guide to Tumblr grabbing tools (I update this as much as I can)

Grabbing tools:
 
TumblThree  
Difficulty: Filly  
Pro: Easy to use, and can grab enough to recreate the tumblr in a basic way (preserves posts, pictures, audios, videos, and texts, with correct links).  
Con: Compared to grab-site, does not fully recreate the tumblr. Compared to bbolli python script, does not generate readily viewable outcome website and requires cleanup later.  
How to install: follow the instructions.  
How to configure: Other than the default settings, do these changes:
 
Set image quality to “Raw” (doesn’t seem to help, since tumblr stopped allowing Raw download, but doesn’t hurt either)  
Login there for access.
 

 
Tick all the boxes on the top, especially the “metadata” part.  
Set metadata format to json  
Click “Dump crawler data” to get the texts in the posts, to preserve the post structures (like, which image appears in which post in what order). This allows future reconstruction of the blog.  

 
bbolli python script  
Difficulty: Teen  
Pro: generates a standalone html page that structures the posts as in the original tumblr.  
Con: There’s an issue with API key limits. See below.  
How to install: follow the instructions.  
How to use: follow the linked guide, or this easier guide.  
Issues: it runs out of API keys fairly quickly, since Tumblr API request rate is limited to 1,000/hour, and 5,000/day. If it throws errors that mentions something about “api”, it’s probably this problem. You would have to make the script use a new API key. I have yet to figure out how to make it change API key.
 
According to @randomGuy 339D, the following command avoids the API issue, by making sure to retry upon API errors, until the API key works again. It reads lines from the url.txt file and runs the backup script on each line.  

 
if you want to archive only non-reblogs, you should add the --no-reblog flag.
 
New Issue: It would sometimes fail to download pictures errors. In such cases, it generates html files that point to the online pictures instead of the offline ones. I’m trying to write a script to fix the problem by grabbing all the error messages and using wget to get the error urls, then modify the local html files so that the urls in them point to the offline pictures instead of the online picturess.
 
ripme  
Difficulty: Filly  
Pro: Easy to use.  
Con: Can only grab pictures. All texts will be lost. Needs API keys to work, so it also risks running out of API keys if you mass download. Personally, I suggest using this only as a last resort.  
How to install: easy.  
How to use: easy.
 
grab-site  
Difficulty: Mare  
Pro: Can grab everything, allowing the tumblr to be fully recreate later.  
Con: Harder to use. Takes more storage space.  
How to install: follow the instructions. Requires working with command line, and Linux/iOS/VM Linux on Windows.  
If you can install it successfully, go to the Discord server of The Pony Archive and look around #announcements, #savetumblr, and good luck. (I still can’t get it to work.)
 
Tip:
 
Always check if the tool actually grabbed the things. Open the folders and check if it’s empty, give a quick glance to see if there are weird broken things.
 
How to get new Tumblr API keys

Some other useful tools:
 
The Internet Archive  
And many other archives
 
Google Cache Grabber  
for when you want to check if there’s google cache of that site

Link to this added to the OP of the other thread.

More detailed information about using bbolli:
 
bbolli does not work with hidden tumblrs as far as I can see. For those, use TumblThree or some other tools.
 
bbolli also has this weird tendency to fail to download some pictures. I found a hacky way to work around it.
 
This assumes you have run the python script without using the -D flag, so that you get a directory structure such that a picture’s address looks like ~/blogs/ask-pony/media/tumblr_inline_nlcdauePXn1rbezzw.png
 
Then in the folder ~/blogs/ask-pony/, you can run the following bash script  
grep -r ./ --include=**.html -Pe 'https?:[^"<>]**?\.(png|jpg|jpeg|bmp|mp3|mp4|swf|gif)' -oh | awk '!x[$0]____' | sort > urls.txt  
which searches through all the html files in the archive and check for external url references. Those are de-duplicated, sorted, and saved to urls.txt
 
Then, you can download them and save them to ~/blogs/ask-pony/media by something like wget:  
wget -N -P media/ -i urls.txt -nv
 
Then, you can run the following python script IN THE FOLDER ~/blogs/, and it will rewrite all the external urls to link to the local files instead.  
Script

Is there any way to download raw images with either TumblThree, or with modifying the image link at Tumblr?
 
I also haven’t been able to get TumblThree to work

@~M0noponi3  
It seems impossible to get raw images, except by reblogging them to a tumblr you have, then archive your own tumblr.

@furrypony  
Alright

Been meaning to post this for days, but just getting around to it now, ugh.
 
If anyone can ask: do you have to be a member of that website to be able to download/get this stuff? I’m not really a coder, so I dunno if doing it for myself (in addition to everyone) would get anything, if I don’t know what I’m doing… as right now, I don’t see anything about how to download these downloader programs that are all part of the same website, apparently.
 
I mean, don’t know how much I can get/relocate with certain files I saved without recording who did it… before the date in question, when everything vanishes. Much less everything that I always hoped I’d save one day, but never got around to… ugh. Heck, even discovering new things in the time that’s left, from wherever I first saw this notice about it yesterday before it finally registered with me on here, heh. Like how apparently, I could always find a “raw” version of the full image(s) if it was bigger than the file size limit on tumblr, even though I dunno if everyone knew this, given how sometimes people would link to another site with the full size image(s). Shame I just learned about this NOW…

And also: from what I’ve gathered from an artist on here who was on tumblr mainly before this “purge” (ponkpank), tumblr is preventing new uploads that’re NSFW before the actual date. But I dunno if along those lines, it’d be possible to create a new blog with less than 3 days left until it occurs, in order to reblog this stuff to and then save it?
 
As I have a primary tumblr that I’ve BARELY used in all the years I’ve had it for posting/reblogging and only reposted one thing NSFW to it, as well as probably under that category a couple of NSFW reblog posts about being NSFW and a clop artist (well, wannabe clop writer), which can be easily removed after the date or before it. And have already edited other such parts of my page that would’ve flagged it perhaps (aside from the reblogs about being NSFW/clop writer) and saved the one pic, even if it was already on another website from years ago at the same resolution I saved it under. So I don’t have to add those ones to it and then remove all of them, even if I don’t foresee using tumblr much after this purge occurs…

I’m sorry I want to help save the blogs but can’t, I have no clue how to use one of those tools

TumblThree would probably be a great tool if it actually worked.

I haven’t gotten it to work too