string.py revision 06ba34c5d47751f9a5ae8e16bf1a7cf12c871609
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 if maxsplit <= 0: maxsplit = n 78 count = 0 79 while i < n: 80 while i < n and s[i] in whitespace: i = i+1 81 if i == n: break 82 if count >= maxsplit: 83 res.append(s[i:]) 84 break 85 j = i 86 while j < n and s[j] not in whitespace: j = j+1 87 count = count + 1 88 res.append(s[i:j]) 89 i = j 90 return res 91 92# Split a list into fields separated by a given string 93# NB: splitfields(s, ' ') is NOT the same as split(s)! 94# splitfields(s, '') returns [s] (in analogy with split() in nawk) 95def splitfields(s, sep=None, maxsplit=0): 96 if sep is None: return split(s, None, maxsplit) 97 res = [] 98 nsep = len(sep) 99 if nsep == 0: 100 return [s] 101 ns = len(s) 102 if maxsplit <= 0: maxsplit = ns 103 i = j = 0 104 count = 0 105 while j+nsep <= ns: 106 if s[j:j+nsep] == sep: 107 count = count + 1 108 res.append(s[i:j]) 109 i = j = j + nsep 110 if count >= maxsplit: break 111 112 else: 113 j = j + 1 114 res.append(s[i:]) 115 return res 116 117# Join words with spaces between them 118def join(words, sep = ' '): 119 return joinfields(words, sep) 120 121# Join fields with optional separator 122def joinfields(words, sep = ' '): 123 res = '' 124 for w in words: 125 res = res + (sep + w) 126 return res[len(sep):] 127 128# Find substring, raise exception if not found 129def index(s, sub, i = 0, last=None): 130 if last is None: last = len(s) 131 res = find(s, sub, i, last) 132 if res < 0: 133 raise ValueError, 'substring not found in string.index' 134 return res 135 136# Find last substring, raise exception if not found 137def rindex(s, sub, i = 0, last=None): 138 if last is None: last = len(s) 139 res = rfind(s, sub, i, last) 140 if res < 0: 141 raise ValueError, 'substring not found in string.index' 142 return res 143 144# Count non-overlapping occurrences of substring 145def count(s, sub, i = 0, last=None): 146 Slen = len(s) # cache this value, for speed 147 if last is None: 148 last = Slen 149 elif last < 0: 150 last = max(0, last + Slen) 151 elif last > Slen: 152 last = Slen 153 if i < 0: i = max(0, i + Slen) 154 n = len(sub) 155 m = last + 1 - n 156 if n == 0: return m-i 157 r = 0 158 while i < m: 159 if sub == s[i:i+n]: 160 r = r+1 161 i = i+n 162 else: 163 i = i+1 164 return r 165 166# Find substring, return -1 if not found 167def find(s, sub, i = 0, last=None): 168 Slen = len(s) # cache this value, for speed 169 if last is None: 170 last = Slen 171 elif last < 0: 172 last = max(0, last + Slen) 173 elif last > Slen: 174 last = Slen 175 if i < 0: i = max(0, i + Slen) 176 n = len(sub) 177 m = last + 1 - n 178 while i < m: 179 if sub == s[i:i+n]: return i 180 i = i+1 181 return -1 182 183# Find last substring, return -1 if not found 184def rfind(s, sub, i = 0, last=None): 185 Slen = len(s) # cache this value, for speed 186 if last is None: 187 last = Slen 188 elif last < 0: 189 last = max(0, last + Slen) 190 elif last > Slen: 191 last = Slen 192 if i < 0: i = max(0, i + Slen) 193 n = len(sub) 194 m = last + 1 - n 195 r = -1 196 while i < m: 197 if sub == s[i:i+n]: r = i 198 i = i+1 199 return r 200 201# Convert string to float 202re = None 203def atof(str): 204 global re 205 if re is None: 206 import re 207 sign = '' 208 s = strip(str) 209 if s and s[0] in '+-': 210 sign = s[0] 211 s = s[1:] 212 if not s: 213 raise ValueError, 'non-float argument to string.atof' 214 while s[0] == '0' and len(s) > 1 and s[1] in digits: s = s[1:] 215 if not re.match('[0-9]*(\.[0-9]*)?([eE][-+]?[0-9]+)?$', s): 216 raise ValueError, 'non-float argument to string.atof' 217 try: 218 return float(eval(sign + s)) 219 except SyntaxError: 220 raise ValueError, 'non-float argument to string.atof' 221 222# Convert string to integer 223def atoi(str, base=10): 224 if base != 10: 225 # We only get here if strop doesn't define atoi() 226 raise ValueError, "this string.atoi doesn't support base != 10" 227 sign = '' 228 s = str 229 if s and s[0] in '+-': 230 sign = s[0] 231 s = s[1:] 232 if not s: 233 raise ValueError, 'non-integer argument to string.atoi' 234 while s[0] == '0' and len(s) > 1: s = s[1:] 235 for c in s: 236 if c not in digits: 237 raise ValueError, 'non-integer argument to string.atoi' 238 return eval(sign + s) 239 240# Convert string to long integer 241def atol(str, base=10): 242 if base != 10: 243 # We only get here if strop doesn't define atol() 244 raise ValueError, "this string.atol doesn't support base != 10" 245 sign = '' 246 s = str 247 if s and s[0] in '+-': 248 sign = s[0] 249 s = s[1:] 250 if not s: 251 raise ValueError, 'non-integer argument to string.atol' 252 while s[0] == '0' and len(s) > 1: s = s[1:] 253 for c in s: 254 if c not in digits: 255 raise ValueError, 'non-integer argument to string.atol' 256 return eval(sign + s + 'L') 257 258# Left-justify a string 259def ljust(s, width): 260 n = width - len(s) 261 if n <= 0: return s 262 return s + ' '*n 263 264# Right-justify a string 265def rjust(s, width): 266 n = width - len(s) 267 if n <= 0: return s 268 return ' '*n + s 269 270# Center a string 271def center(s, width): 272 n = width - len(s) 273 if n <= 0: return s 274 half = n/2 275 if n%2 and width%2: 276 # This ensures that center(center(s, i), j) = center(s, j) 277 half = half+1 278 return ' '*half + s + ' '*(n-half) 279 280# Zero-fill a number, e.g., (12, 3) --> '012' and (-3, 3) --> '-03' 281# Decadent feature: the argument may be a string or a number 282# (Use of this is deprecated; it should be a string as with ljust c.s.) 283def zfill(x, width): 284 if type(x) == type(''): s = x 285 else: s = `x` 286 n = len(s) 287 if n >= width: return s 288 sign = '' 289 if s[0] in ('-', '+'): 290 sign, s = s[0], s[1:] 291 return sign + '0'*(width-n) + s 292 293# Expand tabs in a string. 294# Doesn't take non-printing chars into account, but does understand \n. 295def expandtabs(s, tabsize=8): 296 res = line = '' 297 for c in s: 298 if c == '\t': 299 c = ' '*(tabsize - len(line)%tabsize) 300 line = line + c 301 if c == '\n': 302 res = res + line 303 line = '' 304 return res + line 305 306# Character translation through look-up table. 307def translate(s, table, deletions=""): 308 if type(table) != type('') or len(table) != 256: 309 raise TypeError, "translation table must be 256 characters long" 310 res = "" 311 for c in s: 312 if c not in deletions: 313 res = res + table[ord(c)] 314 return res 315 316# Capitalize a string, e.g. "aBc dEf" -> "Abc def". 317def capitalize(s): 318 return upper(s[:1]) + lower(s[1:]) 319 320# Capitalize the words in a string, e.g. " aBc dEf " -> "Abc Def". 321# See also regsub.capwords(). 322def capwords(s, sep=None): 323 return join(map(capitalize, split(s, sep)), sep or ' ') 324 325# Construct a translation string 326_idmapL = None 327def maketrans(fromstr, tostr): 328 if len(fromstr) != len(tostr): 329 raise ValueError, "maketrans arguments must have same length" 330 global _idmapL 331 if not _idmapL: 332 _idmapL = map(None, _idmap) 333 L = _idmapL[:] 334 fromstr = map(ord, fromstr) 335 for i in range(len(fromstr)): 336 L[fromstr[i]] = tostr[i] 337 return joinfields(L, "") 338 339# Substring replacement (global) 340def replace(str, old, new, maxsplit=0): 341 return joinfields(splitfields(str, old, maxsplit), new) 342 343 344# Try importing optional built-in module "strop" -- if it exists, 345# it redefines some string operations that are 100-1000 times faster. 346# It also defines values for whitespace, lowercase and uppercase 347# that match <ctype.h>'s definitions. 348 349try: 350 from strop import * 351 letters = lowercase + uppercase 352except ImportError: 353 pass # Use the original, slow versions 354