Using Windows Home Server’s Command-Line Tools : Working with the Command-Line Tools (part 2) - Working with File and Folder Management Tools

2. Working with File and Folder Management Tools

Windows Explorer is the GUI tool of choice for most file and folder operations. However, Windows Home Server comes with an impressive collection of command-line file and folder tools that let you perform all the standard operations such as renaming, copying, moving, and deleting, as well as more interesting chores such as changing file attributes and comparing the contents of two files. Table 2 lists the file management tools that you can use with Windows Home Server.

The next few sections take a closer look at a half dozen of these tools: ATTRIB, FIND, REN, REPLACE, SORT, and XCOPY.

Before getting to the tools, I should mention that most of the file and folder management tools work with the standard wildcard characters: ? and *. In a file or folder specification, you use ? to substitute for a single character, and you use * to substitute for multiple characters. Here are some examples:

ATTRIB: Modifying File and Folder Attributes

A file’s attributes are special codes that indicate the status of the file. There are four attributes you can work with:

The ATTRIB command lets you turn these attributes on or off. Here’s the syntax:

ATTRIB [+A | -A] [+H | -H] [+R | -R] [+S | -S] filename [/S [/D]]

For example, if you’d like to hide all the DOC files in the current directory, use the following command:

attrib +h *.doc

As another example, if you’ve ever tried to delete or edit a file and gotten the message Access denied, the file is likely read-only. You can turn off the read-only attribute by running ATTRIB with the -R switch, as in this example:

attrib -r readonly.txt

You can also use ATTRIB for protecting important or sensitive files. When you hide a file, it doesn’t show up in a listing produced by the DIR command. Out of sight is out of mind, so someone taking a casual glance at your files won’t see the hidden ones and, therefore, won’t be tempted to display or erase them.

Although a hidden file is invisible, it’s not totally safe. Someone who knows the name of the file can attempt to modify the file by opening it with the appropriate program. As an added measure of safety, you can also set the file’s read-only attribute. When you do this, the file can’t be modified. You can set both attributes with a single command:

attrib +h +r payroll.xlsx

FIND: Locating a Text String in a File

You use the FIND command to search for a string inside a file. Here’s the syntax:

FIND [/C] [/I] [/N] [/V] [/OFF[LINE]] "string" filename

For example, to find the string DVD in a file named WishList.txt, you use the following command:

find "DVD" WishList.txt

If the string you want to find contains double quotation marks, you need to place two quotation marks in the search string. For example, to find the phrase Dave “The Hammer” Schultz in the file players.doc, use the following command:

find "Dave ""The Hammer"" Schultz" players.doc

FOR %f IN (filespec) DO FIND "string" %f

FOR %f IN (*.doc) DO FIND "Thanksgiving" %f

FOR %f IN (*.doc) DO FIND "Thanksgiving" "%f"

One of the most common uses of the FIND command is as a filter in pipe operations. In this case, instead of a filename, you pipe the output of another command through FIND. In this case, FIND searches this input for a specified string and, if it finds a match, it displays the line that contains the string.

For example, the last line of a DIR listing tells you the number of bytes free on the current drive. Rather than wade through the entire DIR output just to get this information, use this command instead:

dir | find "free"

You’ll see something like the following:

2 Dir(s) 28,903,331,184 bytes free

FIND scours the DIR listing piped to it and looks for the word free. You can use this technique to display specific lines from, say, a CHKDSK report. For example, searching for bad finds the number of bad sectors on the disk.

REN: Renaming a File or Folder

You use the REN (or RENAME) command to change the name of one or more files and folders. Here’s the syntax:

REN old_filename1 new_filename

For example, the following command renamed Budget 2011.xlsx to Budget 2012.xlsx:

ren "Budget 2011.xlsx" "Budget 2012.xlsx"

A simple file or folder rename such as this probably isn’t something you’ll ever fire up a command-line session to do because renaming a single object is faster and easier in Windows Explorer. However, the real power of the REN command is that it accepts wildcards in the file specifications. This enables you to rename several files at once, something you can’t do in Windows Explorer.

For example, suppose you have a folder full of files, many of which contain 2011 somewhere in the filename. To rename all those files by changing 2011 to 2012, you’d use the following command:

ren *2011* *2012*

Similarly, if you have a folder full of files that use the .htm extension and you want to change each extension to .asp, you’d use the following command:

ren *.htm *.asp

Note that for these multiple-file renames to work, in most cases the original filename text and the new filename text must be the same length. For example, digital cameras often supply photos with names such as img_1234.jpg and img_5678.jpg. If you have a number of related photos in a folder, you might want to give them more meaningful names. If the photos are from a vacation in Rome, you might prefer names such as Rome_Vacation1234.jpg and Rome_Vacation5678.jpg. Unfortunately, the REN command can’t handle this. However, it can rename the files to Rome1234.jpg and Rome5678.jpg:

ren img_* Rome*

The exception to the same length rule is if the replacement occurs at the end of the filenames. For example, the following command renames all files with the .jpeg extension to .jpg:

ren *.jpeg *.jpg

REPLACE: Smarter File Copying

If there was such a thing as a Most Underrated Command award, REPLACE would win it hands down. This command, which you almost never hear about, can do three very useful (and very different) things:

• It copies files, but only if their names match those in the target directory.

• It copies files, but only if their names don’t exist in the target directory.

• It copies files, but only if their names match those in the target directory and the matching files in the target directory are older than the files being copied.

Here’s the syntax:

REPLACE source_files target /A /U /P /R /S /W

If you don’t specify switches, REPLACE copies a file from the source folder to the target folder if and only if it finds a file with a matching name in the target.

More useful is the REPLACE command’s updating mode, where it copies a file from the source folder to the target folder if and only if it finds a file with a matching name in the target and that target file is older than the source file. A good example where updating comes in handy is when you copy some files to a disk or memory card so you can use them on another machine (such as taking files from your computer at work to use them at home). When you need to copy the files back to the first machine, the following REPLACE command does the job. (This assumes the disk or memory card is in drive G:.)

replace g:*.* %UserProfile% /s /u

For each file on drive G:, REPLACE looks for matching filenames anywhere in the %UserProfile% folder and its subfolders (thanks to the /S switch) and replaces only the ones that are newer (the /U switch).

What if you created some new files on the other computer? To copy those to the first machine, use the /A switch, like so:

replace g:*.* %UserProfile%\Documents /a

In this case, REPLACE only copies a file from G if it doesn’t exist in the %UserProfile%\Documents folder. (You have to specify a target folder because you can’t use the /S switch with /A.)

SORT: Sorting the Contents of a File

When you obtain a file from the Internet or some other source, the data in the file may not appear in the order you want. What I usually do in such cases is import the file into Word or Excel and then use the program’s Sort feature. This sometimes involves extra steps (such as converting text to a table in Word), so it’s not always an efficient way to work.

If the file is text, it’s often easier and faster to run the SORT command-line tool. By default, SORT takes the content of the file, sorts it in ascending alphanumeric order (0 to 9, then a to z, and then A to Z) starting at the beginning of each line in the file, and then displays the sorted results. You can also run descending order sorts, write the results to the same file or another file, and more. Here’s the syntax:

SORT [input_file] [/+n] [/R] [/L locale] [/M kilobytes] [/REC characters] [/T temp_folder] [/O output_file]

					  

For example, the following SORT command sorts the data in records.txt and stores the results in sorted_records.txt:

sort records.txt sorted_records.txt

XCOPY: Advanced File Copying

The XCOPY command is one of the most powerful of the file management command-line tools, and you can use it for some fairly sophisticated file copying operations.

Here’s the syntax for XCOPY:

XCOPY source destination [/A | /M] [/C] [/D[:mm-dd—yyyy]] [/EXCLUDE:
file1[+file2[+file3]]] [/F] [/G] [/H] [/I] [/K] [/L] [/N] [/O] [/P] [/Q] [/R] [/S [/E]] [/T] [/U] [/V] [/W] [/X] [/Y | -Y] [/Z] [/B] [/J]

					  

In its basic form, XCOPY works just like COPY. So, for example, to copy all the .doc files in the current folder to a folder called Documents in drive G:, use the following command:

xcopy *.doc g:\documents

Besides being faster, XCOPY also contains a number of features not found in the puny COPY command. Think of it as COPY on steroids. (The X in XCOPY means that it’s an extended COPY command.) For example, suppose you want to copy all the .doc files in the current folder and all the .doc files in any attached subfolders to G:\Documents. With COPY, you first have to create the appropriate folders on the destination partition and then perform separate COPY commands for each folder, which is not very efficient, to say the least. With XCOPY, all you do is add a single switch:

xcopy *.doc g:\documents /s
xcopy *.bat d:\batch /s

The /S switch tells XCOPY to copy the current folder and all non-empty subfolders and to create the appropriate folders in the destination, as needed. (If you want XCOPY to copy empty subfolders, include the /E switch, as well.)

Another useful feature of XCOPY is the ability to copy files by date. This is handy for performing incremental backups of files that you modified on or after a specific date. For example, suppose you keep your word processing documents in %UserProfile%\Documents and you want to make backup copies in your Windows Home Server user share of all the .doc files that have changed since August 23, 2011. You can do this with the following command:

xcopy %userprofile%\documents\*.doc \\server\users\%Username%\ /d:08-23-2011

					  

It’s common to use XCOPY in batch files, but take care to handle errors. For example, what if a batch file tries to use XCOPY, but there’s not enough memory? Or what if the user presses Ctrl+C during the copy? It might seem impossible to check for these kinds of errors; yet it is not only possible, it’s really quite easy.

When certain commands finish, they always file a report on the progress of the operation. This report, or exit code, is a number that specifies how the operation went. For example, Table 3 lists the exit codes that the XCOPY command uses.

What does all this mean for your batch files? You can use a variation of the IF command—IF ERRORLEVEL—to test for these exit codes. For example, here’s a batch file called CheckCopy.bat, which uses some of the XCOPY exit codes to check for errors:

@ECHO OFF
XCOPY %1 %2
IF ERRORLEVEL 4 GOTO ERROR
IF ERRORLEVEL 2 GOTO CTRL+C
IF ERRORLEVEL 1 GOTO NO_FILES
GOTO DONE
:ERROR
ECHO Bad news! The copy failed because there wasn't
ECHO enough memory or disk space or because there was
ECHO something wrong with your file specs . . .
GOTO DONE
:CTRL+C
ECHO Hey, what gives? You pressed Ctrl+C to abort . . .
GOTO DONE
:NO_FILES
ECHO Bad news! No files were found to copy . . .
:DONE

As you can see, the ERRORLEVEL conditions check for the individual exit codes and then use GOTO to jump to the appropriate label.

One of the most important things to know about the IF ERRORLEVEL test is how Windows Home Server interprets it. For example, consider the following IF command:

IF ERRORLEVEL 2 GOTO CTRL+C

Windows Home Server interprets this command as “If the exit code from the last command is equal to or greater than 2, jump to the CTRL+C label.” This has two important consequences for your batch files:

• The test IF ERRORLEVEL 0 doesn’t tell you much because it’s always true. If you simply want to find out whether the command failed, use the test IF NOT ERRORLEVEL 0.

• To get the correct results, always test the highest ERRORLEVEL first and then work your way down.