Closed xriss closed 7 years ago
@xriss, I've got 'https://github.com/devinit/digital-platform/blob/master/country-year/poorest20pct-percentages.csv' as not being used in the DH:
ddw_development=# SELECT
ddw_development-# *
ddw_development-# FROM public.di_concept_in_dh
ddw_development-# WHERE id = 'poorest20pct-percentages';
concept_id | series | id | in_dh
------------+--------------+--------------------------+-------
218 | country-year | poorest20pct-percentages | 0
(1 row)
@akmiller01 helped me figure out which files were/are actually being used in the DH here: https://github.com/devinit/ddw-data/issues/124. As this one came out as in_dh = 0 I didn't import the table.
@timstrawson, can you help please? Do we leave 'poorest20pct-percentages.csv' out? Do I need to create a table it?
@xriss, you were right about
UNUSED WAREHOUSE TABLE: fact.gdp_usd_current_2012
Don't use this one. I'm going to update the comment above to correct this.
This is right:
country-year/gdp-usd-current.csv <- fact."gdp_usd_current"
@xriss, just to confirm, this is right:
country-year/gdp-usd-current.csv <- fact."gdp_usd_current"
country-year/gni-pc-usd-current.csv <- fact."gni_pc_usd_current"
country-year/gni-usd-current.csv <- fact."gni_usd_current"
and this is also right:
UNUSED WAREHOUSE TABLE: fact.gdp_usd_current_2012
UNUSED WAREHOUSE TABLE: fact.gni_pc_usd_current_2012
UNUSED WAREHOUSE TABLE: fact.gni_usd_current_2012
I got it wrong earlier today so disregard me trying to mislead you on these. They are UNUSED. I've corrected the information in the above comment.
Hi @dw8547 as discussed just now, I can confirm that we don't need to copy the unused tables across to the data warehouse.
@kriss, @timstrawson has confirmed that the data series/.csv files below are not even used in the DH so do not need to be replaced with DW data.
The following DP csv files are not going to be replaced with DW data
MISSING FROM WAREHOUSE: country-year/adult-literacy
MISSING FROM WAREHOUSE: country-year/domestic-netlending
MISSING FROM WAREHOUSE: country-year/education-pc-transferred-oda
MISSING FROM WAREHOUSE: country-year/employment-agriculture
MISSING FROM WAREHOUSE: country-year/employment-by-sector
MISSING FROM WAREHOUSE: country-year/employment-industry
MISSING FROM WAREHOUSE: country-year/employment-services
MISSING FROM WAREHOUSE: country-year/gdp-current-ncu-fy
MISSING FROM WAREHOUSE: country-year/gdp-growth
MISSING FROM WAREHOUSE: country-year/gdp-pc-usd-2005
MISSING FROM WAREHOUSE: country-year/gdp-pc-usd-current
MISSING FROM WAREHOUSE: country-year/gdp-usd-2005
MISSING FROM WAREHOUSE: country-year/gdp-usd-2012
MISSING FROM WAREHOUSE: country-year/gni-usd-2005
MISSING FROM WAREHOUSE: country-year/govtspend-USD
MISSING FROM WAREHOUSE: country-year/health-pc-transferred-oda
MISSING FROM WAREHOUSE: country-year/income-share-top-10pc
MISSING FROM WAREHOUSE: country-year/infant-mortality
MISSING FROM WAREHOUSE: country-year/in-oda-and-repayments
MISSING FROM WAREHOUSE: country-year/in-oof-and-repayments
MISSING FROM WAREHOUSE: country-year/in-oof-net
MISSING FROM WAREHOUSE: country-year/intl-flows-donors-wide
MISSING FROM WAREHOUSE: country-year/intl-flows-recipients-wide
MISSING FROM WAREHOUSE: country-year/kenya-electricity-avg
MISSING FROM WAREHOUSE: country-year/kenya-electricity-rank
MISSING FROM WAREHOUSE: country-year/kenya-improved-sani-avg
MISSING FROM WAREHOUSE: country-year/kenya-improved-sani-rank
MISSING FROM WAREHOUSE: country-year/kenya-improved-water-avg
MISSING FROM WAREHOUSE: country-year/kenya-improved-water-rank
MISSING FROM WAREHOUSE: country-year/kenya-paved-roads-avg
MISSING FROM WAREHOUSE: country-year/kenya-paved-roads-rank
MISSING FROM WAREHOUSE: country-year/kenya-pov-avg
MISSING FROM WAREHOUSE: country-year/kenya-pov-rank
MISSING FROM WAREHOUSE: country-year/kenya-urban-avg
MISSING FROM WAREHOUSE: country-year/kenya-urban-rank
MISSING FROM WAREHOUSE: country-year/long-term-debt
MISSING FROM WAREHOUSE: country-year/mean-years-of-schooling
Dead issue.
I can see that some of the data is available with slightly different table names, eg
fact.population_by_age_0_14 -> country-year/population-0-14
@xriss, yes. The tables that have names different from the file names in the GitHub repository are (not accounting for possible typos):
In 6 & 7 we've changed uppercase to lowercase only.
(I do a simple replace of _ - and then a string compare of the filename to try and auto find the matching csv table so these are not going to be picked up until I add them in explicitly.) @xriss, OK, got it. The above will need to be added then.
There is the oda and oda_donor which should be split and we should use the _2012 version (right?), @xriss, that's right:
however, it still looks like we are missing some data.
Take a look at all the other MISSING FROM WAREHOUSE lines below and please advise where that data should come from. @xriss, OK, having a look now.
@kriss, that's right, the above DW tables are not needed for your purposes.
@kriss, the above DW tables are needed, see table at the top with summary of name changes.
@kriss, that's right, the above DW tables are not needed for your purposes.
@kriss, that's right. My understanding is that these data series/.csv files are not even used in the DH so do not need to be replaced with DW data. @timstrawson, can you please confirm?
@kriss, these do need to be replaced with DW data. The data for these is in fact.oda_donor_2012. To get at the individual donor data, we need to filer on the di_id for the relevant years so:
@xriss, watch out for:
Delete:
Use:
instead.
@xriss, more to watch out for:
More info about this here: https://github.com/devinit/digital-platform/issues/243
@kriss, that's right. These do not need to be replaced with DW data.
@kriss, this one does need to be replaced with DW data, but is genuinely missing from DW. I will correct this & let you & @notshi know when I have.
@kriss, these do need to be replaced with DW data, see table at the top with summary of name changes.
@kriss, that's right. These do not need to be replaced with DW data.
@xriss, yes, that's right.
@xriss, yes, that's right.
@xriss, no, do (filtering on year BETWEEN 2006 AND 2014):
@xriss, yes, that's right.
@xriss, we need these too, see table at the top with summary of name changes.