331–340 of 360 posts
In the same vein, I made the Recipe Filter extension for browsers: Chrome: https://chrome.google.com/webstore/detail/recipe-filter/ahlc... FF: https://addons.mozilla.org/en-US/firefox/addon/recipe-filter... Source code (there's Safari in there if you don't mind building it yourself): https://github.com/sean-public/RecipeFilter I was spurred into action by a comment here on HN back in 2017: https://news.ycombinator.co…
Earlier quoted context omitted.
Also that.
It's obvious to everyone that content producers who write, format, publish and promote recipes usually have a revenue goal. Any tactic that defeats that is theft, whether you want to point to fineprint somewhere or not. You know that someone made something and you have decided not to give back. Ad free recipe sites are already handled, you pay a subscription fee instead. That being said, the final format of an onlyre…
Earlier quoted context omitted.
The irony of having an ad supported app to scrape recipes from ad supported websites. I'm not sure how I feel about this.
@AwkwardPanda And how does a site opt out of your scraping? Do you have a unique user-agent when you scrape? A set of IPs?
Let me see if I can build a opt-out list. But wouldn't it beat the entire purpose of this app?
Props for building a PWA, but why do I have to copy paste a url into it when the Share Target API would allow me to send the URL directly to you app using the browser's share feature?
I'm curious how you wrote the logic for recipe detection? I've know some others who've tried to solve this and like many pattern recognition tasks it turns out to be harder than you'd think, but not impossible of course. Just curious how you did it.
Since google encourages this and ranks pages with schema higher, all recipe blogs and sites add this schema to their html.
This is structured data. I just parse this json.
Not all recipes follow this though. So you may occasionally see error messages for old sites that have not implemented schema.org structure.
Earlier quoted context omitted.
I use CookBook which scans websites, and OCRs cookbooks with good success! I have all my regular recipes on there now https://thecookbookapp.com/
[dead]
It's fine to share something you founded, just be upfront about it and don't spam. It would have been easy to just post "We're working on something similar over at Pepper! Our approach is .... Would love to hear what people think."
Looks poorly tested. QR code scanner doesn't work and the "how to use?" tip doesn't display anything. Manually entering a url requires the user to include the subdomain AND http/https (instead of defaulting to " http://www ".
Regarding "How to use?", yes this is an issue on android. If you click multiple times, it will work.
I will fix it.
Regarding http/https, I'll implement that.
In the same vein, I made the Recipe Filter extension for browsers: Chrome: https://chrome.google.com/webstore/detail/recipe-filter/ahlc... FF: https://addons.mozilla.org/en-US/firefox/addon/recipe-filter... Source code (there's Safari in there if you don't mind building it yourself): https://github.com/sean-public/RecipeFilter I was spurred into action by a comment here on HN back in 2017: https://news.ycombinator.co…
Funny running into you here! I'm the 'maintainer' of the Firefox version (it doesn't take much work.) I just got a request to do imperial to metric conversion, which I had to politely decline based on the idea of looking up the density of every ingredient involved; not to mention writing an actual parser for recipies proved to be quite a challenge given the wide diversity of how the recipes are layed out.
One feature that would improve the recipe recognition and unlock lots of other features where it's found would be JSON-LD support. I'm seeing lots of recipe pages containing those handy, structured formats lately.
A comment about JSON-LD: https://github.com/sean-public/RecipeFilter/issues/22#issuec...
But all of these things stray from the fundamental simplicity of what it does. The code is just 92 lines of JS and works (or breaks!) transparently. If I started adding features, it would quickly become another complex system to maintain and I have plent of that at work ;)
reminds me of https://www.paprikaapp.com/