Change .xls to .csv without losing columns - csv

Question

I have a folder with around 400 .txt files that I need to convert to .csv. When I batch rename them to .csv, all the columns get smushed together into one. Same thing happens when I convert to .xls then .csv, even though the columns are fine in .xls. If I open the .xls file and save as to .csv, it's fine, but this would require opening all 400 files.
I am working with sed from the mac terminal. After navigating to the folder that contains the files within the terminal, here is some code that did not work:
for file in *.csv; do sed 's/[[:blank:]]+/,/g'
for file in *.csv; do sed -e "s/ /,/g"
for file in *.csv; do s/[[:space:]]/,/g
for file in *.csv; do sed 's/[[:space:]]{1,}/,/g'
Any advice on how to restore the column structure to the csv files would be much appreciated. And it's probably already apparent but I'm a coding newb so please go easy. Thanks!
Edit: here is an example of how the xls columns look, and how they should look in csv format:
Dotsc.exe 2/12/15 1:17 PM 0 Nothing 1 Practice
Everything that is separated by spaces here (except the space between 7 and PM) are separated by columns in the file. Here is what it looks like when I rename the batch rename the file to .csv:
Dotsc.exe 2/12/15 1:17 PM 0 Nothing 1 Practice
Columns have now turned into spaces, and all data is in the same column. Hope that clarifies things.

Answer 1

I think that what you are trying to do only in batch is not possible . I suggest you to use some library in Java.
Take a look here : http://poi.apache.org/spreadsheet/