Greetings!
I have 2 related knowledge bases that I have been working on for a long time which I would like to download for my portfolio. I reduced the size of one by 5/6, and the other by 4/5, in addition to very very heavy ease-of-use editing . It is a substantial improvement, and what I want to market myself to do, if I have to keep working. I actively enjoy doing it.
My company will not approve the third party backup application for zendesk (the host of these kbs) and I don’t know how to write a script to back it up or anything. I don’t know how any of that works, but zendesk recommends a python script.
I checked a few indexers and couldn’t find the pages, but even if it was indexed by url, I can’t really go through and download individual page links, so I’m not sure what to do. I don’t know python, or anyone who knows Python.
For context, the project as a whole has been my world for the last 9 months, it’s not what I was hired to fix explicitly, which is why they dgaf about my portfolio, but it’s what I want to do (and I’m great at it, even have a degree for it!). I want a snapshot of the whole thing on my last day, ideally including the archived articles I got rid of.
Any ideas for someone who doesn’t code?
You must log in or register to comment.
Hoo boy.
So, first off, this might be something you’ve already considered, but you mentioned “your company” a couple of times. Chances are if you’ve contributed to these knowledge bases as an employee, the content on those knowledge bases are owned by your employer. If you don’t get permission to get a copy and incorporate it into your portfolio, you’re potentially asking for legal trouble.
So, I guess what I’m saying is if you’re not 100% sure your employer will be totally cool with it, try not to get caught. Also, depending on how on-the-ball the IT department is, they might notice that you’re scraping specifically because you’re sending a lot of extra traffic to those knowledge bases.
So, that out of the way, wget will most likely do it. But it’ll probably take a lot of tinkering.
A few considerations. First, if you have to be authenticated to access the content you’re wanting to scrape, that’ll definitely add a layer of complexity to this endeavor.
Second, if the Zendesk application interface uses a lot of AJAX, it’s likely to be much more of a bear.
If you’re using wget, a basic version of what you’ll need is a command something like:
wget --mirror --tries 5 -o log --continue --show-progress --wait 2 --waitretry 2 --convert-links --page-requisites https://example.com/If you need to be authenticated to view the data you’re wanting, you’ll have to do some magic with cookies or some such. The way I’d deal with that:
- Log in to the application in a browser.
- Hit F12 to open developer tools.
- Hit the “Storage” tab. (Sorry, I don’t remember if the equivalent tab is called “Storage” or somthing else in Chromium-based browsers. I’m using Firefox.)
- Expand out “Cookies” on the left. (Again, sorry if you’re using Chrome or some such.)
- Make a text file with one line per cookie. This bit might be tricky, but it’s likely to be quite finicky. You’ll have to use a text editor and not one that saves in plaintext, not rich text. (Think Notepad, not Microsoft Word. Notepad++ will work nicely. I don’t think Notepad itself will work because…) You’ll probably need to use Unix-style line endings.
- Each line will need to follow the format outlined here. Note in particular the tab characters, not spaces.
- Save that text to a file named “cookies.txt” in the directory where you’re running wget and add to your wget command
--load-cookies cookies.txt. You’ll need to add that right before the url.
If you’re on Windows, you can get wget here.
And, honestly, all this is a little like programming, really. Unfortunately, I’m not aware of any friendlier kind of apps for this sort of thing. Hopefully this ends up getting you what you’re hoping for.
Second the first bit. I’m a software dev and it’s generally known that anything I write for a company is not mine. I can talk about it for future employers, I can think about how I solved past problems, but what I wrote is the company’s. I do not own it and I claim no ownership.
Op here is not only looking at disciplinary actions but possible legal action taken against them if they try to rip this stuff without written permission.
If you do it for a company, it’s the company’s property. Not yours.
deleted by creator
Ah! Ok. From what you said, it sounded like it was something only available on an internal network. Like a corporate Wiki or some such. It’s totally possible I just missed something in the original post that would have disabused me of that incorrect notion.
(Disclaimer, I am not a lawyer and like to geek out on intellectual property law studies sometimes, but ultimately don’t really know what I’m talking about here. But…)
If it’s publicly available, that probably doesn’t make scraping a copy legal. (It’s called “copyright” because it places restrictions on “copying” (though it also places restrictions on other things like “public performance” and such.))
The knowledge bases being publicly available might make it less likely that they’ll pursue you over anything. Maybe. But that depends on your boss’ state of mind and such. One thought might be to do the scraping through Tor, but if you’re doing this scraping close to the time you’re leaving the company, that may not really give you any significant amount of palusible deniability.
Now, the fact that it’s available on the open internet may open up some other options. If the number of pages in question isn’t too terribly large, you could request one-by-one that the Wayback Machine save pages. (The “save page now” section at the bottom right.) That would mean a) no learning curve b) it’s not you copying things so if there were any legal issues, it would most likely (again, IANAL) be Archive.org fighting them not you. That archive would stay up for a long time. You wouldn’t have a copy locally, but if you had it bookmarked, it’d stay out there on the internet in a frozen state for you to reference any time.
Aslo, if you did go the
wgetroute, you wouldn’t have to deal with the cookies or anything, which makes things easier.
deleted by creator
If you have a linux machine, wget can create a mirror copy of a website but it does require following the URLs. While you are still there, do you have the access to make a direct copy of the database and the back-end code that drives the website?
As an alternative that might help show the changes you have made, if this page is exposed to the internet then you could check the internet archive, which will have snapshots from different dates that could provide a comparison of the changes.
deleted by creator
Yeah I figured the solutions wouldn’t be quite that easy since you were here asking, but sometimes it helps to get the simple ideas out of the way first so others can come along with better methods. Good luck on your search!
I’m curious, since you mentioned it: What degree is it that qualifies you to edit knowledge bases?
deleted by creator
