sammdec / punkapi Goto Github PK

View Code? Open in Web Editor NEW

124.0 7.0 25.0 1.3 MB

DIY Dog meets JSON

License: Other

JavaScript 100.00%

api brewdog beer

punkapi's Introduction

punkapi-db

Crowd Sourcing an API from BrewDog's hugely generous DIY Dog

In the spirit of DIY & giving back I thought I would attempt to collect all the BrewDog Recipes from their DIY Dog pdf and create an api from it!

You can find the API at https://punkapi.com and it's completely free!

If you are looking for the source code for the server that powers https://punkapi.com head over to samjbmason/punkapi-server

What's the deal?

This repo contains all the individual files for each recipe in the DIY Dog pdf (data). It also contains a singular JSON file data.json that contains all the recipes in one single file.

It's the data.json file that actually powers the punkapi API.

Installation

$ npm i --save punkapi-db

Usage

const punkapi = require('punkapi-db')

Libraries

These are user contributed wrappers for the API

The Beers

Status: 325 of 325 completed

If you find an issue with a recipe, please feel free to submit an issue

License

Please see License

punkapi's People

Contributors

Stargazers

Watchers

punkapi's Issues

Change "twist" from a string to an object

The twist property is currently a string. However, there are numerous recipes that have multiple twist/brewhouse additions with various properties.

The twist property should be an array object similar to the hops property to support multiple additions.

In DIY Dog 2018 you can look at AB:24 as an example.

As a workaround for now they could be added under the hops as many of these additions are added during the dry hop stage or to the kettle during flame out or whirlpool.

I can't make a request with parameters

Everytime I make a request with parameters, I get this error:

For the above request, I've used this url:

https://api.punkapi.com/v2/beers?brewed_before=11-2012&abv_gt=6

which is present in the documentation:

Far too much dark crystal in Old World Russian Imperial Stout

In v2 the Dark Crystal amount for Old World Russian Imperial Stout is 0.66kg, not 6.56kg.

Filter by ABV fails if the value is Float

Filtering by abv_gt with a float value breaks the search, here an example:
https://api.punkapi.com/v2/beers?abv_gt=1.2
and this is the current response:
{ "statusCode": 400, "error": "Bad Request", "message": "Invalid query params", "data": [ { "param": "abv_gt", "msg": "Must be a number greater than 0", "value": "1.2" } ] }

added pagination information to the response headers

Transferred issue from sammdec/punkapi-server#12

SRM greater than 40

Hi,

i searched online what the values mean. Would be nice to add this to the documentation.

However according to all sources i've seen the SRM value is between 1 and 40, but in the data set there are also values like 100 or even higher. For example the one with id 92 https://api.punkapi.com/v2/beers/92 .

Is this an issue or is there another reason for that?

[DB] : Updating to latest DIY Dog version

At the time of writing, BrewDog has published it's 8th revision of the DIY Dog book (which you can find here : link to DiyDog.

There are 415 recipes now and the db might need to be updated.
Speaking of which, I haven't found your method nor the sources that you used to parse their PDF file ; can you share some details about that maybe ? I would have liked to try pdf parsing myself and was really interested to see how you did it 😄 !

I believe that as long as the data format in the pdf does not change, we can have a periodic github action that updates the database every once in a while and publish the resulting artifacts (like your data json files).

Results limited to 80 per page

The API V2 documentation page says:

All parameters are optional and without them the API will just return the beers in ascending order from their ID.

You also mentioned in this issue that:

...you can return as many or as few as you like from the api...

Therefore, I did think that simply querying the base URL would return all beers. However, this:

curl https://api.punkapi.com/v2/beers

Returns only twenty five results, which seems to be the default per_page number.

Using the per_page parameter seems to max out at 80

curl https://api.punkapi.com/v2/beers?per_page=81
{"statusCode":400,"error":"Bad Request","message":"Invalid query params","data":[{"location":"query","param":"per_page","msg":"Must be a number greater than 0 and less than 80","value":"81"}]}

I've just started writing a wrapper for the API and am trying to handle the case where the user inputs no parameters.

This could also be linked to Issues #44 and #52

Adding beers

Heya Sam,

Can you let me know how to go about adding new recipes to this? I've forked the repo but I'm concerned about a couple of fields, ID for example, where does that come from?

The schema has some fields marked as required (will an empty string value satisfy that?)

Megs gave out some DIY dog recipes on the BD forum and I went to add them but got stuck immediately, lol.

What filters should the API have?

While adding recipes I'm also trying to build the API at the same time so that when the final recipe is published the API can go live as well. Therefore I need suggestions on what kind of filters you would want from this API. So far I have

If you have any suggestions just add them in so we can discuss!

XML output

Would it be possible to generate XML output? I'm trying to use JSONPath on the JSON output and having trouble selecting ancestors after filtering. Seems this is only possible in XPath but not JSONPath.

Spelling

Yeast spelling Errors:

Wyeast 1056 - Ameican Ale™ (#1 Buzz)
Wyeast 1056 - American Ale (#154 No Label)
Wyeast 1056 American Ale™ (#23 Storm)

name should be "Wyeast 1056 - American Ale™"

Maybe swapped gravities in some items

My knowledge about brewery is about zero and therefore I may be wrong, but it seems to me that some beers have their target gravities swapped.
According to a Wikipedia page "Beer_measurement" final gravity should be less or equal to original gravity. But the beers with ids in (28, 70, 127, 131, 193, 214, 221, 222, 223, 224, 225, 227, 228) have target gravities swapped: original is less than final.

Not an issue a question

Could you tell me how I could pull all the names of the beers? They have put the recipes as beer xml's online https://brewdogrecipes.com/beerxml/ipa-is-dead-mandarina-bavaria.xml for example, and I would like to make a way to download all the xml's, but for that I need all the names they have put recipes out for and came across your project.

Correct the spelling of litres

The recipe data is using the American spelling of liters for the boil_volume and volume units.

  "volume": {
    "value": 20,
    "unit": "liters"
  },
  "boil_volume": {
    "value": 25,
    "unit": "liters"
  }

It should use the international spelling for the measurement which is litres to be consistent with the other international spellings used for measurements in the current data set such as grams, kilograms and celsius.

Create a development branch

Would it be possible to add a development branch or a 3.0 branch? For example there are a couple of major schema changes required and also a fix for all IDs to bring them in line with the IDs in the book. Both of which would be a major bump. Might be a good to isolate these changes to an active development branch for a 3.0 release?

The servers certificate chain is incomplete.

When trying to make a connection to the API an exception is raised

javax.net.ssl.SSLHandshakeException: java.security.cert.CertPathValidatorException: Trust anchor for certification path not found.

The servers certificate chain is incomplete which causes this error.
There are a number of other SSL issues that could be addressed as well.

https://www.ssllabs.com/ssltest/analyze.html?d=punkapi.com

split hops and extras

hello, i've noticed a "trouble" in the ingredients. In fact hops array includes also extras, e.g. beer "Hello My Name Is Vladimir" (id: 40) contains "Blackberry Concentrate" and "Sour Cherry Puree" among different hops' names. Sometimes it could be confusing and it's not possible to count amount of hops properly (in this case, extras always make a statistical deviation). Maybe it's a good idea to split hops and extras into two separate arrays?

Fixing issues and typos in DIY DOG

Do you have any plans to fix up obvious errors in DIY DOG?

For example in 10 Heads High there are no weights against the hop additions in the boil. I've attempted to reverse engineer this to achieve the same IBUs. I'm using 40g Chinook at 60 mins, 15g Chinook at 30 mins and 15g Centennial at 30 mins. The 0 minute additions are a bit harder but you could base them off Hops Kill Nazis, which this recipe is loosely based on, ie. 62.5g each of Chinook and Centennial.

Ideally BrewDog would release a 2nd revision with these errors fixed up.

Add a way to get total number of beers

Currently it is not possible to get the total number of beers/pages from the api.Thus it is also not possible to display the index of the last page to the user. Displaying the page index is a very common thing to do when displaying a paginated list to a user.

Of course you could fetch all the beers in advance to find out the total amount. But that would lead to over-fetching data and is not an option at all (the rate limit is 1req/sec). And even then there is a possibility that a beer could be added/deleted without the client taking notice of that.

Is it possible to add some metadata to the api response that includes the total number of beers?

Something like this:

{
    metadata: {
        "page": 5,
        "per_page": 20,
        "page_count": 27,
        "total_count": 521,
    },
    records: [
        ...
    ],
}

Guide for Contributing

The newest DIY Dog publication, now v8, lists 415 individual recipes. The API data currently only contains 325.

I'm interested in contributing newer recipes, but do not know how.

While it's straightforward enough to produce the necessary JSON file containing the new recipe data, there's no instructions regarding how to make a pull request containing the necessary image for the image_url.