README.md

iterable-search

A lib providing interfaces and tools for searching through iterations of tabular data.

This was created as a code challenge and also to explore ideas for some tools I was thinking of.

This was created for a code challenge and to explore ideas for some tools I was thinking of. It became useful a few times but it is incomplete and at version <= 1, so APIs may still change.

Ideas for future improvement (also see TODO).

• Simpler CLI API around the APIs here (with possible simplifications).
• Create a command so it's not required of the user of this lib to create.

Once you've set things up (which is quick), you can use expressions from symfomy/expression-language via CLI.

At the moment only csv files are supported for input but it's easy enough to add handlers.

Install

composer require johnpancoast/iterable-parser:~0

Usage

All examples here can be seen in docs/example/.

The first things you should do when using data-processor is to create your CLI command that loads this lib's commands. You only need to do this once. This is meant to be improved upon so you don't have to create a command and can just include one from the lib.

bin/your-cli-command

#!/usr/bin/env php
<?php

require __DIR__ . '/../../vendor/autoload.php';

use Pancoast\DataProcessor\ConsoleHelper;

ConsoleHelper::createAndLoad()->run();

Now let's imagine the following csv at ./animals.csv.

id, title, sound
1,dog,bark
2,cat,meow
3,lion,meow
4,dinosaur,rawr

You first need to define a class that represents an iteration of this data.

<?php

namespace YourNamespace;

class Animal
{
    public $id;
    public $title;
    public $sound;
}

Note that the properties are public. They don't have to be but this provides flexibility later.

Now let's add jms/serializer annotations so the data can be (de)serialized and validated.

<?php

namespace YourNamespace;

use JMS\Serializer\Annotation as JMS;

/**
 * @JMS\ExclusionPolicy("all")
 */
class Animal
{
    /**
     * @var int
     *
     * @JMS\Expose
     * @JMS\Type("int")
     */
    public $id;

    /**
     * @var string
     * 
     * @JMS\Expose
     * @JMS\Type("string")
     */
    public $title;

    /**
     * @var string
     *
     * @JMS\Expose
     * @JMS\Type("string")
     */
    public $sound;
}

We can now use your CLI command to filter data using expressions.

Show animals with id greater than 1 (see --expr option)

./bin/your-cli-command csv \
--file="./animals.csv" \
--data-class="YourNamespace\\Animal" \
--expr-root="pet" \
--expr="pet.id > 1"

Show pets having title with length greater than 3 and which meow

./bin/your-cli-command csv \
--file="./animals.csv" \
--data-class="YourNamespace\\Animal" \
--expr-root="pet" \
--expr="len(pet.title) > 3 && pet.sound == 'meow'"

Use a config file. Configs are yaml files and can contain any of the command options. CLI options override those in configs.

# assume config file at config.yaml
$ cat config.yaml
file="./animals.csv"'
data-class="YourNamespace\\Animal"
expr-root="pet"

./bin/your-cli-command csv -c "config.yaml" -e "len(pet.title) && pet.sound == 'meow'"

You can provide an output for an expression which will direct iteration there instead of STDOUT when expression is true.

# assume config file at config.yaml
$ cat config.yaml
file="./animals.csv"'
data-class="YourNamespace\\Animal"
expr-root="pet"

./bin/your-cli-command csv -c "config.yaml" -o ./output-expr1.csv -e "len(pet.title) && pet.sound == 'meow'"

Both --expr and --out are arrays and each are run independently of the other. Meaning, you can create any number of expression and output pairings that each run on their own.

TODO

• Simpler CLI API around the APIs here (possible simplifications).
• Create a command so the user (you) doesn't have to.
• Tests
• A new --group array option for each --expr so that output can be grouped.
• More commands to input more formats (internally, all formats that the serializer supports are supported... it's just our command that doesn't support more yet).