I've been using asd=readcsv(filename) to read a csv file in Julia.
The first row of the csv file contains strings which describe the column contents; the rest of the data is a mix of integers and floats. readcsv reads the numbers just fine, but only reads the first 4+1/2 string entries.
After that, it renders "". If I ask the REPL to display asd[1,:], it tells me it is 1x65 Array{Any,2}.
The fifth column in the first row of the csv file (this seems to be the entry it chokes on) is APP #1 bias voltage [V]; but asd[1,5] is just APP . So it looks to me as though readcsv has choked on the "#" character.
I tried using "quotes=false" keyword in readcsv, but it didn't help.
I used to use xlsread in Matlab and it worked fine.
Has anybody out there seen this sort of thing before?
The comment character in Julia is #, and this applies when reading files from delimited text files.
But luckily, the readcsv() and readdlm() functions have an optional argument to help in these situations.
You should try readcsv(filename; comment_char = '/').
Of course, the example above assumes that you don't have any / characters in your first line. If you do, then you'll have to change that / above to something else.
Related
I am new to Julia so sorry if this question is obvious.
I am trying to use Julia to help me run a series of finite element models, which use a text input file to give instructions to the finite element solver. Basically, I would like to use Julia to read in the base input file, edit some parameters on some lines of the file and then write it as a new file. I am getting hung up on a couple things though.
Currently, I am reading in the file like this
mdl = "fullmodelSVTV"; #name of input file
A = readlines(mdl*".inp")
This read each line from the file in as a separate string in a vector which I like because it makes it easier to edit the sections I want but it also makes things more difficult when I try to write to a new file.
I am writing the file like this.
io = open("name.inp","w")
print(io,A)
close(io)
When I try to write to a new file the output ends up look like this
Output from code
which is ["string at index 1","string at index 2","string at index 3"...].
What I would like to do is output this the exact same way is it is read in with string at each index of the vector on its own line. I would also like to remove the brackets and quotation marks from the file, as they might interfere with the finite element solver.
I think I have found a way to concatenate all of the strings at each index and separated them with a new line like shown below.
for i in 1:length(A)
conc = conc*"\n"*lines[i]
end
The issue with this is that it takes a long time to do given the size of the input files I am working with and I feel like there has to achieve my goal.
I also cannot find a way to remove the brackets or quotation marks when writing the file.
So, I'm wondering if anyone has any advice for a better way to write these text files in terms of both concatenating all of the strings from the vector when outputting as well as outputting without the brackets and quotation marks.
Thanks, any advice is appreciated.
The issue with print(io,A) is that it is printing a representation of the vector, but in fact you want to print each element of the vector. To do so, you can simply print each line in a loop:
open("name.inp", "w") do io
for line in A
println(io, line)
end
end
This avoids the overhead of string concatenation.
Is there a way to read a file into R where I do not know the complete file name. Something like.
read.csv("abc_*")
In this case I do not know the complete file name after abc_
If you have exactly one file matching your criteria, you can do it like this:
read.csv(dir(pattern='^abc_')[1])
If there is more than one file, this approach would just use the first hit. In a more elaborated version you could loop over all matches and append them to one dataframe or something like that.
Note that the pattern uses regular expressions and thus is a bit different from what you did expect (and what I wrongly assumed at my first shot to answer the question). Details can be found using ?regex
If you have a directory you want to submit, you have do modify the dir command accordingly:
read.csv(dir('path/to/your/file', full.names=T, pattern="^abc"))
The submitted path in your case may be c:\\users\\user\\desktop, and then the pattern as above. full.names=T forces dir() to output a whole path and not only the file name. Try running dir(...) without the read.csv to understand what is happening there.
If you want to give your path as a complete string, it again gets a bit more complicated:
filepath <- 'path/to/your/file/abc_'
read.csv(dir(dirname(filepath), full.names=T, pattern=paste("^", basename(filepath), sep='')))
That process will fail if your filename contains any regular expression keywords. You would have to substitute then with their corresponding escape sequences upfront. But that again is another topic.
Setting:
I have (simple) .csv and .dat files created from laboratory devices and other programs storing information on measurements or calculations. I have found this for other languages but nor for R
Problem:
Using R, I am trying to extract values to quickly display results w/o opening the created files. Hereby I have two typical settings:
a) I need to read a priori unknown values after known key words
b) I need to read lines after known key words or lines
I can't make functions such as scan() and grep() work.
c) Finally I would like to loop over dozens of files in a folder and give me a summary (to make the picture complete: I will manage this part)
I woul appreciate any form of help.
ok, it works for the key value (although perhaps not very nice)
variable<-scan("file.csv", what=character(),sep="")
returns a charactor vector of everything
variable[grep("keyword", ks)+2] # + 2 as the actual value is stored two places ahead
returns characters of seaked values.
as.numeric(lapply(variable, gsub, patt=",", replace="."))
for completion: data had to be altered to number and "," and "." problem needed to be solved.
in a line:
data=as.numeric(lapply(ks[grep("Ks_Boden", ks)+2], gsub, patt=",", replace="."))
Perseverence is not to bad of an asset ;-)
The rest isn't finished, yet, I will post once finished.
I am trying to read this file (3.8mb) using its fixed-width structure as described in the following link.
This command:
a <- read.fwf('~/ccsl.txt',c(2,30,6,2,30,8,10,11,6,8))
Produces an error:
line 37 did not have 10 elements
After replicating the issue with different values of the skip option, I figured that the lines causing the problem all contain the "#" symbol.
Is there any way to get around it?
As #jverzani already commented, this problem is probably the fact that the # sign often used as a character to signal a comment. Setting the comment.char input argument of read.fwf to something other than # could fix the problem. I'll leave my answer below as a more general case that you can use on any character that causes problems (e.g. the 's in the Dutch city name 's Gravenhage).
I've had this problem occur with other symbols. The approach I took was to simply replace the # by either nothing, or by a character which does not generate the error. In my case it was no problem to simply replace the character, but this might not be possible in your case.
So my approach would be to delete the symbol that generates the error, or replace by another character. This can be done using a text editor (find and replace), in an R script, or using some linux tools called grep and sed. If you want to do this in an R script, use scan or readLines to read the lines. Once the text is in memory, you can use sub to replace the character.
If you cannot replace the character, I would try the following approach: replace the character by a character that does not generate an error, read it into R using read.fwf, and finally replace the character by the # character.
Following up on the answer above: to get all characters to be read as literals, use both comment.char="" and quote="" (the latter takes care of #PaulHiemstra's problem with single-quotes in Dutch proper nouns) in the call to read.fwf (this is documented in ?read.table).
I have a huge csv file, separated by comma's and I want to do a analysis with glm in R.
In one column there exists data with a comma implied, something like: bla,blabla
When reading the file in R with read.csv.sql there comes a error-message:
RS-DBI driver: (RS_sqlite_import: ./agp.csv line 47612 expected 37 columns of data but found 38)
This is due to the 'extra' comma in some of the data, not the whole column has an extra column.
How can I fix this? I want to remove this extra superfluous comma.
Thanks for the reaction,
André
The CSV format is very simple and can easily be hand edited. In order to include a comma in a value, you must surround the value with quotes quotes. Try this: "bla,blabla". If that data happens to contain any quotes, eg. blah,"thequotedblah",blah, those quotes need to be escaped with another quote, like this: "blah,""thequotedblah"",blah".
Although there is no official standard around it, there isn't much to the CSV format. Wikipedia has a great CSV reference that I have personally used to implement CSV support in applications. Spend 5-10 minutes reading it and you'll know everything you ever need to know to manually create/read/repair CSV data.
Is it just this one line that contains a non-quoted comma - or are there several such lines? Editing the .csv with an editor that can handle large files (e.g. Ultraedit) to sanitize that one record would certainly help. Asaph's suggestion of quoting is also a good 'un.