stringold.py revision 7b7c5786161d9d443d13f56276620519587a2624
1# module 'string' -- A collection of string operations 2 3# Warning: most of the code you see here isn't normally used nowadays. 4# At the end of this file most functions are replaced by built-in 5# functions imported from built-in module "strop". 6 7# Some strings for ctype-style character classification 8whitespace = ' \t\n\r\v\f' 9lowercase = 'abcdefghijklmnopqrstuvwxyz' 10uppercase = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 11letters = lowercase + uppercase 12digits = '0123456789' 13hexdigits = digits + 'abcdef' + 'ABCDEF' 14octdigits = '01234567' 15 16# Case conversion helpers 17_idmap = '' 18for i in range(256): _idmap = _idmap + chr(i) 19_lower = _idmap[:ord('A')] + lowercase + _idmap[ord('Z')+1:] 20_upper = _idmap[:ord('a')] + uppercase + _idmap[ord('z')+1:] 21_swapcase = _upper[:ord('A')] + lowercase + _upper[ord('Z')+1:] 22del i 23 24# Backward compatible names for exceptions 25index_error = ValueError 26atoi_error = ValueError 27atof_error = ValueError 28atol_error = ValueError 29 30# convert UPPER CASE letters to lower case 31def lower(s): 32 res = '' 33 for c in s: 34 res = res + _lower[ord(c)] 35 return res 36 37# Convert lower case letters to UPPER CASE 38def upper(s): 39 res = '' 40 for c in s: 41 res = res + _upper[ord(c)] 42 return res 43 44# Swap lower case letters and UPPER CASE 45def swapcase(s): 46 res = '' 47 for c in s: 48 res = res + _swapcase[ord(c)] 49 return res 50 51# Strip leading and trailing tabs and spaces 52def strip(s): 53 i, j = 0, len(s) 54 while i < j and s[i] in whitespace: i = i+1 55 while i < j and s[j-1] in whitespace: j = j-1 56 return s[i:j] 57 58# Strip leading tabs and spaces 59def lstrip(s): 60 i, j = 0, len(s) 61 while i < j and s[i] in whitespace: i = i+1 62 return s[i:j] 63 64# Strip trailing tabs and spaces 65def rstrip(s): 66 i, j = 0, len(s) 67 while i < j and s[j-1] in whitespace: j = j-1 68 return s[i:j] 69 70 71# Split a string into a list of space/tab-separated words 72# NB: split(s) is NOT the same as splitfields(s, ' ')! 73def split(s, sep=None, maxsplit=0): 74 if sep is not None: return splitfields(s, sep, maxsplit) 75 res = [] 76 i, n = 0, len(s) 77 while i < n: 78 while i < n and s[i] in whitespace: i = i+1 79 if i == n: break 80 j = i 81 while j < n and s[j] not in whitespace: j = j+1 82 res.append(s[i:j]) 83 i = j 84 return res 85 86# Split a list into fields separated by a given string 87# NB: splitfields(s, ' ') is NOT the same as split(s)! 88# splitfields(s, '') returns [s] (in analogy with split() in nawk) 89def splitfields(s, sep=None, maxsplit=0): 90 if sep is None: return split(s, None, maxsplit) 91 res = [] 92 nsep = len(sep) 93 if nsep == 0: 94 return [s] 95 ns = len(s) 96 i = j = 0 97 count = 0 98 while j+nsep <= ns: 99 if s[j:j+nsep] == sep: 100 count = count + 1 101 res.append(s[i:j]) 102 i = j = j + nsep 103 if (maxsplit and (count >= maxsplit)): 104 break 105 106 else: 107 j = j + 1 108 res.append(s[i:]) 109 return res 110 111# Join words with spaces between them 112def join(words, sep = ' '): 113 return joinfields(words, sep) 114 115# Join fields with optional separator 116def joinfields(words, sep = ' '): 117 res = '' 118 for w in words: 119 res = res + (sep + w) 120 return res[len(sep):] 121 122# Find substring, raise exception if not found 123def index(s, sub, i = 0, last=None): 124 if last == None: last = len(s) 125 res = find(s, sub, i, last) 126 if res < 0: 127 raise ValueError, 'substring not found in string.index' 128 return res 129 130# Find last substring, raise exception if not found 131def rindex(s, sub, i = 0, last=None): 132 if last == None: last = len(s) 133 res = rfind(s, sub, i, last) 134 if res < 0: 135 raise ValueError, 'substring not found in string.index' 136 return res 137 138# Count non-overlapping occurrences of substring 139def count(s, sub, i = 0): 140 if i < 0: i = max(0, i + len(s)) 141 n = len(sub) 142 m = len(s) + 1 - n 143 if n == 0: return m-i 144 r = 0 145 while i < m: 146 if sub == s[i:i+n]: 147 r = r+1 148 i = i+n 149 else: 150 i = i+1 151 return r 152 153# Find substring, return -1 if not found 154def find(s, sub, i = 0, last=None): 155 Slen = len(s) # cache this value, for speed 156 if last == None: 157 last = Slen 158 elif last < 0: 159 last = max(0, last + Slen) 160 elif last > Slen: 161 last = Slen 162 if i < 0: i = max(0, i + Slen) 163 n = len(sub) 164 m = last + 1 - n 165 while i < m: 166 if sub == s[i:i+n]: return i 167 i = i+1 168 return -1 169 170# Find last substring, return -1 if not found 171def rfind(s, sub, i = 0, last=None): 172 Slen = len(s) # cache this value, for speed 173 if last == None: 174 last = Slen 175 elif last < 0: 176 last = max(0, last + Slen) 177 elif last > Slen: 178 last = Slen 179 if i < 0: i = max(0, i + Slen) 180 n = len(sub) 181 m = last + 1 - n 182 r = -1 183 while i < m: 184 if sub == s[i:i+n]: r = i 185 i = i+1 186 return r 187 188# Convert string to float 189def atof(str): 190 import regex 191 sign = '' 192 s = str 193 if s and s[0] in '+-': 194 sign = s[0] 195 s = s[1:] 196 if not s: 197 raise ValueError, 'non-float argument to string.atof' 198 while s[0] == '0' and len(s) > 1 and s[1] in digits: s = s[1:] 199 if regex.match('[0-9]*\(\.[0-9]*\)?\([eE][-+]?[0-9]+\)?', s) != len(s): 200 raise ValueError, 'non-float argument to string.atof' 201 try: 202 return float(eval(sign + s)) 203 except SyntaxError: 204 raise ValueError, 'non-float argument to string.atof' 205 206# Convert string to integer 207def atoi(str, base=10): 208 if base != 10: 209 # We only get here if strop doesn't define atoi() 210 raise ValueError, "this string.atoi doesn't support base != 10" 211 sign = '' 212 s = str 213 if s and s[0] in '+-': 214 sign = s[0] 215 s = s[1:] 216 if not s: 217 raise ValueError, 'non-integer argument to string.atoi' 218 while s[0] == '0' and len(s) > 1: s = s[1:] 219 for c in s: 220 if c not in digits: 221 raise ValueError, 'non-integer argument to string.atoi' 222 return eval(sign + s) 223 224# Convert string to long integer 225def atol(str, base=10): 226 if base != 10: 227 # We only get here if strop doesn't define atol() 228 raise ValueError, "this string.atol doesn't support base != 10" 229 sign = '' 230 s = str 231 if s and s[0] in '+-': 232 sign = s[0] 233 s = s[1:] 234 if not s: 235 raise ValueError, 'non-integer argument to string.atol' 236 while s[0] == '0' and len(s) > 1: s = s[1:] 237 for c in s: 238 if c not in digits: 239 raise ValueError, 'non-integer argument to string.atol' 240 return eval(sign + s + 'L') 241 242# Left-justify a string 243def ljust(s, width): 244 n = width - len(s) 245 if n <= 0: return s 246 return s + ' '*n 247 248# Right-justify a string 249def rjust(s, width): 250 n = width - len(s) 251 if n <= 0: return s 252 return ' '*n + s 253 254# Center a string 255def center(s, width): 256 n = width - len(s) 257 if n <= 0: return s 258 half = n/2 259 if n%2 and width%2: 260 # This ensures that center(center(s, i), j) = center(s, j) 261 half = half+1 262 return ' '*half + s + ' '*(n-half) 263 264# Zero-fill a number, e.g., (12, 3) --> '012' and (-3, 3) --> '-03' 265# Decadent feature: the argument may be a string or a number 266# (Use of this is deprecated; it should be a string as with ljust c.s.) 267def zfill(x, width): 268 if type(x) == type(''): s = x 269 else: s = `x` 270 n = len(s) 271 if n >= width: return s 272 sign = '' 273 if s[0] in ('-', '+'): 274 sign, s = s[0], s[1:] 275 return sign + '0'*(width-n) + s 276 277# Expand tabs in a string. 278# Doesn't take non-printing chars into account, but does understand \n. 279def expandtabs(s, tabsize=8): 280 res = line = '' 281 for c in s: 282 if c == '\t': 283 c = ' '*(tabsize - len(line)%tabsize) 284 line = line + c 285 if c == '\n': 286 res = res + line 287 line = '' 288 return res + line 289 290# Character translation through look-up table. 291def translate(s, table, deletions=""): 292 if type(table) != type('') or len(table) != 256: 293 raise TypeError, "translation table must be 256 characters long" 294 res = "" 295 for c in s: 296 if c not in deletions: 297 res = res + table[ord(c)] 298 return res 299 300# Capitalize a string, e.g. "aBc dEf" -> "Abc def". 301def capitalize(s): 302 return upper(s[:1]) + lower(s[1:]) 303 304# Capitalize the words in a string, e.g. " aBc dEf " -> "Abc Def". 305# See also regsub.capwords(). 306def capwords(s, sep=None): 307 return join(map(capitalize, split(s, sep)), sep or ' ') 308 309# Construct a translation string 310_idmapL = None 311def maketrans(fromstr, tostr): 312 if len(fromstr) != len(tostr): 313 raise ValueError, "maketrans arguments must have same length" 314 global _idmapL 315 if not _idmapL: 316 _idmapL = map(None, _idmap) 317 L = _idmapL[:] 318 fromstr = map(ord, fromstr) 319 for i in range(len(fromstr)): 320 L[fromstr[i]] = tostr[i] 321 return joinfields(L, "") 322 323# Try importing optional built-in module "strop" -- if it exists, 324# it redefines some string operations that are 100-1000 times faster. 325# It also defines values for whitespace, lowercase and uppercase 326# that match <ctype.h>'s definitions. 327 328try: 329 from strop import * 330 letters = lowercase + uppercase 331except ImportError: 332 pass # Use the original, slow versions 333