Features

• Matche scene release and episode metadata using YQL Open Data Tables.
• Unrar release.
• Create screens (contact sheet) of media file.
• Upload images to jerking image host.
• Create mediainfo report.
• Add and clean tags based on custom regexes.
• Create torrent file.
• Support for different templates for different sites by site name convention.
• Support for custom site handling (create gif cover, add photo set etc).
• Upload description as a private template with title, tags and cover (default) or direct upload.

Prerequisites

• Linux environment with bash v4+, xmllint and mediainfo.
• Python 3: https://www.python.org/downloads/ (Needed for image upload and torrent maker).
• Python packages beautifulsoup4, html5lib and requests, see scripts/jerking_uploader.py for more info. Make sure you can successfully upload a local image using this script on beforehand.

Installation

• Download and extract, make sure folder structure is preserved.
• Open config/settings.cfg and review the variables needed and their meaning. Make sure you have a working cookie file to the tracker.
• Make sure scripts/maketemplate.sh and scripts/jerking_uploader.py are executable: chmod u+x <file>. Test that you have the installed Python dependencies by running jerking_uploader -help. No errors should be displayed.
• You need a torrent create tool, the current implementation assumes you are using this tool : https://bitbucket.org/rsnitsch/py3createtorrent/downloads/. To use something different, make the necessary changes in createtorrent.sh.
• To create screens you need Movie thumbnailer: http://moviethumbnail.sourceforge.net/ You will also need tahomabd.ttf font file, find it on Google. Put this font file in your mtn folder. (If you need 4K support use this instead https://github.com/thetmk/mtn/releases/tag/0.1 ) To use a total different tool, make the necessary changes in createscreens.sh

Usage

Syntax: ./maketemplate.sh <path to release folder> [url to episode / scene] Example: ./maketemplate.sh ~/downloads/SomeAdultPaysite.18.01.01.Hot.Performer.XXX.1080p.MP4-GRP If core variables are misconfigured, the script should display an error message about the problem and exit. If second paramter url is supplied, any episode matching is skipped and the url will be used directly.

Logging

A log file will be produced in your configured temp folder with name <network>_<sitename>.log or <sitename>.log

Tagging

Cast will be automatically added as tags, as well as sitename domain (<sitename>.com) and if a network/parent site is associated with the site, the network domain (<network>.com) will be added also. You can add static tags for specific sites in tags_sites.txt, each site on a new line. Syntax: <sitename>=tag1 tag2 tag3 etc. Each tag separated by space.

The same goes for actors/performers. You can add static tags for specific sites in tags_actors.txt, each site on a new line. Syntax: actor.name=tag1 tag2 tag3 etc. Each tag separated by space.

Resolution height is also added automatically as <pixel height>p You can also add tags for the following properties:

• Media file extension by setting variable addExtensionAsTag = 1.
• Date by setting addDateAsTag = 1, see settings.cfg for more info about date format.
• FPS by setting addFpsAsTag = 1. This switch is also dependant on values in variable fpsList. By default, only 60fps will be used. When media file FPS is extracted from the mediainfo report, this number is rounded to nearest integer.

Tags will then be split into an array and cleaned. Each tag is trimmed and non alpha-chars are replaced with a dot. Ampersand (&) is replace with the word 'and'. Each tag will be processed by a sed script named tagfix.txt, followed by network specific rules in tagfix_<network>.txt and site specific rules in tagfix_<sitename>.txt. In the end, duplicates are removed, lower cased and sorted alphabetically.

Using preloaded variables

It's possible to feed the script with data that the script otherwise is not able to capture, or you want to prepare some data on before hand to override the script creating the values or if you are debugging the script and don't want to spam the imagehost with reuploads of the same images over and over. In case the script finds a value in these variables, it will usually override (images) or add(tags/description). See file config/scenedata.cfg for examples and explanation on how to use this functionality.

Custom site handling

Scripts that matches config/<network>.sh or config/<sitename>.sh will be included into the main script. These scrips may override or contain custom functions to modify progam behaviour. Site specific scripts will take presedence over network specific scripts. See existing custom scripts in the config folder for examples.

Downloading images before uploading

Some paysites have query parameters in their image urls for a given scene, usually due to content protection measurements. The jerking image host does not support to load such url's directly, so instead they must be downloaded to your temporary folder first. To enable this functionality include variable dlimages="1" and it will be handled automatically.

Custom BB-code templates

To make network or site specific templates, please follow the following naming convention: <network>_template.txt or <sitename>_template.txt Note: filenames must be all lower case. All templates should be stored in the templates folder.

Add more technical media info properties

See script scripts/mediainfofuncs.sh on how to extract more technical properties from the mediainfo report xml and assign them to global variables.

Overriding functions

Some functions in corefuncs.sh are encouraged to be overridden by a custom site config. These are:

• preMergeTemplate

This function can be used to have create additional variables with content referenced in your template definition BEFORE the actual merging and .torrent creation is done. Here you can create folder to include photoset or create .gif files from trailer or media file. Additional image upload must be handled in your function. If you need to change the source for torrent, set the variable TORRENT_INPUT to a custom folder. Default is the extracted media file.

• getQueryOverrideParamsForSceneList and getQueryOverrideParamsForScene

Here you can modify the input parameters for an YQL request to identify your episode or fetching episode metadata. This can for example include:

• Downloading html via phantomjs from sites that heavily depends on javascript to be rendered completely.
• Fetch html from sites the require cookie and certain headers to be set. You can then store the result on a public folder so that YQL can fetch the end result on your web server.

Adding tables for new sites

Documentation on how to develop YQL Open Data Tables: https://developer.yahoo.com/yql/guide/dev-external_tables.html A custom table contains javascript code to parse html and build an xml response back with library E4X: https://developer.mozilla.org/en-US/docs/Archive/Web/E4X/Processing_XML_with_E4X

In order to add support for new site, you need two files, one yql table to list a site's episodes <sitename>.xml and one table to extract episode metadata (<sitename>_scene.xml) The index table should return an xml strcture like this:

<results>
    <scene>
      <url>http://www.kinkysite.com/videos/ep/1234</url>
      <title>A Stepmom's Secrets</title>
      <titlematch>a stepmoms secrets</titlematch>
      <date>17.10.21</url>
    </scene>
    <scene>
    ...
    </scene>
</results>

The where-clause (ex: select * from index where date = '17.10.21') in the query will hopefully narrow down the scene-list down to 1. This means we have a match between release and episode. It is also possible to match on title (from the title in the release name) agains element titlematch in the xml in case the episode list does not provide dates.

Here is an example with output from YQL response from a <sitename>_scene.xml:

<results>
  <scene>
    <title>A Stepmom's Secrets</title>
    <dateiso>2017-10-21</dateiso>
    <poster>http://www.kinkysite.com/videos/ep/1234/cover.jpg</poster>
    <images>
      <image>http://www.kinkysite.com/videos/ep/1234/preview1.jpg</image>
      <image>http://www.kinkysite.com/videos/ep/1234/preview2.jpg</image>
    </images>
    <cast>Cory Hunter, Mike Black &amp; Stephanie Swift</cast>
    <tags>Threesome, BJ, Hardcore</tags>
    <desc>Stepmom Cory catches her step daughter giving sloppy head to her bf, she decides to join in...</desc>
  </scene>
</results>

Please see existing custom tables to various sites as examples on how to develop your own.

Integrate script with rTorrent

To achieve a faster level of automation, tell rTorrent to execute a script when a download finishes. Add this line in your .rtorrent.rc file:
system.method.set_key = event.download.finished, rtorrent_post_script, "execute = /home/MYUSERNAME/utils/rtorrent_post_script.sh, \"$d.name=\"" (Change actual path however you want)

Script contents of rtorrent_post_script.sh: (remember chmod u+x rtorrent_post_script.sh and review the paths in the scripts to fit your setup)

#!/bin/bash

torrentname="$1"
basedir="${HOME}/downloads"

XXX_0DAY_REGEX="(.*?)\.([0-9]{2}\.[0-9]{2}\.[0-9]{2}|E[0-9]{2,4})\.(.*)\.XXX.*"
trimmed=$(echo "${torrentname}" | tr " .-" "___" | tr --delete "()")
if echo "${torrentname}" | grep -qiP "${XXX_0DAY_REGEX}"; then
	if mkdir "${basedir}/temp/${trimmed}" 2>/dev/null ; then
		"${HOME}/utils/emp-auto-template-main/scripts/maketemplate.sh" "${basedir}/torrents/${torrentname}" 
	fi
fi