/hydrus/ - Hydrus Network

Is it possible to remote control Hydrus from another computer running Hydrus?

For example, if I install Hydrus on computer 2, Hydrus 1 would be the host. If I'm on computer 2, I could pass links to have them download on Hydrus 1, I could also view all images and search stuff from Hydrus 1 on Hydrus 2. 

It'd essentially be connecting to Hydrus 1, from Hydrus 2.
Nothing gets downloaded on Hydrus 2, it just is a remote control for Hydrus 1.

I know you can use the API for stuff like this, but I couldn't figure out if I can connect to the API from another Hydrus Network install. Not seeing where I would do that.

I made a ton of changes to various preinstalled url classes and parsers. How do I update hydrus without having them all overwritten with defaults?

how could i use the PTR to tag images that are already in my db?

 >>/983/
Tags are automatically grabbed for images in your db that it finds matches for.
You can also manually tag stuff.

Hey, I am sorry, catching up after a long delay.

 >>/941/
Not yet. Some network job controls let you override error delay from the little cog icon, but there's a domain-level block I have hardcoded atm. Ultimately, I will expose that system in some new UI to govern current domain status across the program and let you reset it.

 >>/942/
I would like to maintain a bunker for those who are uncomfortable. I hope we can keep talking here.

 >>/943/
I have improved some of the UI labels on this dialog. I hope it is more obvious now. basically you put the 'bad' tag(s) on the left, and the 'good' tag on the right.

The list at the top is what is currently stored in your database. When as you enter tags in the bottom, either side, the UI shows what those are already involved in, so you can see if you are overwriting or joining existing relationships.

One of these days I will carve out a few months to write the whole thing and have drag and drop networks of arrows.

 >>/948/
Hey, I think we already spoke elsewhere. If not, there is no way to do this en masse. But if you just have one or two parent relationships to change, just enter in what you actually want, and the dialog should say like 'hey, this conflicts with what is already stored, do you want to overwrite?', and if you hit yes, it will automatically petition for the old stuff to be deleted.

 >>/950/
If you are on an HDD, unfortunately yes. I do not recommend syncing to the PTR on an HDD any more. There are just too many tags. About a billion mappings in the current PTR, which is ~42GB of complicated data on your drive.

If you are on an SSD, most users get around 1,000 to 5,000 rows a second once they are synced. You should see peaks of 50,000+ when you are earlier in sync.

If you are only getting 100 rows a second or similar, please pause your sync. If you are on an HDD, I recommend you pause it permanently and resume whenever you next update your hardware. If you are on SSD with those kinds of speeds, please let me know. Sending me some profiles as under help->debug->profile modes would be helpful so I can see what is going on.

 >>/951/
Thank you, I believe this is now fixed. It turns out before '6+girls' was only working when it was mixed with other tags in a multi-tag query. Let me know if you still have trouble.

 >>/957/
I do not know much about this myself, but I think this is the current guide:
https://github.com/CuddleBear92/Hydrus-Presets-and-Scripts/tree/master/Downloaders/Gelbooru

If you did that and it still doesn't work, I guess hit network->data->review session cookies and then clear everything and try again.

 >>/958/
In hydrus, pend is add, petition is remove. Let's say you open a manage tags dialog on 10 files, 6 of which have 'character:samus aran'. You then double-click that tag. You then have the options:

pend samus to 4 (add that tag to the other four that don't have it yet)
petition samus from 6 (remove it from the six that do have it)

The terms 'pend' and 'petition' are different from 'add' and 'delete', yeah, because when you are talking to the PTR, all changes you make wait in the 'pending' menu up top. They only actually become 'real' when you commit by uploading to the server. You get the little (+) and (-) symbols after the tags, meaning they are 'waiting' to be added/deleted.

Thank you, I will clean up the help on this, and the UI.

 >>/960/
 >>/961/
Encryption is notoriously difficult to get right. I am not confident I can do as good a job as third party programs, so I strongly recommend you figure out a solution using one of them. Hydrus works completely fine out of a mounted encrypted partition as ciphershed-style software will create.

Syncing to a remote server is not as simple in this situation, but perhaps a program that encrypts individual files would fit into your workflow. Or if your remote connection is secure, and the client and server are also operating out of encrypted containers you trust, it may work.

It is generally not safe to touch the database file while the client is running. I know google-cloud backup has corrupted a hydrus database it was watching. I recommend you only do backups while the client is shut down.

 >>/963/
Are you synced with the PTR? That makes your four client.db files about 40-50GB now. The internal backup is basically just a copy of your 'install_dir/db' folder, which is:

The four client*.db files.
256 'fxx' file folders under 'client_files' folder
256 'txx' thumbnail folders under 'client_files' folder

 >>/974/
Not much yet. Everything at the moment is here:

https://hydrusnetwork.github.io/hydrus/help/client_api.html

You can queue up URLs for download and fetch simple searches back, and files/thumbnails.

I hope to keep expanding it, including eventual full session control so you can make new pages and so on.

 >>/977/
Giving them new names should do it. The basic 'new defaults' overwrite I use in my update (whenever I say 'rolled in a new blahbooru url class' in a changelog) works on name.

 >>/983/
 >>/986/
Yep, mostly you just have to wait for sync. It'll take a couple of weeks, and you need to have an SSD with 50GB of space. As long as your files are 'popular', the same as those stored on the big boorus and other gallery sites, you'll see the tags given to them by other anons on hydrus.

this is fine png
(64.32 KB, 1604x534)

 >>/938/

does it have the ability to strictly isolate local file/tag repos within the same client yet? (e.g. separating sfw from nsfw content completely)

poked around at the latest version for awhile this morning and read a lot of the changelog but didnt see anything obvious -- collections look close but not quite? i remember reading months ago about strides dev was making toward this, and i usually do read the release changelogs, but havent heard much abt this in awhile

Could a way to automatically save tabs be added?
For example, if I have a tab for watched threads, and accidentally close it, it seems I lose all my watched threads in there?

For some reason, it seemed to do this earlier, although I don't remember closing it, but yeah. I have to go back in and reinput all the links to be watched now.

 >>/1004/
you can manually save a session of your pages, which can then be appended to an existing session if you accidentally close that tab. see the 'pages' menu -> 'sessions.'

making this automatic is already implemented in two ways:
- there's session restore, and a decent history of previous sessions saved.
- the 'undo' menu. you can reopen a closed tab within the same client session.

does reopening that page from the undo menu not preserve the tab's configuration for for a watched thread page? is that what you're asking for being improved?

 >>/1005/
> see the 'pages' menu -> 'sessions.'
ah okay, this is what I was looking for.
 >>/994/
> You can queue up URLs for download and fetch simple searches back, and files/thumbnails.
I'm confused on your reply here, you're saying it IS possible to use Hydrus Network client on a second computer to remote control it's just limited?

I am familiar with the client api, as I am using Lolisnatcher Droid with Hydrus for mobile.

My question is basically, where do I go to input the client api information on the secondary Hydrus Network client.
And how does it grab the pages? Do I just go to new page, and choose files and search and it pulls everything from my other Hydrus client?

I'm just confused on the setup on the remote machine when using Hydrus Network client.

Also unrelated, but if I may suggest adding Nginx example configurations to the Hydrus docs.
Thanks.

Is there any way to force a download from a server? I'm just using this to sync files across personal computers, so it seems pointless to wait a whole day for updates to sync.

Hydrus is extremely bloated, I think it would be amazing to see it come with the bare minimum, then allow users to add on whatever features they want.
This would allow for better security, better performance, less bloat, and I would imagine a bit easier to maintain and focus on stuff individually during development.

On a sidenote, is it possible to move the PTR to one drive, but keep hydrus and the db on another? I see no options to select ptr location.

I also think the docs need a major revision, and a search box.

Thanks

I'm using some external tools to dedupe files from the client_files folder. Is there a way to have hydrus scan the database for missing files and remove their entries? Right now I'm just going through each file in hydrus, find one throwing an error message, and then manually delete it from the database.

 >>/993/
> I do not know much about this myself, but I think this is the current guide:
> https://github.com/CuddleBear92/Hydrus-Presets-and-Scripts/tree/master/Downloaders/Gelbooru
> If you did that and it still doesn't work, I guess hit network->data->review session cookies and then clear everything and try again.
Back again over this issue, I've tried both methods and still can't get it to pick up my blacklist. Should I try making an issue over on that github?

 >>/993/
> It is generally not safe to touch the database file while the client is running
I don't mean set rsync on the entire database directory.

I mean in Hydrus when you "update database backup", it backs it up to a separate folder of choice. So it's syncing the folder hydrus backs up the db to, not the hydrus db itself. 

Is it safe to set rsync on the folder hydrus backs up to is my question?

Is there a quick way to transfer or send an image with all the tags url info etc to another database?
I wish i could just drang and drop because the current db i have on a usb drive is now full

Is there a way to automatically tag images with the page: namespace? I want to import a 100+ page comic so I really don't want to do it manually

 >>/1024/
Use the add tags before importing button and you can add sequential page tags, or use regular expressions to grab the page number from the filenames

 >>/1025/
I'm going from the site directly to hydrus through a gallery importer, I guess I'll maybe try to script the download or something in that case.

Hi, we talked about supporting the project with translations (of the help files) before and I've only now gotten the glorious idea to ask you in what way submitting them would be the best/easiest to handle to you.

Also once the URL mappings maybe turn hydrus into the next torrent network, do you have any ideas on how to keep the annoying pairing of bad content and lawyers out?

I've tried to log in to sankaku, it seems to login just fine, but I still cannot download lolis, even through Hydrus Companion.

im trying to mass delete files and leave a custom reason/comment.  this turned out to be so slow, is it because its on a usb drive, or can i tweak something in the options to speed things up?

2021-05-06... png
(12.09 KB, 545x431)

any way to limit the gallery downloader window from getting too long when you start adding items?

Is there any documentation at all for the new docker container? The suika one stopped updating

How do I construct an API call that would search the files by tag as number? What about by date?

Is it possible to just grab the tags used in the PTR without syncing to it? I have an HDD and can't sync the entire thing because of it. If I can just grab the tags I don't need to worry about fucking up tags later when I do sync.

can the program import tags from famous booru sites (sankaku, gelbooru, rule34, danbooru...)? i have been using an addon for firefox that downloads images from booru sites into /folder/copywrite/artist+character+website+MD5 hash. and now i have over 40k files divided over 2k folders with very niche copywrite names and artists. all handpicked and downloaded over like 7 years period. its getting way too big for me to consider organising these files into a respectable gallery manually. and i barely open that folder because it makes me depressed and im more used to the booru tag system by now because it does the job and there is more content being added there. but at the same time i dont wanna snap and shift+del the folder because i know i have some good shit there.
i haven't tried this software yet but i'd really love if it can grab the file md5 hash and import all of its tags from other booru sites. such function would be a huge help. then i can go on and manually decide what pics and other files are worth keeping and what is not.

noob question: how do I cancel the backups ?

Is there an undocumented length limit on the system:tag as number comparison queries? The GUI won't let me enter more than 8 digits for the ~  operations, while I need 9 for my custom ids and 10 for dates as numbers. The client api also finds nothing for those queries.

1623780545280 jpg
(18.86 KB, 599x400)

Probably a stupid question, but I haven't found an answer anywhere else in this thread or the documentation. Basically, I want to be able to add tags while at my laptop, while having the massive database on my server. Does the client and database need to be operating on the same computer or no?

1635410342528 png
(83.36 KB, 703x709)

why is it my hydrus not getting the nsfw pics from pixiv? I already log into my account

8CVQVl2D4V png
(19.51 KB, 498x391)

I've changed the install dir from the original install (it was on E:, now it's on C: since it's an SSD)

When using the Windows Installer to do my updates, it remembers the old file path, and when changing to the new file path it wants to overwrite my existing install rather than update.

Is there a way to clear the cache? Or have it detect an existing install?

How do you get the filenames for Deviantart images to have the name of the artist and the title of the image they post with the end numbers you see in the link?
For example this link image on DA:https://www.deviantart.com/expressthechaos/art/Martian-base-895652671 

Numbers at the end Link ArtistName - TitleName image
And this is how it would look for example:

895652671 expressthechaos - Martian base

Is there a way to collect files by tags beyond what Hydrus presents me by default (chapter, creator, etc.)? If not, are there any plans to implement this in the future?

Pretty confused by the vague documentation at the getting started for the server and client setup.

I am making account types, and I made one with the title downloader.

It tells you to give it the get_data permission, but from what I can see there is no way to do that. There is nothing for setting permissions like that anywhere, no further explaination at all on the faq.

Capture PNG
(22.4 KB, 429x556)

 >>/1269/
here's what i mean, the documentation is repeatedly vague and so i just can't find where im supposed to set that permission. It's not here.

secondly, i'm not sure how im supposed to actually connect to this from another machine locally, am i expected to just my OS's native network share function and just have sort of a shared installation across machines accessing my hydrus locally? 


I was hoping for htis to be an easy way to access and manage all my images across PCs

fucking hell this is some trashy ass software
ill just use a samba server

i've been using the PTR for a while, and most of the tagging I've done has been done in the PTR rather than my tags. but i think the PTR is taking up too much space on my SSD

if i use "migrate tags" with the default settings, can i copy the PTR tags to the local my tags for JUST the files that I currently have, so that I can remove the PTR without losing all the tagging I've done up till now?

How do I dl danbooru pools & groups?

The e621 login script seems to be half-broken: when trying to login it consistently gets blocked by cloudflare, but when testing it on the "edit login script" page it works perfectly.
This happens even on the same IP, and with less than a minute between both tests.
Anyone has a clue as to what might be causing this difference?

when using stuff like "collect by title" in the search page, is it possible to sort by random for the overall results page but sort by page number within the individual collection? 

e.g. say I have a few titled image sets with red_hair tag, with 2-10 images in each. when I sort by random I would like to see a random image set, but then as I scroll through the page I want to see page 1, page 2, page 3 etc from that image set. right now, if I sort by random and group by title, I will see pages in a random order like 6,10,4,1,9, etc

Is there way for nitter importer to apply post date as file metadata date and if it is possible for multiple images in one tweet to have relationship between each other?

How could I set Hydrus to download the entirety of e621? I'm new to Hydrus but want to backup the site due to how much seems to get removed from it

Screenshot... png
(231.29 KB, 599x1174)

Database is locked after my pc was force shutdown yesterday. I have a backup from yesterday, but ideally, I'd like to restore all the changes I made after creating the backup. Any way to fix this?

I changed to a new PC and am trying to move my hydrus database/files to the new computer from backup. I want the contents of the database on a secondary drive while the program is on my main drive. I followed the help page to migrate the database location to my secondary drive (https://hydrusnetwork.github.io/hydrus/database_migration.html#intro). But, now I can't import my backup because it says my "database is stored in multiple locations." But there's only 1 location in "migrate database" window - the new location on my secondary drive that I specified. How can I import my backup to the database?

Hello
I lost my db, and so my backup is a few weeks old.  all my files were on another drive.
how do I get the db to search through its client_files and import or find files not in the db?