1.. _tut-structures: 2 3*************** 4Data Structures 5*************** 6 7This chapter describes some things you've learned about already in more detail, 8and adds some new things as well. 9 10.. _tut-morelists: 11 12More on Lists 13============= 14 15The list data type has some more methods. Here are all of the methods of list 16objects: 17 18 19.. method:: list.append(x) 20 :noindex: 21 22 Add an item to the end of the list. Equivalent to ``a[len(a):] = [x]``. 23 24 25.. method:: list.extend(iterable) 26 :noindex: 27 28 Extend the list by appending all the items from the iterable. Equivalent to 29 ``a[len(a):] = iterable``. 30 31 32.. method:: list.insert(i, x) 33 :noindex: 34 35 Insert an item at a given position. The first argument is the index of the 36 element before which to insert, so ``a.insert(0, x)`` inserts at the front of 37 the list, and ``a.insert(len(a), x)`` is equivalent to ``a.append(x)``. 38 39 40.. method:: list.remove(x) 41 :noindex: 42 43 Remove the first item from the list whose value is equal to *x*. It raises a 44 :exc:`ValueError` if there is no such item. 45 46 47.. method:: list.pop([i]) 48 :noindex: 49 50 Remove the item at the given position in the list, and return it. If no index 51 is specified, ``a.pop()`` removes and returns the last item in the list. (The 52 square brackets around the *i* in the method signature denote that the parameter 53 is optional, not that you should type square brackets at that position. You 54 will see this notation frequently in the Python Library Reference.) 55 56 57.. method:: list.clear() 58 :noindex: 59 60 Remove all items from the list. Equivalent to ``del a[:]``. 61 62 63.. method:: list.index(x[, start[, end]]) 64 :noindex: 65 66 Return zero-based index in the list of the first item whose value is equal to *x*. 67 Raises a :exc:`ValueError` if there is no such item. 68 69 The optional arguments *start* and *end* are interpreted as in the slice 70 notation and are used to limit the search to a particular subsequence of 71 the list. The returned index is computed relative to the beginning of the full 72 sequence rather than the *start* argument. 73 74 75.. method:: list.count(x) 76 :noindex: 77 78 Return the number of times *x* appears in the list. 79 80 81.. method:: list.sort(*, key=None, reverse=False) 82 :noindex: 83 84 Sort the items of the list in place (the arguments can be used for sort 85 customization, see :func:`sorted` for their explanation). 86 87 88.. method:: list.reverse() 89 :noindex: 90 91 Reverse the elements of the list in place. 92 93 94.. method:: list.copy() 95 :noindex: 96 97 Return a shallow copy of the list. Equivalent to ``a[:]``. 98 99 100An example that uses most of the list methods:: 101 102 >>> fruits = ['orange', 'apple', 'pear', 'banana', 'kiwi', 'apple', 'banana'] 103 >>> fruits.count('apple') 104 2 105 >>> fruits.count('tangerine') 106 0 107 >>> fruits.index('banana') 108 3 109 >>> fruits.index('banana', 4) # Find next banana starting a position 4 110 6 111 >>> fruits.reverse() 112 >>> fruits 113 ['banana', 'apple', 'kiwi', 'banana', 'pear', 'apple', 'orange'] 114 >>> fruits.append('grape') 115 >>> fruits 116 ['banana', 'apple', 'kiwi', 'banana', 'pear', 'apple', 'orange', 'grape'] 117 >>> fruits.sort() 118 >>> fruits 119 ['apple', 'apple', 'banana', 'banana', 'grape', 'kiwi', 'orange', 'pear'] 120 >>> fruits.pop() 121 'pear' 122 123You might have noticed that methods like ``insert``, ``remove`` or ``sort`` that 124only modify the list have no return value printed -- they return the default 125``None``. [1]_ This is a design principle for all mutable data structures in 126Python. 127 128Another thing you might notice is that not all data can be sorted or 129compared. For instance, ``[None, 'hello', 10]`` doesn't sort because 130integers can't be compared to strings and *None* can't be compared to 131other types. Also, there are some types that don't have a defined 132ordering relation. For example, ``3+4j < 5+7j`` isn't a valid 133comparison. 134 135 136.. _tut-lists-as-stacks: 137 138Using Lists as Stacks 139--------------------- 140 141.. sectionauthor:: Ka-Ping Yee <ping@lfw.org> 142 143 144The list methods make it very easy to use a list as a stack, where the last 145element added is the first element retrieved ("last-in, first-out"). To add an 146item to the top of the stack, use :meth:`append`. To retrieve an item from the 147top of the stack, use :meth:`pop` without an explicit index. For example:: 148 149 >>> stack = [3, 4, 5] 150 >>> stack.append(6) 151 >>> stack.append(7) 152 >>> stack 153 [3, 4, 5, 6, 7] 154 >>> stack.pop() 155 7 156 >>> stack 157 [3, 4, 5, 6] 158 >>> stack.pop() 159 6 160 >>> stack.pop() 161 5 162 >>> stack 163 [3, 4] 164 165 166.. _tut-lists-as-queues: 167 168Using Lists as Queues 169--------------------- 170 171.. sectionauthor:: Ka-Ping Yee <ping@lfw.org> 172 173It is also possible to use a list as a queue, where the first element added is 174the first element retrieved ("first-in, first-out"); however, lists are not 175efficient for this purpose. While appends and pops from the end of list are 176fast, doing inserts or pops from the beginning of a list is slow (because all 177of the other elements have to be shifted by one). 178 179To implement a queue, use :class:`collections.deque` which was designed to 180have fast appends and pops from both ends. For example:: 181 182 >>> from collections import deque 183 >>> queue = deque(["Eric", "John", "Michael"]) 184 >>> queue.append("Terry") # Terry arrives 185 >>> queue.append("Graham") # Graham arrives 186 >>> queue.popleft() # The first to arrive now leaves 187 'Eric' 188 >>> queue.popleft() # The second to arrive now leaves 189 'John' 190 >>> queue # Remaining queue in order of arrival 191 deque(['Michael', 'Terry', 'Graham']) 192 193 194.. _tut-listcomps: 195 196List Comprehensions 197------------------- 198 199List comprehensions provide a concise way to create lists. 200Common applications are to make new lists where each element is the result of 201some operations applied to each member of another sequence or iterable, or to 202create a subsequence of those elements that satisfy a certain condition. 203 204For example, assume we want to create a list of squares, like:: 205 206 >>> squares = [] 207 >>> for x in range(10): 208 ... squares.append(x**2) 209 ... 210 >>> squares 211 [0, 1, 4, 9, 16, 25, 36, 49, 64, 81] 212 213Note that this creates (or overwrites) a variable named ``x`` that still exists 214after the loop completes. We can calculate the list of squares without any 215side effects using:: 216 217 squares = list(map(lambda x: x**2, range(10))) 218 219or, equivalently:: 220 221 squares = [x**2 for x in range(10)] 222 223which is more concise and readable. 224 225A list comprehension consists of brackets containing an expression followed 226by a :keyword:`!for` clause, then zero or more :keyword:`!for` or :keyword:`!if` 227clauses. The result will be a new list resulting from evaluating the expression 228in the context of the :keyword:`!for` and :keyword:`!if` clauses which follow it. 229For example, this listcomp combines the elements of two lists if they are not 230equal:: 231 232 >>> [(x, y) for x in [1,2,3] for y in [3,1,4] if x != y] 233 [(1, 3), (1, 4), (2, 3), (2, 1), (2, 4), (3, 1), (3, 4)] 234 235and it's equivalent to:: 236 237 >>> combs = [] 238 >>> for x in [1,2,3]: 239 ... for y in [3,1,4]: 240 ... if x != y: 241 ... combs.append((x, y)) 242 ... 243 >>> combs 244 [(1, 3), (1, 4), (2, 3), (2, 1), (2, 4), (3, 1), (3, 4)] 245 246Note how the order of the :keyword:`for` and :keyword:`if` statements is the 247same in both these snippets. 248 249If the expression is a tuple (e.g. the ``(x, y)`` in the previous example), 250it must be parenthesized. :: 251 252 >>> vec = [-4, -2, 0, 2, 4] 253 >>> # create a new list with the values doubled 254 >>> [x*2 for x in vec] 255 [-8, -4, 0, 4, 8] 256 >>> # filter the list to exclude negative numbers 257 >>> [x for x in vec if x >= 0] 258 [0, 2, 4] 259 >>> # apply a function to all the elements 260 >>> [abs(x) for x in vec] 261 [4, 2, 0, 2, 4] 262 >>> # call a method on each element 263 >>> freshfruit = [' banana', ' loganberry ', 'passion fruit '] 264 >>> [weapon.strip() for weapon in freshfruit] 265 ['banana', 'loganberry', 'passion fruit'] 266 >>> # create a list of 2-tuples like (number, square) 267 >>> [(x, x**2) for x in range(6)] 268 [(0, 0), (1, 1), (2, 4), (3, 9), (4, 16), (5, 25)] 269 >>> # the tuple must be parenthesized, otherwise an error is raised 270 >>> [x, x**2 for x in range(6)] 271 File "<stdin>", line 1, in <module> 272 [x, x**2 for x in range(6)] 273 ^ 274 SyntaxError: invalid syntax 275 >>> # flatten a list using a listcomp with two 'for' 276 >>> vec = [[1,2,3], [4,5,6], [7,8,9]] 277 >>> [num for elem in vec for num in elem] 278 [1, 2, 3, 4, 5, 6, 7, 8, 9] 279 280List comprehensions can contain complex expressions and nested functions:: 281 282 >>> from math import pi 283 >>> [str(round(pi, i)) for i in range(1, 6)] 284 ['3.1', '3.14', '3.142', '3.1416', '3.14159'] 285 286Nested List Comprehensions 287-------------------------- 288 289The initial expression in a list comprehension can be any arbitrary expression, 290including another list comprehension. 291 292Consider the following example of a 3x4 matrix implemented as a list of 2933 lists of length 4:: 294 295 >>> matrix = [ 296 ... [1, 2, 3, 4], 297 ... [5, 6, 7, 8], 298 ... [9, 10, 11, 12], 299 ... ] 300 301The following list comprehension will transpose rows and columns:: 302 303 >>> [[row[i] for row in matrix] for i in range(4)] 304 [[1, 5, 9], [2, 6, 10], [3, 7, 11], [4, 8, 12]] 305 306As we saw in the previous section, the nested listcomp is evaluated in 307the context of the :keyword:`for` that follows it, so this example is 308equivalent to:: 309 310 >>> transposed = [] 311 >>> for i in range(4): 312 ... transposed.append([row[i] for row in matrix]) 313 ... 314 >>> transposed 315 [[1, 5, 9], [2, 6, 10], [3, 7, 11], [4, 8, 12]] 316 317which, in turn, is the same as:: 318 319 >>> transposed = [] 320 >>> for i in range(4): 321 ... # the following 3 lines implement the nested listcomp 322 ... transposed_row = [] 323 ... for row in matrix: 324 ... transposed_row.append(row[i]) 325 ... transposed.append(transposed_row) 326 ... 327 >>> transposed 328 [[1, 5, 9], [2, 6, 10], [3, 7, 11], [4, 8, 12]] 329 330In the real world, you should prefer built-in functions to complex flow statements. 331The :func:`zip` function would do a great job for this use case:: 332 333 >>> list(zip(*matrix)) 334 [(1, 5, 9), (2, 6, 10), (3, 7, 11), (4, 8, 12)] 335 336See :ref:`tut-unpacking-arguments` for details on the asterisk in this line. 337 338.. _tut-del: 339 340The :keyword:`!del` statement 341============================= 342 343There is a way to remove an item from a list given its index instead of its 344value: the :keyword:`del` statement. This differs from the :meth:`pop` method 345which returns a value. The :keyword:`!del` statement can also be used to remove 346slices from a list or clear the entire list (which we did earlier by assignment 347of an empty list to the slice). For example:: 348 349 >>> a = [-1, 1, 66.25, 333, 333, 1234.5] 350 >>> del a[0] 351 >>> a 352 [1, 66.25, 333, 333, 1234.5] 353 >>> del a[2:4] 354 >>> a 355 [1, 66.25, 1234.5] 356 >>> del a[:] 357 >>> a 358 [] 359 360:keyword:`del` can also be used to delete entire variables:: 361 362 >>> del a 363 364Referencing the name ``a`` hereafter is an error (at least until another value 365is assigned to it). We'll find other uses for :keyword:`del` later. 366 367 368.. _tut-tuples: 369 370Tuples and Sequences 371==================== 372 373We saw that lists and strings have many common properties, such as indexing and 374slicing operations. They are two examples of *sequence* data types (see 375:ref:`typesseq`). Since Python is an evolving language, other sequence data 376types may be added. There is also another standard sequence data type: the 377*tuple*. 378 379A tuple consists of a number of values separated by commas, for instance:: 380 381 >>> t = 12345, 54321, 'hello!' 382 >>> t[0] 383 12345 384 >>> t 385 (12345, 54321, 'hello!') 386 >>> # Tuples may be nested: 387 ... u = t, (1, 2, 3, 4, 5) 388 >>> u 389 ((12345, 54321, 'hello!'), (1, 2, 3, 4, 5)) 390 >>> # Tuples are immutable: 391 ... t[0] = 88888 392 Traceback (most recent call last): 393 File "<stdin>", line 1, in <module> 394 TypeError: 'tuple' object does not support item assignment 395 >>> # but they can contain mutable objects: 396 ... v = ([1, 2, 3], [3, 2, 1]) 397 >>> v 398 ([1, 2, 3], [3, 2, 1]) 399 400 401As you see, on output tuples are always enclosed in parentheses, so that nested 402tuples are interpreted correctly; they may be input with or without surrounding 403parentheses, although often parentheses are necessary anyway (if the tuple is 404part of a larger expression). It is not possible to assign to the individual 405items of a tuple, however it is possible to create tuples which contain mutable 406objects, such as lists. 407 408Though tuples may seem similar to lists, they are often used in different 409situations and for different purposes. 410Tuples are :term:`immutable`, and usually contain a heterogeneous sequence of 411elements that are accessed via unpacking (see later in this section) or indexing 412(or even by attribute in the case of :func:`namedtuples <collections.namedtuple>`). 413Lists are :term:`mutable`, and their elements are usually homogeneous and are 414accessed by iterating over the list. 415 416A special problem is the construction of tuples containing 0 or 1 items: the 417syntax has some extra quirks to accommodate these. Empty tuples are constructed 418by an empty pair of parentheses; a tuple with one item is constructed by 419following a value with a comma (it is not sufficient to enclose a single value 420in parentheses). Ugly, but effective. For example:: 421 422 >>> empty = () 423 >>> singleton = 'hello', # <-- note trailing comma 424 >>> len(empty) 425 0 426 >>> len(singleton) 427 1 428 >>> singleton 429 ('hello',) 430 431The statement ``t = 12345, 54321, 'hello!'`` is an example of *tuple packing*: 432the values ``12345``, ``54321`` and ``'hello!'`` are packed together in a tuple. 433The reverse operation is also possible:: 434 435 >>> x, y, z = t 436 437This is called, appropriately enough, *sequence unpacking* and works for any 438sequence on the right-hand side. Sequence unpacking requires that there are as 439many variables on the left side of the equals sign as there are elements in the 440sequence. Note that multiple assignment is really just a combination of tuple 441packing and sequence unpacking. 442 443 444.. _tut-sets: 445 446Sets 447==== 448 449Python also includes a data type for *sets*. A set is an unordered collection 450with no duplicate elements. Basic uses include membership testing and 451eliminating duplicate entries. Set objects also support mathematical operations 452like union, intersection, difference, and symmetric difference. 453 454Curly braces or the :func:`set` function can be used to create sets. Note: to 455create an empty set you have to use ``set()``, not ``{}``; the latter creates an 456empty dictionary, a data structure that we discuss in the next section. 457 458Here is a brief demonstration:: 459 460 >>> basket = {'apple', 'orange', 'apple', 'pear', 'orange', 'banana'} 461 >>> print(basket) # show that duplicates have been removed 462 {'orange', 'banana', 'pear', 'apple'} 463 >>> 'orange' in basket # fast membership testing 464 True 465 >>> 'crabgrass' in basket 466 False 467 468 >>> # Demonstrate set operations on unique letters from two words 469 ... 470 >>> a = set('abracadabra') 471 >>> b = set('alacazam') 472 >>> a # unique letters in a 473 {'a', 'r', 'b', 'c', 'd'} 474 >>> a - b # letters in a but not in b 475 {'r', 'd', 'b'} 476 >>> a | b # letters in a or b or both 477 {'a', 'c', 'r', 'd', 'b', 'm', 'z', 'l'} 478 >>> a & b # letters in both a and b 479 {'a', 'c'} 480 >>> a ^ b # letters in a or b but not both 481 {'r', 'd', 'b', 'm', 'z', 'l'} 482 483Similarly to :ref:`list comprehensions <tut-listcomps>`, set comprehensions 484are also supported:: 485 486 >>> a = {x for x in 'abracadabra' if x not in 'abc'} 487 >>> a 488 {'r', 'd'} 489 490 491.. _tut-dictionaries: 492 493Dictionaries 494============ 495 496Another useful data type built into Python is the *dictionary* (see 497:ref:`typesmapping`). Dictionaries are sometimes found in other languages as 498"associative memories" or "associative arrays". Unlike sequences, which are 499indexed by a range of numbers, dictionaries are indexed by *keys*, which can be 500any immutable type; strings and numbers can always be keys. Tuples can be used 501as keys if they contain only strings, numbers, or tuples; if a tuple contains 502any mutable object either directly or indirectly, it cannot be used as a key. 503You can't use lists as keys, since lists can be modified in place using index 504assignments, slice assignments, or methods like :meth:`append` and 505:meth:`extend`. 506 507It is best to think of a dictionary as a set of *key: value* pairs, 508with the requirement that the keys are unique (within one dictionary). A pair of 509braces creates an empty dictionary: ``{}``. Placing a comma-separated list of 510key:value pairs within the braces adds initial key:value pairs to the 511dictionary; this is also the way dictionaries are written on output. 512 513The main operations on a dictionary are storing a value with some key and 514extracting the value given the key. It is also possible to delete a key:value 515pair with ``del``. If you store using a key that is already in use, the old 516value associated with that key is forgotten. It is an error to extract a value 517using a non-existent key. 518 519Performing ``list(d)`` on a dictionary returns a list of all the keys 520used in the dictionary, in insertion order (if you want it sorted, just use 521``sorted(d)`` instead). To check whether a single key is in the 522dictionary, use the :keyword:`in` keyword. 523 524Here is a small example using a dictionary:: 525 526 >>> tel = {'jack': 4098, 'sape': 4139} 527 >>> tel['guido'] = 4127 528 >>> tel 529 {'jack': 4098, 'sape': 4139, 'guido': 4127} 530 >>> tel['jack'] 531 4098 532 >>> del tel['sape'] 533 >>> tel['irv'] = 4127 534 >>> tel 535 {'jack': 4098, 'guido': 4127, 'irv': 4127} 536 >>> list(tel) 537 ['jack', 'guido', 'irv'] 538 >>> sorted(tel) 539 ['guido', 'irv', 'jack'] 540 >>> 'guido' in tel 541 True 542 >>> 'jack' not in tel 543 False 544 545The :func:`dict` constructor builds dictionaries directly from sequences of 546key-value pairs:: 547 548 >>> dict([('sape', 4139), ('guido', 4127), ('jack', 4098)]) 549 {'sape': 4139, 'guido': 4127, 'jack': 4098} 550 551In addition, dict comprehensions can be used to create dictionaries from 552arbitrary key and value expressions:: 553 554 >>> {x: x**2 for x in (2, 4, 6)} 555 {2: 4, 4: 16, 6: 36} 556 557When the keys are simple strings, it is sometimes easier to specify pairs using 558keyword arguments:: 559 560 >>> dict(sape=4139, guido=4127, jack=4098) 561 {'sape': 4139, 'guido': 4127, 'jack': 4098} 562 563 564.. _tut-loopidioms: 565 566Looping Techniques 567================== 568 569When looping through dictionaries, the key and corresponding value can be 570retrieved at the same time using the :meth:`items` method. :: 571 572 >>> knights = {'gallahad': 'the pure', 'robin': 'the brave'} 573 >>> for k, v in knights.items(): 574 ... print(k, v) 575 ... 576 gallahad the pure 577 robin the brave 578 579When looping through a sequence, the position index and corresponding value can 580be retrieved at the same time using the :func:`enumerate` function. :: 581 582 >>> for i, v in enumerate(['tic', 'tac', 'toe']): 583 ... print(i, v) 584 ... 585 0 tic 586 1 tac 587 2 toe 588 589To loop over two or more sequences at the same time, the entries can be paired 590with the :func:`zip` function. :: 591 592 >>> questions = ['name', 'quest', 'favorite color'] 593 >>> answers = ['lancelot', 'the holy grail', 'blue'] 594 >>> for q, a in zip(questions, answers): 595 ... print('What is your {0}? It is {1}.'.format(q, a)) 596 ... 597 What is your name? It is lancelot. 598 What is your quest? It is the holy grail. 599 What is your favorite color? It is blue. 600 601To loop over a sequence in reverse, first specify the sequence in a forward 602direction and then call the :func:`reversed` function. :: 603 604 >>> for i in reversed(range(1, 10, 2)): 605 ... print(i) 606 ... 607 9 608 7 609 5 610 3 611 1 612 613To loop over a sequence in sorted order, use the :func:`sorted` function which 614returns a new sorted list while leaving the source unaltered. :: 615 616 >>> basket = ['apple', 'orange', 'apple', 'pear', 'orange', 'banana'] 617 >>> for i in sorted(basket): 618 ... print(i) 619 ... 620 apple 621 apple 622 banana 623 orange 624 orange 625 pear 626 627Using :func:`set` on a sequence eliminates duplicate elements. The use of 628:func:`sorted` in combination with :func:`set` over a sequence is an idiomatic 629way to loop over unique elements of the sequence in sorted order. :: 630 631 >>> basket = ['apple', 'orange', 'apple', 'pear', 'orange', 'banana'] 632 >>> for f in sorted(set(basket)): 633 ... print(f) 634 ... 635 apple 636 banana 637 orange 638 pear 639 640It is sometimes tempting to change a list while you are looping over it; 641however, it is often simpler and safer to create a new list instead. :: 642 643 >>> import math 644 >>> raw_data = [56.2, float('NaN'), 51.7, 55.3, 52.5, float('NaN'), 47.8] 645 >>> filtered_data = [] 646 >>> for value in raw_data: 647 ... if not math.isnan(value): 648 ... filtered_data.append(value) 649 ... 650 >>> filtered_data 651 [56.2, 51.7, 55.3, 52.5, 47.8] 652 653 654.. _tut-conditions: 655 656More on Conditions 657================== 658 659The conditions used in ``while`` and ``if`` statements can contain any 660operators, not just comparisons. 661 662 663The comparison operators ``in`` and ``not in`` are membership tests that 664determine whether a value is in (or not in) a container. The operators ``is`` 665and ``is not`` compare whether two objects are really the same object. All 666comparison operators have the same priority, which is lower than that of all 667numerical operators. 668 669Comparisons can be chained. For example, ``a < b == c`` tests whether ``a`` is 670less than ``b`` and moreover ``b`` equals ``c``. 671 672Comparisons may be combined using the Boolean operators ``and`` and ``or``, and 673the outcome of a comparison (or of any other Boolean expression) may be negated 674with ``not``. These have lower priorities than comparison operators; between 675them, ``not`` has the highest priority and ``or`` the lowest, so that ``A and 676not B or C`` is equivalent to ``(A and (not B)) or C``. As always, parentheses 677can be used to express the desired composition. 678 679The Boolean operators ``and`` and ``or`` are so-called *short-circuit* 680operators: their arguments are evaluated from left to right, and evaluation 681stops as soon as the outcome is determined. For example, if ``A`` and ``C`` are 682true but ``B`` is false, ``A and B and C`` does not evaluate the expression 683``C``. When used as a general value and not as a Boolean, the return value of a 684short-circuit operator is the last evaluated argument. 685 686It is possible to assign the result of a comparison or other Boolean expression 687to a variable. For example, :: 688 689 >>> string1, string2, string3 = '', 'Trondheim', 'Hammer Dance' 690 >>> non_null = string1 or string2 or string3 691 >>> non_null 692 'Trondheim' 693 694Note that in Python, unlike C, assignment inside expressions must be done 695explicitly with the 696:ref:`walrus operator <why-can-t-i-use-an-assignment-in-an-expression>` ``:=``. 697This avoids a common class of problems encountered in C programs: typing ``=`` 698in an expression when ``==`` was intended. 699 700 701.. _tut-comparing: 702 703Comparing Sequences and Other Types 704=================================== 705Sequence objects typically may be compared to other objects with the same sequence 706type. The comparison uses *lexicographical* ordering: first the first two 707items are compared, and if they differ this determines the outcome of the 708comparison; if they are equal, the next two items are compared, and so on, until 709either sequence is exhausted. If two items to be compared are themselves 710sequences of the same type, the lexicographical comparison is carried out 711recursively. If all items of two sequences compare equal, the sequences are 712considered equal. If one sequence is an initial sub-sequence of the other, the 713shorter sequence is the smaller (lesser) one. Lexicographical ordering for 714strings uses the Unicode code point number to order individual characters. 715Some examples of comparisons between sequences of the same type:: 716 717 (1, 2, 3) < (1, 2, 4) 718 [1, 2, 3] < [1, 2, 4] 719 'ABC' < 'C' < 'Pascal' < 'Python' 720 (1, 2, 3, 4) < (1, 2, 4) 721 (1, 2) < (1, 2, -1) 722 (1, 2, 3) == (1.0, 2.0, 3.0) 723 (1, 2, ('aa', 'ab')) < (1, 2, ('abc', 'a'), 4) 724 725Note that comparing objects of different types with ``<`` or ``>`` is legal 726provided that the objects have appropriate comparison methods. For example, 727mixed numeric types are compared according to their numeric value, so 0 equals 7280.0, etc. Otherwise, rather than providing an arbitrary ordering, the 729interpreter will raise a :exc:`TypeError` exception. 730 731 732.. rubric:: Footnotes 733 734.. [1] Other languages may return the mutated object, which allows method 735 chaining, such as ``d->insert("a")->remove("b")->sort();``. 736