# Envresolver
Envresolver is a Python library for parsing [environment variables](https://en.wikipedia.org/wiki/Environment_variable)
to common Python datatypes. Environment variables are only accessible as pure text (string) typed variables and thus need
some manipulation to transform into any other types. `EnvResolver` class provides a nifty way of parsing the current
environment according to given specifications.
Install with pip:
`pip install envresolver`
#### Supported types (at the moment)
- `str`
- `bool`
- `int`
- `float`
- `list` holding any of the supported types
- `Json`
- `XML`
- `datetime`
#### Why Envresolver?
Envresolver is a lightweight alternative to other environment parsers and customizable by-design.
Adding support for new types is ultra-simplified and does not need any source modifications of the library itself.
## Usage
All variables to-be-resolved must be specified with a supported type when using `EnvResolver`. `EnvResolver` then inspects
all the specified variables and tries to parse an environment variable with the same name to the given type format. For example,
to parse simple string values:
```python
from envresolver import EnvResolver
r = EnvResolver()
r.add_variable("var")
r.resolve()
```
If an environment variable `var` was present in the current environment, it can be accessed after resolving by a special
namespace member `ns` or by an explicit method `getr`.
```python
r.ns.var
# Or
r.getr("var")
```
Additionally, variables can also be fetched from the current environment without pre-calculated resolving.
This is suitable for simple variables and values that can change constantly:
```python
from envresolver import EnvResolver
r = EnvResolver()
r.get("var")
```
User can also supply default values to all requests:
```python
r.add_variable("var2", default="default_value")
# Or
r.get("var2", default="default_value")
```
### Environment Variables with Type Conversions
As stated before, `EnvResolver` also supports automated type conversions for environment variables. Variable types can
be specified as shown:
```python
r.add_variable("var", t=int, default=-1)
# Or
r.get("var", t=int, default=-1)
```
Let's imagine the current environment would hold the variable `var` with a value of `"5"`. By running `EnvResolver.resolve`,
it would be automatically parsed. However, if the environment variable `var` would hold an incompatible value, `"_"` as an example,
the parsing would fail and `r.var` would hold the default value, if one was given:
```python
from envresolver import EnvResolver
# export var=5
r = EnvResolver()
r.add_variable("var", t=int, default=-1)
r.resolve()
r.ns.var # 5
# export var=_
r = EnvResolver()
r.add_variable("var", t=int, default=-1)
r.resolve()
r.ns.var # -1
```
### Advanced Types
`EnvResolver` currently supports also some more advanced types of variables, such as lists, Json and XML. Lists have full support
of type hinting and will try to convert all elements accordingly:
```python
from typing import List
from envresolver import EnvResolver
# export my_list="1,2,3,4"
r = EnvResolver()
r.add_variable("my_list", t=List[int])
r.resolve()
r.ns.my_list # [1, 2, 3, 4]
```
Json and XML are supported via custom type notations stored in `envresolver.Types`. Json and XML will be parsed using Pythons built-in
`json` and `xml` modules. Json parsing will output native python lists/dicts and XML results will be of type `xml.etree.ElementTree.Element`.
Here is an example on Json parsing:
```python
from envresolver import EnvResolver, Types
# export json='{"key": "val"}'
r = EnvResolver()
r.add_variable("json", t=Types.Json)
r.resolve()
r.ns.json # {"key": "val"}
```
Date objects are supported via Pythons built-in `datetime` module. User can specify in which format the date strings are expected, with the default being
`%Y-%m-%d %H:%M:%S`. Here is an example:
```python
import datetime
from envresolver import EnvResolver
# export mydate="2021-01-01 12:34:56"
r = EnvResolver()
r.add_variable("mydate", datetime.datetime)
r.resolve()
r.ns.mydate # datetime.datetime -object with the correct time
```
Parsing certain advaced types, such as `datetime` objects or lists, relies on additional information
regarding the data format. List parsing needs to know the list separator character and datetime conversions
rely on certain date formats. These can be configured either at `EnvResolver` initialization or afterwards using
the methods `set_list_separator` and `set_datetime_format`.
### Custom Types
Users can supply `EnvResolver` with custom parsers as well as override existing ones. Below is an example of using a custom parser
for reading data into a user-defined class:
```python
from envresolver import EnvResolver
class MyData:
def __init__(self, a, b):
self.a = a
self.b = b
def my_data_converter(e: str):
s = e.split(".")
# Raise ValueError if the given
# environment variable is in
# wrong format
if len(s) != 2:
raise ValueError
# Return parsed data
return MyData(a=s[0], b=s[1])
# export data="john.smith"
r = EnvResolver()
r.add_converter(MyData, my_data_converter)
r.add_variable("data", t=MyData)
r.resolve()
r.ns.data # MyData(a = "john", b = "smith")
```