-
Notifications
You must be signed in to change notification settings - Fork 28
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
Implementation for global pincodes #34
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.
@abhu-A-J Looks good to me.
@abhu-A-J We need to remove multiple similar fields when one types a pincode. Like I can see multiple "Pune" in the city dropdown and similarly in state. |
@plxity Also lets keep this to a separate testing branch since it is a completely different development for the module. We'll also encourage other participants to work on the possible improvements to this PR. |
@yellowwoods12 The repetition is due to the dataset itself. It's because the dataset has similar fields, if we use a dataset that has no repetition then that issue won't be present. It's completely dependent on the offline data used. |
Yes, definitely. |
@abhu-A-J You can remove duplicate fields in this too. Please make those changes then we are going to merge this PR. |
@plxity The duplicate data fields are present because the dataset itself has duplicate field. If we can get the data set with non-repetitive fields, this issue won't be present. It's not a side effect from code. You can see this here the data coming back from API endpoint created. |
@abhu-A-J Yes we get that. What we are suggesting here is to filter out duplicate entries after you've fetched them from the API. |
@yellowwoods12 @plxity I have removed the problem for repetitive data. Here is the working now. |
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.
@abhu-A-J Good Work!
@abhu-A-J Can you raise this PR to the development branch? |
Sure, I'll do the same, so I'm closing this PR and make a new PR to development branch. |
fixes #2
@plxity here's what the PR offers:
As discussed since the only way to achieve this would be via offline data set, I have implemented the same version in React demonstrated here.
As discussed right now the data set only includes for India and US zip codes and so it'll work only for that.
The issue about the bundle size optimization is addressed via an API as you suggested, the API not not built extremely robust it's just enough to be up and running and I have used the same to minimize bundle size.
Here's a small example for it's working:

Future Improvements which can be done: