Extracting text between quotes

Question

My data:

Topic   Content
Sunny   "Today is a sunny day." 
John     He listened and walked away."Should I visit Dr.Mary today?"
May      May is playing alone.

I want to extract everything that is inside "everything in this quotation".

Also, I would like to create another column to give the sentence the name, such as if the content has the keyword of "Sunny", the new column will input as "Sunny", if the content has the keyword of "visit", "Hospital" will be inputted in the new column for the row.

I would like to get the output below:

Topic  Content                           Tag
Sunny  "Today is a sunny day!"           Sunny 
John   "Should I visit Dr.Mary today?"   Hospital

dput:

structure(list(Topic = structure(c(3L, 1L, 2L), .Label = c("John", 
"May", "Sunny"), class = "factor"), Content = structure(c(3L, 
1L, 2L), .Label = c("He listened and walked away.\"Should I visit Dr.Mary   today?\"", 
"May is playing alone.", "Today is a sunny day. "), class = "factor")),   .Names =   c("Topic", 
"Content"), class = "data.frame", row.names = c(NA, -3L))

[How to make a great R reproducible example?](http://stackoverflow.com/questions/5963269) — zx8754, Nov 02 '15 at 09:09
@RonakShah I decide it myself. Like if the sentence contains "Sunny", the tag will be "Sunny" etc. — , Nov 02 '15 at 09:16
@Goofy what's wrong with providing the output of dput? WE should provide the answer within 1 min after you have posted the dput output. — Avinash Raj, Nov 02 '15 at 09:17
What if the content contain double quotes but no sunny or visit string? — Avinash Raj, Nov 02 '15 at 09:20
@AvinashRaj If it doesn't contain, the cell will be empty. Thanks — , Nov 02 '15 at 09:24

score 1 · Accepted Answer · answered Nov 02 '15 at 09:29

You may try this,

df <-structure(list(Topic = structure(c(3L, 1L, 2L), .Label = c("John", 
                                                           "May", "Sunny"), class = "factor"), Content = structure(c(3L, 
                                                                                                                     1L, 2L), .Label = c("He listened and walked away.\"Should I visit Dr.Mary   today?\"", 
                                                                                                                                         "May is playing alone.", "Today is a sunny day. "), class = "factor")),   .Names =   c("Topic", 
                                                                                                                                                                                                                                "Content"), class = "data.frame", row.names = c(NA, -3L))
x <- df[grepl('"', df$Content),]
x$Content <- sub('.*"(.*)".*', "\\1", x$Content)
x$Tag <- ifelse(grepl("visit",x$Content), "Hospital", ifelse(grepl("sunny",x$Content), "Sunny", ""))
x
#   Topic                         Content      Tag
#   2  John Should I visit Dr.Mary   today? Hospital

If the sentence contain both "visit" and "sunny", it will only choose the first one to tag. Can make it both tag? Like Hospital & Sunny in x$Tag @AvinashRaj — poppp, Nov 03 '15 at 03:03
try `x$Tag <- ifelse(grepl("visit.*sunny|sunny.*visit",x$Content), "Hospital & Sunny", ifelse(grepl("sunny",x$Content), "Sunny", ifelse(grepl("visit",x$Content), "Hospital", "")))` — Avinash Raj, Nov 03 '15 at 04:20

Extracting text between quotes

1 Answers1

Linked