From 9877b139f684724f455f1efb9f2826965ae2ab52 Mon Sep 17 00:00:00 2001 From: jmt-gh Date: Sun, 12 Jun 2022 00:53:56 -0700 Subject: [PATCH] add new parser for overdrive media markers --- server/objects/mediaTypes/Book.js | 104 +----------------- .../parsers/parseOverdriveMediaMarkers.js | 101 +++++++++++++++++ 2 files changed, 104 insertions(+), 101 deletions(-) create mode 100644 server/utils/parsers/parseOverdriveMediaMarkers.js diff --git a/server/objects/mediaTypes/Book.js b/server/objects/mediaTypes/Book.js index 8048911d..c9d4d429 100644 --- a/server/objects/mediaTypes/Book.js +++ b/server/objects/mediaTypes/Book.js @@ -3,6 +3,7 @@ const Logger = require('../../Logger') const BookMetadata = require('../metadata/BookMetadata') const { areEquivalent, copyValue } = require('../../utils/index') const { parseOpfMetadataXML } = require('../../utils/parsers/parseOpfMetadata') +const { getOverdriveMediaMarkersFromFiles, parseOverdriveMediaMarkers } = require('../../utils/parsers/parseOverdriveMediaMarkers') const abmetadataGenerator = require('../../utils/abmetadataGenerator') const { readTextFile } = require('../../utils/fileUtils') const AudioFile = require('../files/AudioFile') @@ -398,116 +399,17 @@ class Book { return wasUpdated } - generateChaptersFromOverdriveMediaMarkers(overdriveMediaMarkers, includedAudioFiles) { - var parseString = require('xml2js').parseString; // function to convert xml to JSON - - var parsedOverdriveMediaMarkers = [] // an array of objects. each object being a chapter with a name and time key. the values are arrays of strings - - overdriveMediaMarkers.forEach(function (item, index) { - var parsed_result - parseString(item, function (err, result) { - // result.Markers.Marker is the result of parsing the XML for the MediaMarker tags for the MP3 file (Part##.mp3) - // it is shaped like this: - // [ - // { - // "Name": [ - // "Chapter 1: " - // ], - // "Time": [ - // "0:00.000" - // ] - // }, - // { - // "Name": [ - // "Chapter 2: " - // ], - // "Time": [ - // "15:51.000" - // ] - // } - // ] - - parsed_result = result.Markers.Marker - - // The values for Name and Time in parsed_results are returned as Arrays from parseString - // update them to be strings - parsed_result.forEach((item, index) => { - Object.keys(item).forEach(key => { - item[key] = item[key].toString() - }) - }) - }) - - parsedOverdriveMediaMarkers.push(parsed_result) - }) - - // go from an array of arrays of objects to an array of objects - // end result looks like: - // [ - // { - // "Name": "Chapter 1: The Worst Birthday", - // "Time": "0:00.000" - // }, - // { - // "Name": "Chapter 2: Dobby's Warning", - // "Time": "15:51.000" - // }, - // { redacted } - // ] - parsedOverdriveMediaMarkers = parsedOverdriveMediaMarkers - - var index = 0 - - var time = 0.0 - - - // actually generate the chapter object - // logic ported over from benonymity's OverdriveChapterizer: - // https://github.com/benonymity/OverdriveChapterizer/blob/main/chapters.py - var length = 0.0 - var newOChapters = [] - const weirdChapterFilterRegex = /([(]\d|[cC]ontinued)/ - includedAudioFiles.forEach((track, track_index) => { - parsedOverdriveMediaMarkers[track_index].forEach((chapter) => { - Logger.debug(`[Book] Attempting regex check for ${chapter.Name}!`) - if (weirdChapterFilterRegex.test(chapter.Name)) { - Logger.debug(`[Book] That shit weird yo`) - return - } - time = chapter.Time.split(":") - time = length + parseFloat(time[0]) * 60 + parseFloat(time[1]) - newOChapters.push( - { - id: index++, - start: time, - end: length, - title: chapter.Name - } - ) - }) - length += track.duration - }) - - Logger.debug(`[Book] newOChapters: ${JSON.stringify(newOChapters)}`) - return newOChapters - } - - getOverdriveMediaMarkers(audioFiles) { - var markers = audioFiles.map((af) => af.metaTags.tagOverdriveMediaMarker).filter(notUndefined => notUndefined !== undefined).filter(elem => { return elem !== null }) || [] - return markers - } - setChapters(preferOverdriveMediaMarker = false) { // If 1 audio file without chapters, then no chapters will be set var includedAudioFiles = this.audioFiles.filter(af => !af.exclude) - var overdriveMediaMarkers = this.getOverdriveMediaMarkers(includedAudioFiles) + var overdriveMediaMarkers = getOverdriveMediaMarkersFromFiles(includedAudioFiles) // If preferOverdriveMediaMarker is set, try and use that first // fallback to non-overdrive chapters if there are no Overdrive Media Markers available if (preferOverdriveMediaMarker && (overdriveMediaMarkers.length > 0)) { Logger.debug(`[Book] preferring overdrive media markers! Lets generate em.`) - this.chapters = this.generateChaptersFromOverdriveMediaMarkers(overdriveMediaMarkers, includedAudioFiles) + this.chapters = parseOverdriveMediaMarkers(overdriveMediaMarkers, includedAudioFiles) } else { if (includedAudioFiles.length === 1) { // 1 audio file with chapters diff --git a/server/utils/parsers/parseOverdriveMediaMarkers.js b/server/utils/parsers/parseOverdriveMediaMarkers.js new file mode 100644 index 00000000..859711ca --- /dev/null +++ b/server/utils/parsers/parseOverdriveMediaMarkers.js @@ -0,0 +1,101 @@ +const Logger = require('../../Logger') + +// given an array of audioFiles, return an array of unparsed MediaMarkers +module.exports.getOverdriveMediaMarkersFromFiles = (audioFiles) => { + var markers = audioFiles.map((af) => af.metaTags.tagOverdriveMediaMarker).filter(notUndefined => notUndefined !== undefined).filter(elem => { return elem !== null }) || [] + return markers +} + +module.exports.parseOverdriveMediaMarkers = (overdriveMediaMarkers, includedAudioFiles) => { + var parseString = require('xml2js').parseString; // function to convert xml to JSON + + var parsedOverdriveMediaMarkers = [] // an array of objects. each object being a chapter with a name and time key. the values are arrays of strings + + overdriveMediaMarkers.forEach(function (item, index) { + var parsed_result + parseString(item, function (err, result) { + // result.Markers.Marker is the result of parsing the XML for the MediaMarker tags for the MP3 file (Part##.mp3) + // it is shaped like this: + // [ + // { + // "Name": [ + // "Chapter 1: " + // ], + // "Time": [ + // "0:00.000" + // ] + // }, + // { + // "Name": [ + // "Chapter 2: " + // ], + // "Time": [ + // "15:51.000" + // ] + // } + // ] + + parsed_result = result.Markers.Marker + + // The values for Name and Time in parsed_results are returned as Arrays from parseString + // update them to be strings + parsed_result.forEach((item, index) => { + Object.keys(item).forEach(key => { + item[key] = item[key].toString() + }) + }) + }) + + parsedOverdriveMediaMarkers.push(parsed_result) + }) + + // go from an array of arrays of objects to an array of objects + // end result looks like: + // [ + // { + // "Name": "Chapter 1: The Worst Birthday", + // "Time": "0:00.000" + // }, + // { + // "Name": "Chapter 2: Dobby's Warning", + // "Time": "15:51.000" + // }, + // { redacted } + // ] + parsedOverdriveMediaMarkers = parsedOverdriveMediaMarkers + + var index = 0 + + var time = 0.0 + + + // actually generate the chapter object + // logic ported over from benonymity's OverdriveChapterizer: + // https://github.com/benonymity/OverdriveChapterizer/blob/main/chapters.py + var length = 0.0 + var newOChapters = [] + const weirdChapterFilterRegex = /([(]\d|[cC]ontinued)/ + includedAudioFiles.forEach((track, track_index) => { + parsedOverdriveMediaMarkers[track_index].forEach((chapter) => { + Logger.debug(`[parseOverdriveMediaMarkers] Attempting regex check for ${chapter.Name}!`) + if (weirdChapterFilterRegex.test(chapter.Name)) { + Logger.debug(`[parseOverdriveMediaMarkers] That shit weird yo`) + return + } + time = chapter.Time.split(":") + time = length + parseFloat(time[0]) * 60 + parseFloat(time[1]) + newOChapters.push( + { + id: index++, + start: time, + end: length, + title: chapter.Name + } + ) + }) + length += track.duration + }) + + Logger.debug(`[parseOverdriveMediaMarkers] newOChapters: ${JSON.stringify(newOChapters)}`) + return newOChapters +} \ No newline at end of file