arkoniak/configenv.jl

Package for populating ENVironment variables.

configenv.jl's Issues

Remove trailing spaces

It looks unreasonable to keep trailing spaces in variable value, taking into account the fact that we remove extra spaces on the left.

So, for variables like this:

foo = abc 

foo should be equal to abc, not abc . If someone requires this extra space, then he should use quotes:

foo = "abc "

template variables

Make possible to use template variables

#.env
FOO=ABC
BAR=ZXC${FOO}

Second should be resolved at the read time to BAR=ZXCABC

It should check for cycling dependencies and left them unresolved.

TagBot trigger issue

This issue is used to trigger TagBot; feel free to unsubscribe.

If you haven't already, you should update your TagBot.yml to include issue comment triggers.
Please see this post on Discourse for instructions and more details.

If you'd like for me to do this for you, comment TagBot fix on this issue.
I'll open a PR within a few hours, please be patient!

Refactor parse function

Current implementation of parse/dotenv is suboptimal. It returns Dict, but it should return EnvProxyDict. Also parse accepts String and IO, but dotenv not.

It should be all brought to a common denominator:

• parse should return EnvProxyDict
• dotenv should accept all possible inputs.

`dotenvx` as an nonoverwriting alias

Since it looks like other languages provide non overwriting version of dotenv function it make sense to have something like that without resorting to constantly use overwrite=false argument.

Proposal:

dotenv() # default is to load with overwrite
dotenvx() # default is to load without overwrite

Documentation of the new features

In docs:

Add documentation

1. Motivation of dotenvx
2. Intended usage of merge features.
3. Variables interpolation
4. use cases for isresolved and unresolved_keys
5. IO data load from the web

Remove documentation

1. parse function. It is part of internal implementation and should not be visible/recommended to use. May be add its description later as a part of contribution guide.

In README.md:

1. Remove parse function description.

Merge function for EnvProxyDict

Proxy dictionaries should have the ability to merge together. I think, * sign is the best operation, since order matters

cfg1 = dotenv(".env1")
cfg2 = dotenv(".env2")
cfg = cfg1 * cfg2

Or even

cfg = maprrduce(dotenv, *, [".env1", ".env2"])

It can be also part of dotenv function

cfg = dotenv(".env1", ".env2")
cfg = dotenv([".env1", "env2"])

Second version can be supported through the multiple dispatch on Vector, and first one through splatting.