Download the PHP package hashbangcode/sitemap_checker without Composer

Sitemap Checker

A PHP library used to download, parse and crawl sitemap.xml files.

If the sitemap.xml file is gzipped then the file is unzipped and the contents read.

Installation

Download the PHP files to a directory and run composer install. This will set up everything needed for the application to run.

Usage

To run the application on the command line use the following.

php application.php sitemap-checker:run https://www.example.com/sitemap.xml

This will download the sitemap.xml file, let you know how many URLs it detected before starting to crawl them.

You can also use a shorthand to run the same command:

php application.php sc:run https://www.example.com/sitemap.xml

Once the crawling has finished the command will print out the results.

Options

A number of options exist for this tool.

Result Output

To change what type of result is returned you can supply the --result-file option, or -r for short.

The following rendering types are available:

• Plain, the default renderer (prints the results line by line).
• CSV
• JSON
• XML
• HTML

For example, to output the results as a csv you can supply the option like this.

php application.php sc:run -r results.csv https://www.example.com/sitemap.xml

This will automatically trigger the csv result rendering pathway and render the results as a csv file.

Without this option in place the tool will print results to the command line using "plain" rendering.

Limit

The --limit option (or -l for short) simply prevents the tool from checking any more than this limit.

For example, this will only process 10 results, regardless of the number of URLs found.

php application.php sc:run -l 10 https://www.example.com/sitemap.xml

Engine

The --engine option (or -e for short) changes the type of checking engine used.

Options are:

• 'guzzle' : (Default) Run the sitemap checker using Guzzle promises.
• 'chrome' : Run the sitemap checker using headless Chrome. To get this running you'll first need to add the chrome binary to the location ./chrome/chrome (i.e. within the package).

For example, to change the sitemap checker engine to use headless Chrome use the following.

php application.php sc:run -e chrome https://www.example.com/sitemap.xml

Exclude

Pass a list of URLs to exclude using the --exclude (or -x for short) flag. This will prevent URLs from being added to the collections and checked. This can be a comma separated list of URLs to exclude. Wildcards can also be used to prevent certain inner URLs from being used.

Some examples:

To prevent the path https://www.example.com/some-page being used.

php application.php sc:run https://www.example.com/ --exclude='https://www.hashbangcode.com/some-page.html'

To prevent anything in https://www.example.com/sub-dir1 and https://www.example.com/sub-dir2 from being used:

php application.php sc:run https://www.example.com/ --exclude='https://www.example.com/sub-dir1/*,https://www.hashbangcode.com/sub-dir2/*'

To prevent anything on the external site https://www.example2.org being used.

php application.php sc:run https://www.example.com/ --exclude='https://www.example2.org/*'

Robots

Pass a robots.txt file to the script using --robots (or -t for short). This will download and parse the robots.txt file into a set of exclusion rules.

php application.php sc:run https://www.example.com/ --robots='https://www.example.org/robots.txt'

Note that the robots.txt parser will only use rules that have been created for the User-agent string *.

This option can be used in conjunction with the exclude flag to add to the spider exclusion rules.

Testing

Run ./vendor/bin/phpunit to run the phpunit tests. All web requests are mocked within the unit tests.

Run composer run test-coverage to run the unit tests and produce a code coverage report. This report is added to the directory .build in the root of the application.

For the coverage report you need to add the following to your xdebug.ini configuration file.

Example Using Classes

To extract the classes out of this project to use independently do the following.

To Do

There's still lots to do.

• Add a way to auto-download the Chrome binary.
• Look into using a database or message queue.
• Add Docker container to wrap application.
• Batching processing of urls (i.e. don't do everything in one go).
• Pick better name for application.
• Add ability to add session cookies for authenticated spidering.