R/gta bulk prep.R
In global-trade-alert/gtalibrary: Contains functions to work with GTA data
Defines functions
gta_bulk_prep
Documented in
gta_bulk_prep
# Roxygen documentation
#' Split a data frame containing GTA entries into the components we need for a bulk import. Returns 2x 3 CSVs i.e. a test and the full set.
#'
#' @param dump.name Specify a name for this data dump. It will be used in the files. Default is 'data dump'.
#' @param master.frame Specify data frame that contains all relevant information. Default is 'master'.
#' @param test.size Number of state acts to include in the test set. Default is 5. There is NO overlap between the test and the full set CSVs i.e. you have to import both eventually.
#' @param author State your name. Default is "Piotr".
#' @param sa.id Name of the column containing the state act ID. Default is 'state.act.id'.
#' @param sa.title Name of the column containing the title of the state act. Default is 'title'.
#' @param sa.description Name of the column containing the announcement description. Default is 'announcement.description'.
#' @param announcement.date Name of the column containing the announcement date. Default is 'date.announced'.
#' @param sa.source Name of the column containing the source. Default is 'source'.
#' @param source.off Name of the column containing whether source is official (T/F). Default is 'source.official'.
#' @param int.id Name of the column containing ther intervention ID. Default is 'intervention.id'.
#' @param implementer Name of the column containing the implementing jurisdiction. Default is 'implementing.jurisdiction'.
#' @param implementer.type State how the implementing countries are specified ('un','name'). Default is 'name'.
#' @param int.type Name of the column containing the intervention type. Default is 'intervention.type'.
#' @param int.description Name of the column containing the intervention description. Default is 'intervention.description'.
#' @param int.flow Name of the column containing the affected flow. Default is 'affected.flow'.
#' @param eval.gta Name of the column containing the GTA evaluation. Default is 'gta.evaluation'.
#' @param eval.research Name of the column containing the research evaluation. Default is 'research.evaluation'.
#' @param inception.date Name of the column containing the inception date. Default is 'date.implemented'.
#' @param removal.date Name of the column containing the removal date. Default is 'date.removed'.
#' @param temporary Name of the column containing whether duration is announced as temporary (T/F). Default is 'duration.limited'.
#' @param firms Name of the column containing the eligible firms. Default is 'eligible.firms'.
#' @param imp.level Name of the column containing the implementation level. Default is 'implementation.level'.
#' @param is.noncommercial Name of the column containing whether there is a higher motive (T/F). Default is 'non.commercial'.
#' @param is.horizontal Name of the column containing whether the intervention is horizontal. Default is 'horizontal'.
#' @param is.fta Name of the column containing whether the affected trading partners may include FTA partners (T/F). Default is 'include.fta'.
#' @param int.prior Name of the column containing the prior level values for the intervention description. Default is 'level.prior'.
#' @param int.new Name of the column containing the new level values for the intervention description. Default is 'level.prior'.
#' @param int.unit Name of the column containing the unit for level values attached to the intervention description. Default is 'level.prior'.
#' @param sectors.nonhs Name of the column containing the affected sectors that are not generated by the affected tariff lines. Default is 'affected.sectors.nonhs'.
#' @param dm.nonsupport Name of the column containing the distorted markets that are not generated from a support table. Default is 'dm.editor.choice'.
#' @param dm.type State how the distorted markets are specified ('un','name'). Default is 'name'.
#' @param dm.freeze Name of the column containing whether the distorted market choice should be frozen as specified (T/F). Default is 'freeze.dm'.
#' @param aj.nonsupport Name of the column containing the affected jurisdictions that are not generated from a support table. Default is 'aj.editor.choice'.
#' @param aj.type State how the affected jurisdictions are specified ('un','name'). Default is 'name'.
#' @param aj.freeze Name of the column containing whether the affected jurisdictions choice should be frozen as specified (T/F). Default is 'freeze.aj'.
#' @param hs.code Name of the column containing the affected tariff line. Default is 'affected.product'.
#' @param hs.prior Name of the column containing the prior level for the affected tariff line. Default is 'atl.prior'.
#' @param hs.new Name of the column containing the new level for the affected tariff line. Default is 'atl.new'.
#' @param hs.unit Name of the column containing the unit for the affected tariff line values. Default is 'atl.unit'.
#' @param hs.official Name of the column containing the whether the affected tariff line is official (T/F). Default is 'atl.official'.
#' @param hs.inception Name of the column containing the inception date for the affected tariff line (if different from intervention date). Default is 'atl.implemented'.
#' @param hs.removal Name of the column containing the removal date for the affected tariff line (if different from intervention date). Default is 'atl.removed'.
#'
#'
#' @return Be up to date with our latest functions.
#' @references www.globaltradealert.org
#' @author Global Trade Alert
gta_bulk_prep = function(
dump.name="data dump",
master.frame="master",
test.size=5,
author="Piotr",
sa.id="state.act.id",
sa.title="title",
sa.description="announcement.description",
announcement.date="date.announced",
sa.source="source",
source.off="source.official",
int.id="intervention.id",
implementer="implementing.jurisdiction",
implementer.type="name",
int.type="intervention.type",
int.description="intervention.description",
int.flow="affected.flow",
eval.gta="gta.evaluation",
eval.research="research.evaluation",
inception.date="date.implemented",
removal.date="date.removed",
temporary="duration.limited",
firms="eligible.firms",
imp.level="implementation.level",
is.noncommercial="non.commercial",
is.horizontal="horizontal",
is.fta="include.fta",
int.prior="level.prior",
int.new="level.new",
int.unit="level.unit",
sectors.nonhs="affected.sectors.nonhs",
dm.nonsupport="dm.editor.choice",
dm.type="name",
dm.freeze="freeze.dm",
aj.nonsupport="aj.editor.choice",
aj.type="name",
aj.freeze="freeze.aj",
hs.code="affected.product",
hs.prior="atl.prior",
hs.new="atl.new",
hs.unit="atl.unit",
hs.official="atl.official",
hs.inception="atl.implemented",
hs.removed="atl.removed"
){
## loading and renaming the master data frame
print("Loading and formatting master data frame ...")
eval(parse(text=paste("master=", master.frame, sep="")))
given.names=c(sa.id, sa.title, sa.description, announcement.date, sa.source, source.off, int.id, implementer, int.type, int.description, int.flow, eval.gta, eval.research, inception.date, removal.date, temporary, firms, imp.level, is.noncommercial, is.horizontal, is.fta, int.prior, int.new, int.unit, sectors.nonhs, dm.nonsupport, dm.freeze, aj.nonsupport, aj.freeze, hs.code, hs.prior, hs.new, hs.unit, hs.official, hs.inception, hs.removed)
fct.names=c("state.act.id", "title", "announcement.description", "date.announced", "source", "source.official", "intervention.id", "implementing.jurisdiction", "intervention.type", "intervention.description", "affected.flow", "gta.evaluation", "research.evaluation", "date.implemented", "date.removed", "duration.limited", "eligible.firms", "implementation.level", "non.commercial", "horizontal", "include.fta", "level.prior", "level.new", "level.unit", "affected.sectors.nonhs", "dm.editor.choice", "freeze.dm", "aj.editor.choice", "freeze.aj", "affected.product", "atl.prior", "atl.new", "atl.unit", "atl.official", "atl.implemented","atl.removed")
master=unique(master[,given.names])
names(master)=fct.names
print("Loading and formatting master data frame ... complete.")
#### state act CSV
print("Generating state act data frame ...")
master$author.id=29
if(author=="Johannes"){
master$author.id=1
}
if(author=="Josse"){
master$author.id=30
}
master$title=as.character(master$title)
master$announcement.description=as.character(master$announcement.description)
master$date.announced=as.factor(master$date.announced)
master$source=as.character(master$source)
state.act=unique(data.frame(import_id=master$state.act.id,
author_id=master$author.id,
title=master$title,
description=master$announcement.description,
announcement_date=master$date.announced,
source=master$source,
is_source_official=master$source.official,
status_id=4))
print("State act data frame is ready.")
#### intervention CSV
## intervention type conversion
print("Generating intervention data frame ...")
int.type=gtalibrary::int.mast.types
print("... adding intervention types.")
master=merge(master, int.type[,c("intervention.type", "intervention.type.id")], by="intervention.type", all.x=T)
if(nrow(subset(master, is.na(intervention.type.id)))>0){
stop(paste("Unkown intervention type(s): ", paste(unique(subset(master, is.na(intervention.type.id))$intervention.type), collapse="; "), sep=""))
}
## affected flow id
print("... adding affected flow IDs.")
master$affected.flow.id=NA
master$affected.flow.id[tolower(master$affected.flow)=="inward"]=1
master$affected.flow.id[tolower(master$affected.flow)=="outward"]=2
master$affected.flow.id[tolower(master$affected.flow)=="outward subsidy"]=3
if(nrow(subset(master, is.na(affected.flow.id)))>0){
stop(paste("Unkown affected flow(s): ", paste(unique(subset(master, is.na(affected.flow.id))$affected.flow), collapse="; "), sep=""))
}
## eligible firms id
print("... adding eligible firm IDs.")
master$eligible.firm.id=NA
master$eligible.firm.id[tolower(master$eligible.firms)=="all"]=1
master$eligible.firm.id[tolower(master$eligible.firms)=="smes"]=2
master$eligible.firm.id[tolower(master$eligible.firms)=="firm-specific"]=3
master$eligible.firm.id[tolower(master$eligible.firms)=="state-controlled"]=4
master$eligible.firm.id[tolower(master$eligible.firms)=="state trading enterprise"]=5
master$eligible.firm.id[tolower(master$eligible.firms)=="sector-specific"]=6
if(nrow(subset(master, is.na(eligible.firm.id)))>0){
stop(paste("Unkown eligible firm(s): ", paste(unique(subset(master, is.na(eligible.firm.id))$eligible.firms), collapse="; "), sep=""))
}
## implementation level id
print("... adding intervention level IDs.")
master$implementation.level.id=NA
master$implementation.level.id[tolower(master$implementation.level)=="supranational"]=1
master$implementation.level.id[tolower(master$implementation.level)=="national"]=2
master$implementation.level.id[tolower(master$implementation.level)=="subnational"]=3
master$implementation.level.id[tolower(master$implementation.level)=="sez"]=4
master$implementation.level.id[tolower(master$implementation.level)=="ifi"]=5
master$implementation.level.id[tolower(master$implementation.level)=="nfi"]=6
if(nrow(subset(master, is.na(implementation.level.id)))>0){
stop(paste("Unkown implementation level(s): ", paste(unique(subset(master, is.na(implementation.level.id))$implementation.level), collapse="; "), sep=""))
}
## evaluation ids
print("... adding evaluation IDs.")
master$gta.evaluation.id=NA
master$gta.evaluation.id[tolower(master$gta.evaluation)=="red"]=1
master$gta.evaluation.id[tolower(master$gta.evaluation)=="amber"]=2
master$gta.evaluation.id[tolower(master$gta.evaluation)=="green"]=3
if(nrow(subset(master, is.na(gta.evaluation.id)))>0){
stop(paste("Unkown GTA evaluation(s): ", paste(unique(subset(master, is.na(gta.evaluation.id))$gta.evaluation), collapse="; "), sep=""))
}
master$research.evaluation.id=NA
master$research.evaluation.id[tolower(master$research.evaluation)=="red"]=1
master$research.evaluation.id[tolower(master$research.evaluation)=="amber"]=2
master$research.evaluation.id[tolower(master$research.evaluation)=="green"]=3
master$research.evaluation.id[is.na(master$research.evaluation)] =""
if(nrow(subset(master, is.na(research.evaluation.id)))>0){
stop(paste("Unkown research evaluation(s): ", paste(unique(subset(master, is.na(research.evaluation.id))$research.evaluation), collapse="; "), sep=""))
}
## levels
unit.df=data.frame(level.unit.id=c(1,1,1:19),
level.unit=c("percent","%","per cent", "total budget (USD)", "firm-specific budget (USD)", "USD/MT", "USD/kg", "USD/tonne", "USD/unit", "count", "USD/litre", "USD/pc", "USD/squaremetre", "USD/pair", "USD/ldt", "USD/lb", "USD/gallon", "USD/hl", "USD/LAL", "USD/tyre", "USD/stick"))
unit.df$level.unit=tolower(as.character(unit.df$level.unit))
master$level.unit=tolower(as.character(master$level.unit))
print("... reformatting levels, dates etc.")
master$level.prior=as.character(master$level.prior)
master$level.new=as.character(master$level.new)
master=merge(master, unit.df, by="level.unit", all.x=T)
if(nrow(subset(master, is.na(level.unit.id) & is.na(level.unit)==F))>0){
stop(paste("Unkown level unit(s): ", paste(unique(subset(master, is.na(level.unit.id) & is.na(level.unit)==F)$level.unit), collapse="; "), sep=""))
}
## dates
master$date.implemented=as.factor(master$date.implemented)
master$date.removed=as.factor(master$date.removed)
## sectors
master$affected.sectors.nonhs=as.character(master$affected.sectors.nonhs)
master$affected.sectors.nonhs[nchar(master$affected.sectors.nonhs)==2 & is.na(master$affected.sectors.nonhs)==F]=paste(0,master$affected.sectors.nonhs[nchar(master$affected.sectors.nonhs)==2 & is.na(master$affected.sectors.nonhs)==F], sep="")
## country ids
gta.jurisdictions=gtalibrary::country.names
## IJ
print("... adding country IDs.")
if(implementer.type=="name"){
gta.jur=gta.jurisdictions[,c("name","jurisdiction.id")]
} else {
if(implementer.type=="un"){
gta.jur=gta.jurisdictions[,c("un_code","jurisdiction.id")]
} else {
stop("Implementer type is neither 'un' nor 'name'.")
}
}
names(gta.jur)=c("implementing.jurisdiction","implementing.jurisdiction.id")
master=merge(master, gta.jur, by="implementing.jurisdiction", all.x=T)
if(nrow(subset(master, is.na(implementing.jurisdiction.id)))>0){
stop(paste("Unkown implementer(s): ", paste(unique(subset(master, is.na(implementing.jurisdiction.id))$implementing.jurisdiction), collapse="; "), sep=""))
}
### CSVs
if(nrow(subset(master, is.na(implementing.jurisdiction.id)==F))>0){
ij=unique(master[,c("intervention.id", "implementing.jurisdiction", "implementing.jurisdiction.id")])
ij=merge(aggregate(implementing.jurisdiction ~intervention.id, ij, function(x) paste(unique(x), collapse=",")),
aggregate(implementing.jurisdiction.id ~intervention.id, ij, function(x) paste(unique(x), collapse=",")),
by="intervention.id")
master$implementing.jurisdiction.id=NULL
master$implementing.jurisdiction=NULL
master=merge(unique(master), ij, by="intervention.id", all.x=T)
}
## AJ
if(aj.type=="name"){
gta.jur=gta.jurisdictions[,c("name","jurisdiction.id")]
} else {
if(aj.type=="un"){
gta.jur=gta.jurisdictions[,c("un_code","jurisdiction.id")]
} else {
stop("AJ type is neither 'un' nor 'name'.")
}
}
names(gta.jur)=c("aj.editor.choice","affected.jurisdiction.id")
master=merge(master, gta.jur, by="aj.editor.choice", all.x=T)
if(nrow(subset(master, is.na(affected.jurisdiction.id) & is.na(aj.editor.choice)==F))>0){
stop(paste("Unkown affected jurisdiction(s): ", paste(unique(subset(master, is.na(affected.jurisdiction.id) & is.na(aj.editor.choice)==F)$aj.editor.choice), collapse="; "), sep=""))
}
### CSVs
if(nrow(subset(master, is.na(affected.jurisdiction.id)==F))>0){
aj=unique(master[,c("intervention.id", "aj.editor.choice", "affected.jurisdiction.id")])
aj=merge(aggregate(aj.editor.choice ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
aggregate(affected.jurisdiction.id ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
by="intervention.id")
master$affected.jurisdiction.id=NULL
master$aj.editor.choice=NULL
master=merge(unique(master), aj, by="intervention.id", all.x=T)
}
## DM
if(dm.type=="name"){
gta.jur=gta.jurisdictions[,c("name","jurisdiction.id")]
} else {
if(dm.type=="un"){
gta.jur=gta.jurisdictions[,c("un_code","jurisdiction.id")]
} else {
stop("DM type is neither 'un' nor 'name'.")
}
}
names(gta.jur)=c("dm.editor.choice","distorted.market.id")
master=merge(master, gta.jur, by="dm.editor.choice", all.x=T)
if(nrow(subset(master, is.na(distorted.market.id) & is.na(dm.editor.choice)==F))>0){
stop(paste("Unkown distorted market(s): ", paste(unique(subset(master, is.na(distorted.market.id) & is.na(dm.editor.choice)==F)$dm.editor.choice), collapse="; "), sep=""))
}
### CSVs
if(nrow(subset(master, is.na(distorted.market.id)==F))>0){
aj=unique(master[,c("intervention.id", "dm.editor.choice", "distorted.market.id")])
aj=merge(aggregate(dm.editor.choice ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
aggregate(distorted.market.id ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
by="intervention.id")
master$distorted.market.id=NULL
master$dm.editor.choice=NULL
master=merge(unique(master), aj, by="intervention.id", all.x=T)
}
# generate intervention frame
print("... pulling it all together.")
intervention=unique(data.frame(import_id=master$intervention.id,
import_measure_id=master$state.act.id,
measure_type_id=master$intervention.type.id,
description=master$intervention.description,
affected_flow_id=master$affected.flow.id,
evaluation_id=master$gta.evaluation.id,
research_evaluation_id=master$research.evaluation.id,
inception_date=master$date.implemented,
removal_date=master$date.removed,
is_duration_limited=master$duration.limited,
eligible_firms_id=master$eligible.firm.id,
implementation_level_id=master$implementation.level.id,
is_non_trade_related_rationale=master$non.commercial,
is_chain_measure=0,
is_horizontal_measure=master$horizontal,
is_jumbo=0,
is_fta_included=master$include.fta,
prior_level=master$level.prior,
new_level=master$level.new,
unit=master$level.unit.id,
tariff_peak=0,
affected_sectors=master$affected.sectors.nonhs,
IJ=master$implementing.jurisdiction.id,
DM=master$distorted.market.id,
dm_freeze=master$freeze.dm,
AJ=master$affected.jurisdiction.id,
aj_freeze=master$freeze.aj))
print("Intervention data frame is ready.")
## affected.TL
## level formatting
print("Generating affected product data frame ...")
master$atl.unit=tolower(as.character(master$atl.unit))
names(unit.df)=c("atl.unit.id","atl.unit")
master=merge(master, unit.df, by="atl.unit", all.x=T)
if(nrow(subset(master, is.na(atl.unit.id) & is.na(atl.unit)==F))>0){
stop(paste("Unkown level unit(s) at the HS code level: ", paste(unique(subset(master, is.na(atl.unit.id) & is.na(atl.unit)==F)$atl.unit), collapse="; "), sep=""))
}
master$atl.peak=as.numeric(master$atl.unit==1 & master$atl.new>=15)
master$atl.prior=as.character(master$atl.prior)
master$atl.new=as.character(master$atl.new)
## dates
master$atl.implemented=as.factor(master$atl.implemented)
master$atl.removed=as.factor(master$atl.removed)
## adding sectors
hs.cpc=gtalibrary::cpc.to.hs
names(hs.cpc)=c("atl.sector","affected.product")
master=merge(master, hs.cpc, by="affected.product", all.x=T)
## adding the zeros
master$affected.product=as.character(master$affected.product)
master$affected.product[nchar(master$affected.product)==5 & is.na(master$affected.product)==F]=paste(0, master$affected.product[nchar(master$affected.product)==5 & is.na(master$affected.product)==F], sep="")
master$atl.sector=as.character(master$atl.sector)
master$atl.sector[nchar(master$atl.sector)==2 & is.na(master$atl.sector)==F]=paste(0, master$atl.sector[nchar(master$atl.sector)==2 & is.na(master$atl.sector)==F], sep="")
print("... pulling it all together.")
affected.tl=unique(data.frame(import_intervention_id=master$intervention.id,
tariff_line_code=master$affected.product,
prior_level=master$atl.prior,
new_level=master$atl.new,
unit=master$atl.unit.id,
tariff_peak=master$atl.peak,
is_tariff_line_official=master$atl.official,
sector_code=master$atl.sector,
inception_date=master$atl.implemented,
removal_date=master$atl.removed))
print("Affected product data frame is ready.")
## exporting CSVs
# test
state.act.test=subset(state.act, import_id %in% unique(state.act$import_id)[1:test.size])
intervention.test=subset(intervention, import_measure_id %in% state.act.test$import_id)
affected.tl.test=subset(affected.tl, import_intervention_id %in% intervention.test$import_id)
print("Writing test CSVs ...")
write.csv(state.act.test, file=paste(dump.name," - ",Sys.Date()," - test - measure.csv",sep=""), row.names=F, na="")
write.csv(intervention.test, file=paste(dump.name," - ",Sys.Date()," - test - intervention.csv",sep=""), row.names=F, na="")
write.csv(affected.tl.test, file=paste(dump.name," - ",Sys.Date()," - test - atl.csv",sep=""), row.names=F, na="")
# remaining full set
state.act.full=subset(state.act, ! import_id %in% state.act.test$import_id)
intervention.full=subset(intervention, ! import_measure_id %in% state.act.test$import_id)
affected.tl.full=subset(affected.tl, ! import_intervention_id %in% intervention.test$import_id)
print("Writing full CSVs ...")
write.csv(state.act.full, file=paste(dump.name," - ",Sys.Date()," - full - measure.csv",sep=""), row.names=F, na="")
write.csv(intervention.full, file=paste(dump.name," - ",Sys.Date()," - full - intervention.csv",sep=""), row.names=F, na="")
write.csv(affected.tl.full, file=paste(dump.name," - ",Sys.Date()," - full - atl.csv",sep=""), row.names=F, na="")
print("Voilà!")
# returning modified DF into global environment
master.bulk<<-master
state.act.csv<<- state.act
intervention.csv<<-intervention
affected.tl.csv<<-affected.tl
}
global-trade-alert/gtalibrary documentation built on Jan. 10, 2025, 7 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions gta_bulk_prep
Documented in gta_bulk_prep
# Roxygen documentation
#' Split a data frame containing GTA entries into the components we need for a bulk import. Returns 2x 3 CSVs i.e. a test and the full set.
#'
#' @param dump.name Specify a name for this data dump. It will be used in the files. Default is 'data dump'.
#' @param master.frame Specify data frame that contains all relevant information. Default is 'master'.
#' @param test.size Number of state acts to include in the test set. Default is 5. There is NO overlap between the test and the full set CSVs i.e. you have to import both eventually.
#' @param author State your name. Default is "Piotr".
#' @param sa.id Name of the column containing the state act ID. Default is 'state.act.id'.
#' @param sa.title Name of the column containing the title of the state act. Default is 'title'.
#' @param sa.description Name of the column containing the announcement description. Default is 'announcement.description'.
#' @param announcement.date Name of the column containing the announcement date. Default is 'date.announced'.
#' @param sa.source Name of the column containing the source. Default is 'source'.
#' @param source.off Name of the column containing whether source is official (T/F). Default is 'source.official'.
#' @param int.id Name of the column containing ther intervention ID. Default is 'intervention.id'.
#' @param implementer Name of the column containing the implementing jurisdiction. Default is 'implementing.jurisdiction'.
#' @param implementer.type State how the implementing countries are specified ('un','name'). Default is 'name'.
#' @param int.type Name of the column containing the intervention type. Default is 'intervention.type'.
#' @param int.description Name of the column containing the intervention description. Default is 'intervention.description'.
#' @param int.flow Name of the column containing the affected flow. Default is 'affected.flow'.
#' @param eval.gta Name of the column containing the GTA evaluation. Default is 'gta.evaluation'.
#' @param eval.research Name of the column containing the research evaluation. Default is 'research.evaluation'.
#' @param inception.date Name of the column containing the inception date. Default is 'date.implemented'.
#' @param removal.date Name of the column containing the removal date. Default is 'date.removed'.
#' @param temporary Name of the column containing whether duration is announced as temporary (T/F). Default is 'duration.limited'.
#' @param firms Name of the column containing the eligible firms. Default is 'eligible.firms'.
#' @param imp.level Name of the column containing the implementation level. Default is 'implementation.level'.
#' @param is.noncommercial Name of the column containing whether there is a higher motive (T/F). Default is 'non.commercial'.
#' @param is.horizontal Name of the column containing whether the intervention is horizontal. Default is 'horizontal'.
#' @param is.fta Name of the column containing whether the affected trading partners may include FTA partners (T/F). Default is 'include.fta'.
#' @param int.prior Name of the column containing the prior level values for the intervention description. Default is 'level.prior'.
#' @param int.new Name of the column containing the new level values for the intervention description. Default is 'level.prior'.
#' @param int.unit Name of the column containing the unit for level values attached to the intervention description. Default is 'level.prior'.
#' @param sectors.nonhs Name of the column containing the affected sectors that are not generated by the affected tariff lines. Default is 'affected.sectors.nonhs'.
#' @param dm.nonsupport Name of the column containing the distorted markets that are not generated from a support table. Default is 'dm.editor.choice'.
#' @param dm.type State how the distorted markets are specified ('un','name'). Default is 'name'.
#' @param dm.freeze Name of the column containing whether the distorted market choice should be frozen as specified (T/F). Default is 'freeze.dm'.
#' @param aj.nonsupport Name of the column containing the affected jurisdictions that are not generated from a support table. Default is 'aj.editor.choice'.
#' @param aj.type State how the affected jurisdictions are specified ('un','name'). Default is 'name'.
#' @param aj.freeze Name of the column containing whether the affected jurisdictions choice should be frozen as specified (T/F). Default is 'freeze.aj'.
#' @param hs.code Name of the column containing the affected tariff line. Default is 'affected.product'.
#' @param hs.prior Name of the column containing the prior level for the affected tariff line. Default is 'atl.prior'.
#' @param hs.new Name of the column containing the new level for the affected tariff line. Default is 'atl.new'.
#' @param hs.unit Name of the column containing the unit for the affected tariff line values. Default is 'atl.unit'.
#' @param hs.official Name of the column containing the whether the affected tariff line is official (T/F). Default is 'atl.official'.
#' @param hs.inception Name of the column containing the inception date for the affected tariff line (if different from intervention date). Default is 'atl.implemented'.
#' @param hs.removal Name of the column containing the removal date for the affected tariff line (if different from intervention date). Default is 'atl.removed'.
#'
#'
#' @return Be up to date with our latest functions.
#' @references www.globaltradealert.org
#' @author Global Trade Alert
gta_bulk_prep = function(
dump.name="data dump",
master.frame="master",
test.size=5,
author="Piotr",
sa.id="state.act.id",
sa.title="title",
sa.description="announcement.description",
announcement.date="date.announced",
sa.source="source",
source.off="source.official",
int.id="intervention.id",
implementer="implementing.jurisdiction",
implementer.type="name",
int.type="intervention.type",
int.description="intervention.description",
int.flow="affected.flow",
eval.gta="gta.evaluation",
eval.research="research.evaluation",
inception.date="date.implemented",
removal.date="date.removed",
temporary="duration.limited",
firms="eligible.firms",
imp.level="implementation.level",
is.noncommercial="non.commercial",
is.horizontal="horizontal",
is.fta="include.fta",
int.prior="level.prior",
int.new="level.new",
int.unit="level.unit",
sectors.nonhs="affected.sectors.nonhs",
dm.nonsupport="dm.editor.choice",
dm.type="name",
dm.freeze="freeze.dm",
aj.nonsupport="aj.editor.choice",
aj.type="name",
aj.freeze="freeze.aj",
hs.code="affected.product",
hs.prior="atl.prior",
hs.new="atl.new",
hs.unit="atl.unit",
hs.official="atl.official",
hs.inception="atl.implemented",
hs.removed="atl.removed"
){
## loading and renaming the master data frame
print("Loading and formatting master data frame ...")
eval(parse(text=paste("master=", master.frame, sep="")))
given.names=c(sa.id, sa.title, sa.description, announcement.date, sa.source, source.off, int.id, implementer, int.type, int.description, int.flow, eval.gta, eval.research, inception.date, removal.date, temporary, firms, imp.level, is.noncommercial, is.horizontal, is.fta, int.prior, int.new, int.unit, sectors.nonhs, dm.nonsupport, dm.freeze, aj.nonsupport, aj.freeze, hs.code, hs.prior, hs.new, hs.unit, hs.official, hs.inception, hs.removed)
fct.names=c("state.act.id", "title", "announcement.description", "date.announced", "source", "source.official", "intervention.id", "implementing.jurisdiction", "intervention.type", "intervention.description", "affected.flow", "gta.evaluation", "research.evaluation", "date.implemented", "date.removed", "duration.limited", "eligible.firms", "implementation.level", "non.commercial", "horizontal", "include.fta", "level.prior", "level.new", "level.unit", "affected.sectors.nonhs", "dm.editor.choice", "freeze.dm", "aj.editor.choice", "freeze.aj", "affected.product", "atl.prior", "atl.new", "atl.unit", "atl.official", "atl.implemented","atl.removed")
master=unique(master[,given.names])
names(master)=fct.names
print("Loading and formatting master data frame ... complete.")
#### state act CSV
print("Generating state act data frame ...")
master$author.id=29
if(author=="Johannes"){
master$author.id=1
}
if(author=="Josse"){
master$author.id=30
}
master$title=as.character(master$title)
master$announcement.description=as.character(master$announcement.description)
master$date.announced=as.factor(master$date.announced)
master$source=as.character(master$source)
state.act=unique(data.frame(import_id=master$state.act.id,
author_id=master$author.id,
title=master$title,
description=master$announcement.description,
announcement_date=master$date.announced,
source=master$source,
is_source_official=master$source.official,
status_id=4))
print("State act data frame is ready.")
#### intervention CSV
## intervention type conversion
print("Generating intervention data frame ...")
int.type=gtalibrary::int.mast.types
print("... adding intervention types.")
master=merge(master, int.type[,c("intervention.type", "intervention.type.id")], by="intervention.type", all.x=T)
if(nrow(subset(master, is.na(intervention.type.id)))>0){
stop(paste("Unkown intervention type(s): ", paste(unique(subset(master, is.na(intervention.type.id))$intervention.type), collapse="; "), sep=""))
}
## affected flow id
print("... adding affected flow IDs.")
master$affected.flow.id=NA
master$affected.flow.id[tolower(master$affected.flow)=="inward"]=1
master$affected.flow.id[tolower(master$affected.flow)=="outward"]=2
master$affected.flow.id[tolower(master$affected.flow)=="outward subsidy"]=3
if(nrow(subset(master, is.na(affected.flow.id)))>0){
stop(paste("Unkown affected flow(s): ", paste(unique(subset(master, is.na(affected.flow.id))$affected.flow), collapse="; "), sep=""))
}
## eligible firms id
print("... adding eligible firm IDs.")
master$eligible.firm.id=NA
master$eligible.firm.id[tolower(master$eligible.firms)=="all"]=1
master$eligible.firm.id[tolower(master$eligible.firms)=="smes"]=2
master$eligible.firm.id[tolower(master$eligible.firms)=="firm-specific"]=3
master$eligible.firm.id[tolower(master$eligible.firms)=="state-controlled"]=4
master$eligible.firm.id[tolower(master$eligible.firms)=="state trading enterprise"]=5
master$eligible.firm.id[tolower(master$eligible.firms)=="sector-specific"]=6
if(nrow(subset(master, is.na(eligible.firm.id)))>0){
stop(paste("Unkown eligible firm(s): ", paste(unique(subset(master, is.na(eligible.firm.id))$eligible.firms), collapse="; "), sep=""))
}
## implementation level id
print("... adding intervention level IDs.")
master$implementation.level.id=NA
master$implementation.level.id[tolower(master$implementation.level)=="supranational"]=1
master$implementation.level.id[tolower(master$implementation.level)=="national"]=2
master$implementation.level.id[tolower(master$implementation.level)=="subnational"]=3
master$implementation.level.id[tolower(master$implementation.level)=="sez"]=4
master$implementation.level.id[tolower(master$implementation.level)=="ifi"]=5
master$implementation.level.id[tolower(master$implementation.level)=="nfi"]=6
if(nrow(subset(master, is.na(implementation.level.id)))>0){
stop(paste("Unkown implementation level(s): ", paste(unique(subset(master, is.na(implementation.level.id))$implementation.level), collapse="; "), sep=""))
}
## evaluation ids
print("... adding evaluation IDs.")
master$gta.evaluation.id=NA
master$gta.evaluation.id[tolower(master$gta.evaluation)=="red"]=1
master$gta.evaluation.id[tolower(master$gta.evaluation)=="amber"]=2
master$gta.evaluation.id[tolower(master$gta.evaluation)=="green"]=3
if(nrow(subset(master, is.na(gta.evaluation.id)))>0){
stop(paste("Unkown GTA evaluation(s): ", paste(unique(subset(master, is.na(gta.evaluation.id))$gta.evaluation), collapse="; "), sep=""))
}
master$research.evaluation.id=NA
master$research.evaluation.id[tolower(master$research.evaluation)=="red"]=1
master$research.evaluation.id[tolower(master$research.evaluation)=="amber"]=2
master$research.evaluation.id[tolower(master$research.evaluation)=="green"]=3
master$research.evaluation.id[is.na(master$research.evaluation)] =""
if(nrow(subset(master, is.na(research.evaluation.id)))>0){
stop(paste("Unkown research evaluation(s): ", paste(unique(subset(master, is.na(research.evaluation.id))$research.evaluation), collapse="; "), sep=""))
}
## levels
unit.df=data.frame(level.unit.id=c(1,1,1:19),
level.unit=c("percent","%","per cent", "total budget (USD)", "firm-specific budget (USD)", "USD/MT", "USD/kg", "USD/tonne", "USD/unit", "count", "USD/litre", "USD/pc", "USD/squaremetre", "USD/pair", "USD/ldt", "USD/lb", "USD/gallon", "USD/hl", "USD/LAL", "USD/tyre", "USD/stick"))
unit.df$level.unit=tolower(as.character(unit.df$level.unit))
master$level.unit=tolower(as.character(master$level.unit))
print("... reformatting levels, dates etc.")
master$level.prior=as.character(master$level.prior)
master$level.new=as.character(master$level.new)
master=merge(master, unit.df, by="level.unit", all.x=T)
if(nrow(subset(master, is.na(level.unit.id) & is.na(level.unit)==F))>0){
stop(paste("Unkown level unit(s): ", paste(unique(subset(master, is.na(level.unit.id) & is.na(level.unit)==F)$level.unit), collapse="; "), sep=""))
}
## dates
master$date.implemented=as.factor(master$date.implemented)
master$date.removed=as.factor(master$date.removed)
## sectors
master$affected.sectors.nonhs=as.character(master$affected.sectors.nonhs)
master$affected.sectors.nonhs[nchar(master$affected.sectors.nonhs)==2 & is.na(master$affected.sectors.nonhs)==F]=paste(0,master$affected.sectors.nonhs[nchar(master$affected.sectors.nonhs)==2 & is.na(master$affected.sectors.nonhs)==F], sep="")
## country ids
gta.jurisdictions=gtalibrary::country.names
## IJ
print("... adding country IDs.")
if(implementer.type=="name"){
gta.jur=gta.jurisdictions[,c("name","jurisdiction.id")]
} else {
if(implementer.type=="un"){
gta.jur=gta.jurisdictions[,c("un_code","jurisdiction.id")]
} else {
stop("Implementer type is neither 'un' nor 'name'.")
}
}
names(gta.jur)=c("implementing.jurisdiction","implementing.jurisdiction.id")
master=merge(master, gta.jur, by="implementing.jurisdiction", all.x=T)
if(nrow(subset(master, is.na(implementing.jurisdiction.id)))>0){
stop(paste("Unkown implementer(s): ", paste(unique(subset(master, is.na(implementing.jurisdiction.id))$implementing.jurisdiction), collapse="; "), sep=""))
}
### CSVs
if(nrow(subset(master, is.na(implementing.jurisdiction.id)==F))>0){
ij=unique(master[,c("intervention.id", "implementing.jurisdiction", "implementing.jurisdiction.id")])
ij=merge(aggregate(implementing.jurisdiction ~intervention.id, ij, function(x) paste(unique(x), collapse=",")),
aggregate(implementing.jurisdiction.id ~intervention.id, ij, function(x) paste(unique(x), collapse=",")),
by="intervention.id")
master$implementing.jurisdiction.id=NULL
master$implementing.jurisdiction=NULL
master=merge(unique(master), ij, by="intervention.id", all.x=T)
}
## AJ
if(aj.type=="name"){
gta.jur=gta.jurisdictions[,c("name","jurisdiction.id")]
} else {
if(aj.type=="un"){
gta.jur=gta.jurisdictions[,c("un_code","jurisdiction.id")]
} else {
stop("AJ type is neither 'un' nor 'name'.")
}
}
names(gta.jur)=c("aj.editor.choice","affected.jurisdiction.id")
master=merge(master, gta.jur, by="aj.editor.choice", all.x=T)
if(nrow(subset(master, is.na(affected.jurisdiction.id) & is.na(aj.editor.choice)==F))>0){
stop(paste("Unkown affected jurisdiction(s): ", paste(unique(subset(master, is.na(affected.jurisdiction.id) & is.na(aj.editor.choice)==F)$aj.editor.choice), collapse="; "), sep=""))
}
### CSVs
if(nrow(subset(master, is.na(affected.jurisdiction.id)==F))>0){
aj=unique(master[,c("intervention.id", "aj.editor.choice", "affected.jurisdiction.id")])
aj=merge(aggregate(aj.editor.choice ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
aggregate(affected.jurisdiction.id ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
by="intervention.id")
master$affected.jurisdiction.id=NULL
master$aj.editor.choice=NULL
master=merge(unique(master), aj, by="intervention.id", all.x=T)
}
## DM
if(dm.type=="name"){
gta.jur=gta.jurisdictions[,c("name","jurisdiction.id")]
} else {
if(dm.type=="un"){
gta.jur=gta.jurisdictions[,c("un_code","jurisdiction.id")]
} else {
stop("DM type is neither 'un' nor 'name'.")
}
}
names(gta.jur)=c("dm.editor.choice","distorted.market.id")
master=merge(master, gta.jur, by="dm.editor.choice", all.x=T)
if(nrow(subset(master, is.na(distorted.market.id) & is.na(dm.editor.choice)==F))>0){
stop(paste("Unkown distorted market(s): ", paste(unique(subset(master, is.na(distorted.market.id) & is.na(dm.editor.choice)==F)$dm.editor.choice), collapse="; "), sep=""))
}
### CSVs
if(nrow(subset(master, is.na(distorted.market.id)==F))>0){
aj=unique(master[,c("intervention.id", "dm.editor.choice", "distorted.market.id")])
aj=merge(aggregate(dm.editor.choice ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
aggregate(distorted.market.id ~intervention.id, aj, function(x) paste(unique(x), collapse=",")),
by="intervention.id")
master$distorted.market.id=NULL
master$dm.editor.choice=NULL
master=merge(unique(master), aj, by="intervention.id", all.x=T)
}
# generate intervention frame
print("... pulling it all together.")
intervention=unique(data.frame(import_id=master$intervention.id,
import_measure_id=master$state.act.id,
measure_type_id=master$intervention.type.id,
description=master$intervention.description,
affected_flow_id=master$affected.flow.id,
evaluation_id=master$gta.evaluation.id,
research_evaluation_id=master$research.evaluation.id,
inception_date=master$date.implemented,
removal_date=master$date.removed,
is_duration_limited=master$duration.limited,
eligible_firms_id=master$eligible.firm.id,
implementation_level_id=master$implementation.level.id,
is_non_trade_related_rationale=master$non.commercial,
is_chain_measure=0,
is_horizontal_measure=master$horizontal,
is_jumbo=0,
is_fta_included=master$include.fta,
prior_level=master$level.prior,
new_level=master$level.new,
unit=master$level.unit.id,
tariff_peak=0,
affected_sectors=master$affected.sectors.nonhs,
IJ=master$implementing.jurisdiction.id,
DM=master$distorted.market.id,
dm_freeze=master$freeze.dm,
AJ=master$affected.jurisdiction.id,
aj_freeze=master$freeze.aj))
print("Intervention data frame is ready.")
## affected.TL
## level formatting
print("Generating affected product data frame ...")
master$atl.unit=tolower(as.character(master$atl.unit))
names(unit.df)=c("atl.unit.id","atl.unit")
master=merge(master, unit.df, by="atl.unit", all.x=T)
if(nrow(subset(master, is.na(atl.unit.id) & is.na(atl.unit)==F))>0){
stop(paste("Unkown level unit(s) at the HS code level: ", paste(unique(subset(master, is.na(atl.unit.id) & is.na(atl.unit)==F)$atl.unit), collapse="; "), sep=""))
}
master$atl.peak=as.numeric(master$atl.unit==1 & master$atl.new>=15)
master$atl.prior=as.character(master$atl.prior)
master$atl.new=as.character(master$atl.new)
## dates
master$atl.implemented=as.factor(master$atl.implemented)
master$atl.removed=as.factor(master$atl.removed)
## adding sectors
hs.cpc=gtalibrary::cpc.to.hs
names(hs.cpc)=c("atl.sector","affected.product")
master=merge(master, hs.cpc, by="affected.product", all.x=T)
## adding the zeros
master$affected.product=as.character(master$affected.product)
master$affected.product[nchar(master$affected.product)==5 & is.na(master$affected.product)==F]=paste(0, master$affected.product[nchar(master$affected.product)==5 & is.na(master$affected.product)==F], sep="")
master$atl.sector=as.character(master$atl.sector)
master$atl.sector[nchar(master$atl.sector)==2 & is.na(master$atl.sector)==F]=paste(0, master$atl.sector[nchar(master$atl.sector)==2 & is.na(master$atl.sector)==F], sep="")
print("... pulling it all together.")
affected.tl=unique(data.frame(import_intervention_id=master$intervention.id,
tariff_line_code=master$affected.product,
prior_level=master$atl.prior,
new_level=master$atl.new,
unit=master$atl.unit.id,
tariff_peak=master$atl.peak,
is_tariff_line_official=master$atl.official,
sector_code=master$atl.sector,
inception_date=master$atl.implemented,
removal_date=master$atl.removed))
print("Affected product data frame is ready.")
## exporting CSVs
# test
state.act.test=subset(state.act, import_id %in% unique(state.act$import_id)[1:test.size])
intervention.test=subset(intervention, import_measure_id %in% state.act.test$import_id)
affected.tl.test=subset(affected.tl, import_intervention_id %in% intervention.test$import_id)
print("Writing test CSVs ...")
write.csv(state.act.test, file=paste(dump.name," - ",Sys.Date()," - test - measure.csv",sep=""), row.names=F, na="")
write.csv(intervention.test, file=paste(dump.name," - ",Sys.Date()," - test - intervention.csv",sep=""), row.names=F, na="")
write.csv(affected.tl.test, file=paste(dump.name," - ",Sys.Date()," - test - atl.csv",sep=""), row.names=F, na="")
# remaining full set
state.act.full=subset(state.act, ! import_id %in% state.act.test$import_id)
intervention.full=subset(intervention, ! import_measure_id %in% state.act.test$import_id)
affected.tl.full=subset(affected.tl, ! import_intervention_id %in% intervention.test$import_id)
print("Writing full CSVs ...")
write.csv(state.act.full, file=paste(dump.name," - ",Sys.Date()," - full - measure.csv",sep=""), row.names=F, na="")
write.csv(intervention.full, file=paste(dump.name," - ",Sys.Date()," - full - intervention.csv",sep=""), row.names=F, na="")
write.csv(affected.tl.full, file=paste(dump.name," - ",Sys.Date()," - full - atl.csv",sep=""), row.names=F, na="")
print("Voilà!")
# returning modified DF into global environment
master.bulk<<-master
state.act.csv<<- state.act
intervention.csv<<-intervention
affected.tl.csv<<-affected.tl
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.