Open Aeilert opened 2 years ago
Thanks @Aeilert My initial thought is not to mess up with the response. A solution that could probably satisfy everyone is to have another endpoint that provides that information (we may already have it) and can easily be merged with the main response. What do you think?
Thanks @tonyfujs
Yep. I like the new endpoint idea!
@randrescastaneda , @danielmahler: What do you think?
That could work. What would that imply in practice if I use, say, the Stata ado?
From: Aeilert @.>
Sent: Wednesday, February 2, 2022 6:45 AM
To: PIP-Technical-Team/pipapi @.>
Cc: Daniel Gerszon Mahler @.>; Mention @.>
Subject: Re: [PIP-Technical-Team/pipapi] Show survey_years
when fill_gaps = TRUE
? (Issue #191)
[External]
Yep. I like the new endpoint idea!
@randrescastanedahttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Frandrescastaneda&data=04%7C01%7Cdmahler%40worldbank.org%7Cbc716ae8947940710b6b08d9e6416688%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637793990839353545%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=iZY2GA8M4R4dW%2FXMK%2FymtA%2B%2ByHCsR3ejf2cENrDn23Y%3D&reserved=0 , @danielmahlerhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdanielmahler&data=04%7C01%7Cdmahler%40worldbank.org%7Cbc716ae8947940710b6b08d9e6416688%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637793990839353545%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=HfAoNOXwYnFdGQsA4Wt7gDT8bS4x1SoxYmE1tljHy%2Fs%3D&reserved=0: What do you think?
- Reply to this email directly, view it on GitHubhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FPIP-Technical-Team%2Fpipapi%2Fissues%2F191%23issuecomment-1027857060&data=04%7C01%7Cdmahler%40worldbank.org%7Cbc716ae8947940710b6b08d9e6416688%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637793990839353545%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=PaW%2Flj2MfvGcH4aJWbXMzCRMc6ykO2JeGcmMn6m1FDk%3D&reserved=0, or unsubscribehttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMVXK3H444JWFKD4EQDWUJDUZEKKNANCNFSM5NLVSUAA&data=04%7C01%7Cdmahler%40worldbank.org%7Cbc716ae8947940710b6b08d9e6416688%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637793990839353545%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=v08fCwel1dm7dsC8wYVOhNEYrPLqm5NCVyioX2QxxH0%3D&reserved=0. Triage notifications on the go with GitHub Mobile for iOShttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fapps.apple.com%2Fapp%2Fapple-store%2Fid1477376905%3Fct%3Dnotification-email%26mt%3D8%26pt%3D524675&data=04%7C01%7Cdmahler%40worldbank.org%7Cbc716ae8947940710b6b08d9e6416688%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637793990839353545%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=5nDnptAem8ehv589F1afI1mFTiLXpcHjsoSYSsnYyDQ%3D&reserved=0 or Androidhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fplay.google.com%2Fstore%2Fapps%2Fdetails%3Fid%3Dcom.github.android%26referrer%3Dutm_campaign%253Dnotification-email%2526utm_medium%253Demail%2526utm_source%253Dgithub&data=04%7C01%7Cdmahler%40worldbank.org%7Cbc716ae8947940710b6b08d9e6416688%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637793990839353545%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=vHRQ%2FE0lP62xtVxKogmXBciogW0H%2FLP9I6o2kLRuDF0%3D&reserved=0. You are receiving this because you were mentioned.Message ID: @.**@.>>
That could work. What would that imply in practice if I use, say, the Stata ado?
@danielmahler: Good question. So I think the easiest would be to just have a new command / function hat pulls that information. I.e in addition to pip
you have a set of other methods. E.g for the R wrapper I'm imagining functions like pipr::pip()
(main response from /pip endpoint), pipr::aux()
(select your AUX table of choice) and maybe a few other functions. The information on survey years used for interpolation could either be in a new "aux" table or have a completely separate function (e.g pipr::something_else
)
Gotcha, that sounds complicated, in that case my view is that you should forget this comment of mine and focus on the other items on your plate 😊
From: Aeilert @.>
Sent: Thursday, February 3, 2022 2:56 AM
To: PIP-Technical-Team/pipapi @.>
Cc: Daniel Gerszon Mahler @.>; Mention @.>
Subject: Re: [PIP-Technical-Team/pipapi] Show survey_years
when fill_gaps = TRUE
? (Issue #191)
[External]
That could work. What would that imply in practice if I use, say, the Stata ado?
@danielmahlerhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdanielmahler&data=04%7C01%7Cdmahler%40worldbank.org%7Cfd53ae5c6fa7490c303d08d9e6eaa78c%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637794717774949229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=tinxvn1UMadg3z%2FpqQxM7Lj3y2q8Q1CIh8MOxvQkRk0%3D&reserved=0: Good question. So I think the easiest would be to just have a new command / function hat pulls that information. I.e in addition to pip you have a set of other methods. E.g for the R wrapper I'm imagining functions like pipr::pip() (main response from /pip endpoint), pipr::aux() (select your AUX table of choice) and maybe a few other functions. The information on survey years used for interpolation could either be in a new "aux" table or have a completely separate function (e.g pipr::something_else)
— Reply to this email directly, view it on GitHubhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FPIP-Technical-Team%2Fpipapi%2Fissues%2F191%23issuecomment-1028697128&data=04%7C01%7Cdmahler%40worldbank.org%7Cfd53ae5c6fa7490c303d08d9e6eaa78c%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637794717774949229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=vLmXipqepWjL4HGsjo9Pi%2BC81JSyTFCV1LM1zahK43E%3D&reserved=0, or unsubscribehttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMVXK3EKKKKXG4P7PSDQN53UZIYJZANCNFSM5NLVSUAA&data=04%7C01%7Cdmahler%40worldbank.org%7Cfd53ae5c6fa7490c303d08d9e6eaa78c%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637794717774949229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=1dq23aP4%2BqciO0ROajZWD%2FSUNCpa2kqApl4GfSpcUb8%3D&reserved=0. Triage notifications on the go with GitHub Mobile for iOShttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fapps.apple.com%2Fapp%2Fapple-store%2Fid1477376905%3Fct%3Dnotification-email%26mt%3D8%26pt%3D524675&data=04%7C01%7Cdmahler%40worldbank.org%7Cfd53ae5c6fa7490c303d08d9e6eaa78c%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637794717774949229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=CCk6uLR%2FNFN5vDiLeIs%2BUxnz02QrdIBguKUhZ60upig%3D&reserved=0 or Androidhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fplay.google.com%2Fstore%2Fapps%2Fdetails%3Fid%3Dcom.github.android%26referrer%3Dutm_campaign%253Dnotification-email%2526utm_medium%253Demail%2526utm_source%253Dgithub&data=04%7C01%7Cdmahler%40worldbank.org%7Cfd53ae5c6fa7490c303d08d9e6eaa78c%7C31a2fec0266b4c67b56e2796d8f59c36%7C0%7C0%7C637794717774949229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=NzSG1YD8houHjvWzb7BPaLYU6TjjRXGUHB8DA8%2Fi0ms%3D&reserved=0. You are receiving this because you were mentioned.Message ID: @.**@.>>
Original issue raised by @danielmahler : "Is there a way to show the survey_years that is used in the interpolation? I think not, but if there is, then that would be incredibly helpful to users (and for us to calculate coverage)".
This is technically possible, but has also has some technical downsides;
survey_year
since it is numeric when fill_gaps = FALSE, but would need to be character if it is a string of collapsed survey years (e.g."2000|2005"
).survey_year
.On the hand maybe this would be so beneficial for users that we should come up with a solutions. Any thoughts @tonyfujs?
We won't need this to calculate coverage however, since that will happen at the pipeline stage (using the full refyear table)