email¶

The usability of theemail package in Python 3 has been mostly fixed bythe extensive efforts of R. David Murray. The problem was that emails aretypically read and stored in the form ofbytes rather thanstrtext, and they may contain multiple encodings within a single email. So, theemail package had to be extended to parse and generate email messages in bytesformat.

• New functionsmessage_from_bytes() andmessage_from_binary_file(), and new classesBytesFeedParser andBytesParserallow binary message data to be parsed into model objects.

• Given bytes input to the model,get_payload()will by default decode a message body that has aContent-Transfer-Encoding of8bit using the charsetspecified in the MIME headers and return the resulting string.

• Given bytes input to the model,Generator willconvert message bodies that have aContent-Transfer-Encoding of8bit to instead have a7bitContent-Transfer-Encoding.

  Headers with unencoded non-ASCII bytes are deemed to beRFC 2047-encodedusing theunknown-8bit character set.

• A new classBytesGenerator produces bytes as output,preserving any unchanged non-ASCII data that was present in the input used tobuild the model, including message bodies with aContent-Transfer-Encoding of8bit.

• ThesmtplibSMTP class now accepts a byte stringfor themsg argument to thesendmail() method,and a new method,send_message() accepts aMessage object and can optionally obtain thefrom_addr andto_addrs addresses directly from the object.

(Proposed and implemented by R. David Murray,bpo-4661 andbpo-10321.)

elementtree¶

Thexml.etree.ElementTree package and itsxml.etree.cElementTreecounterpart have been updated to version 1.3.

Several new and useful functions and methods have been added:

• xml.etree.ElementTree.fromstringlist() which builds an XML documentfrom a sequence of fragments

• xml.etree.ElementTree.register_namespace() for registering a globalnamespace prefix

• xml.etree.ElementTree.tostringlist() for string representationincluding all sublists

• xml.etree.ElementTree.Element.extend() for appending a sequence of zeroor more elements

• xml.etree.ElementTree.Element.iterfind() searches an element andsubelements

• xml.etree.ElementTree.Element.itertext() creates a text iterator overan element and its subelements

• xml.etree.ElementTree.TreeBuilder.end() closes the current element

• xml.etree.ElementTree.TreeBuilder.doctype() handles a doctypedeclaration

有兩個方法已被棄用：

• xml.etree.ElementTree.getchildren() uselist(elem) instead.

• xml.etree.ElementTree.getiterator() useElement.iter instead.

For details of the update, seeIntroducing ElementTreeon Fredrik Lundh's website.

（由 Florent Xicluna 和 Fredrik Lundh 貢獻，bpo-6472。）

functools¶

• Thefunctools module includes a new decorator for caching functioncalls.functools.lru_cache() can save repeated queries to an externalresource whenever the results are expected to be the same.

  For example, adding a caching decorator to a database query function can savedatabase accesses for popular searches:

  >>>importfunctools>>>@functools.lru_cache(maxsize=300)...defget_phone_number(name):...c=conn.cursor()...c.execute('SELECT phonenumber FROM phonelist WHERE name=?',(name,))...returnc.fetchone()[0]

  >>>fornameinuser_requests:...get_phone_number(name)# cached lookup

  To help with choosing an effective cache size, the wrapped function isinstrumented for tracking cache statistics:

  >>>get_phone_number.cache_info()CacheInfo(hits=4805, misses=980, maxsize=300, currsize=300)

  If the phonelist table gets updated, the outdated contents of the cache can becleared with:

  >>>get_phone_number.cache_clear()

  (Contributed by Raymond Hettinger and incorporating design ideas from JimBaker, Miki Tebeka, and Nick Coghlan; seerecipe 498245,recipe 577479,bpo-10586, andbpo-10593.)

• Thefunctools.wraps() decorator now adds a__wrapped__ attributepointing to the original callable function. This allows wrapped functions tobe introspected. It also copies__annotations__ ifdefined. And now it also gracefully skips over missing attributes such as__doc__ whichmight not be defined for the wrapped callable.

  In the above example, the cache can be removed by recovering the originalfunction:

  >>>get_phone_number=get_phone_number.__wrapped__# uncached function

  (By Nick Coghlan and Terrence Cole;bpo-9567,bpo-3445, andbpo-8814.)

• To help write classes with rich comparison methods, a new decoratorfunctools.total_ordering() will use existing equality and inequalitymethods to fill in the remaining methods.

  For example, supplying__eq__ and__lt__ will enabletotal_ordering() to fill-in__le__,__gt__ and__ge__:

  @total_orderingclassStudent:def__eq__(self,other):return((self.lastname.lower(),self.firstname.lower())==(other.lastname.lower(),other.firstname.lower()))def__lt__(self,other):return((self.lastname.lower(),self.firstname.lower())<(other.lastname.lower(),other.firstname.lower()))

  With thetotal_ordering decorator, the remaining comparison methodsare filled in automatically.

  （由 Raymond Hettinger 貢獻。）

• To aid in porting programs from Python 2, thefunctools.cmp_to_key()function converts an old-style comparison function tomodernkey function:

  >>># locale-aware sort order>>>sorted(iterable,key=cmp_to_key(locale.strcoll))

  For sorting examples and a brief sorting tutorial, see theSorting HowTo tutorial.

  （由 Raymond Hettinger 貢獻。）

itertools¶

• Theitertools module has a newaccumulate() functionmodeled on APL'sscan operator and Numpy'saccumulate function:

  >>>fromitertoolsimportaccumulate>>>list(accumulate([8,2,50]))[8, 10, 60]

  >>>prob_dist=[0.1,0.4,0.2,0.3]>>>list(accumulate(prob_dist))# cumulative probability distribution[0.1, 0.5, 0.7, 1.0]

  For an example usingaccumulate(), see theexamples forthe random module.

  （由 Raymond Hettinger 貢獻，並結合了 Mark Dickinson 的設計建議。）

collections¶

• Thecollections.Counter class now has two forms of in-placesubtraction, the existing-= operator forsaturating subtraction and the newsubtract() method for regular subtraction. Theformer is suitable formultisetswhich only have positive counts, and the latter is more suitable for use casesthat allow negative counts:

  >>>fromcollectionsimportCounter>>>tally=Counter(dogs=5,cats=3)>>>tally-=Counter(dogs=2,cats=8)# saturating subtraction>>>tallyCounter({'dogs': 3})

  >>>tally=Counter(dogs=5,cats=3)>>>tally.subtract(dogs=2,cats=8)# regular subtraction>>>tallyCounter({'dogs': 3, 'cats': -5})

  （由 Raymond Hettinger 貢獻。）

• Thecollections.OrderedDict class has a new methodmove_to_end() which takes an existing key andmoves it to either the first or last position in the ordered sequence.

  The default is to move an item to the last position. This is equivalent ofrenewing an entry withod[k]=od.pop(k).

  A fast move-to-end operation is useful for resequencing entries. For example,an ordered dictionary can be used to track order of access by aging entriesfrom the oldest to the most recently accessed.

  >>>fromcollectionsimportOrderedDict>>>d=OrderedDict.fromkeys(['a','b','X','d','e'])>>>list(d)['a', 'b', 'X', 'd', 'e']>>>d.move_to_end('X')>>>list(d)['a', 'b', 'd', 'e', 'X']

  （由 Raymond Hettinger 貢獻。）

• Thecollections.deque class grew two new methodscount() andreverse() thatmake them more substitutable forlist objects:

  >>>fromcollectionsimportdeque>>>d=deque('simsalabim')>>>d.count('s')2>>>d.reverse()>>>ddeque(['m', 'i', 'b', 'a', 'l', 'a', 's', 'm', 'i', 's'])

  （由 Raymond Hettinger 貢獻。）

threading¶

Thethreading module has a newBarriersynchronization class for making multiple threads wait until all of them havereached a common barrier point. Barriers are useful for making sure that a taskwith multiple preconditions does not run until all of the predecessor tasks arecomplete.

Barriers can work with an arbitrary number of threads. This is a generalizationof aRendezvous whichis defined for only two threads.

Implemented as a two-phase cyclic barrier,Barrier objectsare suitable for use in loops. The separatefilling anddraining phasesassure that all threads get released (drained) before any one of them can loopback and re-enter the barrier. The barrier fully resets after each cycle.

Example of using barriers:

fromthreadingimportBarrier,Threaddefget_votes(site):ballots=conduct_election(site)all_polls_closed.wait()# do not count until all polls are closedtotals=summarize(ballots)publish(site,totals)all_polls_closed=Barrier(len(sites))forsiteinsites:Thread(target=get_votes,args=(site,)).start()

In this example, the barrier enforces a rule that votes cannot be counted at anypolling site until all polls are closed. Notice how a solution with a barrieris similar to one withthreading.Thread.join(), but the threads stay aliveand continue to do work (summarizing ballots) after the barrier point iscrossed.

If any of the predecessor tasks can hang or be delayed, a barrier can be createdwith an optionaltimeout parameter. Then if the timeout period elapses beforeall the predecessor tasks reach the barrier point, all waiting threads arereleased and aBrokenBarrierError exception is raised:

defget_votes(site):ballots=conduct_election(site)try:all_polls_closed.wait(timeout=midnight-time.now())exceptBrokenBarrierError:lockbox=seal_ballots(ballots)queue.put(lockbox)else:totals=summarize(ballots)publish(site,totals)

In this example, the barrier enforces a more robust rule. If some electionsites do not finish before midnight, the barrier times-out and the ballots aresealed and deposited in a queue for later handling.

SeeBarrier Synchronization Patternsfor more examples of how barriers can be used in parallel computing. Also, there isa simple but thorough explanation of barriers inThe Little Book of Semaphores,section 3.6.

（在bpo-8777 中由 Kristján Valur Jónsson 貢獻、Jeffrey Yasskin 進行 API 審查。）

datetime 和 time¶

• Thedatetime module has a new typetimezone thatimplements thetzinfo interface by returning a fixed UTCoffset and timezone name. This makes it easier to create timezone-awaredatetime objects:

  >>>fromdatetimeimportdatetime,timezone>>>datetime.now(timezone.utc)datetime.datetime(2010, 12, 8, 21, 4, 2, 923754, tzinfo=datetime.timezone.utc)>>>datetime.strptime("01/01/2000 12:00 +0000","%m/%d/%Y %H:%M %z")datetime.datetime(2000, 1, 1, 12, 0, tzinfo=datetime.timezone.utc)

• Also,timedelta objects can now be multiplied byfloat and divided byfloat andint objects.Andtimedelta objects can now divide one another.

• Thedatetime.date.strftime() method is no longer restricted to yearsafter 1900. The new supported year range is from 1000 to 9999 inclusive.

• Whenever a two-digit year is used in a time tuple, the interpretation has beengoverned bytime.accept2dyear. The default isTrue which means thatfor a two-digit year, the century is guessed according to the POSIX rulesgoverning the%y strptime format.

  Starting with Py3.2, use of the century guessing heuristic will emit aDeprecationWarning. Instead, it is recommended thattime.accept2dyear be set toFalse so that large date rangescan be used without guesswork:

  >>>importtime,warnings>>>warnings.resetwarnings()# remove the default warning filters>>>time.accept2dyear=True# guess whether 11 means 11 or 2011>>>time.asctime((11,1,1,12,34,56,4,1,0))Warning (from warnings module):  ...DeprecationWarning: Century info guessed for a 2-digit year.'Fri Jan  1 12:34:56 2011'>>>time.accept2dyear=False# use the full range of allowable dates>>>time.asctime((11,1,1,12,34,56,4,1,0))'Fri Jan  1 12:34:56 11'

  Several functions now have significantly expanded date ranges. Whentime.accept2dyear is false, thetime.asctime() function willaccept any year that fits in a C int, while thetime.mktime() andtime.strftime() functions will accept the full range supported by thecorresponding operating system functions.

（由 Alexander Belopolsky 和 Victor Stinner 在bpo-1289118、bpo-5094、bpo-6641、bpo-2706、bpo-1777412、bpo-8013 和bpo-10827 中貢獻。）

math¶

Themath module has been updated with six new functions inspired by theC99 standard.

Theisfinite() function provides a reliable and fast way to detectspecial values. It returnsTrue for regular numbers andFalse forNan orInfinity:

>>>frommathimportisfinite>>>[isfinite(x)forxin(123,4.56,float('Nan'),float('Inf'))][True, True, False, False]

Theexpm1() function computese**x-1 for small values ofxwithout incurring the loss of precision that usually accompanies the subtractionof nearly equal quantities:

>>>frommathimportexpm1>>>expm1(0.013671875)# more accurate way to compute e**x-1 for a small x0.013765762467652909

Theerf() function computes a probability integral orGaussianerror function. Thecomplementary error function,erfc(), is1-erf(x):

>>>frommathimporterf,erfc,sqrt>>>erf(1.0/sqrt(2.0))# portion of normal distribution within 1 standard deviation0.682689492137086>>>erfc(1.0/sqrt(2.0))# portion of normal distribution outside 1 standard deviation0.31731050786291404>>>erf(1.0/sqrt(2.0))+erfc(1.0/sqrt(2.0))1.0

Thegamma() function is a continuous extension of the factorialfunction. Seehttps://en.wikipedia.org/wiki/Gamma_function for details. Becausethe function is related to factorials, it grows large even for small values ofx, so there is also algamma() function for computing the naturallogarithm of the gamma function:

>>>frommathimportgamma,lgamma>>>gamma(7.0)# six factorial720.0>>>lgamma(801.0)# log(800 factorial)4551.950730698041

（由 Mark Dickinson 貢獻。）

abc¶

Theabc module now supportsabstractclassmethod() andabstractstaticmethod().

These tools make it possible to define anabstract base class thatrequires a particularclassmethod() orstaticmethod() to beimplemented:

classTemperature(metaclass=abc.ABCMeta):@abc.abstractclassmethoddeffrom_fahrenheit(cls,t):...@abc.abstractclassmethoddeffrom_celsius(cls,t):...

(Patch submitted by Daniel Urban;bpo-5867.)

io¶

Theio.BytesIO has a new method,getbuffer(), whichprovides functionality similar tomemoryview(). It creates an editableview of the data without making a copy. The buffer's random access and supportfor slice notation are well-suited to in-place editing:

>>>REC_LEN,LOC_START,LOC_LEN=34,7,11>>>defchange_location(buffer,record_number,location):...start=record_number*REC_LEN+LOC_START...buffer[start:start+LOC_LEN]=location>>>importio>>>byte_stream=io.BytesIO(...b'G3805  storeroom  Main chassis    '...b'X7899  shipping   Reserve cog     '...b'L6988  receiving  Primary sprocket'...)>>>buffer=byte_stream.getbuffer()>>>change_location(buffer,1,b'warehouse  ')>>>change_location(buffer,0,b'showroom   ')>>>print(byte_stream.getvalue())b'G3805  showroom   Main chassis    'b'X7899  warehouse  Reserve cog     'b'L6988  receiving  Primary sprocket'

（由 Antoine Pitrou 在bpo-5506 中貢獻。）

reprlib¶

When writing a__repr__() method for a custom container, it is easy toforget to handle the case where a member refers back to the container itself.Python's builtin objects such aslist andset handleself-reference by displaying "..." in the recursive part of the representationstring.

To help write such__repr__() methods, thereprlib module has a newdecorator,recursive_repr(), for detecting recursive calls to__repr__() and substituting a placeholder string instead:

>>>classMyList(list):...@recursive_repr()...def__repr__(self):...return'<'+'|'.join(map(repr,self))+'>'...>>>m=MyList('abc')>>>m.append(m)>>>m.append('x')>>>print(m)<'a'|'b'|'c'|...|'x'>

（由 Raymond Hettinger 在bpo-9826 和bpo-9840 中貢獻。）

logging¶

In addition to dictionary-based configuration described above, thelogging package has many other improvements.

The logging documentation has been augmented by abasic tutorial, anadvanced tutorial, and acookbook oflogging recipes. These documents are the fastest way to learn about logging.

Thelogging.basicConfig() set-up function gained astyle argument tosupport three different types of string formatting. It defaults to "%" fortraditional %-formatting, can be set to "{" for the newstr.format() style, orcan be set to "$" for the shell-style formatting provided bystring.Template. The following three configurations are equivalent:

>>>fromloggingimportbasicConfig>>>basicConfig(style='%',format="%(name)s ->%(levelname)s:%(message)s")>>>basicConfig(style='{',format="{name} ->{levelname}{message}")>>>basicConfig(style='$',format="$name -> $levelname: $message")

If no configuration is set-up before a logging event occurs, there is now adefault configuration using aStreamHandler directed tosys.stderr for events ofWARNING level or higher. Formerly, anevent occurring before a configuration was set-up would either raise anexception or silently drop the event depending on the value oflogging.raiseExceptions. The new default handler is stored inlogging.lastResort.

The use of filters has been simplified. Instead of creating aFilter object, the predicate can be any Python callable thatreturnsTrue orFalse.

There were a number of other improvements that add flexibility and simplifyconfiguration. See the module documentation for a full listing of changes inPython 3.2.

csv¶

Thecsv module now supports a new dialect,unix_dialect,which applies quoting for all fields and a traditional Unix style with'\n' asthe line terminator. The registered dialect name isunix.

Thecsv.DictWriter has a new method,writeheader() for writing-out an initial row to documentthe field names:

>>>importcsv,sys>>>w=csv.DictWriter(sys.stdout,['name','dept'],dialect='unix')>>>w.writeheader()"name","dept">>>w.writerows([...{'name':'tom','dept':'accounting'},...{'name':'susan','dept':'Salesl'}])"tom","accounting""susan","sales"

(New dialect suggested by Jay Talbot inbpo-5975, and the new methodsuggested by Ed Abraham inbpo-1537721.)

contextlib¶

There is a new and slightly mind-blowing toolContextDecorator that is helpful for creating acontext manager that does double duty as a function decorator.

As a convenience, this new functionality is used bycontextmanager() so that no extra effort is needed to supportboth roles.

The basic idea is that both context managers and function decorators can be usedfor pre-action and post-action wrappers. Context managers wrap a group ofstatements using awith statement, and function decorators wrap agroup of statements enclosed in a function. So, occasionally there is a need towrite a pre-action or post-action wrapper that can be used in either role.

For example, it is sometimes useful to wrap functions or groups of statementswith a logger that can track the time of entry and time of exit. Rather thanwriting both a function decorator and a context manager for the task, thecontextmanager() provides both capabilities in a singledefinition:

fromcontextlibimportcontextmanagerimportlogginglogging.basicConfig(level=logging.INFO)@contextmanagerdeftrack_entry_and_exit(name):logging.info('Entering:%s',name)yieldlogging.info('Exiting:%s',name)

Formerly, this would have only been usable as a context manager:

withtrack_entry_and_exit('widget loader'):print('Some time consuming activity goes here')load_widget()

Now, it can be used as a decorator as well:

@track_entry_and_exit('widget loader')defactivity():print('Some time consuming activity goes here')load_widget()

Trying to fulfill two roles at once places some limitations on the technique.Context managers normally have the flexibility to return an argument usable byawith statement, but there is no parallel for function decorators.

In the above example, there is not a clean way for thetrack_entry_and_exitcontext manager to return a logging instance for use in the body of enclosedstatements.

（由 Michael Foord 在bpo-9110 中貢獻。）

decimal and fractions¶

Mark Dickinson crafted an elegant and efficient scheme for assuring thatdifferent numeric datatypes will have the same hash value whenever their actualvalues are equal (bpo-8188):

asserthash(Fraction(3,2))==hash(1.5)== \hash(Decimal("1.5"))==hash(complex(1.5,0))

Some of the hashing details are exposed through a new attribute,sys.hash_info, which describes the bit width of the hash value, theprime modulus, the hash values forinfinity andnan, and the multiplierused for the imaginary part of a number:

>>>sys.hash_infosys.hash_info(width=64, modulus=2305843009213693951, inf=314159, nan=0, imag=1000003)

An early decision to limit the interoperability of various numeric types hasbeen relaxed. It is still unsupported (and ill-advised) to have implicitmixing in arithmetic expressions such asDecimal('1.1')+float('1.1')because the latter loses information in the process of constructing the binaryfloat. However, since existing floating-point value can be converted losslesslyto either a decimal or rational representation, it makes sense to add them tothe constructor and to support mixed-type comparisons.

• Thedecimal.Decimal constructor now acceptsfloat objectsdirectly so there in no longer a need to use thefrom_float()method (bpo-8257).

• Mixed type comparisons are now fully supported so thatDecimal objects can be directly compared withfloatandfractions.Fraction (bpo-2531 andbpo-8188).

Similar changes were made tofractions.Fraction so that thefrom_float() andfrom_decimal()methods are no longer needed (bpo-8294):

>>>fromdecimalimportDecimal>>>fromfractionsimportFraction>>>Decimal(1.1)Decimal('1.100000000000000088817841970012523233890533447265625')>>>Fraction(1.1)Fraction(2476979795053773, 2251799813685248)

Another useful change for thedecimal module is that theContext.clamp attribute is now public. This is useful in creatingcontexts that correspond to the decimal interchange formats specified in IEEE754 (seebpo-8540).

（由 Mark Dickinson 和 Raymond Hettinger 貢獻。）

ftp¶

Theftplib.FTP class now supports the context management protocol tounconditionally consumesocket.error exceptions and to close the FTPconnection when done:

>>>fromftplibimportFTP>>>withFTP("ftp1.at.proftpd.org")asftp:        ftp.login()        ftp.dir()'230 Anonymous login ok, restrictions apply.'dr-xr-xr-x   9 ftp      ftp           154 May  6 10:43 .dr-xr-xr-x   9 ftp      ftp           154 May  6 10:43 ..dr-xr-xr-x   5 ftp      ftp          4096 May  6 10:43 CentOSdr-xr-xr-x   3 ftp      ftp            18 Jul 10  2008 Fedora

Other file-like objects such asmmap.mmap andfileinput.input()also grew auto-closing context managers:

withfileinput.input(files=('log1.txt','log2.txt'))asf:forlineinf:process(line)

（由 Tarek Ziadé 和 Giampaolo Rodolà 在bpo-4972 中貢獻，由 Georg Brandl 在bpo-8046 和bpo-1286 中貢獻。）

TheFTP_TLS class now accepts acontext parameter, which is assl.SSLContext object allowing bundling SSL configuration options,certificates and private keys into a single (potentially long-lived) structure.

（由 Giampaolo Rodolà 貢獻；bpo-8806。）

popen¶

Theos.popen() andsubprocess.Popen() functions now supportwith statements for auto-closing of the file descriptors.

（由 Antoine Pitrou 和 Brian Curtin 在bpo-7461 和bpo-10554 中貢獻。）

select¶

Theselect module now exposes a new, constant attribute,PIPE_BUF, which gives the minimum number of bytes which areguaranteed not to block whenselect.select() says a pipe is readyfor writing.

>>>importselect>>>select.PIPE_BUF512

(Available on Unix systems. Patch by Sébastien Sablé inbpo-9862)

gzip 和 zipfile¶

gzip.GzipFile now implements theio.BufferedIOBaseabstract base class (except fortruncate()). It also has apeek() method and supports unseekable as well aszero-padded file objects.

Thegzip module also gains thecompress() anddecompress() functions for easier in-memory compression anddecompression. Keep in mind that text needs to be encoded asbytesbefore compressing and decompressing:

>>>importgzip>>>s='Three shall be the number thou shalt count, '>>>s+='and the number of the counting shall be three'>>>b=s.encode()# convert to utf-8>>>len(b)89>>>c=gzip.compress(b)>>>len(c)77>>>gzip.decompress(c).decode()[:42]# decompress and convert to text'Three shall be the number thou shalt count'

（由 Anand B. Pillai 在bpo-3488 中貢獻；由 Antoine Pitrou、Nir Aides 和 Brian Curtin 在bpo-9962、bpo-1675951、bpo-7471 和bpo-2846 中貢獻。）

Also, thezipfile.ZipExtFile class was reworked internally to representfiles stored inside an archive. The new implementation is significantly fasterand can be wrapped in anio.BufferedReader object for more speedups. Italso solves an issue where interleaved calls toread andreadline gave thewrong results.

(Patch submitted by Nir Aides inbpo-7610.)

tarfile¶

TheTarFile class can now be used as a context manager. Inaddition, itsadd() method has a new option,filter,that controls which files are added to the archive and allows the file metadatato be edited.

The newfilter option replaces the older, less flexibleexclude parameterwhich is now deprecated. If specified, the optionalfilter parameter needs tobe akeyword argument. The user-supplied filter function accepts aTarInfo object and returns an updatedTarInfo object, or if it wants the file to be excluded, thefunction can returnNone:

>>>importtarfile,glob>>>defmyfilter(tarinfo):...iftarinfo.isfile():# only save real files...tarinfo.uname='monty'# redact the user name...returntarinfo>>>withtarfile.open(name='myarchive.tar.gz',mode='w:gz')astf:...forfilenameinglob.glob('*.txt'):...tf.add(filename,filter=myfilter)...tf.list()-rw-r--r-- monty/501        902 2011-01-26 17:59:11 annotations.txt-rw-r--r-- monty/501        123 2011-01-26 17:59:11 general_questions.txt-rw-r--r-- monty/501       3514 2011-01-26 17:59:11 prion.txt-rw-r--r-- monty/501        124 2011-01-26 17:59:11 py_todo.txt-rw-r--r-- monty/501       1399 2011-01-26 17:59:11 semaphore_notes.txt

（由 Tarek Ziadé 提議並由 Lars Gustäbel 實作於bpo-6856。）

hashlib¶

Thehashlib module has two new constant attributes listing the hashingalgorithms guaranteed to be present in all implementations and those availableon the current implementation:

>>>importhashlib>>>hashlib.algorithms_guaranteed{'sha1', 'sha224', 'sha384', 'sha256', 'sha512', 'md5'}>>>hashlib.algorithms_available{'md2', 'SHA256', 'SHA512', 'dsaWithSHA', 'mdc2', 'SHA224', 'MD4', 'sha256','sha512', 'ripemd160', 'SHA1', 'MDC2', 'SHA', 'SHA384', 'MD2','ecdsa-with-SHA1','md4', 'md5', 'sha1', 'DSA-SHA', 'sha224','dsaEncryption', 'DSA', 'RIPEMD160', 'sha', 'MD5', 'sha384'}

（由 Carl Chenet 在bpo-7418 中提議。）

ast¶

Theast module has a wonderful a general-purpose tool for safelyevaluating expression strings using the Python literalsyntax. Theast.literal_eval() function serves as a secure alternative tothe builtineval() function which is easily abused. Python 3.2 addsbytes andset literals to the list of supported types:strings, bytes, numbers, tuples, lists, dicts, sets, booleans, andNone.

>>>fromastimportliteral_eval>>>request="{'req': 3, 'func': 'pow', 'args': (2, 0.5)}">>>literal_eval(request){'args': (2, 0.5), 'req': 3, 'func': 'pow'}>>>request="os.system('do something harmful')">>>literal_eval(request)Traceback (most recent call last):...ValueError:malformed node or string: <_ast.Call object at 0x101739a10>

（由 Benjamin Peterson 和 Georg Brandl 實作。）

os¶

Different operating systems use various encodings for filenames and environmentvariables. Theos module provides two new functions,fsencode() andfsdecode(), for encoding and decodingfilenames:

>>>importos>>>filename='Sehenswürdigkeiten'>>>os.fsencode(filename)b'Sehensw\xc3\xbcrdigkeiten'

Some operating systems allow direct access to encoded bytes in theenvironment. If so, theos.supports_bytes_environ constant will betrue.

For direct access to encoded environment variables (if available),use the newos.getenvb() function or useos.environbwhich is a bytes version ofos.environ.

（由 Victor Stinner 貢獻。）

shutil¶

Theshutil.copytree() function has two new options:

• ignore_dangling_symlinks: whensymlinks=False so that the functioncopies a file pointed to by a symlink, not the symlink itself. This optionwill silence the error raised if the file doesn't exist.

• copy_function: is a callable that will be used to copy files.shutil.copy2() is used by default.

（由 Tarek Ziadé 貢獻。）

In addition, theshutil module now supportsarchiving operations for zipfiles, uncompressed tarfiles, gzipped tarfiles,and bzipped tarfiles. And there are functions for registering additionalarchiving file formats (such as xz compressed tarfiles or custom formats).

The principal functions aremake_archive() andunpack_archive(). By default, both operate on the currentdirectory (which can be set byos.chdir()) and on any sub-directories.The archive filename needs to be specified with a full pathname. The archivingstep is non-destructive (the original files are left unchanged).

>>>importshutil,pprint>>>os.chdir('mydata')# change to the source directory>>>f=shutil.make_archive('/var/backup/mydata',...'zip')# archive the current directory>>>f# show the name of archive'/var/backup/mydata.zip'>>>os.chdir('tmp')# change to an unpacking>>>shutil.unpack_archive('/var/backup/mydata.zip')# recover the data>>>pprint.pprint(shutil.get_archive_formats())# display known formats[('bztar', "bzip2'ed tar-file"), ('gztar', "gzip'ed tar-file"), ('tar', 'uncompressed tar file'), ('zip', 'ZIP file')]>>>shutil.register_archive_format(# register a new archive format...name='xz',...function=xz.compress,# callable archiving function...extra_args=[('level',8)],# arguments to the function...description='xz compression'...)

（由 Tarek Ziadé 貢獻。）

sqlite3¶

Thesqlite3 module was updated to pysqlite version 2.6.0. It has two new capabilities.

• Thesqlite3.Connection.in_transit attribute is true if there is anactive transaction for uncommitted changes.

• Thesqlite3.Connection.enable_load_extension() andsqlite3.Connection.load_extension() methods allows you to load SQLiteextensions from ".so" files. One well-known extension is the fulltext-searchextension distributed with SQLite.

（由 R. David Murray 和 Shashwat Anand 貢獻；bpo-8845。）

html¶

A newhtml module was introduced with only a single function,escape(), which is used for escaping reserved characters from HTMLmarkup:

>>>importhtml>>>html.escape('x > 2 && x < 7')'x &gt; 2 &amp;&amp; x &lt; 7'

socket¶

Thesocket module has two new improvements.

• Socket objects now have adetach() method which putsthe socket into closed state without actually closing the underlying filedescriptor. The latter can then be reused for other purposes.(Added by Antoine Pitrou;bpo-8524.)

• socket.create_connection() now supports the context management protocolto unconditionally consumesocket.error exceptions and to close thesocket when done.(Contributed by Giampaolo Rodolà;bpo-9794.)

ssl¶

Thessl module added a number of features to satisfy common requirementsfor secure (encrypted, authenticated) internet connections:

• A new class,SSLContext, serves as a container for persistentSSL data, such as protocol settings, certificates, private keys, and variousother options. It includes awrap_socket() for creatingan SSL socket from an SSL context.

• A new function,ssl.match_hostname(), supports server identityverification for higher-level protocols by implementing the rules of HTTPS(fromRFC 2818) which are also suitable for other protocols.

• Thessl.wrap_socket() constructor function now takes aciphersargument. Theciphers string lists the allowed encryption algorithms usingthe format described in theOpenSSL documentation.

• When linked against recent versions of OpenSSL, thessl module nowsupports the Server Name Indication extension to the TLS protocol, allowingmultiple "virtual hosts" using different certificates on a single IP port.This extension is only supported in client mode, and is activated by passingtheserver_hostname argument tossl.SSLContext.wrap_socket().

• Various options have been added to thessl module, such asOP_NO_SSLv2 which disables the insecure and obsolete SSLv2protocol.

• The extension now loads all the OpenSSL ciphers and digest algorithms. Ifsome SSL certificates cannot be verified, they are reported as an "unknownalgorithm" error.

• The version of OpenSSL being used is now accessible using the moduleattributesssl.OPENSSL_VERSION (a string),ssl.OPENSSL_VERSION_INFO (a 5-tuple), andssl.OPENSSL_VERSION_NUMBER (an integer).

（由 Antoine Pitrou 在bpo-8850、bpo-1589、bpo-8322、bpo-5639、bpo-4870、bpo-8484 和bpo-8321 中貢獻。）

nntp¶

Thenntplib module has a revamped implementation with better bytes andtext semantics as well as more practical APIs. These improvements breakcompatibility with the nntplib version in Python 3.1, which was partlydysfunctional in itself.

Support for secure connections through both implicit (usingnntplib.NNTP_SSL) and explicit (usingnntplib.NNTP.starttls())TLS has also been added.

（由 Antoine Pitrou 在bpo-9360 中和 Andrew Vant 在bpo-1926 中貢獻。）

certificates¶

http.client.HTTPSConnection,urllib.request.HTTPSHandlerandurllib.request.urlopen() now take optional arguments to allow forserver certificate checking against a set of Certificate Authorities,as recommended in public uses of HTTPS.

(Added by Antoine Pitrou,bpo-9003.)

imaplib¶

Support for explicit TLS on standard IMAP4 connections has been added throughthe newimaplib.IMAP4.starttls method.

（由 Lorenzo M. Catucci 和 Antoine Pitrou 於bpo-4471 貢獻。）

http.client¶

There were a number of small API improvements in thehttp.client module.The old-style HTTP 0.9 simple responses are no longer supported and thestrictparameter is deprecated in all classes.

TheHTTPConnection andHTTPSConnection classes now have asource_addressparameter for a (host, port) tuple indicating where the HTTP connection is madefrom.

Support for certificate checking and HTTPS virtual hosts were added toHTTPSConnection.

Therequest() method on connection objectsallowed an optionalbody argument so that afile object could be usedto supply the content of the request. Conveniently, thebody argument nowalso accepts aniterable object so long as it includes an explicitContent-Length header. This extended interface is much more flexible thanbefore.

To establish an HTTPS connection through a proxy server, there is a newset_tunnel() method that sets the host andport for HTTP Connect tunneling.

To match the behavior ofhttp.server, the HTTP client library now alsoencodes headers with ISO-8859-1 (Latin-1) encoding. It was already doing thatfor incoming headers, so now the behavior is consistent for both incoming andoutgoing traffic. (See work by Armin Ronacher inbpo-10980.)

unittest¶

The unittest module has a number of improvements supporting test discovery forpackages, easier experimentation at the interactive prompt, new testcasemethods, improved diagnostic messages for test failures, and better methodnames.

• The command-line callpython-munittest can now accept file pathsinstead of module names for running specific tests (bpo-10620). The newtest discovery can find tests within packages, locating any test importablefrom the top-level directory. The top-level directory can be specified withthe-t option, a pattern for matching files with-p, and a directory tostart discovery with-s:

  $python-munittestdiscover-smy_proj_dir-p_test.py

  （由 Michael Foord 貢獻。）

• Experimentation at the interactive prompt is now easier because theunittest.TestCase class can now be instantiated withoutarguments:

  >>>fromunittestimportTestCase>>>TestCase().assertEqual(pow(2,3),8)

  （由 Michael Foord 貢獻。）

• Theunittest module has two new methods,assertWarns() andassertWarnsRegex() to verify that a given warning typeis triggered by the code under test:

  withself.assertWarns(DeprecationWarning):legacy_function('XYZ')

  （由 Antoine Pitrou 於bpo-9754 貢獻。）

  Another new method,assertCountEqual() is used tocompare two iterables to determine if their element counts are equal (whetherthe same elements are present with the same number of occurrences regardlessof order):

  deftest_anagram(self):self.assertCountEqual('algorithm','logarithm')

  （由 Raymond Hettinger 貢獻。）

• A principal feature of the unittest module is an effort to produce meaningfuldiagnostics when a test fails. When possible, the failure is recorded alongwith a diff of the output. This is especially helpful for analyzing log filesof failed test runs. However, since diffs can sometime be voluminous, there isa newmaxDiff attribute that sets maximum length ofdiffs displayed.

• In addition, the method names in the module have undergone a number of clean-ups.

  For example,assertRegex() is the new name forassertRegexpMatches() which was misnamed because thetest usesre.search(), notre.match(). Other methods usingregular expressions are now named using short form "Regex" in preference to"Regexp" -- this matches the names used in other unittest implementations,matches Python's old name for there module, and it has unambiguouscamel-casing.

  （由 Raymond Hettinger 貢獻，由 Ezio Melotti 實作。）

• To improve consistency, some long-standing method aliases are beingdeprecated in favor of the preferred names:

  舊名稱	Preferred Name
  assert_()	assertTrue()
  assertEquals()	assertEqual()
  assertNotEquals()	assertNotEqual()
  assertAlmostEquals()	assertAlmostEqual()
  assertNotAlmostEquals()	assertNotAlmostEqual()

  Likewise, theTestCase.fail* methods deprecated in Python 3.1 are expectedto be removed in Python 3.3.

  （由 Ezio Melotti 貢獻；bpo-9424。）

• TheassertDictContainsSubset() method was deprecatedbecause it was misimplemented with the arguments in the wrong order. Thiscreated hard-to-debug optical illusions where tests likeTestCase().assertDictContainsSubset({'a':1,'b':2},{'a':1}) would fail.

  （由 Raymond Hettinger 貢獻。）

random¶

The integer methods in therandom module now do a better job of producinguniform distributions. Previously, they computed selections withint(n*random()) which had a slight bias whenevern was not a power of two.Now, multiple selections are made from a range up to the next power of two and aselection is kept only when it falls within the range0<=x<n. Thefunctions and methods affected arerandrange(),randint(),choice(),shuffle() andsample().

（由 Raymond Hettinger 貢獻；bpo-9025。）

poplib¶

POP3_SSL class now accepts acontext parameter, which is assl.SSLContext object allowing bundling SSL configuration options,certificates and private keys into a single (potentially long-lived)structure.

（由 Giampaolo Rodolà 貢獻；bpo-8807。）

asyncore¶

asyncore.dispatcher now provides ahandle_accepted() methodreturning a(sock,addr) pair which is called when a connection has actuallybeen established with a new remote endpoint. This is supposed to be used as areplacement for oldhandle_accept() and avoidsthe user to callaccept() directly.

（由 Giampaolo Rodolà 貢獻；bpo-6706。）

tempfile¶

Thetempfile module has a new context manager,TemporaryDirectory which provides easy deterministiccleanup of temporary directories:

withtempfile.TemporaryDirectory()astmpdirname:print('created temporary dir:',tmpdirname)

（由 Neil Schedulenauer 和 Nick Coghlan 貢獻；bpo-5178。）

inspect¶

• Theinspect module has a new functiongetgeneratorstate() to easily identify the current state of agenerator-iterator:

  >>>frominspectimportgetgeneratorstate>>>defgen():...yield'demo'...>>>g=gen()>>>getgeneratorstate(g)'GEN_CREATED'>>>next(g)'demo'>>>getgeneratorstate(g)'GEN_SUSPENDED'>>>next(g,None)>>>getgeneratorstate(g)'GEN_CLOSED'

  （由 Rodolfo Eckhardt 和 Nick Coghlan 於bpo-10220 貢獻。）

• To support lookups without the possibility of activating a dynamic attribute,theinspect module has a new function,getattr_static().Unlikehasattr(), this is a true read-only search, guaranteed not tochange state while it is searching:

  >>>classA:...@property...deff(self):...print('Running')...return10...>>>a=A()>>>getattr(a,'f')Running10>>>inspect.getattr_static(a,'f')<property object at 0x1022bd788>

（由 Michael Foord 貢獻。）

pydoc¶

Thepydoc module now provides a much-improved web server interface, aswell as a new command-line option-b to automatically open a browser windowto display that server:

$pydoc3.2-b

（由 Ron Adam 貢獻；bpo-2001。）

dis¶

Thedis module gained two new functions for inspecting code,code_info() andshow_code(). Both provide detailed codeobject information for the supplied function, method, source code string or codeobject. The former returns a string and the latter prints it:

>>>importdis,random>>>dis.show_code(random.choice)Name:              choiceFilename:          /Library/Frameworks/Python.framework/Versions/3.2/lib/python3.2/random.pyArgument count:    2Kw-only arguments: 0Number of locals:  3Stack size:        11Flags:             OPTIMIZED, NEWLOCALS, NOFREEConstants:   0: 'Choose a random element from a non-empty sequence.'   1: 'Cannot choose from an empty sequence'Names:   0: _randbelow   1: len   2: ValueError   3: IndexErrorVariable names:   0: self   1: seq   2: i

In addition, thedis() function now accepts string argumentsso that the common idiomdis(compile(s,'','eval')) can be shortenedtodis(s):

>>>dis('3*x+1 if x%2==1 else x//2')  1           0 LOAD_NAME                0 (x)              3 LOAD_CONST               0 (2)              6 BINARY_MODULO              7 LOAD_CONST               1 (1)             10 COMPARE_OP               2 (==)             13 POP_JUMP_IF_FALSE       28             16 LOAD_CONST               2 (3)             19 LOAD_NAME                0 (x)             22 BINARY_MULTIPLY             23 LOAD_CONST               1 (1)             26 BINARY_ADD             27 RETURN_VALUE        >>   28 LOAD_NAME                0 (x)             31 LOAD_CONST               0 (2)             34 BINARY_FLOOR_DIVIDE             35 RETURN_VALUE

Taken together, these improvements make it easier to explore how CPython isimplemented and to see for yourself what the language syntax doesunder-the-hood.

（由 Nick Coghlan 在bpo-9147 中貢獻。）

dbm¶

All database modules now support theget() andsetdefault() methods.

（由 Ray Allen 在bpo-9523 中建議。）

ctypes¶

A new type,ctypes.c_ssize_t represents the Cssize_t datatype.

site¶

Thesite module has three new functions useful for reporting on thedetails of a given Python installation.

• getsitepackages() lists all global site-packages directories.

• getuserbase() reports on the user's base directory where data canbe stored.

• getusersitepackages() reveals the user-specific site-packagesdirectory path.

>>>importsite>>>site.getsitepackages()['/Library/Frameworks/Python.framework/Versions/3.2/lib/python3.2/site-packages', '/Library/Frameworks/Python.framework/Versions/3.2/lib/site-python', '/Library/Python/3.2/site-packages']>>>site.getuserbase()'/Users/raymondhettinger/Library/Python/3.2'>>>site.getusersitepackages()'/Users/raymondhettinger/Library/Python/3.2/lib/python/site-packages'

Conveniently, some of site's functionality is accessible directly from thecommand-line:

$python-msite--user-base/Users/raymondhettinger/.local$python-msite--user-site/Users/raymondhettinger/.local/lib/python3.2/site-packages

（由 Tarek Ziadé 在bpo-6693 中貢獻。）

sysconfig¶

The newsysconfig module makes it straightforward to discoverinstallation paths and configuration variables that vary across platforms andinstallations.

The module offers access simple access functions for platform and versioninformation:

• get_platform() returning values likelinux-i586 ormacosx-10.6-ppc.

• get_python_version() returns a Python version stringsuch as "3.2".

It also provides access to the paths and variables corresponding to one ofseven named schemes used bydistutils. Those includeposix_prefix,posix_home,posix_user,nt,nt_user,os2,os2_home:

• get_paths() makes a dictionary containing installation pathsfor the current installation scheme.

• get_config_vars() returns a dictionary of platform specificvariables.

There is also a convenient command-line interface:

C:\Python32>python -m sysconfigPlatform: "win32"Python version: "3.2"Current installation scheme: "nt"Paths:        data = "C:\Python32"        include = "C:\Python32\Include"        platinclude = "C:\Python32\Include"        platlib = "C:\Python32\Lib\site-packages"        platstdlib = "C:\Python32\Lib"        purelib = "C:\Python32\Lib\site-packages"        scripts = "C:\Python32\Scripts"        stdlib = "C:\Python32\Lib"Variables:        BINDIR = "C:\Python32"        BINLIBDEST = "C:\Python32\Lib"        EXE = ".exe"        INCLUDEPY = "C:\Python32\Include"        LIBDEST = "C:\Python32\Lib"        SO = ".pyd"        VERSION = "32"        abiflags = ""        base = "C:\Python32"        exec_prefix = "C:\Python32"        platbase = "C:\Python32"        prefix = "C:\Python32"        projectbase = "C:\Python32"        py_version = "3.2"        py_version_nodot = "32"        py_version_short = "3.2"        srcdir = "C:\Python32"        userbase = "C:\Documents and Settings\Raymond\Application Data\Python"

(Moved out of Distutils by Tarek Ziadé.)

pdb¶

Thepdb debugger module gained a number of usability improvements:

• pdb.py now has a-c option that executes commands as given in a.pdbrc script file.

• A.pdbrc script file can containcontinue andnext commandsthat continue debugging.

• 現在Pdb 類別建構函式接受nosigint 引數。

• New commands:l(list),ll(longlist) andsource forlisting source code.

• New commands:display andundisplay for showing or hidingthe value of an expression if it has changed.

• New command:interact for starting an interactive interpreter containingthe global and local names found in the current scope.

• Breakpoints can be cleared by breakpoint number.

（由 Georg Brandl、Antonio Cuni 和 Ilya Sandler 貢獻。）

configparser¶

Theconfigparser module was modified to improve usability andpredictability of the default parser and its supported INI syntax. The oldConfigParser class was removed in favor ofSafeConfigParserwhich has in turn been renamed toConfigParser. Supportfor inline comments is now turned off by default and section or optionduplicates are not allowed in a single configuration source.

Config parsers gained a new API based on the mapping protocol:

>>>parser=ConfigParser()>>>parser.read_string("""...[DEFAULT]...location = upper left...visible = yes...editable = no...color = blue......[main]...title = Main Menu...color = green......[options]...title = Options...""")>>>parser['main']['color']'green'>>>parser['main']['editable']'no'>>>section=parser['options']>>>section['title']'Options'>>>section['title']='Options (editable:%(editable)s)'>>>section['title']'Options (editable: no)'

The new API is implemented on top of the classical API, so custom parsersubclasses should be able to use it without modifications.

The INI file structure accepted by config parsers can now be customized. Userscan specify alternative option/value delimiters and comment prefixes, change thename of theDEFAULT section or switch the interpolation syntax.

There is support for pluggable interpolation including an additional interpolationhandlerExtendedInterpolation:

>>>parser=ConfigParser(interpolation=ExtendedInterpolation())>>>parser.read_dict({'buildout':{'directory':'/home/ambv/zope9'},...'custom':{'prefix':'/usr/local'}})>>>parser.read_string("""...[buildout]...parts =...  zope9...  instance...find-links =...  ${buildout:directory}/downloads/dist......[zope9]...recipe = plone.recipe.zope9install...location = /opt/zope......[instance]...recipe = plone.recipe.zope9instance...zope9-location = ${zope9:location}...zope-conf = ${custom:prefix}/etc/zope.conf...""")>>>parser['buildout']['find-links']'\n/home/ambv/zope9/downloads/dist'>>>parser['instance']['zope-conf']'/usr/local/etc/zope.conf'>>>instance=parser['instance']>>>instance['zope-conf']'/usr/local/etc/zope.conf'>>>instance['zope9-location']'/opt/zope'

A number of smaller features were also introduced, like support for specifyingencoding in read operations, specifying fallback values for get-functions, orreading directly from dictionaries and strings.

(All changes contributed by Łukasz Langa.)

urllib.parse¶

A number of usability improvements were made for theurllib.parse module.

Theurlparse() function now supportsIPv6 addresses as described inRFC 2732:

>>>importurllib.parse>>>urllib.parse.urlparse('http://[dead:beef:cafe:5417:affe:8FA3:deaf:feed]/foo/')ParseResult(scheme='http',            netloc='[dead:beef:cafe:5417:affe:8FA3:deaf:feed]',            path='/foo/',            params='',            query='',            fragment='')

Theurldefrag() function now returns anamed tuple:

>>>r=urllib.parse.urldefrag('http://python.org/about/#target')>>>rDefragResult(url='http://python.org/about/', fragment='target')>>>r[0]'http://python.org/about/'>>>r.fragment'target'

And, theurlencode() function is now much more flexible,accepting either a string or bytes type for thequery argument. If it is astring, then thesafe,encoding, anderror parameters are sent toquote_plus() for encoding:

>>>urllib.parse.urlencode([...('type','telenovela'),...('name','¿Dónde Está Elisa?')],...encoding='latin-1')'type=telenovela&name=%BFD%F3nde+Est%E1+Elisa%3F'

As detailed inParsing ASCII Encoded Bytes, all theurllib.parsefunctions now accept ASCII-encoded byte strings as input, so long as they arenot mixed with regular strings. If ASCII-encoded byte strings are given asparameters, the return types will also be an ASCII-encoded byte strings:

>>>urllib.parse.urlparse(b'http://www.python.org:80/about/')ParseResultBytes(scheme=b'http', netloc=b'www.python.org:80',                 path=b'/about/', params=b'', query=b'', fragment=b'')

(Work by Nick Coghlan, Dan Mahn, and Senthil Kumaran inbpo-2987,bpo-5468, andbpo-9873.)

mailbox¶

Thanks to a concerted effort by R. David Murray, themailbox module hasbeen fixed for Python 3.2. The challenge was that mailbox had been originallydesigned with a text interface, but email messages are best represented withbytes because various parts of a message may have different encodings.

The solution harnessed theemail package's binary support for parsingarbitrary email messages. In addition, the solution required a number of APIchanges.

As expected, theadd() method formailbox.Mailbox objects now accepts binary input.

StringIO and text file input are deprecated. Also, string inputwill fail early if non-ASCII characters are used. Previously it would fail whenthe email was processed in a later step.

There is also support for binary output. Theget_file()method now returns a file in the binary mode (where it used to incorrectly setthe file to text-mode). There is also a newget_bytes()method that returns abytes representation of a message correspondingto a givenkey.

It is still possible to get non-binary output using the old API'sget_string() method, but that approachis not very useful. Instead, it is best to extract messages fromaMessage object or to load them from binary input.

(Contributed by R. David Murray, with efforts from Steffen Daode Nurpmeso and aninitial patch by Victor Stinner inbpo-9124.)

turtledemo¶

The demonstration code for theturtle module was moved from theDemodirectory to main library. It includes over a dozen sample scripts withlively displays. Being onsys.path, it can now be run directlyfrom the command-line:

$python-mturtledemo

(Moved from the Demo directory by Alexander Belopolsky inbpo-10199.)