Python Web Scraping with Beautiful Soup and Regex

[@cetrusbr]

I like your tutorials because u go directly to the content, something rare in youtube these days...

[@kalef1234]

Hey guys what's up before we get started smash that subscribe button, like this share it i am giving away a fucking gift card follow the links to my merch watch my ads really helps thanks okay...roll that intro *45 second intro*

[@sourabhch3044]

So true thank you for putting out the points which matters.

[@mixalismcgamer3188]

Dude i watched over 15 videos+ that was recommended and after hours i found this FULLY EXPLAINED.

[@kalef1234]

I felt so powerful as soon as I pulled an array of strings from a random website. Thank you for your great tutorial

[@justinhamilton8647]

Cheers man i used this tutorial to sort through 310000 embed links you’re so awesome

[@zigginzag584]

It helps so much to have someone that matches your personality when learning stuff. I can't stand when asking someone for instructions on how to do something and they tell me everything that I can expect and every once i a while throw in the thing I'm supposed to do next. None of the fluff here. Just context. Every other creator would/has made this subject a 45min+ video but here I am feeling proficient after just 14 minutes with EM. Thank you, Sir!

[@EngineerMan]

You're welcome buddy!

[@kurdmajid4874]

he makes it so quick and simple

[@bhumikakhiyani4230]

I was struggling to navigate to iterate through second span tag in multiple td tags I.e. (tr[1:]/td[0]/span[1]) I was trying it the whole day. This is the best tutorial I have seen. Thank youuuuu.

[@dilshand.5127]

I was able to do this on another leaderboard site, appreciate your work here.

[@CODTALES-KILLSTREAKS]

Hey man! I watched this and applied the concepts to a weather site and made a csv of all the sunset / sunrises in 2019! Thank you! Please I love the way you explain things keep making videos sir! I have applied your teaching in a couple videos and it’s great! Learning so much!

[@mhalton]

13:52 Happiest man!

[@EngineerMan]

Oh god I'm not gonna be able to unhear that any time soon.

[@estilen69]

Using CSS selectors is the way to go, gets rid of nested for loops and is more robust.

[@matteomannini1205]

how?

[@robertpearson2143]

Been doing something similar for a while but in a much more complicated way. Looking forward to making my life much easier. Thank you!

[@PS3PCDJ]

This is THE best beautifulsoup tutorial on the internet.

[@xrefor]

Love this presentation. Straight to the point with short and specific explanation. Keep it coming! :)

[@kennethmcquade4341]

You're definitely skilled! For anyone watching these videos, don't get discouraged, this takes time. @Engineer Man , Can you talk about the experience of learning how at the beginning of your videos?

[@DrSarge37]

It would be cool to see how to deal with pagination. So you want data from /page=1, /page=2 etc. Etc.

[@joefagan9335]

In your browser go to next page and copy the url of, say, page 2 and go to last to find the last page url. Use that as a template to build the url of each page you want. Loop over them in turn.

[@joefagan9335]

John Keymer nope you’re not parsing the page a second time to find the next button. You scrape the current page and then grab the neat page by creating the string for the next url and accessing the next page - just one grab per page.

[@ViniciusProvenzano]

Real Nice content! Straight to the point. I’ve played around with beautiful soup a few years ago for an small project, and I just wish this video was around at the time....

[@ladyViviaen]

was trying to scrape modarchive for my project, this is way better than writing the name and id down by hand lmao, thank you!

[@johnbecker3116]

I spent forever teaching myself this last week and now you post this. Kill me now

[@YeeYeez]

If only I had this tutorial a few years back. Good stuff.

[@yanggao4878]

Your videos are fast-paced and straight to the point. Thanks!

[@enyoc3d]

in a sea of youtube tutorials yours is the pearl. thanks!

[@impossible441]

This is remarkable, very informative and down to the earth - I really love this concise format of yours which is rather contradictory to what most of ppl on yt are providing

[@TomSilver_42]

Simply brilliantly explained. I have seen few of your videos and I like your style, therefore You have earned another subscriber.

[@worsethanjoerogan8061]

Dude you're helping me out immensely with computer science courses

[@SusiEzhil]

wow.. thats the crisp explnation,,, you're the man!!

[@Lu3ck]

Your videos are fast but glorious! Love your content man! Thank you! Bless 🙏

[@clownboy84]

Thanks for the video. I like how you take the basics and break it down with really good and practical examples.

[@susbedoo]

You are the coolest tech guy I have ever seen on KZbin

[@laxlyfters8695]

Went through a 30 second hillshire farms ad. Great match youtube

[@EngineerMan]

Google knows you're into web scraping and sliced turkey lol.

[@laxlyfters8695]

Engineer Man no lie came back and got an ad for $3 jack box munchie meals. KZbin thinks your fans are stone while watching your videos

[@andriybortnik8310]

This is an awesome video, I actually enjoy the in depth walk through of what your reasoning behind writing code is, step by step. Versus just saying " I did this" and not really explaining anything. On a separate note , I'm looking to get into python, and I have previous code development experience, but It's been a little while, and setting up an environment to start doing some coding is a bit daunting. I'm looking to do more on the machine learning , neural networks side of things. I don't struggle with any of the logic, mathematics, but I know there are many pros/cons of various IDE's . Some have better support for various packages , etc.. I was wondering if you could either make a video on some of this information, or maybe throw a few pointers my way. I would really appreciate that. Otherwise, keep up the great content!!!

[@KingEbolt]

Let me throw some pointers at you. 0x3A738216 0x6B321970 0x88AC172B

[@EluviumMC]

I've found that I really like using Microsoft's VS Code (not to be confused with Visual Studio). The IDE has a good clean interface, lots of extension support, and a built-in terminal.

[@andriybortnik8310]

@@KingEbolt I can't even get mad at that... Well done

[@camaulay]

@@EluviumMC +1 VS Code, switched from Sublime

[@EluviumMC]

Happy that you've chosen this topic. I've been exploring web scraping and have a script that works pretty well on a site that I frequent. Another awesome tool that can be used to also automate web navigation is the selenium package. But on more of a question-related note, I know the script you just made was pretty simple, and the one I have isn't that complicated, but I've been wondering how one would go about writing an object-oriented script for scraping?

[@UchihaAditya]

What are the advantages of selenium over Beautiful Soup?? I have a web-scraping assignment now and was advised to use selenium.

[@EluviumMC]

Selenium can be used as a web scraper, but I use it more for web navigation and then use beautiful soup to actually get the data I need from the pages once they've been navigated to. I just find beautiful soup to be a more intuitive for extracting the data.

[@yixunnnn]

With selenium it is like an automated user, and when you use it, you require a web driver, and you can choose if you want the automated browser to run in the background or not. I recently used selenium because I was trying to request for content behind a microsoft login page, which is loaded using javascript, thus I needed to wait till the content was actually loaded finish before i submit anything. Unlike requests, which instantly retrieves the page content.

[@K2ThaYo]

Beautiful video man! Really valuable information here. As a sysadmin with over 10 years experience, I can state its really clean method of scraping. I was used to use bash scripts for everything but using libraries in python is sooo helpful. It would be a pain in the as in bash with awk, grep, etc. I hope to see more soon

[@supalistmain4882]

@Engineer Man , what is your day job? And how did you get into coding? Do you have a CS degree? and.... well instead of more questions, rather just ask whats your background (ito what lead to you adding so much value with these vids)?

[@rustyelectron]

This video is really a good intro to web scraping.

[@legioner304]

3 searches in the loop - very dirty ) "The speed of software halves every 18 months"

[@axelcano1623]

Really nice content! You explain just enough to be clear but not too much that's perfect. Please continue to remind the type of the elements you create, it's very important for beginners.

[@asdfasdfasdf383]

You go straight to the point. Obviously, you know a lot more in-depth about this topic. Anyway, I like it.

[@ledosilverknight4619]

Some of the best tutors are always straight-forward: down and dirty!

[@arturmangabeira9990]

EM you're awesome. i was studying web scraping and this come up. subscribed yesterday to your channel! lol

[@EngineerMan]

Nice!

[@ddmin3082]

Awesome video! Can you do one on the requests module please?

[@stephenrochester6309]

These videos are brilliant. Thanks for all your hard work.

[@chowfatt38]

Great video again. I've been playing web scraping a while and I find that most of websites nowadays using javascript rendering quite heavy. Will you make a part 2 for talking about how to web scrape javascript rendering website? And what do you think about another web scraping package, Scrapy? thanks Man

[@poidog22]

This would be a great follow on. +1

[@cruzab3153]

Selenium is good and easy....

[@trailrider6844]

+2

[@tayfun6378]

puppeteer does a good job these days I think

[@Megaloplex]

+100

[@Omar-ic3wc]

Exactly what I needed thank you very much!!

[@qettyz]

These were really good examples, thank you!

[@PriZ0nM1ke]

Wow these videos are awesome! Direct and concise but understandable!! Well done!

[@TheEndermanMob]

How does he knows a lib for everything? i'am addicted to his videos.

[@oromis995]

This content is absolute gold.

[@kristiyangerasimov6708]

Great video. Stuff like that makes me want to program and develop software until i die.

[@bennieliu3261]

Awesome tutorial man! Can I suggest scraping dynamic pages as the next tutorial. Would be a sweet follow up

[@EngineerMan]

Thanks. Part 2 of this is being requested a lot, I need to see what is best to do.

[@grantfaith]

ty, saved me an hour of time from all these other videos. holy shit

[@donaldandmijung]

great tutorials! do you have a tutorial on scraping with a function( ) using beautiful soup

[@treybailey6752]

Great vid with fantastic content. Would love to see this where you first login in order to get content. Getting the headers set is a challenge.

[@EluviumMC]

Using Selenium to do the site navigation to get you logged in is how I worked around getting into a site that requires login credentials prior to scraping.

[@Project_OMEG4]

Great video EM, but requests is not a built-in module for python; (does not come with the default python installation), so you will have to install it. For any missing library, the source is usually available at pypi.python.org/pypi/. You can download requests here: pypi.python.org/pypi/requests To install: • OSX/Linux : Use $ sudo pip install requests if you have pip installed. Alternatively you can also use sudo easy_install -U requests if you have easy_install installed. • Windows : From a cmd prompt, use > Path\easy_install.exe requests, where Path is your Python*\Scripts folder, if it was installed (for example: C:\Python32\Scripts\easy_install.exe). If you manually want to add a library to a windows machine, you can download the compressed library, unzip it, and then place it into the Lib folder of your python path. (For example: C:\Python32\Lib) NOTE: Mac OSX and Windows, after downloading the source zip, un-compress it and from the termiminal/cmd run python setup.py install from the uncompressed directory.

[@magicyvan]

loved it ! Efficiency and very clear for a beginner. Would be great to have the login part, and why not sending the extraction into a csv file ;) I subscribe ;)

[@blevenzon]

Wow just found your channel by accident and I’m loving it. Awesome content!! Do you think you can do a vid on Elastic Stack?

[@JeroenTrappers]

Good video. Personally, i like using node with dom module and write css queries to extract what i want.

[@luis96xd]

Wow, I liked this video so much! It was very useful! 😄 You really have helped me a lot, it was well and fully explained, with real life examples Thank you so much for this tutorial! 👏👏

[@alfredleppanen6796]

Hey great video! Lets say in your last leaderboard example, I would like to get notified when the leaderboard has changed, so to say when something changed on the site. I have built a script where I can see the HASH change, but I cant output what actually changed on to website, do you have any tips to how to monitor what actually changed on the website?

[@nicememe999]

Yes! A great tutorial on web scraping! Now I got some ideas on some websites I could scrape for data... What kind of real-world applications could this be used for? With websites providing APIs with the data nicely packaged in JSON format, it seems like getting data via APIs seems to be the better (or at least the most common way) to do this. Are there any situations where web scraping would be better?

[@impossible441]

I guess that any kind of scientific literature databeses use webscraping (i.e. google scholar)

[@EluviumMC]

Webscraping should be a last resort. Getting data via an API is much better.

[@chrisabreu7469]

your videos are a life saver man. keep up the great content

[@molimola3]

Hey I love you videos ! You explain everything so well. I am trying to scrape some websites but they don't allow me because of their bot protection... Do you have any tips about this ? Thanks

[@сашарассадин-щ5ъ]

I got this problem too. In my case I have solved that by changing a type of my request, now it includes *headers*. You need to look up for the data of your headers in your web browser. You should visit google.com page, press ctrl+shift+I, in opened console find "network", and search necessary elements there. In in the other words, the solution is adding "headers", I hope the information will help you. Example: headers = {"accept" : "your accept symbols", "user_agent", "your user agent string" session = requests.Session() request_variable = session.get(url, headers) P.S. I am from Russian, I was not using a translating while typing it, I hope you were able to understand me.

[@kingseekerbackup3085]

I use requests and bs4. Never thought of using regex besides pattern searching

[@ilobuhabib8325]

love your tutorials. I tried following your method to scrape a site, but the output is empty. when I checked the 'tr' throughout the source code, it has values, but I do not understand why the output is empty.

[@royslapped4463]

this is perfect for what I needed thank you!

[@xppaicyber3823]

Great content

[@DirtySocrates]

Excellent! Thank you!! Great vid!

[@MrFrondoso]

Génial. Dieu sait que je galère à utiliser BSoup . Et là j'ai l'impression d'avoir enfin compris.

[@NoorquackerInd]

_I can't believe I used to use Selenium for this_ At least for that project I rewrote it and used raw Requests when I found out my target could return data in JSON

[@NokiaN8Guides]

thank you so much for this amazing tutorial, i would like to ask what do we do if the site i want to scrap require to be logged in btw this got recap

[@joefagan9335]

Usually, you can login first. Leave it open in your browser and scrape away.

[@santiagorivera1562]

What is the advantage to using Beautiful Soup over other webscraper packages with Python?

[@syntaxis5584]

why did you use 'View page source' instead of 'inspect' to find the page structure?

[@EngineerMan]

I did it because view source represents the content that was delivered to the browser on load whereas inspect represents the content currently on the page. Since the scraper doesn't see anything dynamically generated, view source is best.

[@JoseGarciam4t0n]

Hey, I really love your videos man! How about elements within elements, so for example `this link is in a paragraph`. How would you approach that?

[@daltonkraklan2257]

This was so freaking helpful

[@siloenoah]

Teach me your ways

[@Viruhemanth]

carefully he's a hero

[@SeamusHarper1234]

This is awesome for the " old world". Can you use beautifulsoup for scraping data from any of these new shiny JavaScript SPAs made with Vue / Angular / React, where you have to execute JavaScript?

[@EngineerMan]

SPAs are easy, just analyze their web services and get the data that way (often in JSON already).

[@SiegeX1]

Can you go over an example that first requires you to login and then requires you to use a query string with a hash token that changes after every login?

[@jacoboneill3735]

Saw this, instantly thought this would be easy to implement to get amazon prices... bot blockers, who thought captcha would get me 😂

[@wilkinsanchez8737]

Amazing video. How did you decide to select 0,1,3 on your code. For example, on line 14,15,16. place = tr.find_all('td')[0].text.strip() why the Zero, One and Three?

[@defau1tMC]

Because tr.find_all('td') returns a list of tds in each tr, the 0, 1, and 3 correspond to those entries in the list

[@stefandevos1520]

love your tutorials man

[@soldiergaming2722]

That's great and all but what I wanna know is... How the hell did he ctrl + u and get neat html rather than the smooshed together junk I get

[@mixalismcgamer3188]

WHERE IS THE MILLION SUBS?

[@EngineerMan]

Soon. Help spread the word to speed it up :)

[@leninespindola4835]

Hello, first of all thanks for your great video! :) Now I would like to ask you how it can be possible to do the same as you did but on the html that I scrape using Scrapy? And maybe to clean up a little bit the html generated to give it better format to get the data? Many thanks and kindly would be nice to get a hand :)

[@svampebob007]

well shit this would have been nice to know, I scraped 1.3GB using bash and text editor (kate). It got the job done since I saved my brother about a day or two of manually finding, downloading and renaming the files he wanted, but I had to do some crazy sed/grep magic to extract the downloadable links.

[@LarsHolmVV46]

That was beautiful not to say absolutely excellent. Man ,,,,,

[@jeuxdeau2009]

I love your page man.

[@sgttye]

Keep up the good work man!

[@StrangeIndeed]

I wanted to scrape 4channel. I wanted to get all the thread divs. But I got nothing. It took me 15 minutes to realize that all the divs are initialy empty, and JavaScript injects them when the page loads. And when we use request, it just downloads HTML, without running JS. Lesson learned

[@Ashesoftheliving]

Hey E-man Great video! I wanted to ask you this. You said that "you have to know the structure of html before doing scraping from websites" which is true but I need to create a process where I can search a keyword or basically a word on multiple websites and get the content out of those websites. Now from this content, I will create a sentiment value and generate a newsfeed in my application with a sentiment value in it. Can you suggest a way I can do that? Thanks in Advance

[@DevastaingDj]

Awesome! Kudos! Very helpful. Thanks man!

[@virtualize2424]

How do you scrape something like KZbin comments (without using KZbin api)? When I get the html data for a video using requests library, the video's comments are not their in the html data.

What editor are you using for python? I’m a newbie. Thanks.

[@elliottharris7496]

This was a pro tip on how to eat a t-rex 10:47

[@kylemichaelreaves]

Super helpful, thank you.

[@reneepaz8077]

Hi love your channel, would like some advise, I want to scrape dat from a table that is dynamically generated by a website based on a user input, i.e. not static. The website does not have a downloadable pricelist csv file for their products, so what it does is based on the criteria that I enter it will generate a table in html format, also due to the massive amount of data available, the table has multiple pages. All I want from the table is get the UPC number and the price of the items so that I could use that data into my product analyzer software.

[@ShreksSpliff]

Is there a way that this method could work with pages that require authorization? Like my own favourites list on a desktop wallpaper site. I tried retrieving a website using something simple but the only thing that showed up was a redirect to check my credentials. Great content still! As you can see I'm smashing through your content. You have left a legacy, so thank you. Look forward to donating or seeing you on patreon.

[@laalaajonsen]

What you need to do is to do Get request with all necessary parameters (i.e cookies and that kind). I just applied this very technique to an own project of mine. Use Chrome tools. Basically this: login normally with credentials, and from Chrome inspect, copy the cURL of your get request. Then i used an online tool to convert cURL to python. Defined that as a function with the final URL as input parameter. From there you get the credentials-protected content in your regular soup object. Struggled with this for s bit, but it worked out in the end :) Now i can download entire structures of vimeo videos with a single url. Its awesome.

[@dralexhunter]

Can you do a video showing the interactions between sessions and bs4?

[@BrettKromkamp]

Excellent tutorial. Thanks.

[@tristanbellingham6759]

You should probably talk about responsible scraping etc especially if you are publishing for an audience that might not know any better

[@Faszinated]

Thanatos 12321 what do you mein with that? Not too many requests in a short period of time?

[@Faszinated]

do you have any examples

[@EngineerMan]

And so far they've lost that suit.