A parser for email replies. Strip all quotes and signatures and just get reply.

README.md

Requirements

composer require league/html-to-markdown

How to use

Use this line in mail for parsing;

<!-- Target: <p class="data-remove-below">...</p> -->
<p class="data-remove-below">##- Please write your answer above this line. -##</p>

EmailContentService.php

<?php

namespace App\Services;

use DOMDocument;
use DOMElement;
use DOMXPath;
use Illuminate\Mail\Markdown;
use Illuminate\Support\HtmlString;
use League\HTMLToMarkdown\HtmlConverter;

class EmailContentService
{
    private $quoteHeadersRegex = [
        '/^(.+\s<.+>,\s.*\syazdı:)$/m',
        '/^(.+\s<.+>\s.*\syazdı\s[(].+[)]:)$/m',
        '/^(.+,\s([0-1]?[0-9]|2[0-3]):[0-5][0-9],\s".+"\s<.+>:)$/m',
        '/^\s*(On(?:(?!^>*\s*On\b|\bwrote:).){0,1000}wrote:)$/ms', // On DATE, NAME <EMAIL> wrote:
        '/^\s*(Le(?:(?!^>*\s*Le\b|\bécrit:).){0,1000}écrit(\s|\xc2\xa0):)$/ms', // Le DATE, NAME <EMAIL> a écrit :
        '/^\s*(El(?:(?!^>*\s*El\b|\bescribió:).){0,1000}escribió:)$/ms', // El DATE, NAME <EMAIL> escribió:
        '/^\s*(Il(?:(?!^>*\s*Il\b|\bscritto:).){0,1000}scritto:)$/ms', // Il DATE, NAME <EMAIL> ha scritto:
        '/^[\S\s]+ (написа(л|ла|в)+)+:$/msu', // Everything before написал: not ending on wrote:
        '/^\s*(Op\s.+?(schreef|geschreven).+:)$/ms', // Op DATE schreef NAME <EMAIL>:, Op DATE heeft NAME <EMAIL> het volgende geschreven:
        '/^\s*((W\sdniu|Dnia)\s.+?(pisze|napisał(\(a\))?):)$/msu', // W dniu DATE, NAME <EMAIL> pisze|napisał:
        '/^\s*(Den\s.+\sskrev\s.+:)$/m', // Den DATE skrev NAME <EMAIL>:
        '/^\s*(Am\s.+\sum\s.+\sschrieb\s.+:)$/m', // Am DATE um TIME schrieb NAME:
        '/^(在.+写道：)$/ms', // > 在 DATE, TIME, NAME 写道：
        '/^(20[0-9]{2}\..+\s작성:)$/m', // DATE TIME NAME 작성:
        '/^(20[0-9]{2}\/.+のメッセージ:)$/m', // DATE TIME、NAME のメッセージ:
        '/^(.+\s<.+>\sschrieb:)$/m', // NAME <EMAIL> schrieb:
        '/^\s*(From\s?:.+\s?(\[|<).+(\]|>))/mu', // "From: NAME <EMAIL>" OR "From : NAME <EMAIL>" OR "From : NAME<EMAIL>"(With support whitespace before start and before <)
        '/^\s*(发件人\s?:.+\s?(\[|<).+(\]|>))/mu', // "发件人: NAME <EMAIL>" OR "发件人 : NAME <EMAIL>" OR "发件人 : NAME<EMAIL>"(With support whitespace before start and before <)
        '/^\s*(De\s?:.+\s?(\[|<).+(\]|>))/mu', // "De: NAME <EMAIL>" OR "De : NAME <EMAIL>" OR "De : NAME<EMAIL>"  (With support whitespace before start and before <)
        '/^\s*(Van\s?:.+\s?(\[|<).+(\]|>))/mu', // "Van: NAME <EMAIL>" OR "Van : NAME <EMAIL>" OR "Van : NAME<EMAIL>"  (With support whitespace before start and before <)
        '/^\s*(Da\s?:.+\s?(\[|<).+(\]|>))/mu', // "Da: NAME <EMAIL>" OR "Da : NAME <EMAIL>" OR "Da : NAME<EMAIL>"  (With support whitespace before start and before <)
        '/^(20[0-9]{2}\-(?:0?[1-9]|1[012])\-(?:0?[0-9]|[1-2][0-9]|3[01]|[1-9])\s[0-2]?[0-9]:\d{2}\s.+?:)$/ms', // 20YY-MM-DD HH:II GMT+01:00 NAME <EMAIL>:
        '/^\s*([a-z]{3,4}\.\s.+\sskrev\s.+:)$/ms', // DATE skrev NAME <EMAIL>:
    ];

    protected $content;

    protected $document;

    protected $xpath;

    public function init($content)
    {
        $this->content = $this->removeUnallowedTags($content);

        $this->document = new DOMDocument();
        $this->document->loadHTML('<?xml encoding="utf-8" ?>' . $this->content);

        $this->xpath = new DOMXPath($this->document);
    }

    public function format($content): HtmlString
    {
        $this->init($content);

        $this->removeBelow();
        $this->removeQuotes();
        $this->formatForGmail();
        $this->formatForOutlook();
        $this->formatForYandex();

        $body = $this->convertToMarkdown($this->body());

        return Markdown::parse($body);
    }

    private function formatForGmail(): void
    {
        $quote = $this->xpath->query('//div[@class="gmail_quote"]')->item(0);

        if (is_null($quote)) {
            return;
        }

        $quote->parentNode->removeChild($quote);
    }

    private function formatForYandex(): void
    {
        $allDivs = $this->xpath->query('//div');

        foreach ($allDivs as $div) {
            foreach ($this->quoteHeadersRegex as $regex) {
                preg_match($regex, $div->textContent, $matched);

                if (isset($matched[1])) {
                    $div->parentNode->removeChild($div);
                }
            }
        }
    }

    private function formatForOutlook(): void
    {
        $quotes = $this->xpath->query("//div[substring(@id, string-length(@id) - 12) = 'divRplyFwdMsg']");

        foreach ($quotes as $quote) {
            $this->removePrevSibling($quote, 'hr', 3);
            $this->removeAllNextSiblings($quote);

            try {
                if ($quote->parentNode) {
                    $quote->parentNode->removeChild($quote);
                }
            } catch (\ErrorException $e) {
                //
            }
        }
    }

    private function removeBelow(): void
    {
        $belowText = $this->xpath
            ->query("//p[substring(@class, string-length(@class) - 16) = 'data-remove-below']")
            ->item(0);

        $this->removeAllNextSiblings($belowText);
    }

    private function removeQuotes(): void
    {
        $quotes = $this->xpath->query('//blockquote');

        foreach ($quotes as $quote) {
            $quote->parentNode->removeChild($quote);
        }
    }

    private function removeUnallowedTags($content): ?string
    {
        $content = preg_replace('/<meta http-equiv="Content-Type".+?>/', '', $content);
        $content = preg_replace('/<style.*?>.*?<\/style>/ms', '', $content);

        return $content;
    }

    private function convertToMarkdown($content): ?string
    {
        $converter = new HtmlConverter(['strip_tags' => true]);
        $markdown = $converter->convert($content);

        return $this->removeExtraLineBreaks($markdown);
    }

    private function removeExtraLineBreaks($markdown): ?string
    {
        return preg_replace('/(^[\r\n]*|[\r\n]+)[\s\t]*[\r\n]+/', "\n\r", $markdown);
    }

    private function body(): ?string
    {
        $body = $this->document->getElementsByTagName('body')->item(0);

        return preg_replace(['/<body.+?>/', '/<\/body>/'], '', $this->document->saveHTML($body));
    }

    private function removePrevSibling(?DOMElement $element, string $targetNodeName = null, int $try = 100): void
    {
        try {
            $loop = 0;
            $prev = $element->previousSibling;

            while (true) {
                if ($loop >= $try || is_null($prev)) {
                    break;
                }

                if (is_null($targetNodeName) || $prev->nodeName === $targetNodeName) {
                    $prev->parentNode->removeChild($prev);
                    break;
                }

                $prev = $prev->previousSibling;
                $try++;
            }
        } catch (\ErrorException $e) {
            //
        }
    }

    private function removeAllNextSiblings(?DOMElement $element): void
    {
        try {
            while (true) {
                if (is_null($element)) {
                    break;
                }

                $nextSibling = $element->nextSibling;
                $element->parentNode->removeChild($element);

                if (!$nextSibling) {
                    break;
                }

                $element = $nextSibling;
            }
        } catch (\ErrorException $e) {
            //
        }
    }
}