Plugiin not executable, what abiout Debug.lua about lightroomplugin-clarifaitagger HOT 10 CLOSED

Yes, thank you @axelstahlhut, for your feedback. The debugging toolkit was recently added while making some improvements. It should work (as-is) if you rename the plugin from .lrdevplugin to .lrplugin... but I have just removed the debugger and created a new commit to master so that nobody else has to be confused by this. I will not be doing much further development of this plugin. Related development will continue in another project (in collaboration with another Github dev): https://github.com/mcrosson/lr_plugin_computer_vision_tagging In many ways, I think the approach of an "export plugin" is better, all around.

That said, this plugin should work for you now if you update to the latest state.

from lightroomplugin-clarifaitagger.

from lightroomplugin-clarifaitagger.

@axelstahlhut I feel your pain when it comes to the learning curve. It takes a while to get used to. First you should learn the basics of Lua, then you need to learn your way around the basics of the Lightroom SDK/API, which is not very nicely documented. None of it is very easy to find good examples of compared to popular coding languages like C++, Java, PHP, Javascript, etc. That is partly why I wanted to make my contributions here more public (to at least help the "next guys").

The other project already has the Google Vision API in the "roadmap" and your help implementing it would be welcome, I'm sure. 👍

You'll learn by experience and there are online forums where others may provide support if you run into problems, but collaborating really helps a lot.

Cheers,

Lowell

from lightroomplugin-clarifaitagger.

from lightroomplugin-clarifaitagger.

@axelstahlhut — Hi Axel,

If your goal is to have a Lightroom plugin, then it's hard (I won't say impossible) to integrate other technologies (e.g. Java). There may be some tricks with regard to using Java via command-line calls from the Lua/Lightroom SDK, but that would require users all have Java installed. I know other Lightroom plugin developers do make use of some non-Lua tools, though, so it's not impossible. So far, though, there are not a lot of examples for these kinds of novel "tricks". I know it is sometimes done, but then the code is most often compiled, so you won't see how it was achieved.

Normally, on Github, you make a fork (of an existing project), add features in a branch, test and make sure it's all working, then create a "pull request" to the project "owner" to merge the changes. In the case of this plugin, the originator (safx) gave me commit authorization to his project, so I have been able to merge my own pull requests lately, but that's just because he's a bit spread thin, I think, and doesn't normally work with Lua/Lightroom SDK. I think he mostly does Mac/iPhone/Swift development.

For the other "computer vision" plugin I've been collaborating on, @mcrosson is very involved and does a great job of managing his project. I'm sure you'll be able to find a way to be useful if your goal is to have a functional Lightroom plugin that leverages the Google Vision API. Most of the "heavy lifting" has already been achieved by Mike and myself, I think—inasmuch as there are many core functionalities that should not need to be reinvented (very much) for each service, so it should mostly be a matter of figuring out how to communicate with the Google system and how to interpret the responses.

Cheers,

Lowell

from lightroomplugin-clarifaitagger.

Indeed, the Lightroom Vision project I started does have the Google Computer Vision API and Amazon Vision API's as part of the long term roadmap here (see the bottom section). I also started work on the Microsoft Vision API prior to @LowellMontgomery 's awesome work on the hierarchical tagging. The changes I started are here. There is some refactoring that needs to be done in the linked commit but it should serve as a solid base for adding Google's Computer Vision API.

I'm hoping to make some more progress "soon" but real life has been pretty crazy for awhile and I'm working on getting some more free time to resume work on the Microsoft Vision API implementation.

from lightroomplugin-clarifaitagger.

from lightroomplugin-clarifaitagger.

@axelstahlhut Most of these services/agencies have their own plugins for exporting to them; same with print companies. There are also other plugins and services that already do some integration and/or help you track where you have submitted what, etc. That said, if there is something to innovate in this area, I'd be happy to collaborate on that since establishing a stock photo portfolio was my original goal when I started working on these plugins to streamline some of the "metadata stuff".

Looking forward to collaborating... but first sleeping. ;-) (I'm only a few hours south of you—in Frankfurt.)

from lightroomplugin-clarifaitagger.

from lightroomplugin-clarifaitagger.

@axelstahlhut thank you for the offer to help. Right now I mostly need help with the LUA glue for the plugin. The Microsoft APIs are really really straight forward and I had no problems getting the fundamental API calls working. The trick now is getting the branch updated to work with @LowellMontgomery 's latest updates on the master along with polishing the implementation / testing.

I'd welcome any/all help you'd be willing to provide.

Also, if you continue forward with your stock photography work, please let me know. I'm curious what you are able to put together. Much like Lowell, I've been toying with some stock photography type stuff and the Computer Vision plugin was born to help with the metadata grunt work that has been involved.

from lightroomplugin-clarifaitagger.