Skip to main content

A YAML type validator

Project description

yamlval

Python tool used to easily define a schema for a yaml file. Yamlval allows users to define types for yaml fields and will throw clear and will generate clear and transparent logs for debugging

If you're skimming this then read these next few points to save yourself a lot of headache

- Do not use enums in yDict, it will not work, you will get an error
- Do not use mutliple enums in yList, just make one large enum or you will get an error
- yAny is true for everything EXCEPT None
- yNone is true for ONLY None
- yDict takes tuples of yObjects as inputs, i.e 
```
yDict( (yKeyType, yValueType1, yValueType2, ...), (yKeyType, yValueType1, yValueType2, ...), ...  )
```
Happy Validating!

Example

ticker: "AAPL"
fields: 
  - "Close"
  - "Adj Close"
  - "Open"
col_map: 
  2 : "hello"
  3 : "hello plus one"
transform: 
  - lag: 1
  - scalar: 2
  - targets: 
      - "col1"
      - "col2"
      - "col3"
from yamlval import yString, yList, yEnum, yDict, yInt, yFloat, ySchema

from enum import Enum

class Ticker(Enum):
    msft = "MSFT"
    aapl = "AAPL"
    spy = "^GSPC"

class ValuationFields(Enum):
    close = "Close"
    adjclose = "Adj Close"
    opn = "Open"

class Schema(ySchema):
    ticker = yEnum(Ticker)
    fields = yList(yEnum(ValuationFields))
    col_map = yDict(
            (yInt(), yString())
        )
    transform = yList(
                    yDict(
                        (yString(), yInt(), yList(yString(), lower=3, upper=3))
                    )
                )

config = []
with open("example.yml", "r") as f:
    config = Schema.validate_and_load(f)
print(config) 

Longer Example

  • This is an example of a .yml or .yaml file which does not match the config,. The structure is mostly there, but I have simulated common errors which are easy to overlook. You can copy-paste and directly run this code to get a feel for the errors thrown by the system.

  • Notice that most validation errors are logged before an exception is raised, letting you see a large number of validation errors simultaneously, saving you important time.

  • Directory structure for example:

./
    example.py
    example.yml
  • Requirements for example:
    • python >= 3.6
    • yamlval
name: "Michaell"
age: 21
height: "5'11 3/4extrachars"
listOlists: 
  - - - "hello"
      - "sdf"
      - "asdf"
      - "asdf"
    - - "asdfasdf"
      - "asdfasdf"
      - 1.
      - "asdfasdf"
  - - - "hello"
      - "sdf"
      - "asdf"
      - "asdf"
    - - "asdfasdf"
      - "asdfasdf"
      - 1
      - "asdfasdf"
normalList:
  - "hellothere"
  - 4
floatingPoint: 9.
dictOstuff:
  Michael: 
    "string1" : "string2"
  1: 
    - - "string123"
      - "asdf"
      - "asdf"
    - - "asdfasdfasdf"
      - "asdfasdfadsf"
  somewords: "some more words"

anyDict: 
  asdf: 5
  6: "eight"
  NULL: NULL

noneDict:
  asdf: NULL
  5.5: NULL
  5: NULL

anyList: 
  - "asdf"
  - 5
  - 5.5
  - NULL

noneList:
  - NULL
  - NULL
  - NULL

noneListPlus:
  - NULL
  - 5.5
  - 5
  - NULL
  - "string from yamlval here"
from yamlval import yString, yList, yEnum, yDict, yInt, yFloat, ySchema, yAny, yNone

from enum import Enum

class Names(Enum):
    mike = "Michael"

class Schema(ySchema):
    name = yEnum(Names)
    age = yInt(lower=0, upper=120)
    height = yString(lower=3, upper=8)
    listOlists = yList(yList(yList(yString(), yInt())), yInt(), lower=2)
    normalList = yList(yString(lower=2), yInt(upper=10))
    floatingPoint = yFloat(upper=10)
    dictOstuff = yDict(
        (yString(), yInt(), yDict((yString(), yString()))),
        (yInt(upper=2), yList(yList(yString(upper=7), upper=2))),
        (yString(), yString()),
    )
    anyDict = yDict(
        (yAny(), yInt(), yString())
    )
    noneDict = yDict(
        (yAny(), yNone())
    )
    anyList = yList(yAny())
    noneList = yList(yNone())
    noneListPlus = yList(yNone(), yFloat())
    
with open("example.yml") as f:
    config = Schema.validate_and_load(f)

print(config)

You can also call yObject.match() method to validate separately!

  • If you do not want to make an entire schema, want to just validate tehe type of an object, or want to validate to a degree above what yamlval can do by default, you can define a yObject and then call the .matches() method separately!

  • For example, lets say you have a string = stringToCheck, and you want to check if it's part of an enum, you can define checker = yEnum(EnumName) and then call checker.matches(stringToCheck). This will return a tuple matching the following format:

  • Tuple[bool, Optional[List[str]]] = (match, err if not match else None)

  • Translation: a tuple with a true/false value for the match at the 0th index, and a list of strings containing the caught errors (if there are any) or 'None' (if there are no errors) at the 1st index.

from yamlval import yEnum

from enum import Enum

class Names(Enum):
    mike = "Michael"

checker = yEnum(Names)
stringToCheck = "Michaelll"
(match, err) = checker.matches(stringToCheck)
if not match:
    print(err)

will yield

["Input <Michaelll> not in <['Michael']>"]

You can also iterate through all validation errors in err and then print them separately to get a nicer looking output for multiple validation errors.

checker = yEnum(Names)
stringToCheck = "Michaelll"
(match, err) = checker.matches(stringToCheck)
if not match:
    for error in err:
        print(error)

This will work for any of the yObject types (yDict, yList, yString, yInt, ...) and is actually the foundation of the internal implementation of yamlval.

==========================================================================

If you find any errors, bugs, or simply want to contribute, let me know at mjm.digregorio@gmail.com !

You can also initiate a pull request!

- Mike

Happy Validating!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

yamlval-1.0.0.tar.gz (9.3 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page