On Sun, May 8, 2016 at 5:24 PM, DFS <nos...@dfs.com> wrote:
> On 5/8/2016 7:36 AM, Steven D'Aprano wrote:
>>
>> On Sun, 8 May 2016 11:16 am, DFS wrote:
>>
>>> address data is scraped from a website:
>>>
>>> names = tree.xpath()
>>> addr = tree.xpath()
>>
>>
>> Why are you scraping the data twice?
>
>
>
> Because it exists in 2 different sections of the document.
>
> names = tree.xpath('//span[@class="header_text3"]/text()')
> addresses = tree.xpath('//span[@class="text3"]/text()')
>
>
> I thought you were a "master who knew her tools", and I was the apprentice?
>
> So why did "the master" think xpath() was magic?
>
>
>
>
>
>
>> names = addr = tree.xpath()
>>
>> or if you prefer the old-fashioned:
>>
>> names = tree.xpath()
>> addr = names
>>
>> but that raises the question, how can you describe the same set of data as
>> both "names" and "addr[esses]" and have them both be accurate?
>>
>>
>>> I want to store the data atomically,
>>
>>
>> I'm not really sure what you mean by "atomically" here. I know what *I*
>> mean
>> by "atomically", which is to describe an operation which either succeeds
>> entirely or fails.
>
>
> That's atomicity.
>
>
>
>> But I don't know what you mean by it.
>
> http://www.databasedesign-resource.com/atomic-database-values.html
>
>
>
>>> so I parse street, city, state, and
>>> zip into their own lists.
>>
>>
>> None of which is atomic.
>
>
> All of which are atomic.
>
>
>
>>> "1250 Peachtree Rd, Atlanta, GA 30303
>>>
>>> street = [s.split(',')[0] for s in addr]
>>> city = [c.split(',')[1].strip() for c in addr]
>>> state = [s[-8:][:2] for s in addr]
>>> zipcd = [z[-5:] for z in addr]
>>
>>
>> At this point, instead of iterating over the same list four times, doing
>> the
>> same thing over and over again, you should do things the old-fashioned
>> way:
>>
>> streets, cities, states, zipcodes = [], [], [], []
>> for word in addr:
>> items = word.split(',')
>> streets.append(items[0])
>> cities.append(items[1].strip())
>> states.append(word[-8:-2])
>> zipcodes.append(word[-5:])
>
>
>
>
> That's a good one.
>
> Chris Angelico mentioned something like that, too, and I already put it
> place.
>
>
>
>> Oh, and use better names. "street" is a single street, not a list of
>> streets, note plural.
>
>
>
> I'll use whatever names I like.
>
>
>
>
>
> --
> https://mail.python.org/mailman/listinfo/python-list
Starting to look like trolling. Lots of good advice here. If you
ask, and don't like the advice, don't use it.
--
Joel Goldstick
http://joelgoldstick.com/blog
http://cc-baseballstats.info/stats/birthdays
--
https://mail.python.org/mailman/listinfo/python-list
