Once upon a time, I received a request to download musical compositions from the website bananastreet.ru (which mainly features mixes, remixes, remakes, etc.). It seemed like an easy task to complete, but in reality, it turned out quite differently. What made me write this post specifically about this site and data parsing via :ZennoPoster: — read on...
It's a beautifully designed project with no direct way to download music files, BUT! Has that ever stopped anyone...
So, what makes collecting data from such sites so challenging?
Ajax or Flash player
The most common answer to this question is the presence of a notorious Ajax or Flash player on the site. While the first one is relatively straightforward — you just need to parse through an internal browser — the second component is trickier. You either have to decompile the video itself or collect code from various fragments (which is what I did when working with bananastreet.ru).
bananastreet or free music downloads!
If you're reading this post, you've probably already figured out that we want to obtain the data for free, not through a monthly subscription to someone out there. That's why we extract certain data from the code (I won't specify which ones to avoid encouraging more freeloaders :)), then combine them (the track itself) and convert everything into a single file — an mp3 file.
In short, the "hassle" primarily lies in figuring out which number to use for parsing the track (this programmer's trick still needs to be worked around), but the process itself is as follows:
- take the URL with the list of songs,
- then add them one by one to the desired folder on the hard drive,
- and finally, obtain the finished tracks and their list in a text file (so you don't forget what you parsed).
Script update and performance improvements
Later, it was decided to completely rework the script — now it works on a completely different principle:
- select the necessary URLs from the sitemap and insert the list number into the action,
- then parsing is done without a browser, which increases performance and reduces load,
- all mp3 files are taken in their original form, so there's no risk of missing a track,
- parsing can be done across all 32,602!!! albums listed in the text file and available to anyone (who, of course, wants to pay for music downloads).
Attention for those who want to purchase the parsing script!
THE SCRIPT ITSELF IS NOT FOR SALE!
bananastreet results
Almost any project can be parsed using :ZennoPoster:, if you know how. And if you don't, you just need to think and figure it out!
https://www.youtube.com/embed/ydYNYB6vwDc
[maxbutton id="1" text="Order the script" url="/contact" linktitle="External link | External link" window="new" nofollow="true"]
Was this helpful?
