Naming Conventions

• Avoid spaces; use _ or - instead.
• Do not start names with - to prevent confusion with options/flags.
• Stick to characters: 0-9, a-z, ., -, _. Special characters can have different meanings.
• Use single quotes ' ' for names with spaces or special characters.

---

File Movement and Copy Command

Moving / Renaming (mv), copying (cp), creating, deleting (rm) files and directories.
These work by modifying the directory entries of the files they access.

• cp adds an entry to the directory with the name of destination file and inode number that is allotted by the kernel.
• mv replaces the name of an existing directory entry without disturbing its inode number.
• rm removes an entry from the directory.

---

mv

mv is used to move or rename both files and directories.

mv [options] source destination

To rename a file, source is the old name and destination is the new name. Otherwise, both need to be in different directories. If destination is a directory without a file name, the file's name is not changed.

mv fo1.txt ~/temp
# moves to temp

mv fo1.txt ~/temp/fo2.txt
# moves and renames

mv *.txt /home/zapp

To move a file into the current directory, . can be used as the destination.

Renaming Files To rename a file or move it to a new location:

mv [old] [new]  # Moves or renames a file

mv trial/draft.txt trial/quotes.txt  
# Renames draft.txt to quotes.txt

mv draft.txt quotes.txt
# Renames within the same directory

Note: Moving a file to a directory with the same name will silently overwrite the original. Use mv -i to prompt for confirmation.

To move files to a different directory:

mv trial/quotes.txt .  
# Moves to the current directory

mv sucrose.dat maltase.dat ../raw  
# Moves files to the parent directory's raw folder

We are not allowed to rename multiple files using a single mv command as it has only a single destination parameter. mv *.txt *.dat will cause an error because *.dat means multiple destinations, but it needs to be a single directory or file name.

-i and -f are interactive and force modes. The difference occurs when a file being moved has the same name at the destination. If not interactive, the destination file is overwritten.

---

cp

The format is the same as mv:

cp [options] source destination

cp [old] [new]  # Copies a file

cp quotes.txt thesis/quotation.txt  
# Copies to a new location

There are three combinations that can be used:

• Destination is another directory, then the file name remains the same after copying.
• Destination is a directory and a file name, then the file is copied with a new name.
• Destination is a filename, then the file is copied into the current directory with a new name.

cp fo.txt ~zap/fo1.txt
cp *.txt ~
cp ~zapp/foo.txt .

-i and -f remain the same as mv.

• -b: Creates a backup of every destination file.
• -I/-s: Creates a hard or symbolic link rather than a physical copy.
• -u: Copies only if the source is newer than the destination or if the destination is missing.
• -L: If the source is a symbolic link, it follows the link to the actual file and copies that file, not the link.
• -v: Provides output for each file copied. (verbose)
• -p: Preserves the original ownership, permissions, and timestamps. (When cp creates a copy, it's ownership and permissions are updated to that pf the user who issued the command.)
• -r: Recursive copy for directories and their contents.

To copy a directory and all its contents, use -r:

cp -r thesis thesis_backup  
# Copies the entire directory

---

Operations with Multiple Files and Directories

To copy or move multiple files, list the files followed by the target directory. If given more than one file name followed by a directory name, the directory must be the last argument.

cp file1.txt file2.txt target_directory/  
# Copy multiple files

mv file1.txt file2.txt target_directory/   
# Move multiple files

Using wildcards simplifies this process:

cp *.txt backup/  # Copies all .txt files to backup/

$ mkdir backup

$ cp cretures/minotaur.dat creatures/unicorn.dat backup/

$ cp minotaur.dat unicorn.dat basilisk.dat
# Error occurs with multiple files; use wildcards instead

Wildcards

Wildcards represent unknown characters in commands.

Common wildcards include:

*: Represents zero or more characters.

• *.pdb matches all files ending with .pdb.
• *ethane.pdb matches ethane and methane.
• p*.pdb matches files starting with p and ending with .pdb.

?: Represents exactly one character.

• ?ethane.pdb matches methane.pdb.

Using Wildcards in Commands

ls *t*ane.pdb   
# Lists files with 't' and 'ane' in their names

cp *dataset* backup/datasets  
# Copies all files with 'dataset' in the name

ls *t?ne.*     # matches octane, pentane
ls *t??ne.pdb  # matches ethane, methane
ls ethane.*    # only ethane

Wildcards can be combined for more specific patterns:

ls ???ane.pdb  
# Matches any three characters followed by 'ane.pdb'

When a shell sees a wildcard, it expands the wildcard to create a list of matching filenames before running the preceding command.

Note: Be cautious when using wildcards, as errors can occur if not handled properly (e.g., *.pdf in a directory with .pdb files).

Using Wildcards for Copying

$ cp *dataset* backup/datasets
# Copy anything with 'dataset' in the name to the datasets directory inside backups

$ cp *calibration.txt backup/calibration
# Copy all calibration files

$ cp 2015-11-* send_to_bob/all_november_files/
# Copy just November files

$ cp *-23-dataset* send_to_bob/all_datasets_created_on_23rd/
# Copy only 23rd files

---

rm

rm is the delete command.

rm [options] file(s)

It can work on multiple files by listing, using wildcards, or both.

rm my_file.txt  
# Deletes the specified file

Use rm -i to prompt for confirmation before deletion. rm permanently deletes files, so it's advisable to use -i to ask for confirmation before deleting. The alias for rm can be set as rm -i

If too many files are being deleted and you want to override the prompts, use -f, but this can be dangerous: rm -f *.txt.

rm cannot remove directories directly. To remove a directory and its contents, use the recursive -r option:

rm -r directory_name  
# Deletes the directory and all its contents

To delete a complete sub-hierarchy and override the prompts, use rm -fr *.

• To remove files matching a pattern (e.g., all .txt files) with confirmation:

rm -i *.txt  # Prompts for each .txt file

Note: The shell does not have a trash bin, so deleted files are permanently removed.

rmdir is the only way to delete an empty directory. Use rmdir -p to recursively delete parent and current directories.

mkdir Creating Directories

Used to create a directory, in the current directory or a specified path.

• -m --mode: Specify the initial permissions of the directory.
• -z --context: Specify the SELinux context.

mkdir [name]  # Create a directory

To create multiple directories at once:

mkdir north south pacific  
# Creates three separate directories

Creating a Directory Tree, by creating the main directory first, then child directories inside.

mkdir place place/one place/two

The -p option is used to create multiple directories at once:

mkdir -p [path/to/nested/directories]  
# Creates nested directories

mkdir -p ../project/data ../project/results

To list all nested subdirectories within a directory, use ls -R.

Example:

To create a directory 2016 with a sub-directory data that contains two directories, processed and raw:

Method 1: Step-by-Step Creation

mkdir 2016
mkdir 2016/data
mkdir 2016/data/processed
mkdir 2016/data/raw

Method 2: Navigating and Creating

mkdir 2016
cd 2016
mkdir data
cd data
mkdir processed raw

Method 3: Using -p

mkdir -p 2016/data/{processed,raw}  
# Creates the full structure in one command

---