Fusionner plusieurs tables

Bonjour,
En fait j'ai dans mon répertoire un certain nombre de tables de tailles différentes. Par exemple avec 3 tables:

Code:

---

1 2 3 4 5 6 7 8

tab1 <- structure(list( time= structure(1:10, .Label = c("2017-04-27 19:39:32.036","2017-04-28 09:20:13.222","2017-04-28 09:20:54.613", "2017-04-28 11:30:58.015","2017-04-28 11:31:28.334","2017-04-28 11:32:25.411", "2017-04-28 11:33:50.444","2017-04-28 14:47:42.557","2017-04-28 14:49:37.136", "2017-04-28 14:49:46.070"), class = "factor"), value = structure(c(2L,1L,2L,1L,2L,1L,2L,1L,2L,1L), .Label = c("ON", "OFF"), class = "factor")), .Names = c("time","value"), class = "data.frame", row.names = c(NA,-10L)) tab1$time<- as.POSIXct(tab1$time, format = "%Y-%m-%d %H:%M:%S")

---

Code:

---

1 2 3 4 5 6 7 8

tab2 <- structure(list( time= structure(1:10, .Label = c("2017-05-02 15:06:38.546","2017-05-02 15:09:13.441","2017-05-02 15:11:12.512", "2017-05-02 15:11:45.824","2017-05-02 15:13:15.581","2017-05-02 15:13:18.476", "2017-05-02 15:15:13.689","2017-05-02 15:15:36.948","2017-05-02 15:16:21.762", "2017-05-02 15:21:11.759"), class = "factor"), value = structure(c(2L,1L,2L,1L,2L,1L,2L,1L,2L,1L), .Label = c("ON", "OFF"), class = "factor")), .Names = c("time","value"), class = "data.frame", row.names = c(NA,-10L)) tab2$time<- as.POSIXct(tab2$time, format = "%Y-%m-%d %H:%M:%S")

---

Code:

---

1 2 3 4 5 6 7 8

tab3 <- structure(list( time= structure(1:10, .Label = c("2017-04-27 15:39:35.349","2017-04-28 09:20:08.501","2017-04-28 09:20:25.020", "2017-04-28 11:30:52.417","2017-04-28 11:30:56.314","2017-04-28 11:33:26.238", "2017-04-28 11:33:29.545","2017-04-28 14:47:30.948","2017-04-28 14:47:35.278", "2017-04-28 14:55:54.692"), class = "factor"), value = structure(c(2L,1L,2L,1L,2L,1L,2L,1L,2L,1L), .Label = c("OPEN", "CLOSED"), class = "factor")), .Names = c("time","value"), class = "data.frame", row.names = c(NA,-10L)) tab3$time<- as.POSIXct(tab3$time, format = "%Y-%m-%d %H:%M:%S")

---

Au final je souhaiterai les fusionner pour obtenir une seule table tout sans perte d'aucune information sur la date, car à chaque timestamps doit correspondre un événement.
Exemple de la table finale que je souhaite obtenir:

Code:

---

1 2 3 4 5 6 7 8 9 10

tab=structure(list( time= structure(1:13, .Label = c("2017-04-27 15:39:35","2017-04-27 19:39:32","2017-04-28 09:20:08", "2017-04-28 09:20:13","2017-04-28 09:20:25","2017-04-28 09:20:54", "2017-04-28 11:30:52","2017-04-28 11:30:56","2017-04-28 11:30:58", "2017-04-28 11:31:28","2017-04-28 11:32:25","2017-04-28 11:33:26", "2017-04-28 11:33:29"), class = "factor"), capteur1 = structure(c(3L,2L,3L,1L,3L,2L,3L,3L,1L,2L,1L,3L,3L), .Label = c("ON", "OFF","NA"), class = "factor"), capteur2 = structure(c(3L,3L,3L,3L,3L,3L,3L,3L,3L,3L,3L,3L,3L), .Label = c("ON", "OFF","NA"), class = "factor"), capteur3 = structure(c(2L,3L,1L,3L,2L,3L,1L,2L,3L,3L,3L,1L,2L), .Label = c("OPEN", "CLOSED","NA"), class = "factor")), .Names = c("time","capteur1","capteur2","capteur3"), class = "data.frame", row.names = c(NA,-13L))

---

Merci,
Korera

Bonjour,

une possibilité :

Code:

---

1 2 3

for (i in 1:3) eval(parse(text = sprintf("colnames(tab%1$d)[2] <- 'capteur%1$d'", i))) Reduce(function(x, y) merge(x, y, all = T), list(tab1, tab2, tab3))

---

cdlt

Merci tototode !
J'essaie d'adapter ton code aux 36 tables que j'ai :)

Bonjour,

une possibilité, si tes noms de tables ont un pattern commun, tab1, tab2, tabn+1, etc.

Code:

---

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

# pour l'exemple tab1 <- structure(list( time= structure(1:10, .Label = c("2017-04-27 19:39:32.036","2017-04-28 09:20:13.222","2017-04-28 09:20:54.613", "2017-04-28 11:30:58.015","2017-04-28 11:31:28.334","2017-04-28 11:32:25.411", "2017-04-28 11:33:50.444","2017-04-28 14:47:42.557","2017-04-28 14:49:37.136", "2017-04-28 14:49:46.070"), class = "factor"), value = structure(c(2L,1L,2L,1L,2L,1L,2L,1L,2L,1L), .Label = c("ON", "OFF"), class = "factor")), .Names = c("time","value"), class = "data.frame", row.names = c(NA,-10L)) tab1$time<- as.POSIXct(tab1$time, format = "%Y-%m-%d %H:%M:%S")   for (i in 2:36) assign(sprintf("tab%d", i), tab1)   # debut du code nms <- ls(pattern = "tab\\d+") nms <- nms[order(sprintf("%2s", gsub("\\D+", "", nms)))]   for (i in nms) eval(parse(text = sprintf("colnames(%s)[2] <- 'capteur%s'", i, gsub("\\D+","",i))))   l1 <- lapply(nms, get) # ou : eval(parse(text = sprintf("list(%s)", paste(nms, collapse = ",")))) Reduce(function(x, y) merge(x, y, all = T), l1)

---

cdlt