Mailman 3 August 2013 - docs

venv 3.3.2 docs - "create_directories" should be "ensure_directories"
by Michael Rand Oct. 6, 2013

Oct. 6, 2013

To whom it may concern, When subclassing venv.EnvBuilder and customizing create(), the docs say that one should use "self.create_directories(env_dir)". This call raises an AttributeError. When looking at the subclass, I don't see "create_directories" but instead I see the method "ensure_directories", which has the following docstring: "Create the directories for the environment". It seems like either the docs should be fixed to reflect "ensure_directories" as the appropriate method name or the method should be renamed "create_directories". Apologies if I'm mistaken on this ... this is my first doc bug report. Thanks, Mike Rand

2 1

Wrong toolkit for PythonCAD
by Helge Stenström Oct. 6, 2013

Oct. 6, 2013

According to http://docs.python.org/2/library/othergui.html, PythonCAD is written in pygtk, but according to the project page, http://sourceforge.net/projects/pythoncad/, it uses PyQt. Kind regardss, Helge Stenström

2 1

Bug in last httplib example in Python 2.7.5 docs in section 20.7.3
by Alex MacAulay Oct. 6, 2013

Oct. 6, 2013

On the following page in the Python 2.7.5 docs: http://docs.python.org/2/library/httplib.html In section 20.7.3, the last example has: >>> response = conn.getresponse()>>> print resp.status, response.reason However it should be: >>> response = conn.getresponse()>>> print response.status, response.reason Hope it helps! Alex.

2 1

binascii.b2a_base64 can convert data length more than 57
by Chen, Liying Oct. 6, 2013

Oct. 6, 2013

Hi, In the document binascii.b2a_base64(data) said the data length should less than 57.But I can use it for more than 57.Can you give me an explain? >>> import struct >>> import binascii >>> payload = struct.pack('>hhiii',4,0,0,0,0) >>> payload = payload*10 >>> len(payload) 160 >>> binascii.b2a_base64(payload) 'AAQAAAAAAAAAAAAAAAAAAAAEAAAAAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAAAQAAAAAAAAAAAAAAAAAAAAEAAAAAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAAAQAAAAAAAAAAAAAAAAAAAAEAAAAAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAAAQAAAAAAAAAAAAAAAAAAA==\n' >>> a = binascii.b2a_base64(payload) >>> len(a) 217 >>> Thanks Chen Liying

2 1

Should say 'REAL' not 'FLOAT' as the SQLite datatype in 11.13.1 of sqlite3.
by Richard Kelsall Oct. 6, 2013

Oct. 6, 2013

Hello Python Docs, TLDR: I think it should probably say 'REAL' not 'FLOAT' as the SQLite datatype in 11.13.1 of sqlite3. I am reading the sqlite3 page http://docs.python.org/2/library/sqlite3.html for Python v2.7.5. It says in 11.13.1: "SQLite natively supports only the types TEXT, INTEGER, FLOAT, BLOB and NULL." This made me wonder, is this a double or single FLOAT? (I have not used this database or this Python module before.) And I looked in the SQLite documentation here http://sqlite.org/datatype3.html This says that SQLite version 3 uses an 8 byte (i.e. double) float and has this as type REAL. So I go back to the python page thinking maybe this is an earlier version of SQLite - maybe Python 2 provides SQLite 2 for backwards compatibility and the module name 'sqlite3' does not follow the SQLite version number for some reason. I cannot see on this page mention of what version of SQLite this module provides, maybe I missed that? But I see in 11.13.5.1: "SQLite natively supports the following types: NULL, INTEGER, REAL, TEXT, BLOB." So this internal contradiction makes me think section 11.13.1 is incorrect. I feel posting to this public list, of unknown proportions, may be a rather loud way of reporting such a minor problem, but the documentation page seems to ask me to do this. A wiki might have been an easier and less noisy way to suggest corrections. I expect many people would not go to the trouble of emailing a public mailing list to get a detail like this corrected and that there are many similar bugs which are found but not immediately reported for this reason. Richard.

2 1

Typo in collections.Counter class example
by Михаил Голубев Oct. 6, 2013

Oct. 6, 2013

Hello, It seems that I found micro bug in Python documentation describing collections.Counter class<http://docs.python.org/3.3/library/collections.html#counter-objects> . There is the following snippet of code c.most_common()[:-n:-1] # n least common elements Actually this expression returns not *n* least common elements, but rather *n - 1,* beacuse second boundary is not included (as usual with slices). You probably meant c.most_common()[-n:] or c.most_common()[:-n-1:-1] if reverse ordering is nesessary. -- Best regards Mikhail Golubev

2 1

I found bugs in the Queue documentation
by 오진균 Oct. 6, 2013

Oct. 6, 2013

http://docs.python.org/2/library/queue.html There are two typing errors in this document http://docs.python.org/2/library/queue.html#Queue.Queue.put http://docs.python.org/2/library/queue.html#Queue.Queue.get there is a same sentence within this two function descriptions, If optional args *block* is true and *timeout* is None (the default) In this sentence, I think it should be "True" not "true" because it stands for actual python Boolean value. Thanks.

2 1

bytearray and language usage.
by Mark Janssen Oct. 6, 2013

Oct. 6, 2013

Looking at the Python docs for v2.7.5. Section 2. bytearray([source[, encoding[, errors]]]) " Return a new array of bytes. The bytearray type is a mutable sequence of integers in the range 0 <= x < 256." This should not be called a "mutable sequence of integers", because to call them "integers" already casts raw bits into a particular *type*. They could, hypothetically be very short floats, for example. More accurately, they are merely "uninterpreted bits presented to the user in 8-bit chunks, with a consistent starting boundary" (held by the machine storage address). -- MarkJ Tacoma, Washington

2 2

About loops
by Blanca Mancilla Oct. 6, 2013

Oct. 6, 2013

Hi, In section 4.4 break and continue Statements and else Clauses in loops, the first program has an issue: the second for: for x in range(2,n) is not entered in the first iteration of the n for (for n in range(2,10)) as range (2,2) give an empty range: >>> range(2,2) [] therefore 2 is never "declared" prime. Also the break is in the wrong place. bm

2 2

Possible error
by Blanca Mancilla Oct. 6, 2013

Oct. 6, 2013

Hi, I'm going through the documentation and came across the following error: in http://docs.python.org/2.7/tutorial/interpreter.html, in the Source Encoding Section, there is a program: # -*- coding: iso-8859-15 -*- currency = u"€"print ord(currency) which presents a problem when running: Traceback (most recent call last): File "encoding.py", line 6, in <module> print ord(currency) TypeError: ord() expected a character, but string of length 3 found but if I change the encoding to UTF-8 it works: # -*- coding: UTF -*- it give the desired results. Not sure if it is a real bug. Regards, bm

2 1