API, getfame for exporting fame data as json for further use and analysis Use the API to get one or more series from a Fame database, into your favorite programming language such as Python, javascript, R other.. Using getfame -e fameexpressions you can evaluate series-expressions, converting, summing and take use of all the functionality and power FAME offers.
getfame -e can be used with custom functions inside functions.pro ie common basis year: cb(seriesname,yyyy)
1. Create a directory mkdir api ie ../../../api where the application and files can be stored
2. copy all files to this folder first time.
3. start FAME and compile the 2 .pro files: getfame and functions
4. All users should have (rx) access to files in this folder) If log file stored here users need w access to common log file
5. getfame needs to be in PATH or you need to prefix with the path when executing
5b.Test to execute ie getfame-n from a xterm before jupiterlab or other to get better control errormessage
5c.New users may use getfame -n getfame -s and getfame -e instead of the 3 identical scripts getfamenames getfamenames getfameexpr as they are to be phased out
6. Explore the directory $HOME/.GetFAME for error messages and output (this will also be listed as part of the execution)
7. The new getfame script isd different from getfamenames, getfameseries & getfameexpr as it takes different num of arguments.
8. Use getfame -n getfame -s and the most advanced getfame -e
Note
Idea: The 3 IDENTICAL scripts ( could/should have been written as one ) each produce a getfameexpr.json, getfamenames.json and getfameseries.json for external programs to have the possibility to read and connect the metadata from getfamenames and getfameseries in one qry /map. The scripts are identical. The script getfame can be used instead in attition to the 3 identical scripts, then use getfame -n getfame -s getfame -e
Note
Use double quotes around parameters when special characters is in use as the parameters getfameexpr "mydb.db" "pct(myseries.A)" which is similar to getfame -e "mydb.db" "pct(myseries.A)"
Note
Idea: To Use epoch date representation for comparison of timeseries of unequal frequencies and length, and use the json format in python R & not Excel. See sample.py for an starting template using python in jupiterlab and/or R-n-fameexpr.ipynb for a R-sample template.
Important
1)getfamenames == getfame -n, 2)getfameseries == getfame -s & 3)getfameexpr == getfame -e
1. getfamenames / getfame -n: wildcard/objectname search for famenames -> gets metadata info ( no date param)
xterm : getfame -n $REFERTID/data/fornavn.db "ERIK"
xterm : getfame -n $REFERTID/data/fornavn.db "?ERIK?,JIM,JAN?"
xterm : getfame -n $REFERTID/data/fornavn.db "?RIK,KRISTIN,JIM,HE?"
xterm : getfame -n "/ssb/bruker/refertid/data/kpi_publ.db, /ssb/bruker/refertid/data/fornavn.db" "?T?I?" | more
jupiterlab : !ssh sl-fame-1.ssb.no 'getfamenames $REFERTID/data/fornavn.db "?ERIK" '
=> sample getfamenames json result:
[{"GetFameJsonApi": "ErikS",
"ApiVersion": "20240721",
"Executed": "24-Jul-24 19:09:49",
"Famever": "11.53",
"Database": "/ssb/bruker/refertid/data/fornavn.db",
"Wildcard": "?ERIK?",
"Found": 4,
"Series":[
{"Name":"ERIK","Class":"SERIES","Desc":"Erik Popularitet i prosent","Updated":"16MAY18"},
{"Name":"ERIKA","Class":"SERIES","Desc":"Erika Popularitet i prosent","Updated":"16MAY18"},
{"Name":"FREDERIK","Class":"SERIES","Desc":"Frederik Popularitet i prosent","Updated":"16MAY18"},
{"Name":"JAN_ERIK","Class":"SERIES","Desc":"Jan_erik Popularitet i prosent","Updated":"16MAY18"}
]
}]
2. getfameseries / getfame -s -> gets observations for >=1 series/wildcards. No functions. Convert will use observed settings from FAME definition
xterm: getfame -s $REFERTID/data/kpi_publ.db "total.ipr"
xterm: getfame -s /ssb/bruker/refertid/data/kpi_publ.db "total.ipr" "date 2024"
xterm: getfame -s /ssb/bruker/refertid/data/kpi_publ.db "total.ipr" "freq m; date thisday(m)-5 to *""
xterm: getfame -s $REFERTID/data/fornavn.db "?ERIK,KRISTIN,JIM?}" "date 2010 to 2012"
xterm: getfame -s /ssb/bruker/refertid/data/fornavn.db "?JAN?" "date 2000 to 2005"
xterm: getfame -s "mynames.db, /ssb/bruker/refertid/data/fornavn.db" "JI?" "date 2000 to *; deci 2"
jupiterlab: !ssh sl-fame-1.ssb.no 'getfame -s $REFERTID/data/fornavn.db "ERIK?" "date 2000 to *" '
=> sample getfameseries json result:
[{"GetFameJsonApi": "ErikS",
"ApiVersion": "20240721",
"Executed": "24-Jul-24 19:27:46",
"Famever": "11.53",
"Database": "/ssb/bruker/refertid/data/fornavn.db",
"Series": [
{"Name": "JILL",
"Desc": "Jill Popularitet i prosent",
"Daterange": "2000 TO *",
"Frequency": "ANNUAL",
"Observations":[
{"Date":"2000-01-01", "Value":0.006154793, "Epo":[946684800000, 0.006154793]},
{"Date":"2001-01-01", "Value":0.00958834, "Epo":[978307200000, 0.00958834]},
{"Date":"2002-01-01", "Value":null, "Epo":[1009843200000, null]},
{"Date":"2003-01-01", "Value":0.01281312, "Epo":[1041379200000, 0.01281312]},
{"Date":"2004-01-01", "Value":0.003196829, "Epo":[1072915200000, 0.003196829]},
{"Date":"2005-01-01", "Value":0.01283326, "Epo":[1104537600000, 0.01283326]},
{"Date":"2006-01-01", "Value":0.006285158, "Epo":[1136073600000, 0.006285158]},
{"Date":"2007-01-01", "Value":null, "Epo":[1167609600000, null]},
{"Date":"2008-01-01", "Value":0.006213496, "Epo":[1199145600000, 0.006213496]},
{"Date":"2009-01-01", "Value":0.003051292, "Epo":[1230768000000, 0.003051292]},
{"Date":"2010-01-01", "Value":0.003093772, "Epo":[1262304000000, 0.003093772]},
{"Date":"2011-01-01", "Value":0.006469979, "Epo":[1293840000000, 0.006469979]},
{"Date":"2012-01-01", "Value":0.003376325, "Epo":[1325376000000, 0.003376325]}
] } ,
{"Name": "JIM",
"Desc": "Jim Popularitet i prosent",
"Daterange": "2000 TO *",
"Frequency": "ANNUAL",
"Observations":[
{"Date":"2000-01-01", "Value":0.02642551, "Epo":[946684800000, 0.02642551]},
{"Date":"2001-01-01", "Value":0.02763109, "Epo":[978307200000, 0.02763109]},
{"Date":"2002-01-01", "Value":0.02512405, "Epo":[1009843200000, 0.02512405]},
{"Date":"2003-01-01", "Value":0.03990913, "Epo":[1041379200000, 0.03990913]},
{"Date":"2004-01-01", "Value":0.03663787, "Epo":[1072915200000, 0.03663787]},
{"Date":"2005-01-01", "Value":0.0215504, "Epo":[1104537600000, 0.0215504]},
{"Date":"2006-01-01", "Value":0.0119976, "Epo":[1136073600000, 0.0119976]},
{"Date":"2007-01-01", "Value":0.02411018, "Epo":[1167609600000, 0.02411018]},
{"Date":"2008-01-01", "Value":0.008790693, "Epo":[1199145600000, 0.008790693]},
{"Date":"2009-01-01", "Value":0.02022186, "Epo":[1230768000000, 0.02022186]},
{"Date":"2010-01-01", "Value":0.01475361, "Epo":[1262304000000, 0.01475361]},
{"Date":"2011-01-01", "Value":0.006076072, "Epo":[1293840000000, 0.006076072]},
{"Date":"2012-01-01", "Value":0.003228097, "Epo":[1325376000000, 0.003228097]}
] } ] } ]
3. getfameexpr / getfame -e: ->gets observations, series-expression as a result, based on a fame expression. recommended. Full flexibility for converting.
xterm: getfame -e $REFERTID/data/fornavn.db "ERIK"
xterm: getfame -e $REFERTID/data/fornavn.db "mave(ERIK,2)" "date 2000 to 2010"
xterm: getfame -e $REFERTID/data/fornavn.db "Lsum(ERIK,EIRIK)" "date 2000 to *"
xterm: getfame -e $REFERTID/data/kpi_publ.db "convert(total.ipr,annual,constant)" "date 2020 to *"
--use eof custom common basis to set base-year to 2010 (instead of current 2015:) more custom functions are available
xterm: getfame -e /ssb/bruker/refertid/data/kpi_publ.db "cb(total.ipr,2010)" "date 2010 to 2020"
xterm: getfame -e $REFERTID/data/kpi_publ.db "total.ipr" "freq m; date jan20 to feb20;deci 0"
xterm: getfame -e "mynames.db, $REFERTID/data/fornavn.db" "ERIK+EIRIK" "date 2000 to 2010;deci 1"
jupiterlab : !ssh sl-fame-1.ssb.no 'getfame -e $REFERTID/data/fornavn.db "pct(ERIK)" "date 2000 to *" '
=> sample getfameexpr json result:
[{"GetFameJsonApi": "ErikS",
"ApiVersion": "20240721",
"Executed": "24-Jul-24 19:12:00",
"Famever": "11.53",
"Database": "/ssb/bruker/refertid/data/fornavn.db",
"Series": [
{"Name": "ERIK",
"Desc": "ERIK",
"Daterange": "2000 TO 2010",
"Frequency": "ANNUAL",
"Observations":[
{"Date":"2000-01-01", "Value":0.786893, "Epo":[946684800000, 0.786893]},
{"Date":"2001-01-01", "Value":0.8105121, "Epo":[978307200000, 0.8105121]},
{"Date":"2002-01-01", "Value":0.731738, "Epo":[1009843200000, 0.731738]},
{"Date":"2003-01-01", "Value":0.8473015, "Epo":[1041379200000, 0.8473015]},
{"Date":"2004-01-01", "Value":0.6900131, "Epo":[1072915200000, 0.6900131]},
{"Date":"2005-01-01", "Value":0.7388707, "Epo":[1104537600000, 0.7388707]},
{"Date":"2006-01-01", "Value":0.6718656, "Epo":[1136073600000, 0.6718656]},
{"Date":"2007-01-01", "Value":0.7022091, "Epo":[1167609600000, 0.7022091]},
{"Date":"2008-01-01", "Value":0.6299997, "Epo":[1199145600000, 0.6299997]},
{"Date":"2009-01-01", "Value":0.6326554, "Epo":[1230768000000, 0.6326554]},
{"Date":"2010-01-01", "Value":0.649159, "Epo":[1262304000000, 0.649159]}
] } ] } ]