Open-source live-chat, email support, omni-channel desk. An alternative to Intercom, Zendesk, Salesforce Service Cloud etc. 🔥💬
Ruby actioncablechat-widgetconversationcustomer-supportdashboarddesigndockerdocker-imageherokuintercomjavascriptlivechatopensourcerailsrubysassvuejswhatsappzendesk
module MailboxInlineAttachmentHelper
private
def load_email_content
@html_content = processed_mail.serialized_data[:html_content][:full]
@text_content = processed_mail.serialized_data[:text_content][:reply]
end
def inline_attachment?(attachment)
# Only process images as potential inline attachments
return false unless mail_content.present? && attachment[:original].content_type.to_s.start_with?('image/')
# Without an HTML body there is nothing that could reference the image, so
# keep the historical behavior and leave explicitly inline-marked images inline.
return attachment[:original].inline? if @html_content.blank?
cid = attachment[:original].cid
return false if cid.blank?
# With an HTML body present, treat the image as inline only when that body
# references its CID; inline-marked images the body never references fall
# through to become regular attachments.
body_references_cid?(cid)
end
def body_references_cid?(cid)
# Check if CID is referenced in HTML content
return false if @html_content.blank?
cid_url_patterns_for(cid).any? { |cid_url_pattern| @html_content.match?(cid_url_pattern) }
end
def upload_inline_image(mail_attachment)
content_id = mail_attachment[:original].cid
image_url = inline_image_url(mail_attachment[:blob]).to_s
cid_url_patterns_for(content_id).each do |cid_url_pattern|
@html_content = @html_content.gsub(cid_url_pattern, image_url)
end
end
def cid_url_patterns_for(cid)
# RFC 2392 defines cid URLs for referencing MIME body parts and allows URL-encoded Content-ID values:
# https://www.rfc-editor.org/rfc/rfc2392.html#section-2
# URI schemes are case-insensitive per RFC 3986, but the Content-ID remains case-sensitive:
# https://www.rfc-editor.org/rfc/rfc3986.html#section-3.1
encoded_cid = ERB::Util.url_encode(cid)
lowercase_encoded_cid = encoded_cid.gsub(/%[0-9A-F]{2}/, &:downcase)
[cid, encoded_cid, lowercase_encoded_cid].uniq.map do |cid_value|
/(?i:cid):#{Regexp.escape(cid_value)}/
end
end
end