Baptiste Nusaibah validation #11
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,183 @@ | ||
| # function that compares the summrised tables in R versus those in SAS | ||
| # output: if difference, dataframe with each row details of difference | ||
| # if no difference, dataframe displaying table R and table SAS side by side. | ||
| # author: Nusaibah | ||
|
|
||
|
|
||
| compare_grade <- function(tabR,tabSAS){ | ||
|
|
||
| tab=list() # list containing all difference between the 2 tables | ||
| tabRraw=tabR | ||
| tabSASraw=tabSAS | ||
|
|
||
| #warning("Different number of grade levels")------------ | ||
| if (nrow(tabR)!=nrow(tabSAS)){ | ||
|
|
||
| if (nrow(tabR)<nrow(tabSAS)){ | ||
| tab$grade=c(tab$grade,tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull) | ||
| tab$level=c(tab$level,rep("Mineur",length(tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull))) | ||
| tab$table=c(tab$table,rep("R",length(tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull))) | ||
| tab$main=c(tab$main,rep("Missing grade level",length(tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull))) | ||
| tab$valueR=c(tab$valueR,rep(NA,length(tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull))) | ||
| tab$valueSAS=c(tab$valueSAS,rep("Filled",length(tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull))) | ||
| tab$arm=c(tab$arm,rep("All",length(tabSAS[which(!(tabSAS$grade%in% tabR$grade)),"grade"]%>%pull))) | ||
|
|
||
| }else if (nrow(tabR)>nrow(tabSAS)){ | ||
| tab$grade=c(tab$grade,tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull) | ||
| tab$level=c(tab$level,rep("Mineur",length(tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull))) | ||
| tab$table=c(tab$table,rep("SAS",length(tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull))) | ||
| tab$main=c(tab$main,rep("Missing grade level",length(tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull))) | ||
| tab$valueR=c(tab$valueR,rep("Filled",length(tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull))) | ||
| tab$valueSAS=c(tab$valueSAS,rep(NA,length(tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull))) | ||
| tab$arm=c(tab$arm,rep("All",length(tabR[which(!(tabR$grade%in% tabSAS$grade)),"grade"]%>%pull))) | ||
|
|
||
|
|
||
|
|
||
| } | ||
| } | ||
| # warning("Different number of arm")---------- | ||
| if (ncol(tabR)!=ncol(tabSAS)){ | ||
|
|
||
| if (ncol(tabR)<ncol(tabSAS)){ | ||
| tab$level=c(tab$level,"Mineur") | ||
| tab$arm=c(tab$arm,paste(unique(str_extract(colnames(tabSAS)[which(!(colnames(tabSAS)%in% colnames(tabR)))],"[:digit:]")),collapse = " & ")) | ||
| tab$table=c(tab$table,"R") | ||
| tab$main=c(tab$main,"Missing arm") | ||
| tab$valueR=c(tab$valueR,NA) | ||
| tab$valueSAS=c(tab$valueSAS,"Filled") | ||
| tab$grade=c(tab$grade,"All") | ||
|
|
||
|
|
||
| }else if (ncol(tabR)>ncol(tabSAS)){ | ||
| tab$level=c(tab$level,"Mineur") | ||
| tab$arm=c(tab$arm,paste(unique(str_extract(colnames(tabR)[which(!(colnames(tabR)%in% colnames(tabSAS)))],"[:digit:]")),collapse = " & ")) | ||
| tab$table=c(tab$table,"SAS") | ||
| tab$main=c(tab$main,"Missing arm") | ||
| tab$valueR=c(tab$valueR,"Filled") | ||
| tab$valueSAS=c(tab$valueSAS,NA) | ||
| tab$grade=c(tab$grade,"All") | ||
|
|
||
| }} | ||
|
|
||
| #compare the commun elements--------------- | ||
| tabR=tabR[,colnames(tabR) %in% colnames(tabSAS)] | ||
| tabSAS=tabSAS[,colnames(tabSAS) %in% colnames(tabR)] | ||
|
|
||
| df=tabR%>%arrange(grade)%>%full_join(tabSAS,by="grade",suffix = c(".r",".sas")) | ||
|
|
||
|
|
||
| indice=which(df[,paste0(tabR%>%select(-grade)%>%colnames(),paste=".r")]!=df[,paste0(tabSAS%>%select(-grade)%>%colnames(),paste=".sas")], | ||
| arr.ind=TRUE) # difference numérique number or percentage | ||
| indice[,"col"]=indice[,"col"]+1 # parce qu'on avait retiré le grade | ||
|
|
||
| indice=rbind2(indice, | ||
| which(is.na(df),arr.ind = T)%>%as.data.frame()) | ||
| # difference NA quand 2 grades ne se correspondent pas dans les 2 tables. je garde NA | ||
|
|
||
| if (nrow(indice)!=0){ | ||
| for (i in 1: nrow(indice)){ | ||
|
|
||
| if (is.na(df[indice[i,"row"],indice[i,"col"]]) & grepl(".r",colnames(df)[indice[i,"col"]])){ | ||
|
|
||
| tab$level=c(tab$level,"Mineur") | ||
| tab$grade=c(tab$grade,df[indice[i,"row"],"grade"]%>%pull) | ||
| tab$table=c(tab$table,"R") | ||
| tab$main=c(tab$main,"Missing grade level") | ||
| tab$valueR=c(tab$valueR,NA) | ||
| if (indice[i,"col"]%%2==0){ | ||
| tab$valueSAS=c(tab$valueSAS, | ||
| df[indice[i,"row"],indice[i,"col"]+2] %>%pull ) | ||
| }else{ | ||
| tab$valueSAS=c(tab$valueSAS, | ||
| paste0(df[indice[i,"row"],indice[i,"col"]+2] %>%pull ,"%")) | ||
| } | ||
| tab$arm = c(tab$arm,str_extract(string=colnames(df)[indice[i,"col"]],pattern = "[:digit:]")) | ||
|
|
||
|
|
||
| }else if (is.na(df[indice[i,"row"],indice[i,"col"]]) & grepl(".sas",colnames(df)[indice[i,"col"]])){ | ||
|
|
||
| tab$level=c(tab$level,"Mineur") | ||
| tab$grade=c(tab$grade,df[indice[i,"row"],"grade"]%>%pull) | ||
| tab$table=c(tab$table,"SAS") | ||
| tab$main=c(tab$main,"Missing grade level") | ||
| if (indice[i,"col"]%%2==0){ #means number | ||
| tab$valueR=c(tab$valueR, df[indice[i,"row"],indice[i,"col"]-2] %>% pull) | ||
| }else{ #mean percentage | ||
| tab$valueR=c(tab$valueR, paste0(df[indice[i,"row"],indice[i,"col"]-2] %>% pull,"%")) | ||
| } | ||
| tab$valueSAS=c(tab$valueSAS,NA) | ||
| tab$arm = c(tab$arm,str_extract(string=colnames(df)[indice[i,"col"]],pattern = "[:digit:]")) | ||
|
|
||
| }#else | ||
|
|
||
| if(indice[i,"col"]%%2==0){ | ||
|
|
||
| tab$level=c(tab$level,"MAJEUR") | ||
| tab$grade=c(tab$grade,df[indice[i,"row"],"grade"]%>%pull) | ||
| tab$table=c(tab$table,"Both") | ||
| tab$main=c(tab$main,"Different number") | ||
| tab$arm = c(tab$arm,str_extract(string=colnames(df)[indice[i,"col"]],pattern = "[:digit:]")) | ||
| if (grepl(".sas",colnames(df)[indice[i,"col"]])){ | ||
| tab$valueR=c(tab$valueR,df[indice[i,"row"],indice[i,"col"]-2]%>%pull) | ||
| tab$valueSAS=c(tab$valueSAS,df[indice[i,"row"],indice[i,"col"]]%>%pull) | ||
| }else{ | ||
| tab$valueR=c(tab$valueR,df[indice[i,"row"],indice[i,"col"]]%>%pull) | ||
| tab$valueSAS=c(tab$valueSAS,df[indice[i,"row"],indice[i,"col"]+2]%>%pull) | ||
| } | ||
| }else if (indice[i,"col"]%%2==1){ | ||
|
|
||
| tab$level=c(tab$level,"MAJEUR") | ||
| tab$grade=c(tab$grade,df[indice[i,"row"],"grade"]%>%pull) | ||
| tab$table=c(tab$table,"Both") | ||
| tab$main=c(tab$main,"Different percentage") | ||
| tab$arm = c(tab$arm,str_extract(string=colnames(df)[indice[i,"col"]],pattern = "[:digit:]")) | ||
| if (grepl(".sas",colnames(df)[indice[i,"col"]])){ | ||
| tab$valueR=c(tab$valueR,df[indice[i,"row"],indice[i,"col"]-2]%>%pull) | ||
| tab$valueSAS=c(tab$valueSAS,df[indice[i,"row"],indice[i,"col"]]%>%pull) | ||
| }else{ | ||
| tab$valueR=c(tab$valueR,df[indice[i,"row"],indice[i,"col"]]%>%pull) | ||
| tab$valueSAS=c(tab$valueSAS,df[indice[i,"row"],indice[i,"col"]+2]%>%pull) | ||
| } | ||
|
|
||
| } | ||
|
|
||
|
|
||
| }} | ||
|
|
||
|
|
||
|
|
||
| if (nrow(tabRraw)==nrow(tabSASraw) & ncol(tabRraw)==ncol(tabSASraw) & nrow(indice)==0 ){ | ||
| # warning("Comparison result: same outputs") | ||
|
|
||
| tablo = tabRraw%>% | ||
| full_join(tabSASraw, | ||
| by="grade", | ||
| suffix = c("_r","_sas"))%>% | ||
| flextable()%>% | ||
| add_footer_lines( "Comparison result: same outputs", top=TRUE) %>% | ||
| autofit() %>% | ||
| add_header_row( values = c("grade","R table", "SAS table"), colwidths = c(1,ncol(tabRraw)-1, ncol(tabSASraw)-1)) %>% | ||
| align(part="header", align="center",i=1)%>% | ||
| bold(part="header") %>% | ||
| # bold(i=~N1_r!=N1_sas,j=c(2,4))%>% | ||
| vline(j=ncol(tabRraw), part = "body") %>% | ||
|
|
||
| set_header_labels(values= c(colnames(tabRraw),colnames(tabSASraw)[-1])) %>% | ||
| merge_at(i=c(1,2),j=1, part = "header")%>% | ||
| align(part="body", align="center",j=1) | ||
| }else{ | ||
| tablo=as.data.frame(tab)%>% | ||
| #distinct(level,grade,table,main,.keep_all = TRUE) %>% | ||
| flextable()%>% | ||
| bold(part="header") %>% | ||
| align(part="body", align="right",i=~str_detect(valueR,pattern="[:digit:]"),j="valueR")%>% | ||
| align(part="body", align="right",i=~str_detect(valueSAS,pattern="[:digit:]"),j="valueSAS")%>% | ||
| autofit() %>% | ||
| bg(i=~level=="MAJEUR",bg="#f06d4d") | ||
| } | ||
|
|
||
|
|
||
| return(tablo) | ||
| } | ||
|
|
||
|
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,194 @@ | ||
| # function that compares the summurised tables in R versus those in SAS | ||
| # output: if difference, dataframe with each row details of difference | ||
| # if no difference, dataframe displaying table R and table SAS side by side. | ||
| # author: Nusaibah | ||
|
|
||
| compare_soc <- function(tabR,tabSAS){ | ||
|
|
||
| # ae_table_soc() affiche toutes les colonnes grade 1à 5 NA et total(option) (pour chaque bras si le bras est indiqué) | ||
| # supprimée les colonnes et lignes vides pour optimiser l'algorithme | ||
| tabR = tabR[,!apply(is.na(tabR), 2, all)] | ||
| tabR = tabR[!apply(is.na(tabR%>%select(-soc)), 1, all),] | ||
|
|
||
| tabRraw=tabR | ||
| tabSASraw=tabSAS | ||
| tab=list() | ||
|
|
||
| if ("term_"%in% c(colnames(tabR),colnames(tabSAS))){ | ||
| tabR = tabR %>% arrange(soc,term_) | ||
| }else{ | ||
| tabR = tabR %>% arrange(soc) | ||
| } | ||
|
|
||
| ## warning("Different number of SOC")---------------- | ||
| if (nrow(tabR)!=nrow(tabSAS)){ | ||
|
|
||
| socdiff=table(tabR$soc) %>% | ||
| as.data.frame() %>% | ||
| full_join(table(tabSAS$soc) %>% | ||
| as.data.frame(), | ||
| by="Var1",suffix = c(".r",".sas")) %>% | ||
| filter(Freq.r !=Freq.sas | (is.na(Freq.r) |is.na(Freq.sas))) | ||
|
|
||
| if (nrow(tabR)<nrow(tabSAS)){ | ||
|
|
||
| tab$level=c(tab$level,"Mineur") | ||
| tab$soc=c(tab$soc,paste(socdiff%>%filter(Freq.r<Freq.sas| is.na(Freq.r))%>%select(Var1)%>%pull,collapse = " & ")) | ||
| tab$table=c(tab$table,"R") | ||
| if(any(grepl("level",colnames(tabR)))){ | ||
| tab$arm=c(tab$arm,NA)} | ||
| if ("term_"%in% c(colnames(tabR),colnames(tabSAS))){ | ||
| tab$term_=c(tab$term_,paste(tabSAS[which(!(tabSAS$term_%in% tabR$term_)),"term_"]%>%pull,collapse = " & ")) | ||
| tab$main=c(tab$main,"Missing PT item") | ||
| }else{ | ||
| tab$term_=c(tab$term_,NA) | ||
| tab$main=c(tab$main,"Missing SOC item") | ||
| } | ||
| tab$valueR=c(tab$valueR,NA) | ||
| tab$valueSAS=c(tab$valueSAS,"Filled") | ||
|
|
||
| }else if (nrow(tabR)>nrow(tabSAS)){ | ||
|
|
||
|
|
||
| tab$level=c(tab$level,"Mineur") | ||
| tab$soc=c(tab$soc,paste(socdiff%>%filter(Freq.r>Freq.sas | is.na(Freq.sas))%>%select(Var1)%>%pull,collapse = " & ")) | ||
| tab$table=c(tab$table,"SAS") | ||
| tab$grade=c(tab$grade,NA) | ||
| if(any(grepl("bras",colnames(tabR)))){ | ||
| tab$arm=c(tab$arm,NA)} | ||
| if ("term_"%in% c(colnames(tabR),colnames(tabSAS))){ | ||
| tab$term_=c(tab$term_,paste(tabR[which(!(tabR$term_%in% tabSAS$term_)),"term_"]%>%pull,collapse = " & ")) | ||
| tab$main=c(tab$main,"Missing PT item") | ||
| }else{ | ||
| tab$term_=c(tab$term_,NA) | ||
| tab$main=c(tab$main,"Missing SOC item") | ||
| } | ||
| tab$valueR=c(tab$valueR,"Filled") | ||
| tab$valueSAS=c(tab$valueSAS,NA) | ||
| } | ||
| } | ||
| ## #warning("Different number of column (arm or grade)")-------------- | ||
| if (ncol(tabR)!=ncol(tabSAS)){ | ||
| if (ncol(tabR)<ncol(tabSAS)){ | ||
|
|
||
| tab$level=c(tab$level,"Mineur") | ||
| tab$grade=c(tab$grade,paste(colnames(tabSAS)[-(which(colnames(tabSAS) %in% colnames(tabR)))],collapse=" & ")) | ||
| tab$table=c(tab$table,"R") | ||
| tab$soc=c(tab$soc,NA) | ||
| tab$main=c(tab$main,"Missing grade") | ||
| tab$valueR=c(tab$valueR,NA) | ||
| tab$valueSAS=c(tab$valueSAS,"Filled") | ||
| tab$term_=c(tab$term_,NA) | ||
|
|
||
| if(any(grepl("bras",colnames(tabR)))){ | ||
| tab$arm=c(tab$arm,paste(unique(str_extract(colnames(tabSAS)[which(!(colnames(tabSAS)%in% colnames(tabR)))],"bras[:alnum:]+")),collapse = " & ")) | ||
| } | ||
| }else if (ncol(tabR)>ncol(tabSAS)){ | ||
|
|
||
| tab$level=c(tab$level,"Mineur") | ||
| tab$soc=c(tab$soc,NA) | ||
| tab$grade=c(tab$grade,paste(colnames(tabR)[-(which(colnames(tabR) %in% colnames(tabSAS)))],collapse = " & ")) | ||
| if(any(grepl("bras",colnames(tabR)))){ | ||
| tab$arm=c(tab$arm,paste(unique(str_extract(colnames(tabR)[which(!(colnames(tabR)%in% colnames(tabSAS)))],"bras[:alnum:]+")),collapse = " & ")) | ||
| } | ||
| tab$table=c(tab$table,"SAS") | ||
| tab$main=c(tab$main,"Missing grade") | ||
| tab$valueR=c(tab$valueR,"Filled") | ||
| tab$valueSAS=c(tab$valueSAS,NA) | ||
| tab$term_=c(tab$term_,NA) | ||
|
|
||
| } | ||
| } | ||
| ## commun element validation ------------ | ||
| if ("term_"%in% c(colnames(tabR),colnames(tabSAS))){ | ||
| dfsas=tabSAS%>% | ||
| pivot_longer(-c("soc","term_"),names_to = "grade",values_to = "count")%>% | ||
| mutate(table="SAS") | ||
|
|
||
|
|
||
| df=tabR%>%arrange(soc,term_)%>% | ||
| pivot_longer(-c("soc","term_"),names_to = "grade",values_to = "count")%>% | ||
| mutate(table="R")%>% | ||
| full_join(dfsas, #instead of full_join | ||
| by=c("soc","term_","grade"), | ||
| suffix = c(".r",".sas")) | ||
|
|
||
|
|
||
| if (any(grepl("bras",df$grade))){ | ||
|
|
||
| df=df%>%separate(col="grade",into=c("arm","grade"),sep="_")} #instead of _ | ||
|
|
||
| indice= df%>%filter(count.r!=count.sas | is.na(count.r)!=is.na(count.sas)) | ||
|
|
||
|
|
||
| }else{ | ||
| df=tabR%>%arrange(soc)%>% | ||
| pivot_longer(-c("soc"),names_to = "grade",values_to = "count")%>% | ||
| mutate(table="R")%>% | ||
| full_join(tabSAS%>% #instead of full | ||
| pivot_longer(-c("soc"),names_to = "grade",values_to = "count")%>% | ||
| mutate(table="SAS"), | ||
| by=c("soc","grade"), | ||
| suffix = c(".r",".sas")) | ||
| if (any(grepl("bras",df$grade))){ df=df%>%mutate("arm"=str_extract(grade,"bras[:alnum:]+"))} | ||
| indice=df[which(df$count.r!=df$count.sas | (is.na(df$count.r) & !is.na(df$count.sas)) | | ||
| (!is.na(df$count.r) & is.na(df$count.sas)) | ||
| ,arr.ind = T),] | ||
| } | ||
|
|
||
| if (nrow(indice)!=0){ | ||
|
|
||
| tab =rbind.fill(as.data.frame(tab),indice%>% | ||
|
There was a problem hiding this comment. La fonction |
||
| mutate(level="MAJEUR", | ||
| table="Both", | ||
| main="Different value")%>% | ||
| dplyr::rename(valueR="count.r", | ||
| valueSAS="count.sas")%>% | ||
| select(-c("table.r","table.sas"))) | ||
| } | ||
|
|
||
|
|
||
| if (nrow(tabRraw)==nrow(tabSASraw) & ncol(tabRraw)==ncol(tabSASraw) & nrow(indice)==0 ){ | ||
| # warning("Comparison result: same outputs") | ||
|
|
||
| if ("term_" %in% colnames(tabRraw)){ jointure=c("soc","term_") | ||
| colsize=c(1,1,ncol(tabRraw)-2,ncol(tabSASraw)-2)}else{jointure="soc" | ||
| colsize=c(1,ncol(tabRraw)-1,ncol(tabSASraw)-1) } | ||
| tab = tabRraw%>% | ||
| full_join(tabSASraw, | ||
| by=jointure, | ||
| suffix = c("_r","_sas"))%>% | ||
| flextable()%>% | ||
| add_footer_lines( "Comparison result: same outputs", top=TRUE) %>% | ||
| autofit() %>% | ||
| add_header_row( values = c(jointure,"R table", "SAS table"), colwidths = colsize) %>% | ||
| align(part="header", align="center",i=1)%>% | ||
| bold(part="header") %>% | ||
| # bold(i=~N1_r!=N1_sas,j=c(2,4))%>% | ||
| vline(j=ncol(tabRraw), part = "body") %>% | ||
|
|
||
| set_header_labels(values= c(colnames(tabRraw),colnames(tabSASraw)[-c(1:length(jointure))])) %>% | ||
| merge_at(i=c(1,2),j=1, part = "header")%>% | ||
| align(part="body", align="left",j=1) | ||
|
|
||
| if ("term_" %in% colnames(tabRraw)){ | ||
| tab = tab %>% | ||
| merge_at(i=c(1,2), j=2, part = "header")%>% | ||
| align(part="body", align="left",j=2)} | ||
| }else{ | ||
| tab=as.data.frame(tab)%>% | ||
| # distinct(level,grade,table,main,soc,.keep_all = TRUE)%>% | ||
| flextable()%>% | ||
| bold(part="header") %>% | ||
| align(part="body", align="right",i=~str_detect(valueR,pattern="[:digit:]"),j="valueR")%>% | ||
| align(part="body", align="right",i=~str_detect(valueSAS,pattern="[:digit:]"),j="valueSAS")%>% | ||
| autofit() %>% | ||
| bg(i=~level=="MAJEUR",bg="#f06d4d") | ||
| } | ||
|
|
||
|
|
||
|
|
||
| return(tab) | ||
|
|
||
|
|
||
| } | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ce full join émet un warning :
Il est émit quand
termest renseigné, et je crois qu'il fautdrait écrirepivot_longer(-any_of(c("soc", "term"), ...), ou quelque chose du genre (et dansbyaussi).J'ai l'impression que le problème d'output vient de là.