) == 0:
return ""
if token.startswith("-"):
minus = True
token = token[1:]
token = token.replace('-', "")
ends = ""
if token.endswith("'s"):
ends = "'s"
token = token[:-2]
elif token.endswith("A") or token.endswith("M"):
token = token[:-1].strip()
if token.isdigit():
res = numstr2word(token)
else:
# 可能是罗马数字
try:
num = roman.fromRoman(token)
res = numstr2word(num)
except roman.InvalidRomanNumeralError:
res = token
if minus:
res = "minus " + res
res += ends
return res
def arabicOrRomanToInt(s):
m = re.match(r'\d+$', s)
if m:
i = int(s)
else:
i = roman.fromRoman(s)
return i
def numeric(x):
if x[0].isdigit():
return int(x) + Util.INTEGER_OFFSET
try:
return roman.fromRoman(x)
except roman.RomanError:
return 0
def testSanity(self):
"""fromRoman(toRoman(n)) == n for all n"""
for integer in range(1, 5000):
numeral = roman.toRoman(integer)
result = roman.fromRoman(numeral)
self.assertEqual(integer, result)
def _ordinal2word(wordsList, indice):
"""Convert an ordinal number to a written
word.
i.e. 1er --> premier
param strNumber: an utf-8 ordinal number
return a 'written' ordinal number
"""
strNumber = NumberFormula._normalizeNumber(wordsList[indice])
if strNumber.encode('utf-8') == u"1st".encode('utf-8'):
return u"first"
strNewNumber = re.sub(u"[ndstrh]", "", strNumber)
# print strNewNumber
# if NumberFormula._isCardinalNumber(strNewNumber):
if strNewNumber.isdigit():
strNewNumber = num2words(int(strNewNumber), ordinal=True)
# print(strNewNumber)
elif NumberFormula._isRomanNumber(strNewNumber):
#Roman to cardinal
strNewNumber = strNewNumber.encode('utf-8')
# print strNewNumber
cardinalNumber = fromRoman(strNewNumber)
#Digits to ordinal
strNewNumber = num2words(cardinalNumber, ordinal=True)
else:
print("newnumberis not digit!!!")
strNewNumber = strNumber
return strNewNumber
def get_unit(title):
pattern = patterns.makeTitlePatterns()
title = title.lower()
unit = title.replace(" ", "_")
unit = re.sub("[ \[\]\(\)]", "_", title)
number = "1"
for unitForm, unitPattern in pattern.iteritems():
m = unitPattern.match(title)
if m:
if unitForm == "title_preamble":
unit = m.group(1)
number = "1"
elif unitForm == "title_number":
unit = "unspecified"
number = m.group(1)
elif unitForm == "title_unit_roman":
unit = m.group(1)
try:
number = str(roman.fromRoman(m.group(2).upper()))
except roman.InvalidRomanNumeralError:
print "Invalid Roman title number:", title, "unit:", unit, "number:", number
number = m.group(2)
else:
unit = m.group(1)
number = m.group(2)
if unit == "art":
unit = "article"
return unit, number
def tb_heading_features(y):
x=y.strip()
parts=x.split('.',2)
pattern = re.compile('[1-9][0-9]*(\.)?([1-9][0-9]*(\.([1-9][0-9]*)?)?)?')
m = pattern.match(x)
if x=="Table" or x=="TABLE" or x=="Figure" or x=="FIGURE" or x=="Fig." or x=="FIG.":
return "0"
if m and m.span()==(0, len(x)):
return "1"
try:
roman.fromRoman(parts[0].upper())
return "1"
except:
if x[0].isupper():
return "2"
return "3"
def maybe_normalize(self,
value,
mapping='',
roman_normalization=True,
mapping_append=True):
if mapping == '':
mapping = self.default_mapping
else:
mapping = mapping + self.default_mapping if mapping_append else self.default_mapping + mapping
for norm in mapping:
if type(norm[0]) == str:
value = value.replace(norm[0], norm[1])
elif isinstance(norm[0], Pattern):
value = norm[0].sub(norm[1], value)
else:
print('UNEXPECTED', type(norm[0]), norm[0])
if roman_normalization:
for ro_before, ro_after, ro in self.get_roman_numbers(value):
try:
value = value.replace(
ro_before + ro + ro_after,
ro_before + str(roman.fromRoman(ro)) + ro_after)
except roman.InvalidRomanNumeralError as ex:
print(ex)
pass
value = self.clean_single_line(value)
return value.replace(' ', " ")
def token_features(y):
x=y.strip()
parts=x.split('.')
if(x=="$$$"):
return "5"
if x=="Table" or x=="TABLE" or x=="Figure" or x=="FIGURE" or x=="Fig.":
return "0"
p_len = len(parts)
if(p_len==1):
if(x.isdigit() and 1<=int(x)<=20):
return "1"
if(p_len==2 or p_len==3):
if(parts[0].isdigit() and 1<=int(parts[0])<=20):
if(parts[1]=='' or (parts[1].isdigit() and int(parts[1])<=20)):
if(p_len==2):
return "1"
if(parts[1].isdigit() and int(parts[1])<=20 and parts[2]==''):
return "1"
if(p_len==1 or (p_len==2 and parts[1]=='')):
try:
val = roman.fromRoman(parts[0].upper())
if(val<=20):
return "1"
except:
None
if((len(parts[0])==1 and 'A'<=parts[0]<='Z') or (len(parts[0])==3 and parts[0][0]=='(' and parts[0][2]==')' and parts[0][1].isalpha() and parts[0][1].isupper()) or (len(parts[0])==2 and parts[0][1]==')' and parts[0][0].isalpha() and parts[0][0].isupper())):
if(p_len==1 or (p_len==2 and parts[1]=='')):
return "1"
if x[0].isupper():
return "2"
if (not(parts[0].isalpha() or parts[0].isdigit())):
return "3"
return "4"
def get_spenser(self):
'''
Returns a dictionary of sonnets.
Each element of the dict is one of the sonnets, keyed by sonnet number
Each sonnet is a list of lists of lines
'''
filename = 'data/spenser.txt'
num = 0
sonnets = {}
sonnet_lines = []
with open(filename) as f:
for line in f:
line = re.sub('[:;.()?!]', '', line.strip()).lower()
try:
num = int(roman.fromRoman(line.upper()))
sonnets[num] = []
sonnet_lines = []
continue
except ValueError:
pass
except roman.InvalidRomanNumeralError:
pass
next_line = line.split()
if len(next_line) > 0:
sonnets[num].append(next_line)
return sonnets
def getSortedList_easy(lst):
"""
Uses pre-made module Roman for speed comparision/ to check if conversion is correct
:param lst: List of Stringed RoyalNames
:return: List of Sorted RoyalNames
"""
if all(isinstance(x, str) for x in lst):
roman_to_int_convertedlst = []
int_to_roman_convertedlst = []
for name,ordinal in [royalname.split() for royalname in lst]: #split the name and roman number
newnumber = roman.fromRoman(ordinal)
royalname = name, newnumber
roman_to_int_convertedlst.append(royalname)
print("Converted from", lst, "\n"
"to", roman_to_int_convertedlst)
sortedlst = sorted(roman_to_int_convertedlst)
print("Sorted", sortedlst)
for name,integer in sortedlst:
roman_numeral = roman.toRoman(integer)
int_to_roman_convertedlst.append(name+" "+str(roman_numeral))
return int_to_roman_convertedlst
def replace_roman_numerals_in_match(m):
s = m.group(2)
s = s.upper()
try:
if s:
return "%s%s:%s" % (m.group(1), roman.fromRoman(s), m.group(7))
except:
return m.group(0)
def ps_event(numeral):
try:
number = roman.fromRoman(numeral.upper())
except roman.InvalidRomanNumeralError:
return "Invalid roman numeral!", 400
event = ps_data.get_ps_event_by_number(number)
return flask.render_template('event.html', event=event)
def parse_token(token):
if token in '()+-*/':
return token
if len(token.strip('0123456789,')) == 0:
return token.replace(',', '')
if len(token.strip('IVXLCDM')) == 0:
return str(roman.fromRoman(token))
log.warning('Unknown token: ' + token)
return raw_input()
def _extract_charge(arg: str):
"""
Receive a `str` representing an element, isotope, or ion.
Return a `tuple` containing a `str` that should represent an
element or isotope, and either an `int` representing the
charge or `None` if no charge information is provided. Raise
an `~plasmapy.utils.InvalidParticleError` if charge information
is inputted incorrectly.
"""
invalid_charge_errmsg = (f"Invalid charge information in the particle string '{arg}'.")
if arg.count(' ') == 1: # Cases like 'H 1-' and 'Fe-56 1+'
isotope_info, charge_info = arg.split(' ')
sign_indicator_only_on_one_end = (
charge_info.endswith(('-', '+')) ^
charge_info.startswith(('-', '+')))
just_one_sign_indicator = (
(charge_info.count('-') == 1 and
charge_info.count('+') == 0) or
(charge_info.count('-') == 0 and
charge_info.count('+') == 1))
if not sign_indicator_only_on_one_end and just_one_sign_indicator:
raise InvalidParticleError(invalid_charge_errmsg) from None
charge_str = charge_info.strip('+-')
try:
if roman.romanNumeralPattern.match(charge_info):
Z_from_arg = roman.fromRoman(charge_info) - 1
elif '-' in charge_info:
Z_from_arg = - int(charge_str)
elif '+' in charge_info:
Z_from_arg = int(charge_str)
else:
raise InvalidParticleError(invalid_charge_errmsg) from None
except ValueError:
raise InvalidParticleError(invalid_charge_errmsg) from None
elif arg.endswith(('-', '+')): # Cases like 'H-' and 'Pb-209+++'
char = arg[-1]
match = re.match(f"[{char}]*", arg[::-1])
Z_from_arg = match.span()[1]
isotope_info = arg[0:len(arg) - match.span()[1]]
if char == '-':
Z_from_arg = -Z_from_arg
if isotope_info.endswith(('-', '+')):
raise InvalidParticleError(invalid_charge_errmsg) from None
else:
isotope_info = arg
Z_from_arg = None
return isotope_info, Z_from_arg
def _roman2word(strNumber):
"""Convert a roman number to a written
word.
param strNumber: an utf-8 roman number
return a 'written' roman number
"""
strNumber = strNumber.encode('utf-8')
cardinalNumber = fromRoman(strNumber)
return NumberFormula._cardinal2word(cardinalNumber)
def markup(self):
all = descend(self.dom,stackOfTags=[],seenSoFar=dict(),depth=1)
returnable = []
for key in all:
line = dict(key)
line["textString"] = all[key]
if re.search(r"\w+",line["textString"]):
try:
line["actNumber"] = roman.fromRoman(re.findall(r"(?:ACT|Act) +(\w+)",line["ACT"])[0])
except KeyError:
pass
try:
line["sceneNumber"] = roman.fromRoman(re.findall(r"(?:SCENE|Scene) +([IXVL]+)",line["SCENE"])[0])
except KeyError:
pass
except:
pass
returnable.append(line)
return returnable
def _roman2word(strNumber):
"""Convert a roman number to a written
word.
param strNumber: an utf-8 roman number
return a 'written' roman number
"""
strNumber = strNumber.encode('utf-8')
cardinalNumber = fromRoman(strNumber)
strNewNumber = num2words(cardinalNumber)
return strNewNumber
def _to_int(self, value):
if isinstance(value, int):
return value
elif isinstance(value, basestring) and self.is_roman(value.upper()):
return roman.fromRoman(value.upper())
elif isinstance(value, RomanNumeral):
return value._val
else:
raise ValueError("Value must be a valid roman numeral, a string"
" representing one or an integer: '{0}'"
.format(value))
def parse_worked_on(self, text_or_node, submodule):
submodule = roman.fromRoman(submodule) - 1
worked_strings = []
if isinstance(text_or_node, basestring):
worked_strings = [text_or_node]
else:
enumerations = text_or_node.traverse(utils.is_list_or_enumeration)
for enumeration in enumerations:
worked_strings += utils.parse_list_items(text_or_node)
self._parse_worked_strings(worked_strings, submodule)
def _compute_sortkey(self):
match = _REGEX_MAPPING[self.type].search(self.name)
key = "4"
latin_number = 0
if self.name != "R":
latin_number = roman.fromRoman(match.group(1))
if self.type == VolumeType.FIRST_SERIES:
key = f"1_{latin_number:02d}_{match.group(2)}"
elif self.type == VolumeType.SECOND_SERIES:
key = f"2_{latin_number:02d}_{match.group(2)}"
elif self.type == VolumeType.SUPPLEMENTS:
key = f"3_{latin_number:02d}"
return key
def split_name_and_submodule(name):
res = submodule_split_re.match(name)
parts = list(res.groups())
parts[0] = parts[0].strip()
if len(parts[1]) == 0:
parts = [parts[0]]
else:
submodules = submodules_split_re.split(
parts[1])
parts = [parts[0]] + [
roman.fromRoman(submodule.upper()) - 1
for submodule in submodules]
return parts
def rewrite_roman_numerals(self, text):
if self.refUnits == []:
return text
romanRef = re.compile(r"\b(" + "|".join(self.refUnits) + ") " + self.romanPattern + r"\b")
matches = romanRef.findall(text)
for unit, romanNumeral in matches:
try:
transform = roman.fromRoman(romanNumeral.upper())
number = str(transform)
text = re.sub(r"\b" + romanNumeral + r"\b", number, text)
#print "unit:", unit, "Roman Numeral:", romanNumeral, "number:", number, "context:", text
except roman.InvalidRomanNumeralError:
pass
return text
def traiter_nombres(chaine):
chaine = chaine.replace(' ','')
global seps
i = -1
sep = 0 # position du separateur
#1. convertir le nombre romain en entier
for c in chaine: #parcourir la chaine
i = i + 1
if c in seps:
morceau = chaine[sep:i].upper() #le morceau à traiter
try:
chaine = chaine[:sep] + str(fromRoman(morceau)) + chaine[i:]
except:
pass
sep = i+1
if sep == 0:
try:
chaine = str(fromRoman(chaine))
except:
pass
chaine = supprimer_sep_nombre(chaine)
return chaine
def parse_dependencies(self, text_or_node, submodule):
submodule = roman.fromRoman(submodule) - 1
dep_strings = []
if isinstance(text_or_node, basestring):
dep_strings = [text_or_node]
else:
enumerations = text_or_node.traverse(utils.is_list_or_enumeration)
for enumeration in enumerations:
dep_strings += utils.parse_list_items(text_or_node)
if submodule not in self.dependencies:
self.dependencies[submodule] = []
self.dependencies[submodule] += dep_strings
def token_features(tok_input):
tok = tok_input.strip()
parts = tok.split('.')
# Mark special sections
if tok == "Abstract" or tok == "ABSTRACT" or tok == "Acknowledgement" or tok == "ACKNOWLEDGEMENT" or tok == "References" or tok == "Reference" or tok == "REFERENCE" or tok == "REFERENCES" or tok == "Acknowledgements" or tok == "ACKNOWLEDGEMENTs":
return "6"
# Mark special symbol '$$$'
if tok == "$$$":
return "5"
# Mark table/figure headings
if tok == "Table" or tok == "TABLE" or tok == "Figure" or tok == "FIGURE" or tok == "Fig.":
return "0"
p_len = len(parts)
# Digit tokens
if p_len == 1:
if tok.isdigit() and 1 <= int(tok) <= 20:
return "1"
if p_len == 2 or p_len == 3:
if parts[0].isdigit() and 1 <= int(parts[0]) <= 20:
if parts[1] == '' or (parts[1].isdigit() and int(parts[1]) <= 20):
if p_len == 2:
return "1"
if parts[1].isdigit() and int(parts[1]) <= 20 and parts[2] == '':
return "1"
# Handling Roman Numericals
if p_len == 1 or (p_len == 2 and parts[1] == ''):
try:
val = roman.fromRoman(parts[0].upper())
if val <= 20:
return "1"
except:
pass
if ((len(parts[0]) == 1 and 'A' <= parts[0] <= 'Z') or (
len(parts[0]) == 3 and parts[0][0] == '(' and parts[0][2] == ')' and parts[0][
1].isalpha() and parts[0][1].isupper()) or (
len(parts[0]) == 2 and parts[0][1] == ')' and parts[0][0].isalpha() and parts[0][
0].isupper())):
if p_len == 1 or (p_len == 2 and parts[1] == ''):
return "1"
# Handling Capitalizations
if tok[0].isupper():
return "2"
if not (parts[0].isalpha() or parts[0].isdigit()):
return "3"
return "4"
def solve(ques):
# We only need lhs
ques = ques.split("=")[0]
# We don't need commas
ques = ques.repace(',', '')
parsed = ''
# Let's separate parts between operators
groups = re.split('([\-\+\*\/\(\)])', ques)
# Convert each part to a number
for g in groups:
g = g.strip()
# If part a roman numeral, convert it to decimal
if re.match('([A-Z]+)', g):
parsed += "%s" % roman.fromRoman(g)
# If part is in plain english, convert it to decimal
elif re.match('([a-z]+)', g):
parsed += "%s" % text2int(g)
else: parsed += g
# Now eval can do the rest
return "%s\r\n" % (eval(parsed))
def decide_REIA_or_REWL(re_sub):
test1 = re_sub.group(0)
searcher = FIT.search(re_sub.group(0))
(arg1, arg2, arg3) = build_arg(searcher)
if searcher.group(1) == 'archive': # archive
return '{{REIA|%s|%s}}' % (arg1, arg2)
else: # wikilivre
arabic = roman.fromRoman(searcher.group(3))
subl = searcher.group(2)
app = searcher.group(4)
halfband = searcher.group(5)
if (app is None) and (subl is None) and (arabic < 12): # all between I and XI
return '{{REIA|%s|%s}}' % (arg1, arg2)
# I A,1 and II A,1
elif (subl is None) and (app == 'A') and (arabic < 3) and (halfband == '1'):
return '{{REIA|%s|%s}}' % (arg1, arg2)
# S I, S II, S III
elif (subl == 'S') and (app is None) and (arabic < 4) and (halfband is None):
return '{{REIA|%s|%s}}' % (arg1, arg2)
else: # rest
return '{{REWL|%s|%s}}' % (arg1, arg2)
