Reference

Command overview

Whereas the Unix toolkit is made of the separate executables cat, tail, cut, sort, etc., Miller has subcommands, invoked as follows:

mlr tac *.dat
mlr cut --complement -f os_version *.dat
mlr sort -f hostname,uptime *.dat

These fall into categories as follows:

Commands Description
cat, cut, grep, head, join, sort, tac, tail, top, uniq Analogs of their Unix-toolkit namesakes, discussed below as well as in Miller features in the context of the Unix toolkit
filter, put, sec2gmt, sec2gmtdate, step, tee awk-like functionality
bar, bootstrap, decimate, histogram, least-frequent, most-frequent, sample, shuffle, stats1, stats2 Statistically oriented
group-by, group-like, having-fields Particularly oriented toward Record-heterogeneity, although all Miller commands can handle heterogeneous records
check, count-distinct, label, merge-fields, nest, nothing, rename, rename, reorder, reshape, seqgen These draw from other sources (see also How original is Miller?): count-distinct is SQL-ish, and rename can be done by sed (which does it faster: see Performance).

I/O options

Formats

Options:

  --dkvp    --idkvp    --odkvp
  --nidx    --inidx    --onidx
  --csv     --icsv     --ocsv
  --csvlite --icsvlite --ocsvlite
  --pprint  --ipprint  --opprint  --right
  --xtab    --ixtab    --oxtab
  --json    --ijson    --ojson

These are as discussed in File formats, with the exception of --right which makes pretty-printed output right-aligned:

$ mlr --opprint cat data/small
a   b   i x                   y
pan pan 1 0.3467901443380824  0.7268028627434533
eks pan 2 0.7586799647899636  0.5221511083334797
wye wye 3 0.20460330576630303 0.33831852551664776
eks wye 4 0.38139939387114097 0.13418874328430463
wye pan 5 0.5732889198020006  0.8636244699032729

$ mlr --opprint --right cat data/small
  a   b i                   x                   y
pan pan 1  0.3467901443380824  0.7268028627434533
eks pan 2  0.7586799647899636  0.5221511083334797
wye wye 3 0.20460330576630303 0.33831852551664776
eks wye 4 0.38139939387114097 0.13418874328430463
wye pan 5  0.5732889198020006  0.8636244699032729

Additional notes:

  • Use --csv, --pprint, etc. when the input and output formats are the same.
  • Use --icsv --opprint, etc. when you want format conversion as part of what Miller does to your data.
  • DKVP (key-value-pair) format is the default for input and output. So, --oxtab is the same as --idkvp --oxtab.

In-place mode

Use the mlr -I flag to process files in-place. For example, mlr -I --csv cut -x -f unwanted_column_name mydata/*.csv will remove unwanted_column_name from all your *.csv files in your mydata/ subdirectory.

By default, Miller output goes to the screen (or you can redirect a file using > or to another process using |). With -I, for each file name on the command line, output is written to a temporary file in the same directory. Miller writes its output into that temp file, which is then renamed over the original. Then, processing continues on the next file. Each file is processed in isolation: if the output format is CSV, CSV headers will be present in each output file; statistics are only over each file's own records; and so on.

Please see here for examples.

Compression

Options:

  --prepipe {command}

The prepipe command is anything which reads from standard input and produces data acceptable to Miller. Nominally this allows you to use whichever decompression utilities you have installed on your system, on a per-file basis. If the command has flags, quote them: e.g. mlr --prepipe 'zcat -cf'. Examples:

# These two produce the same output:
$ gunzip < myfile1.csv.gz | mlr cut -f hostname,uptime
$ mlr --prepipe gunzip cut -f hostname,uptime myfile1.csv.gz
# With multiple input files you need --prepipe:
$ mlr --prepipe gunzip cut -f hostname,uptime myfile1.csv.gz myfile2.csv.gz
$ mlr --prepipe gunzip --idkvp --oxtab cut -f hostname,uptime myfile1.dat.gz myfile2.dat.gz

# Similar to the above, but with compressed output as well as input:
$ gunzip < myfile1.csv.gz | mlr cut -f hostname,uptime | gzip > outfile.csv.gz
$ mlr --prepipe gunzip cut -f hostname,uptime myfile1.csv.gz | gzip > outfile.csv.gz
$ mlr --prepipe gunzip cut -f hostname,uptime myfile1.csv.gz myfile2.csv.gz | gzip > outfile.csv.gz

# Similar to the above, but with different compression tools for input and output:
$ gunzip < myfile1.csv.gz | mlr cut -f hostname,uptime | xz -z > outfile.csv.xz
$ xz -cd < myfile1.csv.xz | mlr cut -f hostname,uptime | gzip > outfile.csv.xz
$ mlr --prepipe 'xz -cd' cut -f hostname,uptime myfile1.csv.xz myfile2.csv.xz | xz -z > outfile.csv.xz

... etc.

Record/field/pair separators

Miller has record separators IRS and ORS, field separators IFS and OFS, and pair separators IPS and OPS. For example, in the DKVP line a=1,b=2,c=3, the record separator is newline, field separator is comma, and pair separator is the equals sign. These are the default values.

Options:

  --rs --irs --ors
  --fs --ifs --ofs --repifs
  --ps --ips --ops
  • You can change a separator from input to output via e.g. --ifs = --ofs :. Or, you can specify that the same separator is to be used for input and output via e.g. --fs :.
  • The pair separator is only relevant to DKVP format.
  • Pretty-print and xtab formats ignore the separator arguments altogether.
  • The --repifs means that multiple successive occurrences of the field separator count as one. For example, in CSV data we often signify nulls by empty strings, e.g. 2,9,,,,,6,5,4. On the other hand, if the field separator is a space, it might be more natural to parse 2 4 5 the same as 2 4 5: --repifs --ifs ' ' lets this happen. In fact, the --ipprint option above is internally implemented in terms of --repifs.
  • Just write out the desired separator, e.g. --ofs '|'. But you may use the symbolic names newline, space, tab, pipe, or semicolon if you like.

Number formatting

The command-line option --ofmt {format string} is the global number format for commands which generate numeric output, e.g. stats1, stats2, histogram, and step, as well as mlr put. Examples:

--ofmt %.9le  --ofmt %.6lf  --ofmt %.0lf

These are just C printf formats applied to double-precision numbers. Please don’t use %s or %d. Additionally, if you use leading width (e.g. %18.12lf) then the output will contain embedded whitespace, which may not be what you want if you pipe the output to something else, particularly CSV. I use Miller’s pretty-print format (mlr --opprint) to column-align numerical data.

To apply formatting to a single field, overriding the global ofmt, use fmtnum function within mlr put. For example:

$ echo 'x=3.1,y=4.3' | mlr put '$z=fmtnum($x*$y,"%08lf")'
x=3.1,y=4.3,z=13.330000

$ echo 'x=0xffff,y=0xff' | mlr put '$z=fmtnum(int($x*$y),"%08llx")'
x=0xffff,y=0xff,z=00feff01

Input conversion from hexadecimal is done automatically on fields handled by mlr put and mlr filter as long as the field value begins with "0x". To apply output conversion to hexadecimal on a single column, you may use fmtnum, or the keystroke-saving hexfmt function. Example:

$ echo 'x=0xffff,y=0xff' | mlr put '$z=hexfmt($x*$y)'
x=0xffff,y=0xff,z=0xfeff01

Data transformations (verbs)

Expression language for filter and put

then-chaining

In accord with the Unix philosophy, you can pipe data into or out of Miller. For example:

mlr cut --complement -f os_version *.dat | mlr sort -f hostname,uptime

You can, if you like, instead simply chain commands together using the then keyword:

mlr cut --complement -f os_version then sort -f hostname,uptime *.dat

(You can precede the very first verb with then, if you like, for symmetry.) Here’s a performance comparison:

% cat piped.sh
mlr cut -x -f i,y data/big | mlr sort -n y > /dev/null

% time sh piped.sh
real 0m2.828s
user 0m3.183s
sys  0m0.137s


% cat chained.sh
mlr cut -x -f i,y then sort -n y data/big > /dev/null

% time sh chained.sh
real 0m2.082s
user 0m1.933s
sys  0m0.137s

There are two reasons to use then-chaining: one is for performance, although I don’t expect this to be a win in all cases. Using then-chaining avoids redundant string-parsing and string-formatting at each pipeline step: instead input records are parsed once, they are fed through each pipeline stage in memory, and then output records are formatted once. On the other hand, Miller is single-threaded, while modern systems are usually multi-processor, and when streaming-data programs operate through pipes, each one can use a CPU. Rest assured you get the same results either way.

The other reason to use then-chaining is for simplicity: you don’t have re-type formatting flags (e.g. --csv --fs tab) at every pipeline stage.

Auxiliary commands

There are a few nearly-standalone programs which have nothing to do with the rest of Miller, do not participate in record streams, and do not deal with file formats. They might as well be little standalone executables but they’re delivered within the main Miller executable for convenience.

$ mlr aux-list
Available subcommands:
  aux-list
  lecat
  termcvt
  hex
  unhex
  netbsd-strptime
For more information, please invoke mlr {subcommand} --help

Examples:

$ echo 'Hello, world!' | mlr hex
00000000: 48 65 6c 6c  6f 2c 20 77  6f 72 6c 64  21 0a       |Hello, world!.|

$ echo 'Hello, world!' | mlr hex -r
48 65 6c 6c  6f 2c 20 77  6f 72 6c 64  21 0a

$ echo 'Hello, world!' | mlr hex -r | mlr unhex
Hello, world!

$ echo 'Hello, world!' | mlr lecat --mono
Hello, world![LF]

Data types

Miller’s input and output are all string-oriented: there is (as of August 2015 anyway) no support for binary record packing. In this sense, everything is a string in and out of Miller. During processing, field names are always strings, even if they have names like "3"; field values are usually strings. Field values’ ability to be interpreted as a non-string type only has meaning when comparison or function operations are done on them. And it is an error condition if Miller encounters non-numeric (or otherwise mistyped) data in a field in which it has been asked to do numeric (or otherwise type-specific) operations.

Field values are treated as numeric for the following:

  • Numeric sort: mlr sort -n, mlr sort -nr.
  • Statistics: mlr histogram, mlr stats1, mlr stats2.
  • Cross-record arithmetic: mlr step.

For mlr put and mlr filter:

  • Miller’s types for function processing are empty-null (empty string), absent-null (reads of unset right-hand sides, or fall-through non-explicit return values from user-defined functions), error, string, float (double-precision), int (64-bit signed), and boolean.
  • On input, string values representable as numbers, e.g. "3" or "3.1", are treated as int or float, respectively. If a record has x=1,y=2 then mlr put '$z=$x+$y' will produce x=1,y=2,z=3, and mlr put '$z=$x.$y' does not give an error simply because the dot operator has been generalized to stringify non-strings. To coerce back to string for processing, use the string function: mlr put '$z=string($x).string($y)' will produce x=1,y=2,z=12.
  • On input, string values representable as boolean (e.g. "true", "false") are not automatically treated as boolean. (This is because "true" and "false" are ordinary words, and auto string-to-boolean on a column consisting of words would result in some strings mixed with some booleans.) Use the boolean function to coerce: e.g. giving the record x=1,y=2,w=false to mlr put '$z=($x<$y) || boolean($w)'.
  • Functions take types as described in mlr --help-all-functions: for example, log10 takes float input and produces float output, gmt2sec maps string to int, and sec2gmt maps int to string.
  • All math functions described in mlr --help-all-functions take integer as well as float input.

Null data: empty and absent

One of Miller’s key features is its support for heterogeneous data. For example, take mlr sort: if you try to sort on field hostname when not all records in the data stream have a field named hostname, it is not an error (although you could pre-filter the data stream using mlr having-fields --at-least hostname then sort ...). Rather, records lacking one or more sort keys are simply output contiguously by mlr sort.

Miller has two kinds of null data:

  • Empty (key present, value empty): a field name is present in a record (or in an out-of-stream variable) with empty value: e.g. x=,y=2 in the data input stream, or assignment $x="" or @x="" in mlr put.
  • Absent (key not present): a field name is not present, e.g. input record is x=1,y=2 and a put or filter expression refers to $z. Or, reading an out-of-stream variable which hasn’t been assigned a value yet, e.g. mlr put -q '@sum += $x'; end{emit @sum}' or mlr put -q '@sum[$a][$b] += $x'; end{emit @sum, "a", "b"}'.

You can test these programatically using the functions is_empty/is_not_empty, is_absent/is_present, and is_null/is_not_null. For the last pair, note that null means either empty or absent.

Rules for null-handling:

  • Records with one or more empty sort-field values sort after records with all sort-field values present:

    $ mlr cat data/sort-null.dat
    a=3,b=2
    a=1,b=8
    a=,b=4
    x=9,b=10
    a=5,b=7
    

    $ mlr sort -n  a data/sort-null.dat
    a=1,b=8
    a=3,b=2
    a=5,b=7
    a=,b=4
    x=9,b=10
    

    $ mlr sort -nr a data/sort-null.dat
    a=,b=4
    a=5,b=7
    a=3,b=2
    a=1,b=8
    x=9,b=10
    

  • Functions/operators which have one or more empty arguments produce empty output: e.g.

    $ echo 'x=2,y=3' | mlr put '$a=$x+$y'
    x=2,y=3,a=5
    

    $ echo 'x=,y=3' | mlr put '$a=$x+$y'
    x=,y=3,a=
    

    $ echo 'x=,y=3' | mlr put '$a=log($x);$b=log($y)'
    x=,y=3,a=,b=1.098612
    

    with the exception that the min and max functions are special: if one argument is non-null, it wins:

    $ echo 'x=,y=3' | mlr put '$a=min($x,$y);$b=max($x,$y)'
    x=,y=3,a=3,b=3
    

  • Functions of absent variables (e.g. mlr put '$y = log10($nonesuch)') evaluate to absent, and arithmetic/bitwise/boolean operators with both operands being absent evaluate to absent. Arithmetic operators with one absent operand return the other operand. More specifically, absent values act like zero for addition/subtraction, and one for multiplication: Furthermore, any expression which evaluates to absent is not stored in the left-hand side of an assignment statement :

    $ echo 'x=2,y=3' | mlr put '$a=$u+$v; $b=$u+$y; $c=$x+$y'
    x=2,y=3,b=3,c=5
    

    $ echo 'x=2,y=3' | mlr put '$a=min($x,$v);$b=max($u,$y);$c=min($u,$v)'
    x=2,y=3,a=2,b=3
    

  • Likewise, for assignment to maps, absent-valued keys or values result in a skipped assignment.
The reasoning is as follows:

Since absent plus absent is absent (and likewise for other operators), accumulations such as @sum += $x work correctly on heterogenous data, as do within-record formulas if both operands are absent. If one operand is present, you may get behavior you don’t desire. To work around this — namely, to set an output field only for records which have all the inputs present — you can use a pattern-action block with is_present:

$ mlr cat data/het.dkvp
resource=/path/to/file,loadsec=0.45,ok=true
record_count=100,resource=/path/to/file
resource=/path/to/second/file,loadsec=0.32,ok=true
record_count=150,resource=/path/to/second/file
resource=/some/other/path,loadsec=0.97,ok=false

$ mlr put 'is_present($loadsec) { $loadmillis = $loadsec * 1000 }' data/het.dkvp
resource=/path/to/file,loadsec=0.45,ok=true,loadmillis=450.000000
record_count=100,resource=/path/to/file
resource=/path/to/second/file,loadsec=0.32,ok=true,loadmillis=320.000000
record_count=150,resource=/path/to/second/file
resource=/some/other/path,loadsec=0.97,ok=false,loadmillis=970.000000

$ mlr put '$loadmillis = (is_present($loadsec) ? $loadsec : 0.0) * 1000' data/het.dkvp
resource=/path/to/file,loadsec=0.45,ok=true,loadmillis=450.000000
record_count=100,resource=/path/to/file,loadmillis=0.000000
resource=/path/to/second/file,loadsec=0.32,ok=true,loadmillis=320.000000
record_count=150,resource=/path/to/second/file,loadmillis=0.000000
resource=/some/other/path,loadsec=0.97,ok=false,loadmillis=970.000000

If you’re interested in a formal description of how empty and absent fields participate in arithmetic, here’s a table for plus (other arithmetic/boolean/bitwise operators are similar):

$ mlr --print-type-arithmetic-info
(+)    | error  absent empty  string int    float  bool
------ + ------ ------ ------ ------ ------ ------ ------
error  | error  error  error  error  error  error  error
absent | error  absent absent error  int    float  error
empty  | error  absent empty  error  empty  empty  error
string | error  error  error  error  error  error  error
int    | error  int    empty  error  int    float  error
float  | error  float  empty  error  float  float  error
bool   | error  error  error  error  error  error  error

String literals

You can use the following backslash escapes for strings such as between the double quotes in contexts such as mlr filter '$name =~ "..."', mlr put '$name = $othername . "..."', mlr put '$name = sub($name, "...", "..."), etc.:

  • \a: ASCII code 0x07 (alarm/bell)
  • \b: ASCII code 0x08 (backspace)
  • \f: ASCII code 0x0c (formfeed)
  • \n: ASCII code 0x0a (LF/linefeed/newline)
  • \r: ASCII code 0x0d (CR/carriage return)
  • \t: ASCII code 0x09 (tab)
  • \v: ASCII code 0x0b (vertical tab)
  • \\: backslash
  • \": double quote
  • \123: Octal 123, etc. for \000 up to \377
  • \x7f: Hexadecimal 7f, etc. for \x00 up to \xff

See also https://en.wikipedia.org/wiki/Escape_sequences_in_C.

These replacements apply only to strings you key in for the DSL expressions for filter and put: that is, if you type \t in a string literal for a filter/put expression, it will be turned into a tab character. If you want a backslash followed by a t, then please type \\t.

However, these replacements are not done automatically within your data stream. If you wish to make these replacements, you can do, for example, for a field named field, mlr put '$field = gsub($field, "\\t", "\t")'. If you need to make such a replacement for all fields in your data, you should probably simply use the system sed command.

Regular expressions

Miller lets you use regular expressions (of type POSIX.2) in the following contexts:

  • In mlr filter with =~ or !=~, e.g. mlr filter '$url =~ "http.*com"'
  • In mlr put with sub or gsub, e.g. mlr put '$url = sub($url, "http.*com", "")'
  • In mlr having-fields, e.g. mlr having-fields --any-matching '^sda[0-9]'
  • In mlr cut, e.g. mlr cut -r -f '^status$,^sda[0-9]'
  • In mlr rename, e.g. mlr rename -r '^(sda[0-9]).*$,dev/\1'
  • In mlr grep, e.g. mlr --csv grep 00188555487 myfiles*.csv

Points demonstrated by the above examples:

  • There are no implicit start-of-string or end-of-string anchors; please use ^ and/or $ explicitly.
  • Miller regexes are wrapped with double quotes rather than slashes.
  • The i after the ending double quote indicates a case-insensitive regex.
  • Capture groups are wrapped with (...) rather than \(...\); use \( and \) to match against parentheses.

For filter and put, if the regular expression is a string literal (the normal case), it is precompiled at process start and reused thereafter, which is efficient. If the regular expression is a more complex expression, including string concatenation using ., or a column name (in which case you can take regular expressions from input data!), then regexes are compiled on each record which works but is less efficient. As well, in this case there is no way to specify case-insensitive matching.

Example:

$ cat data/regex-in-data.dat
name=jane,regex=^j.*e$
name=bill,regex=^b[ou]ll$
name=bull,regex=^b[ou]ll$

$ mlr filter '$name =~ $regex' data/regex-in-data.dat
name=jane,regex=^j.*e$
name=bull,regex=^b[ou]ll$

Regex captures

Regex captures of the form \0 through \9 are supported as follows:

  • Captures have in-function context for sub and gsub. For example, the first \1,\2 pair belong to the first sub and the second \1,\2 pair belong to the second sub:

    mlr put '$b = sub($a, "(..)_(...)", "\2-\1"); $c = sub($a, "(..)_(.)(..)", ":\1:\2:\3")'
    
  • Captures endure for the entirety of a put for the =~ and !=~ operators. For example, here the \1,\2 are set by the =~ operator and are used by both subsequent assignment statements:

    mlr put '$a =~ "(..)_(....); $b = "left_\1"; $c = "right_\2"'
    
  • The captures are not retained across multiple puts. For example, here the \1,\2 won’t be expanded from the regex capture:

    mlr put '$a =~ "(..)_(....)' then {... something else ...} then put '$b = "left_\1"; $c = "right_\2"'
    
  • Captures are ignored in filter for the =~ and !=~ operators. For example, there is no mechanism provided to refer to the first (..) as \1 or to the second (....) as \2 in the following filter statement:

    mlr filter '$a =~ "(..)_(....)'
    
  • Up to nine matches are supported: \1 through \9, while \0 is the entire match string; \15 is treated as \1 followed by an unrelated 5.

Arithmetic

Input scanning

Numbers in Miller are double-precision float or 64-bit signed integers. Anything scannable as int, e.g 123 or 0xabcd, is treated as an integer; otherwise, input scannable as float (4.56 or 8e9) is treated as float; everything else is a string.

If you want all numbers to be treated as floats, then you may use float() in your filter/put expressions (e.g. replacing $c = $a * $b with $c = float($a) * float($b)) — or, more simply, use mlr filter -F and mlr put -F which forces all numeric input, whether from expression literals or field values, to float. Likewise mlr stats1 -F and mlr step -F force integerable accumulators (such as count) to be done in floating-point.

Conversion by math routines

For most math functions, integers are cast to float on input, and produce float output: e.g. exp(0) = 1.0 rather than 1. The following, however, produce integer output if their inputs are integers: + - * / // % abs ceil floor max min round roundm sgn. As well, stats1 -a min, stats1 -a max, stats1 -a sum, step -a delta, and step -a rsum produce integer output if their inputs are integers.

Conversion by arithmetic operators

The sum, difference, and product of integers is again integer, except for when that would overflow a 64-bit integer at which point Miller converts the result to float.

The short of it is that Miller does this transparently for you so you needn’t think about it.

Implementation details of this, for the interested: integer adds and subtracts overflow by at most one bit so it suffices to check sign-changes. Thus, Miller allows you to add and subtract arbitrary 64-bit signed integers, converting only to float precisely when the result is less than -263 or greater than 263-1. Multiplies, on the other hand, can overflow by a word size and a sign-change technique does not suffice to detect overflow. Instead Miller tests whether the floating-point product exceeds the representable integer range. Now, 64-bit integers have 64-bit precision while IEEE-doubles have only 52-bit mantissas — so, there are 53 bits including implicit leading one. The following experiment explicitly demonstrates the resolution at this range:

   64-bit integer     64-bit integer     Casted to double           Back to 64-bit
       in hex           in decimal                                    integer
0x7ffffffffffff9ff 9223372036854774271 9223372036854773760.000000 0x7ffffffffffff800
0x7ffffffffffffa00 9223372036854774272 9223372036854773760.000000 0x7ffffffffffff800
0x7ffffffffffffbff 9223372036854774783 9223372036854774784.000000 0x7ffffffffffffc00
0x7ffffffffffffc00 9223372036854774784 9223372036854774784.000000 0x7ffffffffffffc00
0x7ffffffffffffdff 9223372036854775295 9223372036854774784.000000 0x7ffffffffffffc00
0x7ffffffffffffe00 9223372036854775296 9223372036854775808.000000 0x8000000000000000
0x7ffffffffffffffe 9223372036854775806 9223372036854775808.000000 0x8000000000000000
0x7fffffffffffffff 9223372036854775807 9223372036854775808.000000 0x8000000000000000

That is, one cannot check an integer product to see if it is precisely greater than 263-1 or less than -263 using either integer arithmetic (it may have already overflowed) or using double-precision (due to granularity). Instead Miller checks for overflow in 64-bit integer multiplication by seeing whether the absolute value of the double-precision product exceeds the largest representable IEEE double less than 263, which we see from the listing above is 9223372036854774784. (An alternative would be to do all integer multiplies using handcrafted multi-word 128-bit arithmetic. This approach is not taken.)

Pythonic division

Division and remainder are pythonic:

  • Quotient of integers is floating-point: 7/2 is 3.5.
  • Integer division is done with //: 7//2 is 3. This rounds toward the negative.
  • Remainders are non-negative.

On-line help

Examples:

$ mlr --help
Usage: mlr [I/O options] {verb} [verb-dependent options ...] {zero or more file names}

Command-line-syntax examples:
  mlr --csv cut -f hostname,uptime mydata.csv
  mlr --tsv --rs lf filter '$status != "down" && $upsec >= 10000' *.tsv
  mlr --nidx put '$sum = $7 < 0.0 ? 3.5 : $7 + 2.1*$8' *.dat
  grep -v '^#' /etc/group | mlr --ifs : --nidx --opprint label group,pass,gid,member then sort -f group
  mlr join -j account_id -f accounts.dat then group-by account_name balances.dat
  mlr --json put '$attr = sub($attr, "([0-9]+)_([0-9]+)_.*", "\1:\2")' data/*.json
  mlr stats1 -a min,mean,max,p10,p50,p90 -f flag,u,v data/*
  mlr stats2 -a linreg-pca -f u,v -g shape data/*
  mlr put -q '@sum[$a][$b] += $x; end {emit @sum, "a", "b"}' data/*
  mlr --from estimates.tbl put '
  for (k,v in $*) {
    if (is_numeric(v) && k =~ "^[t-z].*$") {
      $sum += v; $count += 1
    }
  }
  $mean = $sum / $count # no assignment if count unset'
  mlr --from infile.dat put -f analyze.mlr
  mlr --from infile.dat put 'tee > "./taps/data-".$a."-".$b, $*'
  mlr --from infile.dat put 'tee | "gzip > ./taps/data-".$a."-".$b.".gz", $*'
  mlr --from infile.dat put -q '@v=$*; dump | "jq .[]"'
  mlr --from infile.dat put  '(NR % 1000 == 0) { print > stderr, "Checkpoint ".NR}'

Data-format examples:
  DKVP: delimited key-value pairs (Miller default format)
  +---------------------+
  | apple=1,bat=2,cog=3 | Record 1: "apple" => "1", "bat" => "2", "cog" => "3"
  | dish=7,egg=8,flint  | Record 2: "dish" => "7", "egg" => "8", "3" => "flint"
  +---------------------+

  NIDX: implicitly numerically indexed (Unix-toolkit style)
  +---------------------+
  | the quick brown     | Record 1: "1" => "the", "2" => "quick", "3" => "brown"
  | fox jumped          | Record 2: "1" => "fox", "2" => "jumped"
  +---------------------+

  CSV/CSV-lite: comma-separated values with separate header line
  +---------------------+
  | apple,bat,cog       |
  | 1,2,3               | Record 1: "apple => "1", "bat" => "2", "cog" => "3"
  | 4,5,6               | Record 2: "apple" => "4", "bat" => "5", "cog" => "6"
  +---------------------+

  Tabular JSON: nested objects are supported, although arrays within them are not:
  +---------------------+
  | {                   |
  |  "apple": 1,        | Record 1: "apple" => "1", "bat" => "2", "cog" => "3"
  |  "bat": 2,          |
  |  "cog": 3           |
  | }                   |
  | {                   |
  |   "dish": {         | Record 2: "dish:egg" => "7", "dish:flint" => "8", "garlic" => ""
  |     "egg": 7,       |
  |     "flint": 8      |
  |   },                |
  |   "garlic": ""      |
  | }                   |
  +---------------------+

  PPRINT: pretty-printed tabular
  +---------------------+
  | apple bat cog       |
  | 1     2   3         | Record 1: "apple => "1", "bat" => "2", "cog" => "3"
  | 4     5   6         | Record 2: "apple" => "4", "bat" => "5", "cog" => "6"
  +---------------------+

  XTAB: pretty-printed transposed tabular
  +---------------------+
  | apple 1             | Record 1: "apple" => "1", "bat" => "2", "cog" => "3"
  | bat   2             |
  | cog   3             |
  |                     |
  | dish 7              | Record 2: "dish" => "7", "egg" => "8"
  | egg  8              |
  +---------------------+

  Markdown tabular (supported for output only):
  +-----------------------+
  | | apple | bat | cog | |
  | | ---   | --- | --- | |
  | | 1     | 2   | 3   | | Record 1: "apple => "1", "bat" => "2", "cog" => "3"
  | | 4     | 5   | 6   | | Record 2: "apple" => "4", "bat" => "5", "cog" => "6"
  +-----------------------+

Help options:
  -h or --help                 Show this message.
  --version                    Show the software version.
  {verb name} --help           Show verb-specific help.
  --help-all-verbs             Show help on all verbs.
  -l or --list-all-verbs       List only verb names.
  -L                           List only verb names, one per line.
  -f or --help-all-functions   Show help on all built-in functions.
  -F                           Show a bare listing of built-in functions by name.
  -k or --help-all-keywords    Show help on all keywords.
  -K                           Show a bare listing of keywords by name.

Verbs:
   bar bootstrap cat check count-distinct cut decimate filter fraction grep
   group-by group-like having-fields head histogram join label least-frequent
   merge-fields most-frequent nest nothing put regularize rename reorder repeat
   reshape sample sec2gmt sec2gmtdate seqgen shuffle sort stats1 stats2 step
   tac tail tee top uniq unsparsify

Functions for the filter and put verbs:
   + + - - * / // % ** | ^ & ~ << >> == != =~ !=~ > >= < <= && || ^^ ! ? : .
   gsub strlen sub substr tolower toupper abs acos acosh asin asinh atan atan2
   atanh cbrt ceil cos cosh erf erfc exp expm1 floor invqnorm log log10 log1p
   logifit madd max mexp min mmul msub pow qnorm round roundm sgn sin sinh sqrt
   tan tanh urand urand32 urandint dhms2fsec dhms2sec fsec2dhms fsec2hms
   gmt2sec hms2fsec hms2sec sec2dhms sec2gmt sec2gmt sec2gmtdate sec2hms
   strftime strptime systime is_absent is_bool is_boolean is_empty is_empty_map
   is_float is_int is_map is_nonempty_map is_not_empty is_not_map is_not_null
   is_null is_numeric is_present is_string asserting_absent asserting_bool
   asserting_boolean asserting_empty asserting_empty_map asserting_float
   asserting_int asserting_map asserting_nonempty_map asserting_not_empty
   asserting_not_map asserting_not_null asserting_null asserting_numeric
   asserting_present asserting_string boolean float fmtnum hexfmt int string
   typeof depth haskey joink joinkv joinv leafcount length mapdiff mapexcept
   mapselect mapsum splitkv splitkvx splitnv splitnvx

Please use "mlr --help-function {function name}" for function-specific help.

Data-format options, for input, output, or both:
  --idkvp   --odkvp   --dkvp      Delimited key-value pairs, e.g "a=1,b=2"
                                  (this is Miller's default format).

  --inidx   --onidx   --nidx      Implicitly-integer-indexed fields
                                  (Unix-toolkit style).

  --icsv    --ocsv    --csv       Comma-separated value (or tab-separated
                                  with --fs tab, etc.)

  --itsv    --otsv    --tsv       Keystroke-savers for "--icsv --ifs tab",
                                  "--ocsv --ofs tab", "--csv --fs tab".

  --ipprint --opprint --pprint    Pretty-printed tabular (produces no
                                  output until all input is in).
                      --right     Right-justifies all fields for PPRINT output.
                      --barred    Prints a border around PPRINT output
                                  (only available for output).

            --omd                 Markdown-tabular (only available for output).

  --ixtab   --oxtab   --xtab      Pretty-printed vertical-tabular.
                      --xvright   Right-justifies values for XTAB format.

  --ijson   --ojson   --json      JSON tabular: sequence or list of one-level
                                  maps: {...}{...} or [{...},{...}].
    --json-map-arrays-on-input    JSON arrays are unmillerable. --json-map-arrays-on-input
    --json-skip-arrays-on-input   is the default: arrays are converted to integer-indexed
    --json-fatal-arrays-on-input  maps. The other two options cause them to be skipped, or
                                  to be treated as errors.  Please use the jq tool for full
                                  JSON (pre)processing.
                      --jvstack   Put one key-value pair per line for JSON
                                  output.
                      --jlistwrap Wrap JSON output in outermost [ ].
                    --jknquoteint Do not quote non-string map keys in JSON output.
                     --jvquoteall Quote map values in JSON output, even if they're
                                  numeric.
              --jflatsep {string} Separator for flattening multi-level JSON keys,
                                  e.g. '{"a":{"b":3}}' becomes a:b => 3 for
                                  non-JSON formats. Defaults to :.

  -p is a keystroke-saver for --nidx --fs space --repifs

  Examples: --csv for CSV-formatted input and output; --idkvp --opprint for
  DKVP-formatted input and pretty-printed output.

Format-conversion keystroke-saver options, for input, output, or both:
As keystroke-savers for format-conversion you may use the following:
  --c2t --c2d --c2n --c2j --c2x --c2p --c2m
  --t2c       --t2d --t2n --t2j --t2x --t2p --t2m
  --d2c --d2t       --d2n --d2j --d2x --d2p --d2m
  --n2c --n2t --n2d       --n2j --n2x --n2p --n2m
  --j2c --j2t --j2d --j2n       --j2x --j2p --j2m
  --x2c --x2t --x2d --x2n --x2j       --x2p --x2m
  --p2c --p2t --p2d --p2n --p2j --p2x       --p2m
The letters c t d n j x p m refer to formats CSV, TSV, DKVP, NIDX, JSON, XTAB,
PPRINT, and markdown, respectively. Note that markdown format is available for
output only.

Compressed-data options:
  --prepipe {command} This allows Miller to handle compressed inputs. You can do
  without this for single input files, e.g. "gunzip < myfile.csv.gz | mlr ...".
  However, when multiple input files are present, between-file separations are
  lost; also, the FILENAME variable doesn't iterate. Using --prepipe you can
  specify an action to be taken on each input file. This pre-pipe command must
  be able to read from standard input; it will be invoked with
    {command} < {filename}.
  Examples:
    mlr --prepipe 'gunzip'
    mlr --prepipe 'zcat -cf'
    mlr --prepipe 'xz -cd'
    mlr --prepipe cat
  Note that this feature is quite general and is not limited to decompression
  utilities. You can use it to apply per-file filters of your choice.
  For output compression (or other) utilities, simply pipe the output:
    mlr ... | {your compression command}

Separator options, for input, output, or both:
  --rs     --irs     --ors              Record separators, e.g. 'lf' or '\r\n'
  --fs     --ifs     --ofs  --repifs    Field separators, e.g. comma
  --ps     --ips     --ops              Pair separators, e.g. equals sign

  Notes about line endings:
  * Default line endings (--irs and --ors) are "auto" which means autodetect from
    the input file format, as long as the input file(s) have lines ending in either
    LF (also known as linefeed, '\n', 0x0a, Unix-style) or CRLF (also known as
    carriage-return/linefeed pairs, '\r\n', 0x0d 0x0a, Windows style).
  * If both irs and ors are auto (which is the default) then LF input will lead to LF
    output and CRLF input will lead to CRLF output, regardless of the platform you're
    running on.
  * The line-ending autodetector triggers on the first line ending detected in the input
    stream. E.g. if you specify a CRLF-terminated file on the command line followed by an
    LF-terminated file then autodetected line endings will be CRLF.
  * If you use --ors {something else} with (default or explicitly specified) --irs auto
    then line endings are autodetected on input and set to what you specify on output.
  * If you use --irs {something else} with (default or explicitly specified) --ors auto
    then the output line endings used are LF on Unix/Linux/BSD/MacOSX, and CRLF on Windows.

  Notes about all other separators:
  * IPS/OPS are only used for DKVP and XTAB formats, since only in these formats
    do key-value pairs appear juxtaposed.
  * IRS/ORS are ignored for XTAB format. Nominally IFS and OFS are newlines;
    XTAB records are separated by two or more consecutive IFS/OFS -- i.e.
    a blank line. Everything above about --irs/--ors/--rs auto becomes --ifs/--ofs/--fs
    auto for XTAB format. (XTAB's default IFS/OFS are "auto".)
  * OFS must be single-character for PPRINT format. This is because it is used
    with repetition for alignment; multi-character separators would make
    alignment impossible.
  * OPS may be multi-character for XTAB format, in which case alignment is
    disabled.
  * TSV is simply CSV using tab as field separator ("--fs tab").
  * FS/PS are ignored for markdown format; RS is used.
  * All FS and PS options are ignored for JSON format, since they are not relevant
    to the JSON format.
  * You can specify separators in any of the following ways, shown by example:
    - Type them out, quoting as necessary for shell escapes, e.g.
      "--fs '|' --ips :"
    - C-style escape sequences, e.g. "--rs '\r\n' --fs '\t'".
    - To avoid backslashing, you can use any of the following names:
      cr crcr newline lf lflf crlf crlfcrlf tab space comma pipe slash colon semicolon equals
  * Default separators by format:
      File format  RS       FS       PS
      dkvp         auto     ,        =
      json         auto     (N/A)    (N/A)
      nidx         auto     space    (N/A)
      csv          auto     ,        (N/A)
      csvlite      auto     ,        (N/A)
      markdown     auto     (N/A)    (N/A)
      pprint       auto     space    (N/A)
      xtab         (N/A)    auto     space

Relevant to CSV/CSV-lite input only:
  --implicit-csv-header Use 1,2,3,... as field labels, rather than from line 1
                     of input files. Tip: combine with "label" to recreate
                     missing headers.
  --headerless-csv-output   Print only CSV data lines.

Double-quoting for CSV output:
  --quote-all        Wrap all fields in double quotes
  --quote-none       Do not wrap any fields in double quotes, even if they have
                     OFS or ORS in them
  --quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
                     in them (default)
  --quote-numeric    Wrap fields in double quotes only if they have numbers
                     in them
  --quote-original   Wrap fields in double quotes if and only if they were
                     quoted on input. This isn't sticky for computed fields:
                     e.g. if fields a and b were quoted on input and you do
                     "put '$c = $a . $b'" then field c won't inherit a or b's
                     was-quoted-on-input flag.

Numerical formatting:
  --ofmt {format}    E.g. %.18lf, %.0lf. Please use sprintf-style codes for
                     double-precision. Applies to verbs which compute new
                     values, e.g. put, stats1, stats2. See also the fmtnum
                     function within mlr put (mlr --help-all-functions).
                     Defaults to %lf.

Other options:
  --seed {n} with n of the form 12345678 or 0xcafefeed. For put/filter
                     urand()/urandint()/urand32().
  --nr-progress-mod {m}, with m a positive integer: print filename and record
                     count to stderr every m input records.
  --from {filename}  Use this to specify an input file before the verb(s),
                     rather than after. May be used more than once. Example:
                     "mlr --from a.dat --from b.dat cat" is the same as
                     "mlr cat a.dat b.dat".
  -n                 Process no input files, nor standard input either. Useful
                     for mlr put with begin/end statements only. (Same as --from
                     /dev/null.) Also useful in "mlr -n put -v '...'" for
                     analyzing abstract syntax trees (if that's your thing).
  -I                 Process files in-place. For each file name on the command
                     line, output is written to a temp file in the same
                     directory, which is then renamed over the original. Each
                     file is processed in isolation: if the output format is
                     CSV, CSV headers will be present in each output file;
                     statistics are only over each file's own records; and so on.

Then-chaining:
Output of one verb may be chained as input to another using "then", e.g.
  mlr stats1 -a min,mean,max -f flag,u,v -g color then sort -f color

For more information please see http://johnkerl.org/miller/doc and/or
http://github.com/johnkerl/miller. This is Miller version v5.2.0.

$ mlr sort --help
Usage: mlr sort {flags}
Flags:
  -f  {comma-separated field names}  Lexical ascending
  -n  {comma-separated field names}  Numerical ascending; nulls sort last
  -nf {comma-separated field names}  Numerical ascending; nulls sort last
  -r  {comma-separated field names}  Lexical descending
  -nr {comma-separated field names}  Numerical descending; nulls sort first
Sorts records primarily by the first specified field, secondarily by the second
field, and so on.  (Any records not having all specified sort keys will appear
at the end of the output, in the order they were encountered, regardless of the
specified sort order.) The sort is stable: records that compare equal will sort
in the order they were encountered in the input record stream.

Example:
  mlr sort -f a,b -nr x,y,z
which is the same as:
  mlr sort -f a -f b -nr x -nr y -nr z