HDXDSYS-742 global coverage #160
Conversation
Test Results1 tests - 16 0 ✅ - 17 8s ⏱️ - 15m 4s For more details on these errors, see this check. Results for commit 2e11555. ± Comparison against base commit a19efdd. ♻️ This comment has been updated with latest results. |
Pull Request Test Coverage Report for Build 10838185780Details
💛 - Coveralls |
| @@ -70,12 +88,22 @@ def populate(self) -> None: | |||
| add_missing_value_message( | |||
| errors, identifier, "admin 2 code", adm2_name | |||
| ) | |||
| continue | |||
| if adm1_code is None: | |||
| # Map units that cannot be p-coded to national level | |||
There was a problem hiding this comment.
Will the units that cannot be p-coded be mapped to UNSPECIFIED admin 2 codes of the form CD41-XXX or COD-XXX-XXX?
Should this change be added to food security (not necessarily in this PR)?
There was a problem hiding this comment.
Units that can't be p-coded at adm1 will be mapped to unspecified units like COD-XXX-XXX. If we can p-code them to adm1 but not adm2, then they'll be mapped to units like CD41-XXX.
I think it would be good to add this kind of logic to food security! The more data we can get into HAPI the better, even if we can only p-code it to a certain level.
| self.hapi_countries = countries | ||
| else: | ||
| self.hapi_countries = list( | ||
| Country.countriesdata()["countries"].keys() |
There was a problem hiding this comment.
I think this should be added as a method to HDX Python Country at some point
|
The only pipeline I'm concerned about is conflict events, which ran ok on my computer. I'll test the database export before releasing! |
Hdxdsys 1024 provider admin names
|
Incorporated into develop-global branch |
I've added global coverage for all of the standardized data categories. The ones left are non-standard data: operational presence, population, and poverty rate. I left food security out of this branch because I think it's worth waiting to incrementally p-code the data if we can. If we need to p-code it first I'll make a new branch.