rasbt / python_reference Goto Github PK

I realize this post is fairly old, however I was wondering if you could please clarify.

It seems to me that in the functions you define in this section, you use the variable c for cursor within the function even though you pass the cursor object to the functions.

Could you please address this?

On http://nbviewer.ipython.org/github/rasbt/python_reference/blob/master/not_so_obvious_python_stuff.ipynb an anchor is closed in wrong place:

Chapter "Only the first clause of generators is evaluated immediately" is highlighted with the code altogether on mouse over.

See screenshot:
http://h.ctf.su/img/tmp.79rYIPXEi7Ua.png

Also chapter "Assigning types to variables as values" is the only chapter title highlighted on mouse over.

Checked on Chrome 34.0.1847.132 and firefox (linux).

PS: Wanted to patch it myself, but .ipynb syntax is mad. Also it seems HTML generated isn't good at all:

...
<a name="variable_types"></p>
</div>
</div>
</div>
...

I found a workaround for an issue you describe in tutorials/table_of_contents_ipython.ipynb when converting internal links to an HTML format. Instead of being forced to put the id-anchor tag in a separate cell like you did in your notebook:

You can simply place a line break between the id-anchor and whatever text you'd like to include:

Which will render properly without actually including any line break visual:

In the readme file, there is a hyperlink for the IPython notebook for Sorting Algorithms, but it's pointing to "404: Not Found".

Thanks.

One thing I keeps telling people about range/xrange difference is that, the range in python3 implement the __contains__ method, which makes it very easy to test if a value is in range.

For example, in Python2, the following code is super slow:

x = 10000000 # a very large number
if x/2 in xrange(x):
    pass

But in Python3, the following code returns immediately:

x = 10000000 # a very large number
if x/2 in range(x):
    pass

To archive the same performance in Python2, one have to write the code like this:

x = 10000000 # a very large number
if 0 <= x/2 <= x-1:
    pass

https://github.com/rasbt/python_reference/blob/master/not_so_obvious_python_stuff.ipynb#L1118

this function and the next one are iterating over wrong parameter - listcompr instead of generator or generator_yield

I would like to suggest to create a section "development tools".
There are many interesting tools in this area, for example PyBuilder

This is great stuff!

In the not-so-obvious python tutorial, one thing caught me eye - you say that the lambda-closure problem doesn't apply to generators, but it actually does. Your example appears to work because each function returned by the generator is going to return the current value of n, which indicates how far along in generator you've gotten. The list example fails because the whole list is evaluated; the generator example "succeeds" because when you call the ith function, n is still only equal to i. To see this in action:

>>> my_gen = (lambda:n for n in range(5))
>>> a = next(my_gen)
>>> a() # a() returns 0 because that's what n is right now
0
>>> a()
0
>>> # Advancing the generator increases n by one
>>> b = next(my_gen)
>>> b() # b returns 1, as expected
1
>>> a() # but now a returns 1, too
1
>>> # calling list(my_gen) will advance the generator to the end
>>> _ = list(my_gen)
>>> a(), b() # now n is 4
(4, 4)

Thanks for this very helpful intro. Hope this isn't just an annoying nitpick. The sentence:

"As we go up in Python's L -> E -> G -> B hierarchy, the function a_func() finds len() already in the global scope first before it attempts"

is missing an ending such as "to search the Built-in namespace."
I was referred to your tutorial by MITx 6.00.1x Intro to CS & Programming Using Python

A nice addition to your "not so obvious" notebook would be the modifying a list while looping through it pitfall. For instance, you might be tempted into believing that the following will remove all even values from the list a

>>> a = [1, 2, 3, 4, 5]
>>> for i in a:
...     if not i % 2:
...         a.remove(i)
...
>>> a
[1, 3, 5]

But if you try a different example:

>>> a = [2, 4, 5, 6]
>>> for i in a:
...     if not i % 2:
...         a.remove(i)
...
>>> a
[4, 5]

I'll leave it as an exercise for you to figure out exactly what is going on here.

Dictionaries will protect you from this by raising an exception if they change during iteration, but for lists, you should use a copy (for i in a[:]), or convert the for loop into a while loop.

Great notebook by the way!

FYI, IPython.sys_info() can be of help and give you commit hash or IPython and other stuff.

In [1]: import IPython
In [3]: print(IPython.sys_info())
{'commit_hash': b'4aea4a2',
 'commit_source': 'repository',
 'default_encoding': 'UTF-8',
 'ipython_path': '/Users/bussonniermatthias/ipython/IPython',
 'ipython_version': '3.0.0-dev',
 'os_name': 'posix',
 'platform': 'Darwin-11.4.2-x86_64-i386-64bit',
 'sys_executable': '/usr/local/opt/python3/bin/python3.4',
 'sys_platform': 'darwin',
 'sys_version': '3.4.0 (default, Apr  9 2014, 11:57:22) \n'
                '[GCC 4.2.1 Compatible Apple LLVM 4.2 (clang-425.0.28)]'}

Read here : http://nbviewer.ipython.org/github/rasbt/python_reference/blob/master/tutorials/table_of_contents_ipython.ipynb

Do you know that header cell are made for that and will give anchor if you use them ?

In the following README:
https://github.com/rasbt/python_reference/tree/master/tutorials/sqlite3_howto
the link in the line "Download the script: create_new_db.py" appears to be broken, it refers to https://raw.githubusercontent.com/rasbt/python_reference/master/tutorials/code/create_new_db.py which no longer exists.

The link URL should probably be replaced by
https://github.com/rasbt/python_reference/blob/master/tutorials/sqlite3_howto/code/create_new_db.py