Basic File Handling

The next few questions check your understanding of the sort command. Sort can read from a pipe or from a specified filename. The basic way sort works is to sort on a column where the columns are separated by spaces, so a file names.txt like:

gordon  russell
jim     jackson
bob     ludwiniak

sort names.txt

sort -k 2,2 names.txt

To practice, press the button below to create a file "stuff.txt" in /home/caine.

Sort stuff.txt into alphanumeric order based on column 1 (the md5 value). Save this into /home/caine/sort1.txt.

Now sort stuff.txt by column 2, and save the output as sort2.txt.

Introduce the "-u" flag to sort, which removes any duplicate entries (such as /bin/ls in out example file). The 'u' stands for 'unique'. So now sort stuff.txt by column 2, but make sure all rows are unique. Save the output to sort3.txt.

In these examples the character which split column 1 from column 2 was the space character. This is called the column delimiter. Sometimes columns will be separated with other characters, such as "," for comma separated files. So for instance if names.txt was:

gordon,russell
jim,jackson
bob,ludwiniak

sort -t"," names.txt

sort -t"," -k 2,2 names.txt

Sort /etc/passwd (delimited by the ":" character) by column 5 and save that output into sort4.txt.

If you are sorting a column which is composed of numbers, you need to decide if you want alphanumeric or numeric ordering. In alphanumeric ordering, "10" comes before "9" as the first character of "10" has a lower ascii value than "9". If you want numeric ordering where "10" is higher than "9", you must specify this in the "-k" command after each column number using the trailing character "n". So to sort column 10 numerically you need to do: "-k 10n,10n"

Firstly sort /etc/passwd on column 3 alphanumerically, saving the output to sort5.txt. Confirm that the last line of sort5.txt has a 9 in column 3.

Repeat the sort, but this time sort numerically and save the output into sort6.txt. Confirm that the last line of sort6.txt has the number 65534 in column 3.

Use grep on /usr/share/dict/words to find the first word that contains the three letter sequence wta
wta word:

Use grep piped through wc on file /usr/share/dict/words to find the number of words that contain the letter x.
ber words:

Use grep to find all lines in /etc/passwd that do not have nologin one the line. Make grep include the line numbers of the matching lines and send the output to file s5

Use ls -l and grep to find all the files in the directory /etc that were last modified in August (hint: try looking for the case sensitive string "Aug"). Send this list to s6.

Use ls -l and grep and sort to find all the files in /etc that were last modified in Jan. Sort this list in descending order of size and then alphabetically by name (so 2 files with the same size will appear in alphabetic order). Send the output to s7. Sorting using other techniques will probably not get the same answer...