创建一个 data.table 块,其中一列包含原始组,另一列包含分隔的 id
d <- list( c("SD1:LUSH", "SD44:CANCEL", "SD384:FR563", "SD32:TRUMPET"), c("SD23:SWITCH", "SD1:LUSH", "SD567:TREK"), c("SD42:CRAYON", "SD345:FOX", "SD183:WIRE"), c("SD345:HOLE", "SD340:DUST", "SD387:ROLL"), c("SD455:TOMATO", "SD39:MATURE"), c("SD12:PAINTING", "SD315:MONEY31", "SD387:SPRING"), c("SD32:TRUMPET", "SD1:FIELD"))
d2 <- lapply(d, function(x) sapply(strsplit(x, ":"), "[", 1))
d <- lapply(d, paste0, collapse=", ")
d2 <- lapply(d2, paste0, collapse=", ")
d <- as.data.frame(as.matrix(lapply(d, paste0, collapse=", ")))
d2 <- as.data.frame(as.matrix(lapply(d2, paste0, collapse=", ")))
d <- as.data.frame(cbind(d,d2))
colnames(d) <- c("sdw", "sd")
d$sd <- as.character(d$sd)
d$sdw <- as.character(d$sdw)
require(data.table)
Bloc <- data.table( d , key = "sd" )
获取 Bloc 中的所有 id 以及相应的数据
Bloc <- Bloc[ , list( ID = unlist( strsplit( sd , "," ) ) ) , by = list(sdw, sd) ]
Bloc$ID <- gsub("^\\s+|\\s+$", "", Bloc$ID)
Bloc <- data.table( Bloc , key = "ID" )
循环合并 id 相交的向量
Bloc <- as.data.frame(Bloc)
M <- nrow(Bloc)
#create blankd data.frame
G <- data.frame(matrix(ncol=3), stringsAsFactors=FALSE)
G[,1:3] <- as.character(G[,1:3])
#G <- data.frame(sdw=character(), sd=character(), ID= character())
colnames(G) <- c("sdw", "sd", "ID")
N <- M
mch <- as.data.frame(Bloc)
#Loop to sequentially fill data.frame
for (i in 1:M) {
# test if ID already in previous groups
if(Bloc[i,"ID"] %in% G$ID == FALSE) {
# convert element to vector to check for intersect
tm <- strsplit(x=Bloc[i, "sd"], split=", ")
mch$t <- numeric(length=M)
}
for (j in 1:N){
#if intersect exists apply code as 1 mch$t column
ff <- strsplit(x=mch[j, "sd"], split=", ")[[1]]
dd <- intersect (tm[[1]], ff)
if (identical(dd, character(0))== FALSE) mch[j,"t"] = 1
}
submch <- subset(mch, t == 1 )
ID <- submch$ID
Group1 <- sort((unlist(strsplit(paste0(submch$sdw, collapse=","), ","))))
Group1 <- unique(gsub(" ","", Group1))
sdw <- rep(paste0(Group1, collapse=", "), nrow(submch))
Group2 <- sort((unlist(strsplit(paste0(submch$sd, collapse=","), ","))))
Group2 <- unique(gsub(" ","", Group2))
sd <- rep(paste0(Group2, collapse=", "), nrow(submch))
G1 <- cbind(sdw, sd, ID)
G1 <- unique(G1)
G <- rbind(G, G1)
mch$t <- NULL
}
G <- unique(G)
G2 <- data.table(G, key="ID")
G2 <- G2[, list(sdw = paste0(sort(unique(unlist(strsplit(sdw, split=", ")))), collapse=", "),
sd = paste0(sort(unique(unlist(strsplit(sd, split=", ")))), collapse=", ")) , by = "ID"]
G2 <- data.table( G2, key=c("sd", "sdw"))
G2 <- unique(G2)
获取输出为 data.table
Bloc <- G2[-1,]
Bloc$ID <- NULL
重复上面的循环,直到不再有交集为止
repeat
{
N1 <- nrow(Bloc)
Bloc <- Bloc[ , list( ID = unlist( strsplit( sd , "," ) ) ) , by = list(sdw, sd) ]
Bloc$ID <- gsub("^\\s+|\\s+$", "", Bloc$ID)
Bloc <- data.table( Bloc , key = "ID" )
Bloc <- as.data.frame(Bloc)
M <- nrow(Bloc)
#create blankd data.frame
G <- data.frame(matrix(ncol=3), stringsAsFactors=FALSE)
G[,1:3] <- as.character(G[,1:3])
#G <- data.frame(sdw=character(), sd=character(), ID= character())
colnames(G) <- c("sdw", "sd", "ID")
N <- M
mch <- as.data.frame(Bloc)
#Loop to sequentially fill data.frame
for (i in 1:M) {
# test if ID already in previous groups
if(Bloc[i,"ID"] %in% G$ID == FALSE) {
# convert element to vector to check for intersect
tm <- strsplit(x=Bloc[i, "sd"], split=", ")
mch$t <- numeric(length=M)
}
for (j in 1:N){
#check if intersect exists and code accordingly
ff <- strsplit(x=mch[j, "sd"], split=", ")[[1]]
dd <- intersect (tm[[1]], ff)
if (identical(dd, character(0))== FALSE) mch[j,"t"] = 1
}
submch <- subset(mch, t == 1 )
ID <- submch$ID
Group1 <- sort((unlist(strsplit(paste0(submch$sdw, collapse=","), ","))))
Group1 <- unique(gsub(" ","", Group1))
sdw <- rep(paste0(Group1, collapse=", "), nrow(submch))
Group2 <- sort((unlist(strsplit(paste0(submch$sd, collapse=","), ","))))
Group2 <- unique(gsub(" ","", Group2))
sd <- rep(paste0(Group2, collapse=", "), nrow(submch))
G1 <- cbind(sdw, sd, ID)
G1 <- unique(G1)
G <- rbind(G, G1)
mch$t <- NULL
}
G <- unique(G)
G2 <- data.table(G, key="ID")
G2 <- G2[, list(sdw = paste0(sort(unique(unlist(strsplit(sdw, split=", ")))), collapse=", "),
sd = paste0(sort(unique(unlist(strsplit(sd, split=", ")))), collapse=", ")) , by = "ID"]
G2 <- data.table( G2, key=c("sd", "sdw"))
G2 <- unique(G2)
Bloc <- G2[-1,]
Bloc$ID <- NULL
N2 <- nrow(Bloc)
if (N1 == N2)
break
}
Output
Bloc$sdw
[1] "SD1:FIELD, SD1:LUSH, SD23:SWITCH, SD32:TRUMPET, SD384:FR563, SD44:CANCEL, SD567:TREK"
[2] "SD12:PAINTING, SD183:WIRE, SD315:MONEY31, SD340:DUST, SD345:FOX, SD345:HOLE, SD387:ROLL, SD387:SPRING, SD42:CRAYON"
[3] "SD39:MATURE, SD455:TOMATO"