How to import recipes from files (i.e. HTML)? #3644
-
|
Let me explain: Sometimes, I find a recipe published without recipe markup so mealie cannot import it. I found that tasking ChatGPT with this prompt works wonders. Please scrape this site https://site.tld/recipe-example for the recipe and give me the scraped recipe formatted with recipe markup as a HTML file for download. Now I am just wondering about the easiest way to import it into mealie. I don't really want to fire up a local web server to serve the file so mealie can scrape it. Any ideas? |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 8 replies
-
|
Just to keep this discussion updated, I found partial solutions here: https://www.reddit.com/r/selfhosted/comments/11ptvpp/python_script_that_imports_recipes_from_a_url/ and here: #2264 Still playing around with those ideas. |
Beta Was this translation helpful? Give feedback.
-
|
FYI, the recipe scraper now falls back to OpenAI if it fails to parse normally: #3690 The next step is to enable this for uploaded files, too |
Beta Was this translation helpful? Give feedback.
-
|
Oh, wow! I have to check what version I am on and what version this is merged into, but that is great news. |
Beta Was this translation helpful? Give feedback.
-
|
Guess I'm not there yet. OpenAI is enabled and works for ingredients parsing.
|
Beta Was this translation helpful? Give feedback.
-
|
I replied too soon. My second try with a page without recipe markup mealie failed and just said:
URL: https://ottolenghi.co.uk/recipes/pork-souvlaki-with-cucumber-salad-and-tzatziki Prompting ChatGPT directly works:
|
Beta Was this translation helpful? Give feedback.
-
|
Ah yep that's a bug |
Beta Was this translation helpful? Give feedback.
-
|
See: #3749 |
Beta Was this translation helpful? Give feedback.
-
|
Not sure its the same problem as I don't see any errors except this one in the docker logs:
Here is another URL which gives the exact same error as in my last screenshot. https://www.thekitchn.com/arayes-lebanese-stuffed-pita-recipe-23664768 |
Beta Was this translation helpful? Give feedback.
-
|
The 403 error is unrelated to Mealie; this means the website is blocking scraping traffic (the website is returning 403, not Mealie). If we can't fetch the content, we can't do anything with it |
Beta Was this translation helpful? Give feedback.
FYI, the recipe scraper now falls back to OpenAI if it fails to parse normally: #3690
The next step is to enable this for uploaded files, too