new line chars added to csv file after ftp.storbinary()

Question

I am attempting to store a csv file on an ftp server using python's ftplib module. Right now, I have about 30 lines of code which generates probabilities of weather values in a 2-d array. I then write this 2-d array to a csv file.

When I write the csv file onto my local drive, the file displays as expected within excel. However, when I view the file after I uploaded it to an ftp server, I see that a new line character has been added after every row.

I've done some minor testing to see what the problem may be, and I have been able to upload the csv file with coreftp. The csv file displays correctly after I do that. So I am pretty sure the file is fine, its something that is happening when python uploads it onto an ftp server.

I was originally creating a text file with a .csv extension file then reopening it as a binary file and uploading it. I thought that may be the issue so I tried using the csv module, but same issue.

Here is my code at the moment...

TEMPSHEADER = [i-50 for i in range(181)]#upper bounds exclusive
WINDSHEADER = [i for i in range(101)]#upper bounds exclusive
HEADER = TEMPSHEADER + WINDSHEADER
for site in ensmosdic:
    ensmos = ensmosdic.get(site)
    with open(utcnow.strftime("%Y-%m-%d") + "-" +site+"-prob.csv","w",newline='') as csvfile:
        writer = csv.writer(csvfile, delimiter=",")
        writer.writerow(["CODE        ","F","ForecastDate","HOUR"]+HEADER)
        siteTable =[[0 for x in range(286)] for y in range(24,169)]#upper bounds exclusive
        ###########
        #other code here, but not important with regards to post
        ###########
        for i in siteTable:
            writer.writerow(i)

        csvfile.close()#not sure if you have to close csv file, not in csv module docs
    f = open(utcnow.strftime("%Y-%m-%d") + "-" +site+"-prob.csv","rb")
    ftpInno.storbinary("STOR " + utcnow.strftime("%Y-%m-%d-") + site +"-prob.csv",f)
    f.close()
ftpInno.close()

Thanks in advance

Answer 1

After an hour or so of trouble shooting, the answer is fairly simple, although I am not entirely sure why it works.

what i did was create a text file instead of a csv file which I was doing in my original question

with open(FILELOCATION + utcnow.strftime("%Y-%m-%d") + "-" +site+"-prob.txt","w") as f:
    #write to the file below
    f.close()
#open file again as a txt file
f = open(FILELOCATION + utcnow.strftime("%Y-%m-%d") + "-" +site+"-prob.txt","rb")
ftp.storlines("STOR " + utcnow.strftime("%Y-%m-%d-") + site +"-prob.csv",f)
f.close()

reading the file as a binary file and then storing it with the storlines method removed the extra lines I was seeing within the file after I uploaded it to an ftp server.

Answer 2

This might shed some light on your issue. I had a project where I was using windows command line and also windows powershell to transfer .csv files with the ftp get and mget commands. And like you said I was getting a extra between each row. It seems like switching to binary transfer mode fixed my issue. For example once you are in the ftp dialog just type "binary" and hit enter and it switches the mode.