@Romshark Thirteen Thank you The challenge isn't parallelizing. The problem is

Question

Пользователь 61931

@Romshark Thirteen Thank you The challenge isn't parallelizing. The problem is

the difference between different websites!

For example, consider all of my websites are eCommerce and I want to grab the product price from these websites.
So what should I do?

Of course, Amazon has its own template, and Alibaba has its own! So I have to make it clear for my scrapper (the CSS selectors or regex or json unmarshal or...)

Now my current idea is: Add a function for each website, which accepts HTML doc and returns expected information from the website (for example product price)
But as I said, I don't think that it's a good plan!

Bcuz I have to add many functions to my program!
And when there is a new website, I should stop my program, then add a new function for a new website, build my program and run again!
Each time a new website needs to be added I should rebuild my program!

0

11.08.2020

2 ответов

12 просмотров

Пользователь 61931 Автор вопроса

Any idea?

0

12.08.2020

Roman Sharkov · Accepted Answer

Roman Sharkov

or you could spend 10 years building a machine learning model which automatically scrapes websites no matter what layout it uses why build something for 30 minutes if you can automate it for 10 hours? 😂

0

11.08.2020

Похожие чаты

@Romshark Thirteen Thank you The challenge isn't parallelizing. The problem is

2 ответов

Похожие вопросы