Errata

http : / / python.itcarlow.ie & http : / / python.itcarlow.ie/ed2 are not accessible. Trying to get access to html/css templates for webapp project.

Note from the Author or Editor:
Isn't progress a wonderful thing...? Nowadays, it's "https".

2nd Edition from Amazon, Running on the newest version in a ubuntu LTS, virtual environment as of 5/29/23:

req.user_agent or req.user_agent.browser crashed when "Do it" was pressed. If substitute 'test', the data is written to the mysql database.

req.remote_addr seems okay, so don't know why req.user_agent does not work. Why?

Note from the Author or Editor:
Due to an update in the underlying library, req.headers.get('User-Agent') should now be used.

I found it confusing that the renaming from odd.py to odd2.py is made on page 35 after finishing the last two tasks instead of renaming the file after the first task (the for loop) is inserted to the file.

Would be more consequent to have an odd.py file without loop etc. and an odd2.py file with the changes of all three tasks.

Note from the Author or Editor:
Noted. Addressing this has been added to the list of tasks for the 3rd edition.

In the explanation of the for loop there is "... from 99 down to none." but I would explain it with counting "... from 99 down to 1." since the value 0 is not included in the loop.

Note from the Author or Editor:
Yes, the description could be clearly here. Better to use the words "down to but not including zero" instead of "none".

I am challenging the idea of introducing annotation syntax but immediately ignoring PEP 8 style guidelines, first introduced on page 153, (even though that is indeed covered later). This is unnecessarily confusing for someone who read PEP 8 before getting to page 162.

Using ":str" here instead of ": str" creates a less desirable habit right away, and it isn't addressed until page 192.

Likewise, on page 170, PEP 8 specifies to add space around the assignment operator for an annotated default value, where "letters:str='aeiou'" ought to be "letters: str = 'aeiou'".

Maybe there is a different way to approach the value of using PEP 8 compliance tools without having the learner start off on the wrong foot?

Note from the Author or Editor:
This is something (i.e., annotations) that were relatively new when the second edition was being written and - in hindsight - was not handled at all well in the book. The 3rd edition is being actively worked on now (Sept 2021) and I'm toying with the idea of removing annotations altogether.

When I run py.test --pep8 vsearch.py from a shell command my results are quite different. It doesn't seem to be running the test, and I can't figure out why.

PS C:\-pathsnipped-> py.test --pep8 vsearch.py
c:\python38\lib\site-packages\pep8.py:110: FutureWarning: Possible nested set at position 1
EXTRANEOUS_WHITESPACE_REGEX = re.compile(r'[[({] | []}),;:]')
========================= test session starts =====
platform win32 -- Python 3.8.5, pytest-6.0.1, py-1.9.0, pluggy-0.13.1
rootdir: C:\-pathsnipped-
plugins: pep8-1.0.6
collected 0 items / 1 error

================================ ERRORS ==========
______________ ERROR collecting test session ______________
c:\python38\lib\site-packages\pluggy\hooks.py:286: in __call__
return self._hookexec(self, self.get_hookimpls(), kwargs)
c:\python38\lib\site-packages\pluggy\manager.py:93: in _hookexec
return self._inner_hookexec(hook, methods, kwargs)
c:\python38\lib\site-packages\pluggy\manager.py:84: in <lambda>
self._inner_hookexec = lambda hook, methods, kwargs: hook.multicall(
c:\python38\lib\site-packages\pytest_pep8.py:38: in pytest_collect_file
return Pep8Item(path, parent, pep8ignore, config._max_line_length)
c:\python38\lib\site-packages\_pytest\nodes.py:95: in __call__
warnings.warn(NODE_USE_FROM_PARENT.format(name=self.__name__), stacklevel=2)
E pytest.PytestDeprecationWarning: Direct construction of Pep8Item has been deprecated, please use Pep8Item.from_parent.
E See https://docs.pytest.org/en/stable/deprecations.html#node-construction-changed-to-node-from-parent for more details.
======================= short test summary info =============
ERROR - pytest.PytestDeprecationWarning: Direct construction of Pep8Item has been deprecated, please use Pep8Item.from_parent.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
=========================== 1 error in 0.26s ============

Note from the Author or Editor:
Here's some notes from another reader on this issue:

"...However, after going through the internet looking for answers I found a solution that produced the same output as the one in your book! A forum user posted that the pep8 part of pytest is being renamed to pycodestyle, so when I type

sudo -H python3 -m install pytest-pycodestyle

it installs the module to check for pep8 compliance, and then call it by

pytest —pycodestyle *.py

I’m not sure if it is a separate module all together, or it actually is the new name for the pep8 module, but pycodestyle seems to give the same terminal output as pep8 would... "

Note that the published book targets Python 3.5 and that time marches on. 3.9 is just around the corner and pytest 6 recently appeared. Some things change and stuff in the book breaks. This is hard to avoid.

--Paul.

In the code above TEST DRIVE on page 258 you write:
contents=log.read()
but in the code at the bottom of page 268 you write:
contents=log.readlines()

Is there a difference between the read and readlines functions?

Note from the Author or Editor:
The distinction needs to be clearer here: "read" grabs the entire file into a string (series of bytes), whereas "readlines" grabs the entire file into a list of lines (i.e., a list of strings).

In prior examples changes to code have been highlighted in a light-blue. The last example of the return statement for this function I could find is on page 258: "return escape(contents)".

My question is: did I miss where I should have changed the statement to "return escape(''.join(contents))" or should this statement be highlighted.

I suspect the join example should be inserted somewhere earlier (or I missed it) because I could not get the early examples of log_request() to work (p. 261). I kept getting an exception when attempting to print (ImmutableMultiDict is not a string). I corrected it with a "str(req.form)". But seeing this page makes me think you meant to supply "join" earlier.

Note from the Author or Editor:
Yes, I should be highlighting that return statement in blue (the last line of code on page 268) to note that it has changed. The very next paragraph of text does discuss it (for what it's worth).

The link to the MySQL-Connector/Python driver page works fine, but when you arrive at the site there is no longer a 'Platform Independent' option from the dropdown menu when you want to download. You must either select an OS or go with the Source Code option.

Note from the Author or Editor:
This is correct. Oracle changed how this works (per this reader's note). Scheduled to be fixed in the 3rd edition, but a quick fix for now is to simply run "pip install mysql-connector-python" as Oracle have improved support here too.

I have followed steps given in PDF, however, getting some issues as follows. Am I making any mistake here

---------------------------------------------------------------------------------------------------------------------------

Python 3.8.0 (default, Oct 14 2019, 21:29:03)
[GCC 7.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> class CountFromBy:
... pass
...
>>> a = CountFromBy()
>>> b = CountFromBy()
>>> c = CountFromBy()
>>> c
<__main__.CountFromBy object at 0x7f9b3a082a90>
>>> a
<__main__.CountFromBy object at 0x7f9b3a052460>
>>> b
<__main__.CountFromBy object at 0x7f9b3a037520>
>>> c.increase()
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
AttributeError: 'CountFromBy' object has no attribute 'increase'
>>> c
<__main__.CountFromBy object at 0x7f9b3a082a90>
>>>


---------------------------------------------------------------------------------------------------------------------------

Note from the Author or Editor:
The error I made was to show (on pages 312ff) the increase() method in use before the code for it is defined. See the top of page 320 for what you have to add.

The beginning of chapter 8 did a good job preparing the reader for how Python deals with OOP, but suddenly, out of nowhere, appears the "increase" method for the CountFromBy class. I've had some experience with OOP and was confused where this method came from!? The reader should have been warned that the def for this class would be forthcoming and not left scratching their head wondering how it got there. Better yet, give the def for the increase function within the CountFromBy class and then proceed with the dot notation, etc... Don't just pull it out of the air!

Note from the Author or Editor:
Noted for reworking during next update. The text needs to be clearer in its intention to work up to creating the "increase" method.

I have done the step by step procedure until this point but these next few lines are ringing error on my IDLE window:
=======Head First Python (Code) ============
>>> c = CountFromBy()
>>> c
0
>>> c.increase()
>>> c.increase()
>>> c.increase()
>>> c
3
=======================================


My IDLE window error:

>>> class CountFromBy:
pass

>>> c = CountFromBy()
>>> c
<__main__.CountFromBy object at 0x02DFE990>

======================================
I have even added the increase function as described in errata page but the same issue prevails:

>>> class CountFromBy:
def increase(self) ->None:
self.val += self.incr


>>> class CountFromBy:
pass

>>> c = CountFromBy()
>>> c
<__main__.CountFromBy object at 0x02DFE990>
>>> c.increase()
Traceback (most recent call last):
File "<pyshell#23>", line 1, in <module>
c.increase()
File "<pyshell#20>", line 3, in increase
self.val += self.incr
AttributeError: 'CountFromBy' object has no attribute 'val'


==============================================

Requesting assistance on the same.

Regards
ABHAY

Note from the Author or Editor:
The more I look at this material, the more I realise that I did a pretty poor job of presenting this topic. On page 325 (bottom half) you'll find the code required to make this all work - my mistake was to show the code running before I'd presented the code in it's totality.

"Here’s the code from the last chapter, rewritten to use a with statement, which itself…"

refers to Ch 8, but the code presented below was created in Ch 7.

Suggested correction:
"Here’s the code from chapter 7, rewritten to use a with statement, which itself…"

Note from the Author or Editor:
Change second sentence to read: "Here's the code from Chapter 7, rewritten..."

"It’s now time to return to the questions first posed in Chapter 7:"

should actually be

It’s now time to return to the questions first posed in Chapter 6:

## (on page 244)

Note from the Author or Editor:
On page 358: change the first sentence to read: "It’s now time to return to the questions first posed at the end of Chapter 6:".

Also, on page 359: change first half of first sentence to read: "Let’s take our data questions one by one, ..."

You list 4 features that a decorator must have. But, the decorators that we've used already, such as one first defined on page 207, only include the first of these features. How can the decorators we've used be decorators with these features missing?

Note from the Author or Editor:
Note for editor: There is *no* change required to the text.

However, a clarification is needed here: In order to use an existing decorator, all you need to know is what it does, and how to use it (as per the example on page 207). To create your own decorator (which can later be used by you and/or by others) you need to know about and understand the four things listed on page 395. When it comes to creating a decorator, these four things are prerequisites from a knowledge point-of-view, not absolute requirements from an implementation point-of-view. That is, knowing about them helps you get your head around what's possible with decorators. What your decorator ends up doing is an implementation-dependent detail as the decorator machinery is a generic mechanism which can be put to many uses.

Your description of decorators needs more meat. For example,

1) You've used the word decorator in 2 ways, e.g as the thing on the line
starting with a '@' character, and as the function that must meet the
requirements stated on page 395. Is this correct?

2) What's the syntax for invoking a decorator? We've seen 3 examples

@app.route('/page1')
@wraps(func)
@check_logged_in

The first 2 use a similar syntax, but only the 2nd explicitly passes the name of the function being decorated. The 3rd says nothing. When does a decorator require arguments?

3) What does

@check_logged_in

actually do at runtime?

4) Apparently the decorated function is magically passed as an argument to the decorator. How does this work?

Note from the Author or Editor:
There's a lot more to decorators than space allowed me to cover in the 2nd edition. What's shown in Chapter 10 is by way of introduction, as doing a deep-dive with decorators would be considered one of the more advanced Python topics. What's shown is designed to prime a reader to tackle more detailed coverages in other texts (for instance: the Python Cookbook, 3rd edition, has some very good material). Your comments/questions are noted for inclusion/consideration in an updated edition. Thanks.

"if exc_type is mysql.connector.errors.ProgrammingError:"

You need to say something about why you used "is" rather than "==".

Note from the Author or Editor:
Noted for (potential) third edition.

"you can add an elif suite to the end of the __exit__ method that reraises the unexpected exception"

You need to say something about why it is useful to reraise the unexpected exception. After all, it was unexpected before, why won't it be unexpected again?

Note from the Author or Editor:
I did start this paragraph off with the statement "If you want to be extra safe..." which was meant to hint that this was taking things to a bit of an extreme. I've made a note to extend this description in the next edition.