RGBCube/uutils-coreutils
1
Fork 0
mirror of https://github.com/RGBCube/uutils-coreutils synced 2025-07-30 12:37:49 +00:00
Code Issues Activity

cut: switch to using new coreopts syntax

Browse source
This commit is contained in:
Nathan Ross 2016-08-11 12:37:59 -04:00
parent f8509240df
commit e019b2657e
1 changed files with 82 additions and 93 deletions
Download patch file Download diff file Expand all files Collapse all files

175
src/cut/cut.rs
View file

@ -25,8 +25,87 @@ mod buffer;
mod ranges;
mod searcher;
static NAME: &'static str = "cut";
static VERSION: &'static str = env!("CARGO_PKG_VERSION");
static SYNTAX: &'static str = "[-d] [-s] [-z] [--output-delimiter] ((-f|-b|-c) {{sequence}}) {{sourcefile}}+";
static SUMMARY: &'static str = "Prints specified byte or field columns from each line of stdin or the input files";
static LONG_HELP: &'static str = "
Each call must specify a mode (what to use for columns),
a sequence (which columns to print), and provide a data source
Specifying a mode
Use --bytes (-b) or --characters (-c) to specify byte mode
Use --fields (-f) to specify field mode, where each line is broken into
fields identified by a delimiter character. For example for a typical CSV
you could use this in combination with setting comma as the delimiter
Specifying a sequence
A sequence is a group of 1 or more numbers or inclusive ranges separated
by a commas.
cut -f 2,5-7 some_file.txt
will display the 2nd, 5th, 6th, and 7th field for each source line
Ranges can extend to the end of the row by excluding the the second number
cut -f 3- some_file.txt
will display the 3rd field and all fields after for each source line
The first number of a range can be excluded, and this is effectively the
same as using 1 as the first number: it causes the range to begin at the
first column. Ranges can also display a single column
cut -f 1,3-5 some_file.txt
will display the 1st, 3rd, 4th, and 5th field for each source line
The --complement option, when used, inverts the effect of the sequence
cut --complement -f 4-6 some_file.txt
will display the every field but the 4th, 5th, and 6th
Specifying a data source
If no sourcefile arguments are specified, stdin is used as the source of
lines to print
If sourcefile arguments are specified, stdin is ignored and all files are
read in consecutively if a sourcefile is not successfully read, a warning
will print to stderr, and the eventual status code will be 1, but cut
will continue to read through proceeding sourcefiles
To print columns from both STDIN and a file argument, use - (dash) as a
sourcefile argument to represent stdin.
Field Mode options
The fields in each line are identified by a delimiter (separator)
Set the delimiter
Set the delimiter which separates fields in the file using the
--delimiter (-d) option. Setting the delimiter is optional.
If not set, a default delimiter of Tab will be used.
Optionally Filter based on delimiter
If the --only-delimited (-s) flag is provided, only lines which
contain the delimiter will be printed
Replace the delimiter
If the --output-delimiter option is provided, the argument used for
it will replace the delimiter character in each line printed. This is
useful for transforming tabular data - e.g. to convert a CSV to a
TSV (tab-separated file)
Line endings
When the --zero-terminated (-z) option is used, cut sees \\0 (null) as the
'line ending' character (both for the purposes of reading lines and
separating printed lines) instead of \\n (newline). This is useful for
tabular data where some of the cells may contain newlines
echo 'ab\\0cd' | cut -z -c 1
will result in 'a\\0c\\0'
";
struct Options {
out_delim: Option<String>,
@ -408,8 +487,7 @@ fn cut_files(mut filenames: Vec<String>, mode: Mode) -> i32 {
}
pub fn uumain(args: Vec<String>) -> i32 {
let mut opts = uucore::coreopts::CoreOptions::new(NAME);
let mut opts = new_coreopts!(SYNTAX, SUMMARY, LONG_HELP);
opts.optopt("b", "bytes", "filter byte columns from the input source", "sequence");
opts.optopt("c", "characters", "alias for character mode", "sequence");
opts.optopt("d", "delimiter", "specify the delimiter character that separates fields in the input source. Defaults to Tab.", "delimiter");
@ -419,95 +497,6 @@ pub fn uumain(args: Vec<String>) -> i32 {
opts.optflag("s", "only-delimited", "in field mode, only print lines which contain the delimiter");
opts.optflag("z", "zero-terminated", "instead of filtering columns based on line, filter columns based on \\0 (NULL character)");
opts.optopt("", "output-delimiter", "in field mode, replace the delimiter in output lines with this option's argument", "new delimiter");
let usage = opts.usage("Prints specified byte or field columns from each line of stdin or the input files");
opts.help(format!("
{0} {1}
{0} [-d] [-s] [-z] [--output-delimiter] ((-f|-b|-c) {{sequence}}) {{sourcefile}}+
{2}
Reference
Each call must specify a mode (what to use for columns),
a sequence (which columns to print), and provide a data source
Specifying a mode
Use --bytes (-b) or --characters (-c) to specify byte mode
Use --fields (-f) to specify field mode, where each line is broken into
fields identified by a delimiter character. For example for a typical CSV
you could use this in combination with setting comma as the delimiter
Specifying a sequence
A sequence is a group of 1 or more numbers or inclusive ranges separated
by a commas.
cut -f 2,5-7 some_file.txt
will display the 2nd, 5th, 6th, and 7th field for each source line
Ranges can extend to the end of the row by excluding the the second number
cut -f 3- some_file.txt
will display the 3rd field and all fields after for each source line
The first number of a range can be excluded, and this is effectively the
same as using 1 as the first number: it causes the range to begin at the
first column. Ranges can also display a single column
cut -f 1,3-5 some_file.txt
will display the 1st, 3rd, 4th, and 5th field for each source line
The --complement option, when used, inverts the effect of the sequence
cut --complement -f 4-6 some_file.txt
will display the every field but the 4th, 5th, and 6th
Specifying a data source
If no sourcefile arguments are specified, stdin is used as the source of
lines to print
If sourcefile arguments are specified, stdin is ignored and all files are
read in consecutively if a sourcefile is not successfully read, a warning
will print to stderr, and the eventual status code will be 1, but cut
will continue to read through proceeding sourcefiles
To print columns from both STDIN and a file argument, use - (dash) as a
sourcefile argument to represent stdin.
Field Mode options
The fields in each line are identified by a delimiter (separator)
Set the delimiter
Set the delimiter which separates fields in the file using the
--delimiter (-d) option. Setting the delimiter is optional.
If not set, a default delimiter of Tab will be used.
Optionally Filter based on delimiter
If the --only-delimited (-s) flag is provided, only lines which
contain the delimiter will be printed
Replace the delimiter
If the --output-delimiter option is provided, the argument used for
it will replace the delimiter character in each line printed. This is
useful for transforming tabular data - e.g. to convert a CSV to a
TSV (tab-separated file)
Line endings
When the --zero-terminated (-z) option is used, cut sees \\0 (null) as the
'line ending' character (both for the purposes of reading lines and
separating printed lines) instead of \\n (newline). This is useful for
tabular data where some of the cells may contain newlines
echo 'ab\\0cd' | cut -z -c 1
will result in 'a\\0c\\0'
", NAME, VERSION, usage));
let matches = opts.parse(args);
let complement = matches.opt_present("complement");