I chose to make my life much harder than it needed to be by trying to do the first few problems in R code instead of find/replace in Notepad++. As you’ll see, this got tiresome and I eventually abandoned the effort (for now!)
#put the data in a file in the current directory called input_data.r
FileInput = readLines("input_data.r")
print(FileInput)
test_expression <- " [ ]+"
print(test_expression)
replacement <- ","
print(replacement)
output <- gsub(test_expression, replacement, FileInput)
print(output)or!
#take the input as a string
Input = "First String Second 1.22 3.4
Second More Text 1.555555 2.2220
Third x 3 124"
print(Input)
test_expression <- " [ ]+"
print(test_expression)
replacement <- ","
print(replacement)
output <- gsub(test_expression, replacement, FileInput)
print(output)input_string <- "Ballif, Bryan, University of Vermont
Ellison, Aaron, Harvard Forest
Record, Sydne, Bryn Mawr"
print(input_string)
find_expression <- "(.*), (.*), (.*)"
writeLines(find_expression) #check how R will read it
output_expression <- "\\2 \\1 (\\3)"
writeLines(output_expression) #check how R will read it
output_string <- NULL #make sure we're starting clean
output_string <- str_replace_all(input_string,find_expression,output_expression)
print(output_string)input_string <- "0001 Georgia Horseshoe.mp3 0002 Billy In The Lowground.mp3 0003 Winder Slide.mp3 0004 Walking Cane.mp3"
find_expression <- ".mp3 *"
writeLines(find_expression)
replace_expression <- ".mp3\n"
writeLines(replace_expression)
output_string <- str_replace_all(input_string,find_expression,replace_expression)
print(output_string)#in notepad++: find (\d+) (.*).mp3$
#replace with \2_\1.mp3It works! Unlike this:
#the below (intended to effect the same as above) doesn't work
find_expression <- "(\\d+) (.*).mp3$" #this works perfectly in notepad++ but here it seems like maybe the .* is matching even new lines (against documentation?), so it's reading the whole thing as a single match?
writeLines(find_expression)
output_expression <- "\\2_\\1.mp3"
writeLines(output_expression)
output_string <- str_replace_all(input_string,find_expression,output_expression)
print(output_string)
#giving up on the above - too much time spent#Find a letter at the start of a line and grab it, ignore until you see a comma, grab everything til the next comma, ignore everything til the next comma, grab what's left to the end of the line.
#find: ^(\w).*,(.*),.*,(.*)$
#replace: \1_\2,\3# similar to above, but just grab the first 4 letters after the first comma in your second grab
#find: ^(\w).*,(.{4}).*,.*,(.*)$
#replace: \1_\2,\3# trivial after the last one
# find: ^(.{3}).*,(.{3}).*,(.*),(.*)$
# replace: \1\2, \4, \3# Q: Look at the pathogen_binary column-what input are you expecting to see in this column?
# A1: When the pathogen_load is zero, I would expect pathogen_binary always to be 0, not NA. So let's find places a row ends in zero and replace the NA with a 0
# find: (.*)(,NA,)(.*)0$
# replace: \1,0,\30
# A2: When the pathogen_load is non-zero, I would expect pathogen_binary always to be 1, not NA. So then let's find places a row ends in anything but 0 and replace NA with a 1
# find: (.*)(,NA,)(.*)[^0]$
# replace: \1,1,\30
# Q: What are the main errors in bombus_spp and host_plant?
# A: Stray characters like #, @, etc.
# it was almost easy, but the damn underscore is a \w character, so we'll do it the wordy way
# find: [^\[a-z\]\[A-Z\]\[0-9\],\./ \n\r] #Windows needs the \n\r not just \n
# replace: #nothing
# Q: Is there anything visibly wrong with bee_caste?
# A: Extra spaces at the end of some. this doesn't seem to appear anywhere else, so let's just kill all " ,"
# find: " ," #omit the parens - just need you to see there's a space
# replace: ,
#
# Looking good!