-
Notifications
You must be signed in to change notification settings - Fork 6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
add logic to BooleanParser to allow for optional truthy arguments #132
base: main
Are you sure you want to change the base?
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@oroth8 some really clever changes and refactors / reusability! Some more trivial questions for you.
Also correct me if I'm wrong @chawes13 but I think we generally update versions in each PR even if there are a handful open. I suppose you could update right after approval prior to merge to mitigate multiple commits around versioning. |
`:boolean` will parse values of `true` and `1` as truthy. If another value is expected to be truthy, use the option `truth_value` to assign a custom truthy case. | ||
|
||
```ruby | ||
input :checkbox, :boolean, truth_value: 'yes' | ||
``` | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is interesting. It looks like there isn't really a place in the README to document which options are accepted for each parser. I think we can leave this here for now, but I'll create another issue to holistically address adding this missing documentation.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
`:boolean` will parse values of `true` and `1` as truthy. If another value is expected to be truthy, use the option `truth_value` to assign a custom truthy case. | ||
|
||
```ruby | ||
input :checkbox, :boolean, truth_value: 'yes' |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Out of curiosity, why truth
instead of truthy
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
just personal preference, but can switch it to truthy
end | ||
|
||
def self.normalize(value) | ||
raise Decanter::ParseError.new 'Expects a single value' if value.is_a? Array |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why is this part of the normalize method? I'd prefer to see the two decoupled, so that normalize
can more closely follow the Single Responsibility Principle (i.e., be responsible for one thing)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I thought about pulling it out but then we would have to check both val
and the option_val
separately making it less "dry". I opted for the dry trade off. However, looking at your comments below seems like we wont need to do this if opting for multiple option values.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
On a separate note I was flipping through all the parsers and I wonder if there's a bigger refactor available here: ValueParser
could handle the array and empty checks. HashParser
doesn't inherit from ValueParser
(not totally sure why) but perhaps by the same logic ArrayParser
could inherit from Base
. All the other parsers do these same two checks. I haven't done a deep dive to test that out but wanted to surface
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I like where your head is at. I'd be open to that refactor
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
option_val = options.fetch(:true_value, nil) | ||
normalized_option = normalize(option_val) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Out of curiosity, why are you only allowing one additional option to be specified? What if we wanted to allow "on" or "T" as well?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the scope of the original issue seemed to indicate adding the ability for one optional value. That was my interpretation of the issue, but I can work on a solution that will allow for additional optional values if desired.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I also took this as a design decision (perhaps implicitly by me and adopted by Owen). I don't wholeheartedly disagree with it as this option is per input and is meant to handle the nuance of a given input. Flexibility vs intentionality, and I can see both options. Either way that implementation detail can probably also be documented for the user.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
taking the more flexible route is definitely fine, as it obviously also covers the single case
|
||
true_values = ['1', 'true'] | ||
|
||
option_val = options.fetch(:true_value, nil) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In the README, you named the option as truth_value
. We'll need to update the code or the README to be consistent (and accurate)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yep, this is a mistake. Do you think truthy_value
or truthy_values
is a better solution?
normalized_option = normalize(option_val) | ||
|
||
true_values << normalized_option if normalized_option | ||
true_values.find {|tv| !!/#{tv}/i.match(normalized_val)}.present? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🍔 (food for thought): This can be refactored a bit to deal directly with booleans.
true_values.any? { |tv| /#{tv}/i.match?(normalized_val) }
https://ruby-doc.org/core-2.5.1/Regexp.html#method-i-match-3F
https://ruby-doc.org/core-2.7.0/Array.html#method-i-any-3F
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Seeing this code has made me realize that there is a defect here (which at this point would probably be considered a breaking change): we aren't matching the entire string (i.e., looking for an exact match) 😱 .
This means,
!!/1/.match("false1") # true
!!/true/.match("this is not true") # true
I'll raise an issue for this separately.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
good catch
context 'returns true with options for' do | ||
trues_with_options.each do |cond| | ||
it "#{cond[0]}: #{cond[1]}, option: {#{cond[2]}: #{cond[3]}}" do | ||
expect(parser.parse(name, cond[1], true_value: cond[3])).to match({name => true}) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Out of curiosity, is the older hashrocket syntax required?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
no it is not, but it seems that this spec file and some others use it. Might be worth updating all tests in a separate PR?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Makes sense to me! Happy to consider that out of scope here
context 'with empty string and empty options' do | ||
it 'returns nil' do | ||
expect(parser.parse(name, '', true_value: '')).to match({name => nil}) | ||
end | ||
end |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is an interesting edge case. Would you not expect setting true_value: ""
to return true
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In the normalize
method we return nil
if blank
return if (value.nil? || value.blank?) # line 22
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this was part of the originally functionality
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
agreed - this also related to the them of a level of opinion in the parser
end | ||
|
||
context 'returns false with options for' do | ||
falses_with_options.each do |cond| |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do you need an array for this or could it be just a single assertion? It looks like you arrived to that conclusion by only having one entry in the array. I think you could make this test more explicit if you kept it just to one entry.
@oroth8 Yikes, it's only been a year since I last looked at this 🙈 🙊 . Would you mind resolving the conflicts? Are there any other updates required from the code review feedback? |
curious what's open on this one? kinda want to use it 😅 |
Items Addressed
feature request
true_value
option to:boolean
parsertrue_value
can be any single valueNotes for Reviewers
true_value
but open for suggestions