fgroup

Fgroup is a java utility to seek and capture for groups of files using GLOB-like patterns.

✨ Features

Feature	Implemented
Current point location	Yes 👍
Parent starting point	Yes 👍
Root starting point	Yes 👍
Tree starting point	Yes 👍
Allow 2-spaces indent	Not yet 👎
Capture patterns on different indents	Not yet 👎

Install

Prerequisites

Java 11 or newer.

Using Maven

<dependency>
    <groupId>io.peasoup</groupId>
    <artifactId>fgroup</artifactId>
    <version>0.1</version>
</dependency>

Using the CLI (available)

fgroup, version:  0.1.

Usage:
    fgroup [-d <dir>] <config>

Options:
    -d, --directory Root directory. 
                    By default it uses the current 
                    process directory

Parameters:
    <config>        Configuration file wherein instructions are
                    given to fgroup.
    <dir>           Specific root directory.
                    It will override the default value.

How it works

Fgroup uses a file text as its configurations and then tries to find file matching them.

The configuration text file work like this:

Each line implies a new file pattern using GLOB-like syntax;
Each indent means a subdirectory. An indent is a \t character OR 4 spaces.
Only starting points are allowed at the start of a line; IMPORTANT starting point token can be written at the first indent.
There are 4 starting points:
- Current starting point, token is "./". It starts at the current directory where *fgroup is being used at. Equivalent to System.getProperty("user.dir");
- Parent starting point, token is "../". It starts from the parent directory of the current one;
- Root starting point, token is "^/". It starts at the root of the current drive;
- Tree starting point, token is "*/". From the current directory, it will use all its parent directory, including the root.
To capture matching files, use a capture pattern with the following syntax: (PATTERN as 'PATTERNNAME');
You can have has many capture pattern as you want;
IMPORTANT: All capture patterns has to be on the same indent, on individual lines.
IMPORTANT 2: You can't capture from multiple indents (yet). Each capture patterns has to happen on the same indent for the same capture group.
A capture group includes only a single starting point with multiple files and captures patterns.
You can have has many capture groups as you want;
Commenting is available using the # character.

Here is an example (from INV):

# Capture Groovy class files from the /src directory.
./
    src
        (*.groovy as 'groovyFiles')

# Capture Groovy test (Junit) files from the /test directory.
./
    test
        (*.groovy as 'groovyTestFiles')

# Capture different flavors of INV files.
./
    vars
        (*.groovy as 'invFiles')
        (*.yml as 'invFiles')
        (*.yaml as 'invFiles')

# Capture different flavors of REPO files. Expect only one.
./
    (repo.groovy as 'repoFile')
    (repo.yml as 'repoFile')
    (repo.yaml as 'repoFile')

There are 4 capturing groups;

Each capture group uses the current starting point using the ./ token;

Since src and test are two different directories, even if the capture pattern is similar, they have to be defined in two separate capture group.

./, and vars/ are using multiple capture patterns with the same pattern name.

Here is how you would the configuration above (used as fgroup.txt):

import io.peasoup.fgroup.FileMatches;
import io.peasoup.fgroup.FileSeeker;

Path configFile = Path.of(tmpFolder, "fgroup.txt");
FileSeeker fileSeeker = new FileSeeker(configFile.toString());

for (FileMatches.FileMatchRecord match : fileMatches.get("groovyFiles")) {
    System.out.println("Here is one of my Groovy files: " + match.getCurrent());
}

peasoupio / fgroup