Understanding openaddresses data format

I have downloaded us-west geolocation data (postal addresses) from openaddresses.io. Some of the addresses in the datasets are not complete i.e., some of them doesn't have info like zip_code. Is there a way to retrieve it or is the data incomplete?

I have tried to search other files hoping to find any related info. The complete dataset doesn't contain any info relate to it. City of Mesa, AZ has multiple zip codes, so it is hard to assign one to the address. Is there any way to address this problem?

This is how data looks like (City of Mesa, AZ)

LON,LAT,NUMBER,STREET,UNIT,CITY,DISTRICT,REGION,POSTCODE,ID,HASH
-111.8747353,33.456605,790,N DOBSON RD,,SRPMIC,,,,,dc0c53196298eb8d
-111.8886227,33.4295194,2630,W RIO SALADO PKWY,,MESA,,,,,c38b700309e1e9ce
-111.8867018,33.4290795,2401,E RIO SALADO PKWY,,TEMPE,,,,,9b912eb2b1300a27
-111.8832045,33.4232903,700,S EVERGREEN RD,,TEMPE,,,,,3435b99ab3f4f828
-111.8761202,33.4296416,2100,W RIO SALADO PKWY,,MESA,,,,,b74349c833f7ee18
-111.8775844,33.4347782,1102,N RIVERVIEW,,MESA,,,,,17d0cf1542c66083

Solution

Short Answer: The data is incomplete.

The data in OpenAddresses.io is only as complete as the datasource it pulls from. OpenAddresses is just an aggregation of publicly available datasets. There's no real consistency between government agencies that make their data available. As a result, other sections of the OpenAddresses dataset might have city names or zip codes, but there's often something missing.

If you're looking to fill in the missing data, take a look at how projects like Pelias use multiple data sources to augment missing data.

Personally, I always end up going back to OpenStreetMaps (OSM). One could argue that OpenAddresses is better quality because it comes from official sources and doesn't try to fill in data using approximations, but the large gaps of missing data make it far less useful, at least on its own.