/hydrus/ - Hydrus Network

Asked this in the faq, because the answer I got is no, ill post it here

"Is there a way to maintain zoom across multiple non same resolution images in duplicate mode?

I have many images that are questionable if they were resized up or down, and with the difference in resolution its difficult to tell because I need to flip back and forth fast."

I think this is a needed feature a relative/maintain equal size flip between images zoom option.

>>5947

You can use the page of images downloader to download from all the archives like warosu and desuarchive. No need to do it manually.

>>5950

As of the latest version you can for files with the same ratio.

>>5952

It won't just download the thumbnails then? Thanks anon!

May have been a thing in 256 already, but I can't now scroll to the next image in media viewer if it isn't the active window.

Same person who noticed speed boosts in running from source here.

I'm seeing the same improvements in v257.exe

Seems like the Pillow upgrade made a major improvement in handling large quantities of thumbnails etc.

Thanks!

First of all, thank you for making this program. I just found out about it, and it's the best thing I've downloaded in years. Absolutely everything I need. I'm only really having a couple of issues. I searched the booru and the FAQ to try and find the answer, and I found that using regex and setting pages, chapter whatever works to put the files together when I click the group sort box. I'm wondering if there's an easier way to fix this problem that I am unaware of. I tried reading the regex guide and it confused the hell out of me, so I just copied the regex code from your images on the booru. Anyway, here's my problem.

I have many images saved and many of the images come from multi-part collections. Like the multi-image posts on pixiv where they have different variations or a small comic or something that's in a sequence of images. Sometimes as few as 3 sometimes a whole lot of them. Anyway, I need a way to group them so they aren't scattered all over when I search for a tag. Let's say there's a 3 image sequence of donkey kong eating a banana. I search for donkey kong, is there a way to see the normal donkey kong images but have the 3 image set grouped together and not scattered all over the other images so I have to search around for it or add more tags to the box before I can find them? This seems like a big problem right now for me because I have probably 30 of those tiny image sets in each themed folder full of various other random images. I want the ones that go together to appear together.

Sorry if I'm terrible at explaining things. I'm shit at programming and my computer skills have gone into the shitter over the years. Any help is very much appreciated. For now I'm just saving each small image sequence into a separate folder and just importing the single images that don't have any others in sets.

>>5960

Add the title of the imageset as a tag, search for Donkey Kong, then add the title to your search. For example. Or, if you have everything tagged, go up to the top, under the tabs, go to the drop-down list and select "sort by series-creator-title-volume-chapter-page", and even if you just have the artist tagged, they'll be grouped together.

>>5961

>>5965

Thanks, that helps a lot! I think I have the basics of it down now. It just felt weird having a useless tag for hundreds of different image sets just to tell them apart, but I guess they won't get in the way. Like having a page tag with multiple different things having the page 2 tag. I guess that's the way most people do it though?

By the way this is the best program ever.

Another question from a noob. For boorus like e621, when using a parser, is there a way to automatically tag for pools? Or do I just have to go in after they are downloaded and select the files to add a tag for each image pool?

>>5968

There is no system currently for pools.

Some people have added a pool namespace but it's seems a soso temporary workaround/solution.

>>5967

You can use a "collection" sorting to arrange by set your image.

It's located above the search box and should indicate "no collection"

You should hang out in the discord chat and search arounf, there's a lot of useful discussion there.

Are you going to make the thumbnail view of the dupe page a bit more useful? At least add next/previous pair buttons instead of the random pairs button. But maybe it's a waste of effort since everyone seems to be using the dupe filter instead. I just don't see why it's there if it's gonna stay useless.

>>5971

>Are you going to make the thumbnail view of the dupe page a bit more useful?

He said he's gonna put dupe options in the right click menu so you can mass-mark files as alternates/dupes.

>>5957

you can set it to download images and download the links to images.

Update on this bug: >>5886

It happens in the full screen media viewer as well, not just in the duplicate filter.

It only happens if you try to remove a tag from the PTR and the "Enter a reason" window pops up. Once this window has popped up, and you close the manage tags window, Hydrus will focus on and bring to the front the main window instead of the fullscreen window you were using.

Can confirm that known urls still don't show up on page B/white background in the dupe filter.

Also I have a situational question, let's say image A is an alternate of images B and C, and that image B and C are dupes of each other. If the dupe filter compares A to B and I mark it as alternate, and then it compares A to C and I do the same, how would I get it to compare B to C? Or will this happen eventually since it's somewhat random which ones show up first?

>>5997

>Also I have a situational question, let's say image A is an alternate of images B and C, and that image B and C are dupes of each other. If the dupe filter compares A to B and I mark it as alternate, and then it compares A to C and I do the same, how would I get it to compare B to C? Or will this happen eventually since it's somewhat random which ones show up first?

If A is alternate of B and C, hydrus does not make assumptions on the relationship between B and C, so they'll eventually be shown to you so you can make a choice like normal.

This program is amazing. I have many suggestions for improvement, but quite a few of them are just basic visual things and I'm sure you're not wanting to spend much time in that area when you have the meat and potatoes of the program to deal with. Things like complete dark ui toggle which makes the entire UI dark mode, I think this might actually be a bit complicated, but I'm not a programmer. Also the psi icon needs an upgrade. I'm using a black taskbar in windows 10 and I can't see the icon at all. I can try and make one with the same character if you want. I'll just upload it here when I do I guess and whoever wants to use it can use it.

I have so many image files saved all over my PC right now I am going insane. This is the best thing ever, but at the same time it's getting me to realize just how much I collected over the years and just stuck into random folders in folders in folders.

I have a couple of questions, let's say I want to download a single file off of let's say gelbooru, and I want all the tags to automatically be applied like it does when I download a gallery/booru. Only instead of an entire gallery, I just want to download a single image quickly without setting up a subscription to the tag or anything, just download and go. What is my best course of action for this?

One more question. If I dump tons of non-tagged images into the program and then I download a bunch of files from boorus using the gallery downloader, it will automatically apply tags to those images that I already have if it finds the image in the gallery, or am I setting myself up for hell? I have hundreds of thousands of images and frankly I don't have enough time left on this earth to tag them all. I was going to mess with the tag repositories, but I don't want to be connected to the network, I just want to use my local stuff and a few tags from the boorus when I download a gallery.

Thanks to anyone who read that.

Will it be possible to do whatever things you can do with the filter without the filter?

>>6000

The icon is stored in /hydrus network/static/ so if it really bothers you you can change it yourself.

As far as I know there is no way to download single images from a booru. You'll have to download it manually and apply the tags manually I think.

I think it is possible to download just tags, but I don't know how. Don't use the public tag repository, I've heard it has tons of terrible tags.

>>6007

Thanks for that awesome reply, I got everything to work perfectly. The tag fetch for single files is going to be very useful for me. I downloaded the scrips also and installed it. Do I delete the old gelbooru md5 GET script, or should I keep them both there? It looks like it also fetches the file ratings, is that what it does differently?

Also I'll check out the discord when I get a chance, thanks!

>>6008

No need, you can keep all the scripts you want. The different thing it does is that it fetches tags through IQDB rather than directly from gelbooru. IQDB is an aggregator, which means it'll reverse-search the picture rather than searching for the precise hash, and it will find tags even if you don't have the exact same picture as the one on gelbooru.

Trying to download https://www.youtube.com/watch?v=ioaex1oSw7Y

Exception
Could not fetch video info from youtube!
  File "/home/hydrus/Desktop/hydrus/build/client/out00-PYZ.pyz/include.ClientGUIMenus", line 122, in event_callable
  File "/home/hydrus/Desktop/hydrus/build/client/out00-PYZ.pyz/include.ClientGUI", line 2488, in _StartYoutubeDownload
  File "/home/hydrus/Desktop/hydrus/build/client/out00-PYZ.pyz/include.ClientDownloading", line 218, in GetYoutubeFormats

unknown url type: unknown

>>5977

I posted that >>5950 right after the q&a thread told me it didn't work, as I asked it shortly before or after the update I forget which.

If in duplicate more there was a relative zoom, as in either the height or width is matched that would be perfect

On a side note I have been looking at other features and seeing what more I can do with the program.

With thread watcher, I have one thread that 404ed, because 4chan does not keep an archive of the board, and hydrus knows that as if I check now, it tells me it leads to a 404

I'm not sure what it does to threads that hit an archive status, as im not useing it for any that have yet.

However, if the thread had a 'change name when 404/over' option so instead of 'thread watcher' it starts as 'thread watcher' till you give it a thread, then it changes to 'watching thread' and when it hits a 404 or sees text on the page, 4chan gives a

'Thread archived.

You cannot reply anymore.'

Message for me, if it sees either of those, it turns from 'watching thread' to 'thread done' that would be immensely helpful.

With thread watcher too, if there was an option to just keep checking instead of a check X times limit that would also be helpful, currently I have it set for once an hour and 7 days worth of checking times, I could probably set that to 999 and be done with it, but looking at some boards and some sites like 8 chan can span multiple months, a button to definitely check would be very welcome.

Then with the download page of images, as i'm going though a fairly massive back catalogue right now this hits me harder then it probably normally would, but when I add a page I check it to make sure its on the list, and I add lets say I add 8 pages, I now have to scroll to the bottom of the list to make sure the page I added was in fact added. being able to invert that list would be helpful, or the scroll automatically going to the bottom to see the most recently added thing would be helpful.

On downloading page of images, I personally find the process bars useless once you have more then 1 page in the queue, the single image download one because either I have that fast of an internet, or because the program is laggy and sticks, usually only see that bar move for a fraction of a second. the total bar, well, right now i am at 6200-6400 and when a new page adds more images the bar is basically full green so I have a suggestion for that, because I don't want to get an image editor just imagine editor installed right now (new computer) imagine this is all centered or takes up a full full line

———————————–

Imports

Current page urls - ####

#### successful - #### failed - #### failed to timeout - #### in db

Processing ####/####

(Current image) [———————-##%————————–]

(Current page) [———————-##%————————–]

(total queue) [———————-##%————————–]

———————————–

With this, adding the ability to pause the download (already there) and process the pending page urls all at once instead of when they come up, this would allow the queue to instead of be 125, then process a page and jump to 523 images, would instead reflect the total process better.

With that, the 'failed to timeout' is something I would like to address, I probably processed 100 or so links before I went through the log, found out that a number were failing due to time out, and inside the log, you can already retry the images by selecting them and telling them retry, a button for this outside of the log would be convenient.

>>6013

on a side note, and I ask this because where i'm currently pulling from seems to have little issue with me attempting to load entire archived threads at once, an override function that lets me download more than one image at once would be nice, using qtorrent has spoiled me a bit on i get to set 2 different download/upload rule sets at the click of a button. having a button like that so I can activate it when sites don't care. I don't know if its a program limit to only be able to download one at a time, or if it's a limit you imposed, but it would be nice.

Now, there is one more thing that I can think of that would not be programing based that you could do, as you know the program best and would be the best person to do this. the whole download/thread watch/ other things I have yet to do are things that are not explained in the help, or at least are not jumping out at me in the help. Im also going to be honest, reading text about what to do when there is no 'look at this, you did it' effect, such as the download url, its a bit hard to understand the program, and with how seemingly easy it can/could be to set it off on a download ALL THE WEBSITE kick, a video series where you/someone you trust to explain shit right for youtube would be amazing.

Baring a video series, a revamp of the help to a wiki like format where anyone willing to contribute could step in, or a wiki like format as an aside for the help would likely be helpful to everyone.

Sorry if that was a lot or assish, just had a lot of thoughts after using it for more then just my personal image archive programs for a few days

>>6034

Thanks, I bet that's it! I will look into it this week.

I'm going to take some time off next week.

>>6030

>Let me know how it works for you.

Yeah it's fixed now. Thanks!