Mrcrowley NPM | npm.io

MrCrowley

Retrieve data from different websites using html elements to gather the information you need.

Installation

Install node

npm install -g mrcrowley
mrcrowley --config="/home/user/.crawl.json" --save="/home/user/crawlResults.json"

Usage

Core usage

I still have to document how you can require and use the core directly but just so that you know, you can do it and the results are based on promises.

CLI

Set a .crawl.json and run all the tasks you want when you pass it to mrcrowley.

Note: Any kind of path should be absolute or relative to the place the script is called.

mrcrowley --config=<config_json_src> --output=<file_to_save_src> --force=<false|true>

Notes:

<config_json_src>: Path to the config json for crawling. It is required
<file_to_save_src>: Path for the file you want to have the results. For now, only json is supported. It is required
force: It forces to create a new output. If false and the output file exists, it will just update. It will default to false

Configuration

{
    "projectId": "<project_id>",
    "projectName": "<project_name>",
    "data": [{
        "src": "<url_path>",
        "name": "<request_name>",
        "throttle": 2000,
        "enableJs": false,
        "waitFor": "<html_selector>",
        "wait": {
            "selector": "<html_selector>",
            "for": 5000
        },
        "modifiers": {
            "<query_var_in_url>": ["<var_to_replace>"]
        },
        "retrieve": {
            "<name>": {
                "selector": "<html_selector>",
                "attribute": "<attribute_to_retrieve>",
                "ignore": ["<regex_pattern_to_ignore>"]
            }
        }
    }]
}

Notes:

retrieve: Besides the simplified version, you may also nest it to get contained data

{
    "src": "...",
    "retrieve": {
        "<name>": {
            "selector": "<parent_html_selector>",
            "retrieve": {
                "<name>": {
                    "selector": "<child_html_selector>",
                    "attribute": "<attribute_to_retrieve>",
                    "ignore": ["<regex_pattern_to_ignore>"]
                }
            }
        }
    }
}

attribute: If not provided, text content will be returned. Optional key.
ignore: Ignore results with a regex pattern. Optional key.
enableJs: Javascript isn't enable by default for security reasons. Use this if you really need it
wait: Usually used with enableJs. If the sources uses javascript to render, you may wait for the selector to be present

<var_to_replace>: It can also be an object with keys min (it will default to 0) and max (it will default to 10)

{
    "src": "...",
    "modifiers": {
        "<query_var_in_url>": ["<var_to_replace>"],
        "<limit_var_in_url>": [{
            "min": 0,
            "max": 10
        }]
    },
    "retrieve": {}
}