Récupérer des chaines de caractères dans une chaine de caractère

Bonjour à tous,

J'ai un string très très long sans espace de ce type là:
[1] "PO644420PO656002294pour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

En gros ce que je voudrais faire c'est récupérer dans ce string tous les identifiants commençant par PA ou PM et finissant par un P mais sans le P.
Par exemple pour la chaine suivante : PA328PM645184MGPA381
je voudrais récupérer dans un vecteur
1) PA328
2) PM645184MG
3) PA381

Je vous remercie par avance.

Bien cordialement,

Raphaël

Bonjour,

Une proposition avec le package stringr, voir aussi cette page.

Code:

---

> V0 <- "PO644420PO656002294pour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

---

1) Découpage de la chaîne de caractères en utilisant le caractère "P" comme séparateur :

Code:

---

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

> library(stringr) > V1 <- unlist(str_split(V0,"P")) > V1 [1] "" "O644420" "O656002294pour2670024" "A328" "M645184MG" [6] "A381" "M645125" "SE" "A1268" "M645112MG" [11] "A2912" "M645085MG" "A1816" "M645051MG" "A1893" [16] "M645245MG" "A334160" "M645133MG" "A588886" "M645620MG" [21] "A590032" "M645180MG" "A2825" "M645568MG" "A333285" [26] "M645365MG" "A590034" "M645519MG" "A706" "M645268MG" [31] "A785" "M645345MG" "A2240" "M645077MG" "A267622" [36] "M645033MG" "A268048" "M645257MG" "A590031" "M645352MG" [41] "A333611" "M645389MG" "A332364" "M645525MG" "A335532" [46] "M645082MG" "A335543" "M645096MG" "A331924" "M645417MG" [51] "A335999" "M645219MG" "A356" "M645358" "A394" [56] "M645447" "A543" "M645064" "A719" "M645083" [61] "A765" "M645138" "A810" "M645557" "A840" [66] "M645617" "A606040" "M645144" "A1116" "M645530" [71] "A1155" "M6452"

---

2) Sélection des éléments commençant par "A" ou "M"

Code:

---

1 2 3 4 5 6 7 8

> V2 <- V1[which(str_sub(V1[],1,1) %in% c("A","M"))] > V2 [1] "A328" "M645184MG" "A381" "M645125" "A1268" "M645112MG" "A2912" "M645085MG" "A1816" "M645051MG" "A1893" "M645245MG" [13] "A334160" "M645133MG" "A588886" "M645620MG" "A590032" "M645180MG" "A2825" "M645568MG" "A333285" "M645365MG" "A590034" "M645519MG" [25] "A706" "M645268MG" "A785" "M645345MG" "A2240" "M645077MG" "A267622" "M645033MG" "A268048" "M645257MG" "A590031" "M645352MG" [37] "A333611" "M645389MG" "A332364" "M645525MG" "A335532" "M645082MG" "A335543" "M645096MG" "A331924" "M645417MG" "A335999" "M645219MG" [49] "A356" "M645358" "A394" "M645447" "A543" "M645064" "A719" "M645083" "A765" "M645138" "A810" "M645557" [61] "A840" "M645617" "A606040" "M645144" "A1116" "M645530" "A1155" "M6452"

---

3) Ajout du caractère "P" au début de chaque élément :

Code:

---

1 2 3 4 5 6 7 8 9

> V3 <- str_c("P", V2[]) > V3 [1] "PA328" "PM645184MG" "PA381" "PM645125" "PA1268" "PM645112MG" "PA2912" "PM645085MG" "PA1816" "PM645051MG" "PA1893" [12] "PM645245MG" "PA334160" "PM645133MG" "PA588886" "PM645620MG" "PA590032" "PM645180MG" "PA2825" "PM645568MG" "PA333285" "PM645365MG" [23] "PA590034" "PM645519MG" "PA706" "PM645268MG" "PA785" "PM645345MG" "PA2240" "PM645077MG" "PA267622" "PM645033MG" "PA268048" [34] "PM645257MG" "PA590031" "PM645352MG" "PA333611" "PM645389MG" "PA332364" "PM645525MG" "PA335532" "PM645082MG" "PA335543" "PM645096MG" [45] "PA331924" "PM645417MG" "PA335999" "PM645219MG" "PA356" "PM645358" "PA394" "PM645447" "PA543" "PM645064" "PA719" [56] "PM645083" "PA765" "PM645138" "PA810" "PM645557" "PA840" "PM645617" "PA606040" "PM645144" "PA1116" "PM645530" [67] "PA1155" "PM6452"

---

NB : Si vous avez beaucoup de variables à découper ainsi, il faudra prévoir le cas particulier où la chaîne initiale commence par le caractère "A" ou "M" et supprimer le premier élément.

Cordialement,