First commit

Author: hgouchet

Dockerfile

@@ -0,0 +1,10 @@
+FROM bash:4.4
+
+RUN mkdir -p /opt/public-suffix
+WORKDIR /opt/public-suffix
+
+ADD . /opt/public-suffix
+
+RUN apk add --no-cache curl sed mysql-client
+
+RUN curl -s "https://raw.githubusercontent.com/publicsuffix/list/master/public_suffix_list.dat" | sed -e '/^\/\//d' -e '/^\s*$/d' > registry.dat

README.md

@@ -0,0 +1,60 @@
+# Public Suffix for MySql
+
+Mysql's functions to parse domain names based on the Public Suffix List.
+
+
+## Installation
+
+```bash
+$ git clone https://github.com/rvflash/publicsuffix-sql
+```
+
+
+## Testing
+
+Go into the source directory and starts the Docker.
+
+```bash
+$ cd publicsuffix-sql
+$ docker-compose up
+$ docker exec -ti $(docker ps -lq) bash
+```
+
+In the new prompt, opens the MySql client: 
+
+```bash
+$ mysql -h db -u root -ps3cr3tp4ssw0rd ps
+```
+
+
+## Features
+
+* The function named `hostname` extracts the hostname in a URL.
+* Function named `domain` returns the hostname without the sub-domain (ex: google.com.au).
+* The `sld` and `tld` functions returns respectively the second level domain (ex: com.au) and top level domain (ex: au). 
+
+```
+> select hostname("https://www.google.com.au/search?q=golang");
++-------------------------------------------------------+
+| hostname("https://www.google.com.au/search?q=golang") |
++-------------------------------------------------------+
+| www.google.com.au                                     |
++-------------------------------------------------------+
+1 row in set (0.00 sec)
+
+> select domain("https://www.google.com.au/");
++--------------------------------------+
+| domain("https://www.google.com.au/") |
++--------------------------------------+
+| google.com.au                        |
++--------------------------------------+
+1 row in set (0.00 sec)
+
+> select sld(domain("https://www.google.com.au/"));
++-------------------------------------------+
+| sld(domain("https://www.google.com.au/")) |
++-------------------------------------------+
+| com.au                                    |
++-------------------------------------------+
+1 row in set (0.00 sec)
+```

add-functions.sh

@@ -0,0 +1,3 @@
+#!/usr/bin/env bash
+
+mysql -h db -u root -ps3cr3tp4ssw0rd ps < functions.sql

docker-compose.yml

@@ -0,0 +1,18 @@
+version: '3'
+services:
+  db:
+    image: mariadb
+    environment:
+      MYSQL_ROOT_PASSWORD: s3cr3tp4ssw0rd
+      MYSQL_DATABASE: ps
+    ports:
+      - 3306:3306
+  app:
+    depends_on:
+      - db
+    build: .
+    command: ["./wait-for-it.sh", "db:3306", "--", "./add-functions.sh"]
+    volumes:
+      - .:/opt/public-suffix
+    links:
+      - db

functions.sql

@@ -0,0 +1,75 @@
+-- Creates the workspace.
+CREATE TABLE IF NOT EXISTS public_suffix (
+  suffix varchar(255) not null,
+  primary key (suffix)
+) engine=InnoDB default charset=utf8 collate=utf8_general_ci;
+
+-- Loads the public suffix list.
+LOAD DATA LOCAL INFILE '/opt/public-suffix/registry.dat'
+IGNORE INTO TABLE public_suffix
+LINES TERMINATED BY '\n';
+
+DELIMITER $$
+
+-- Extracts the hostname, like `www.rv.net`
+-- Sample: http://username:password@hostname:9090/path?arg=value#anchor
+DROP FUNCTION IF EXISTS HOSTNAME;
+CREATE FUNCTION HOSTNAME (uri VARCHAR(255) CHARSET utf8)
+  RETURNS VARCHAR(255) CHARSET utf8 NO SQL
+  BEGIN
+    DECLARE host VARCHAR(255) CHARSET utf8;
+    -- removes the scheme
+    SET host = SUBSTRING_INDEX(uri, '://', -1);
+    -- removes all chars before [at]
+    SET host = SUBSTRING_INDEX(host, '@', -1);
+    -- removes all before the protocol
+    SET host = SUBSTRING_INDEX(host, ':', 1);
+    -- removes the path
+    SET host = SUBSTRING_INDEX(host, '/', 1);
+    RETURN host;
+  END $$
+
+-- Returns only the second level domain, like `co.uk`
+DROP FUNCTION IF EXISTS SLD;
+CREATE FUNCTION SLD (host VARCHAR(255) CHARSET utf8)
+  RETURNS VARCHAR(50) CHARSET utf8 READS SQL DATA
+  BEGIN
+    DECLARE sld VARCHAR(50) CHARSET utf8;
+    SELECT suffix INTO sld
+      FROM public_suffix
+      WHERE suffix = SUBSTRING_INDEX(host, '.',  -2);
+    RETURN sld;
+  END $$
+
+-- Returns only the top level domain, like `fr`
+DROP FUNCTION IF EXISTS TLD;
+CREATE FUNCTION TLD (host VARCHAR(255) CHARSET utf8)
+  RETURNS VARCHAR(5) CHARSET utf8 READS SQL DATA
+  BEGIN
+    DECLARE tld VARCHAR(5) CHARSET utf8;
+    SELECT suffix INTO tld
+      FROM public_suffix
+      WHERE suffix = SUBSTRING_INDEX(host, '.',  -1);
+    RETURN tld;
+  END $$
+
+-- Returns the root domain like `rv.net` for `www.rv.net`
+DROP FUNCTION IF EXISTS DOMAIN;
+CREATE FUNCTION DOMAIN (uri VARCHAR(255) CHARSET utf8)
+  RETURNS VARCHAR(255) CHARSET utf8 READS SQL DATA
+  BEGIN
+    DECLARE root, host VARCHAR(255) CHARSET utf8;
+    SET host = HOSTNAME(uri);
+    SET root = TLD(host);
+    IF root != '' THEN
+      SET root = SLD(host);
+      IF root != '' THEN
+        SET root = SUBSTRING_INDEX(host, '.',  -3);
+      ELSE
+        SET root = SUBSTRING_INDEX(host, '.',  -2);
+      END IF;
+    END IF;
+    RETURN root;
+  END $$
+
+DELIMITER ;