>>1715
I can't build venv, error on building psd-tools.
Looks like now I need VC++ build tools, even though I didn't before.
After some effort I instal 140 and 141 version.
Now while building psd-tools:
...\python38\include\pyconfig.h(205): fatal error C1083: basetsd.h: No such file or directory

stackoverflow says I need Windows 10 SDK, which won't install.
stackoverflow says I need VC++ Compiler for Python 2.7, which can't be downloaded from microsoft anymore. I find it on wayback machine, Install, it does nothing.

I try other solutions, find this BaseTsd.h in program files and put it in python folder, errors become unreadable.

Damn, I am sorry to say we might just not support Win 7 any more. It sounds like setuptools or some other library that updated in the background now uses python 3.9 tech, or similar, that Win 7 won't do any more. We have known this was coming for some time, so perhaps it has finally hit us. I will update the help to talk about this.

It might be possible to get things to work on old libraries if you were to build your own requirements.txt and go through the advanced/manual version of the 'running from source', as here: https://hydrusnetwork.github.io/hydrus/running_from_source.html#what_you_need but I am quite experienced at this and I would find doing that quite a pain in the ass, so I think your best bet would be to just roll back to v578 and halt there for now.

I am not the world's best git expert, but (I'm assuming you used git to download the hydrus source) if you open a terminal in your source folder and run this:

git checkout tags/v578

I think that will rewind you, undoing any git pull you just did. It may be that v582 or later works for you too--the tag should line up with the tag in the URL of each release, as here: https://github.com/hydrusnetwork/hydrus/releases Sometimes it will be v123a with the 'a' on the end when I had to do a hotfix.

Good luck, and let me know how you get on.

>>1717
Oh, actually, since your problem is with psd-tools, which I am pretty sure is optional, if you navigate to install_dir/static/requirements/advanced and edit 'requirements_core.txt', try removing the 'psd-tools' line and try running setup_venv again. It will skip this library.

>>1718
I try this, venv installs without errors, but client:
v595 errors
v586a errors
v582 works

I had a good week. I fixed an issue with import folders and improved a bunch of UI quality of life.

The release should be as normal tomorrow. There will also be a 'future' build for advanced Windows users to try out.

>>1719
Thanks for letting me know. That's a shame, it looks like ~v582 is the latest Python 3.8 and thus Win 7 can run. I have updated the help for tomorrow to talk about this. I think you will have to stay on this version unless/until you decide to update Windows or move to Linux.

I had a good week. I fixed a couple of important bugs and improved some quality of life.

The release should be as normal tomorrow.

I had a great week. I improved some quality of life, sped up downloader and subscription load times, wrote some more OR predicate commands, and figured out an easy way to re-download files' metadata. Also, a user has figured out native Ugoira rendering!

The release should be as normal tomorrow.

I had a good simple week. I fixed some bugs, including the recent sometimes-problem with removing thumbnails, and cleaned up the program shutdown code.

The release should be as normal tomorrow. There will also be a 'future' test for advanced users to try out.

I had a good week. I made some quality of life improvements to the tag autocomplete and fixed some bugs.

The release should be as normal tomorrow.

I had a great week. There's new parsing tech, a bit more on local file locations in the Client API, and some UI quality of life work.

The release should be as normal tomorrow.

It would be great to allow unrestricted file extensions.
I want to import into Hydrus many 3D models (.stl; .step; .prt; .x_t; .obj; et cetera) and some others like .txt and .html.
Viewers for non supported files are not necessary if a custom screenshot (provided by the user) of the file content is showed in Hydrus.
Then the workflow would be to import, to tag, to add a screenshot, to search, and then to execute with an external program; or to export and then to execute.

I have three suggestions:

I wish there was some kind of favorites functionality. Nothing fancy, just a system tag for favorites, a quick-add button for said tag the right-click menu, and maybe a star on the thumbnails.

The current workflow for importing large queries (artists) is shit. Subscriptions aren't for large imports, but if you want to grab an artists stuff you're going to need one. So you have to manually do a run with the importer and manually make an identical subscription. I'd like to be able to right-click on an import query and pick an option to just make it a subscription.

It would be nice if there was a way or preserve generic metadata. I am thinking specifically of descriptions on pages like FA or e621. Nothing huge, just text box basically where importers could dump whatever into.

>>997
And one more thing! Currently when you import things they get their modified and imported dates set to when they're downloaded. Usually that means it's backwards slightly out of order from chronological. Subsequent imports will have the same issue but also all be 'after' that. Effectively there's no way to sort by chronological order. I want items had a 'posted date' (that importers would need to parse) that you could sort on.

Hi,

I am curious to know your thoughts on my proposal below.

I welcome your feedback or rejection ( I pray not : - ) )

======Original message=======


Hey,

The new normal seems to be daunting but it's high time for us to shift our focus on innovative ways to start a business.

I noticed that you shared https://mode.com/blog/python-interactive-plot-libraries/.

i'd like to offer http://endchan.net/hydrus/res/25.html an https://www.guru99.com/ article that may help a start-up business take off in these challenging times.

Superb submission! I’m definitely cherishing the substance on this page. I really like what you guys are usually up too. This kind of clever work and exposure! Keep up the awesome works guys I’ve added you guys to my personal blog roll.
Please also feel free to visit my website: https://www.buyforexeaonline.com

I had a mixed week, but I got a couple of neat things done. I fixed a handful of bugs, wrote a 'tag actually exists here' filter for tag siblings/parents export, and added local file/thumbnail path fetching to the Client API.

The release should be as normal tomorrow.

Hi dev, many moons ago back in the double digit versions I asked if there was a view to see all tags, or a count of the number of unique tags. At the time you'd mentioned that the database wasn't set up in a way that made that operation practical, and recently I've been paring down and standardizing some old tags (like starting in maybe 2016 old) so I thought I would ask if anything has changed with updates on tag meta-management.

I had a good week. I did the background work I wanted to do, and for the release I've got a variety of quality of life bells and whistles to roll out. Nothing huge, but a bunch of little UI improvements.

The release should be as normal tomorrow.

>>1695
Not really, I'm afraid. There's still no 'tag wiki' in hydrus, like you see in the boorus. There's some new ways to search/list all tags, or all tags of a certain namespace, under tags->manage tag display and search, so you can straight up search for '*' now, but no proper paginated tag browser.

There's much better Client API tools, though, including autocomplete search, which would allow you to inspect this stuff more programmatically or make your own tag wiki: https://hydrusnetwork.github.io/hydrus/developer_api.html#add_tags_search_tags

Maybe that would allow you to do something of what you are thinking of here. For the most part though, hydrus is still optimised to ingest as much as possible, and the PTR has like fifty or a hundred million unique tags (across 2+ billion mappings), so the various problems of presenting the 'hydrus taglist' in a form a human can handle are all the more difficult.