mk

Text preprocessor for content segmentation.

Installation

Via go install

With go 1.18 or higher:

go install github.com/x0k/mk@latest

Examples

Basic

mkfile file

TARGET="mk"
build:
  go build -o ./bin/${TARGET}

clean:
  go clean
  rm -f ./bin/${TARGET}

mk build output

TARGET="mk"
go build -o ./bin/${TARGET}

mk clean output

TARGET="mk"
go clean
rm -f ./bin/${TARGET}

X suffix

If file contains x suffix, then result of preprocessing will be saved as tmp file and executed.

mkfilex file

#!/usr/bin/bash
echo "zero"
one:
  echo "one"
echo $1
two:
  echo "two"

mk one output

zero
one

mk two my-one output

zero
my-one
two

Targets

Each segment can define its targets.

If the segment has a target that is a prefix of current target segment, it will be included in the result.

mkfilex file

#!/usr/bin/bash
prepare: test build
  echo "prepare"
build:
  echo "build"
test:
  echo "test"
clean:
  echo "clean"

mk build output

prepare
build

mk clean output

clean

The equivalent mkfilex

#!/usr/bin/bash
clean:
  echo "clean"
echo "prepare"
build:
  echo "build"
test:
  echo "test"

Default target

By default target segment is default and all segments are implicitly declare default as their target. So $ mk command will include all segments in the result.

If you want to exclude some segment from $ mk result, then:

• define segment with target (any non-empty string/strings not beginning with default)
• define segment after default segment.

mkfilex file

#!/usr/bin/bash
build:
  echo "build"
test: build
  echo "test"
default:
clean:
  echo "clean"

mk output

build

Details

• The segment defined by a label (and optional targets) that satisfies this regular expression ^([A-z][0-9A-z\._-]*):(.*)$ and by the presence of equal indentation ^([ \t]+) on the subsequent lines.

  By default segment is not defined.

• Segment targets are a whitespace-separated list of segment prefixes that follow the : character.

• The end of a segment is determined by indentation changes or the end of the file.

• If segment is not defined for a line, the line will be added to each segment defined below.

• The filename must match the regular expression ^(M|m)kfile.*$, all matching files are processed in reverse lexicographic order until the target segment is found.