few - a file viewer with regular expression filters.
few [--regex '/REGEX/flags']* [--search '/REGEX/flags'] [--tabwidth 'NUM'] [--goto 'NUM'] [-v] [--color] [-h|-?|--help] ['FILE']
few is a program to interactively filter log files with regular expressions and view the result. It is similar to the less(1) program, but has the ability to filter what is displayed with regular expressions.
--regex '/REGEX/flags': preset regular expressions to filter the file.
--search '/REGEX/flags': preset search regular expression.
--tabwidth 'NUM': set the width of a tab character in spaces.
--goto 'NUM': go to line number NUM. If NUM is not included in the currently filtered lines, go to the previous filtered line.
-v: increase verbosity for certain operations.
--color: enable color if the terminal supports color.
-h, -?, --help: show help text.
'FILE': the file name to use. If it is "-" the standard input is used. If no file name was specified standard input is used.
The actions resulting from a key press are modelled after the less(1) program.
- q, Q: quit the program..
- cursor down: scroll down one line.
- n: scroll down one line if no search regex is set.
- cursor up, p: scroll up one line.
- space, page down: scroll down one page.
- b, page up: scroll up one page.
- g, <, home: go to to first line.
- G, >, end: go to last line.
- 1 .. 9, 0: edit regular expressions 1 to 10.
- F1 .. F12: edit regular expressions 11 to 22.
- A: abort any running regular expression evaluations.
- d: scroll down half a screen
- u: scroll up half a screen
- P: goto line
- %: goto percentage of lines
- R: repaint the screen
- h: show help text. This will show the few man page in the few editor. To return to your file, press q. Currently supported on Unix.
- /: enter a search regular expression
- n: go to next search match if a search regex is set.
- N: go to previous search match.
- M: maximize the window. Currently supported on Windows.
- S: save the currently filtered lines to a new file.
- !: execute a shell command. Use !! to execute the last shell command. A % is replaced with the file name currently displayed by few. Currently supported on Unix.
FILTER REGULAR EXPRESSIONS
The standard form of a filter regular expression has the following format:
The regular expression is surrounded by two forward slash characters. The flags are optional.
See http://www.cplusplus.com/reference/regex/ECMAScript/ for details on the regular expression syntax. It is similar to perl regular expressions.
Regular Expression Flags
The filter regular expression can be modified with optional flag characters, the following characters are currently supported:
- i: case insensitive match
- !: negative match
By default the filter regular expression is a positive match. Only lines in the file that (partially) match the regular expression will be displayed. If you use the '!' flag the regular expression becomes a negative match. Only lines that do not match the regular expression are displayed.
Filter Regular Expression Short Form
If you do not need to use flags for the filter regular expression and the filter regular expression does not start with a forward slash, you can omit the surrounding slashes.
If you want a negative match and your filter regular expression does not start with a '!' character and you don't need additional flags, you can start your filter regular expression with a '!' character.
The few program will convert the short forms to the regular form.
Replace Display Filter Regular Expressions
A Replace Display Filter changes the way the lines are displayed. They take the form of a perl regular expression substitute:
Because the parsing of this format is currently naive, the strings regex and replace can not contain any forward slash characters. The replace display filter allows the optional 'i' flag to indicate case insensitive matching. The replace string can contain back references to the regular expression. The following back references or available:
- $1 .. $99: numbered back reference to a match group.
- $&: the entire match.
- $`: prefix to the match.
- $': suffix to the match.
- $$: a single $ character.
See http://www.cplusplus.com/reference/regex/regex_replace/ for details on how the C++ regex library is matching and replaceing.
Attribute Display Filter Regular Expression
You can use a regular expression to change the character attributes and the colors used to display the line. This is called an Attribute Display Filter. This uses the following form:
The following attributes are supported by few. Note that your curses implementation or terminal may not support all of these attributes.
The optional color specification has the following form:
The first color specifies the foreground color. The second color specifies the background color. The following colors are supported:
To use colors, you have to start the few program with the --color command line argument.
Example attribute display filter:
|\d|italic,bold,yellow on blue
Will print all digits in italic and bold with yellow foreground and blue background color.
Regular Expressions and wide characters
The few program is able to display wide characters and files encoded in UTF-8. However because it is designed to work with large files, it will not attempt to match the filter regular expressions on the wide character representations of the file processed. The replace display filter regular expression is matched on the raw byte sequence of the file as well. This is usually not a problem if you match ASCII characters.
The attribute display filter regular expression however is matched against the wide character representation of the displayed lines.
LINKS AND URLs
few will attempt to detect URLs and links. If it detects a link, URL or email address few will underline the match. You can click the underlined text to have the link opened in a web browser or email client. On Unix you need to set the BROWSER environment variable if you don't like the default web browser firefox.
When you start few you'll see an inverted column of number on the left side. You'll see the line number that is displayed right aligned. If there's enough room on the left side of that column you'll see the line size in characters, excluding any newline or carriage return characters left aligned. The main part of your screen is used to show the file contents. The bottom right corner of the main display is used to show information on the file or results from internal processing. If you've entered a filter, display or search regular expression they are shown in the bottom part of the screen in an inverted block. The lower right side
When you edit a regular expression for display filters or search, you can use the tab key to auto complete words that are currently displayed. Auto completion also works on file and directory names when you save the filtered lines to a file.
BROWSER: On Unix: name of web browser to launch; If not set defaults to firefox. On Windows this variable is not used and few uses the system default web browser.
MAILER: On Unix: name of the mail application; If not set defaults to thunderbird. On Windows this variable is not used and few uses the system default mail application.
FEWOPTIONS: default command line arguments.
few is written by Dirk Jagdmann firstname.lastname@example.org. You can write him an email and ask questions about this program.