matroska-subtitles - npm Package Compare versions

Comparing version 1.1.2 to 2.0.0

index.js

@@ -0,130 +1,135 @@
+const Writable = require('stream').Writable
 const ebml = require('ebml')
 const ebmlBlock = require('ebml-block')
-const through = require('through2')
+const readElement = require('./lib/read-element')
 
 // track elements we care about
 const TRACK_ELEMENTS = ['TrackNumber', 'TrackType', 'Language', 'CodecID', 'CodecPrivate']
-
+const SUBTITLE_TYPES = ['S_TEXT/UTF8', 'S_TEXT/SSA', 'S_TEXT/ASS']
 const ASS_KEYS = ['readOrder', 'layer', 'style', 'name', 'marginL', 'marginR', 'marginV', 'effect', 'text']
 
-module.exports = function () {
-  const subtitleTracks = new Map()
-  const decoder = new ebml.Decoder()
+class MatroskaSubtitles extends Writable {
+  constructor (prevInstance) {
+    super()
 
-  var timecodeScale = 1
+    var currentTrack = null
+    var currentSubtitleBlock = null
+    var currentClusterTimecode = null
 
-  var currentTrack
-  var currentSubtitleBlock
-  var currentClusterTimecode
+    this.decoder = new ebml.Decoder()
 
-  decoder.on('data', function (chunk) {
-    // Segment Information //
-
-    if (chunk[1].name === 'TimecodeScale') {
-      timecodeScale = readData(chunk) / 1000000
+    if (prevInstance instanceof MatroskaSubtitles) {
+      prevInstance.end()
+      // copy previous metadata
+      this.subtitleTracks = prevInstance.subtitleTracks
+      this.timecodeScale = prevInstance.timecodeScale
+      this.decoder.on('data', _onClusterData)
+    } else {
+      this.subtitleTracks = new Map()
+      this.timecodeScale = 1
+      this.decoder.on('data', _onMetaData)
     }
 
-    // Clusters //
+    var self = this
 
-    // TODO: assuming this is a Cluster `Timecode`
-    if (chunk[1].name === 'Timecode') {
-      currentClusterTimecode = readData(chunk)
-    }
+    function _onMetaData (chunk) {
+      // Segment Information
+      if (chunk[1].name === 'TimecodeScale') {
+        self.timecodeScale = readElement(chunk[1]) / 1000000
+      }
 
-    // Tracks //
+      // Tracks
+      if (chunk[0] === 'start' && chunk[1].name === 'TrackEntry') {
+        currentTrack = {}
+      }
 
-    if (chunk[0] === 'start' && chunk[1].name === 'TrackEntry') {
-      currentTrack = {}
-    }
+      if (currentTrack && chunk[0] === 'tag') {
+        // save info about track currently being scanned
+        if (TRACK_ELEMENTS.includes(chunk[1].name)) {
+          currentTrack[chunk[1].name] = readElement(chunk[1])
+        }
+      }
 
-    if (chunk[0] === 'end' && chunk[1].name === 'TrackEntry') {
-      // 0x11: Subtitle Track, S_TEXT/UTF8: SRT format
-      if (currentTrack.TrackType === 0x11) {
-        if (currentTrack.CodecID === 'S_TEXT/UTF8' || currentTrack.CodecID === 'S_TEXT/ASS') {
-          subtitleTracks.set(currentTrack.TrackNumber, currentTrack.CodecID)
-          var info = {
-            track: currentTrack.TrackNumber,
-            language: currentTrack.Language,
-            type: currentTrack.CodecID.substring(7)
+      if (chunk[0] === 'end' && chunk[1].name === 'TrackEntry') {
+        if (currentTrack.TrackType === 0x11) { // Subtitle Track
+          if (SUBTITLE_TYPES.includes(currentTrack.CodecID)) {
+            var track = {
+              number: currentTrack.TrackNumber,
+              language: currentTrack.Language,
+              type: currentTrack.CodecID.substring(7).toLowerCase()
+            }
+
+            if (currentTrack.CodecPrivate) {
+              // only SSA/ASS
+              track.header = currentTrack.CodecPrivate.toString('utf8')
+            }
+
+            self.subtitleTracks.set(currentTrack.TrackNumber, track)
           }
-          if (currentTrack.CodecPrivate) {
-            // only SSA/ASS
-            info.header = currentTrack.CodecPrivate.toString('utf8')
-          }
-          stream.push(['new', info])
         }
+        currentTrack = null
       }
-      currentTrack = null
-    }
 
-    if (currentTrack && chunk[0] === 'tag') {
-      // save info about track currently being scanned
-      if (TRACK_ELEMENTS.includes(chunk[1].name)) {
-        currentTrack[chunk[1].name] = readData(chunk)
+      if (chunk[0] === 'end' && chunk[1].name === 'Tracks') {
+        self.decoder.removeListener('data', _onMetaData)
+
+        if (self.subtitleTracks.size <= 0) return self.end()
+
+        self.decoder.on('data', _onClusterData)
+        self.emit('tracks', Array.from(self.subtitleTracks.values()))
       }
     }
 
-    // Blocks //
+    function _onClusterData (chunk) {
+      // TODO: assuming this is a Cluster `Timecode`
+      if (chunk[1].name === 'Timecode') {
+        currentClusterTimecode = readElement(chunk[1])
+      }
 
-    if (chunk[1].name === 'Block') {
-      var block = ebmlBlock(chunk[1].data)
+      if (chunk[1].name === 'Block') {
+        var block = ebmlBlock(chunk[1].data)
 
-      if (subtitleTracks.has(block.trackNumber)) {
-        var type = subtitleTracks.get(block.trackNumber)
+        if (self.subtitleTracks.has(block.trackNumber)) {
+          var type = self.subtitleTracks.get(block.trackNumber).type
 
-        // TODO: would a subtitle track ever use lacing? We just take the first (only) frame.
-        var subtitle = {
-          text: block.frames[0].toString('utf8'),
-          time: (block.timecode + currentClusterTimecode) * timecodeScale
-        }
+          var subtitle = {
+            text: block.frames[0].toString('utf8'),
+            time: (block.timecode + currentClusterTimecode) * self.timecodeScale
+          }
 
-        if (type === 'S_TEXT/ASS') {
-          // extract ASS keys
-          var values = subtitle.text.split(',')
-          // ignore read-order
-          for (var i = 1; i < 9; i++) {
-            subtitle[ASS_KEYS[i]] = values[i]
+          if (type === 'ass' || type === 'ssa') {
+            // extract SSA/ASS keys
+            var values = subtitle.text.split(',')
+            // ignore read-order, and skip layer if ssa
+            var i = type === 'ssa' ? 2 : 1
+            for (; i < 9; i++) {
+              subtitle[ASS_KEYS[i]] = values[i]
+            }
+            // re-append extra text that might have been splitted
+            for (i = 9; i < values.length; i++) {
+              subtitle.text += ',' + values[i]
+            }
           }
-          // re-append extra text that might have been splitted
-          for (i = 9; i < values.length; i++) {
-            subtitle.text += ',' + values[i]
-          }
+
+          currentSubtitleBlock = [subtitle, block.trackNumber]
         }
-
-        currentSubtitleBlock = [block.trackNumber, subtitle]
       }
-    }
 
-    // TODO: assuming `BlockDuration` exists and always comes after `Block`
-    if (currentSubtitleBlock && chunk[1].name === 'BlockDuration') {
-      currentSubtitleBlock[1].duration = readData(chunk) * timecodeScale
+      // TODO: assuming `BlockDuration` exists and always comes after `Block`
+      if (currentSubtitleBlock && chunk[1].name === 'BlockDuration') {
+        currentSubtitleBlock[0].duration = readElement(chunk[1]) * self.timecodeScale
 
-      stream.push(currentSubtitleBlock)
+        self.emit('subtitle', ...currentSubtitleBlock)
 
-      currentSubtitleBlock = null
+        currentSubtitleBlock = null
+      }
     }
-  })
+  }
 
-  // create object stream
-  var stream = through.obj(function write (chunk, _, callback) {
-    decoder.write(chunk)
-    callback()
-  })
-
-  return stream
+  _write (chunk, _, callback) {
+    this.decoder.write(chunk)
+    callback(null)
+  }
 }
 
-function readData (chunk) {
-  switch (chunk[1].type) {
-    case 'b':
-      return chunk[1].data
-    case 's':
-      return chunk[1].data.toString('ascii')
-    case '8':
-      return chunk[1].data.toString('utf8')
-    case 'u':
-      return chunk[1].data.readUIntBE(0, chunk[1].dataSize)
-    default:
-      console.error('Unsupported data:', chunk)
-  }
-}
+module.exports = MatroskaSubtitles

package.json

 {
   "name": "matroska-subtitles",
-  "version": "1.1.2",
-  "description": "Transform stream for parsing embedded .mkv subtitles.",
+  "version": "2.0.0",
+  "description": "Writable stream for parsing embedded .mkv subtitles.",
   "main": "index.js",
   "dependencies": {
     "ebml": "^2.2.0",
-    "ebml-block": "^1.0.0",
-    "through2": "^2.0.1"
+    "ebml-block": "^1.0.0"
   },
@@ -32,3 +31,6 @@ "devDependencies": {},
   },
-  "homepage": "https://github.com/mathiasvr/matroska-subtitles#readme"
+  "homepage": "https://github.com/mathiasvr/matroska-subtitles#readme",
+  "directories": {
+    "example": "examples"
+  }
 }

README.md

@@ -10,102 +10,62 @@ # matroska-subtitles [![npm][npm-img]][npm-url] [![dependencies][dep-img]][dep-url] [![license][lic-img]][lic-url]
 
-Transform stream for parsing embedded .mkv subtitles.
+Writable stream for parsing embedded .mkv subtitles.
 
-> Currently supports extraction of the .srt and .ass format.
+Supported formats: `.srt`, `.ssa`, `.ass`.
 
 ## install
 
-```
+```bash
 npm install matroska-subtitles
 ```
 
-## documentation
+## example
 
-The `data` event of the stream will emit an array that determines the type of the data.
-When a new subtitle track is encountered the *track number*, *language*, *type* and optionally a *header* is emitted:
-
-```
-data = [ 'new', { track: <track number>, language: <string>, type: <string>, header: <string> } ]
-```
-
-Subsequently a specific subtitle track will emit data of this form:
-```
-data = [ <track number>, { text: <string>, time: <ms>, duration: <ms> } ]
-```
-
-## examples
-
-### dump all subtitles
-
 ```javascript
 const fs = require('fs')
-const matroskaSubtitles = require('matroska-subtitles')
+const MatroskaSubtitles = require('matroska-subtitles')
 
-var subs = matroskaSubtitles()
+var parser = new MatroskaSubtitles()
 
-subs.on('data', function (data) {
-  console.log(data)
+// first an array of subtitle track information is emitted
+parser.once('tracks', function (tracks) {
+  console.log(tracks)
 })
 
-fs.createReadStream('Sintel.2010.720p.mkv').pipe(subs)
+// afterwards each subtitle is emitted
+parser.on('subtitle', function (subtitle, trackNumber) {
+  console.log('Track ' + trackNumber + ':', subtitle)
+})
+
+fs.createReadStream('Sintel.2010.720p.mkv').pipe(parser)
 ```
 
-### group subtitle tracks
+### `tracks` event response format
 
-The following is an example of extracting subtitle tracks of an mkv:
-
 ```javascript
-const fs = require('fs')
-const matroskaSubtitles = require('matroska-subtitles')
-
-var tracks = new Map()
-var subs = matroskaSubtitles()
-
-subs.on('data', function (data) {
-  if (data[0] === 'new') {
-    var key = data[1].track
-    tracks.set(key, {
-      language: data[1].language,
-      subtitles: []
-    })
-  } else {
-    var key = data[0]
-    var subtitle = data[1]
-    tracks.get(key).subtitles.push(subtitle)
-  }
-})
-
-subs.on('end', function () {
-  tracks.forEach((track) => console.log(track))
-})
-
-fs.createReadStream('Sintel.2010.720p.mkv').pipe(subs)
+[
+  { number: 3, language: 'eng', type: 'utf8' },
+  { number: 4, language: 'jpn', type: 'ass', header: '[Script Info]\r\n...' }
+]
 ```
 
-> Notice that this example doesn't take advantage of streaming since the subtitles first are being outputted when the stream ends.
+> Note that the `language` may be `undefined` if the mkv track doesn't specify it.
 
-### response
+### `subtitle` event response format
 
-The response of this example would look like this:
 ```javascript
-{ language: 'eng',
-  subtitles: 
-   [ { text: 'This blade has a dark past.',
-       time: 107250,
-       duration: 1970 },
-     { text: 'It has shed much innocent blood.',
-       time: 111800,
-       duration: 4000 },
-     { text: 'You\'re a fool for traveling alone,\r\nso completely unprepared.',
-       time: 118000,
-       duration: 3450 } ] }
+{
+  text: 'This blade has a dark past.',
+  time: 107250,  // ms
+  duration: 1970 // ms
+}
 ```
 
-> Note that the `language` might be `undefined` if the mkv track has not specified it.
+> May also contain additional `.ass` specific values
 
-## contributing
+## random access
+The parser must obtain the `tracks` metadata event before it can begin to emit subtitles.
+To read subtitles from a specific position in the stream,
+you can pass in a previous instance as parameter: `parser = new MatroskaSubtitles(parser)`
+after the `tracks` event and pipe from a given position. See `examples/random-access.js` for an example.
 
-This is still a work in progress.
-
-If you find a bug or have suggestions feel free to create an issue or a pull request!
-
 ## see also 
@@ -112,0 +72,0 @@

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Major refactor

Supply chain risk

Package has recently undergone a major refactor. It may be unstable or indicate significant internal changes. Use caution when updating to versions that include significant changes.

Found 1 instance in 1 package

Filesystem access

Supply chain risk

Accesses the file system, and could potentially read sensitive data.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

10428

increased by21.72%

Dependency count

2

decreased by-33.33%

Number of package files

8

increased by100%

Lines of code

170

increased by61.9%

Worsened metrics

Number of lines in readme file

77

decreased by-34.19%

Number of low supply chain risk alerts

3

increased byInfinity%

Number of medium supply chain risk alerts

1

increased byInfinity%

Dependency changes

• - Removedthrough2@^2.0.1

• - Removedcore-util-is@1.0.3(transitive)

• - Removedinherits@2.0.4(transitive)

• - Removedisarray@1.0.0(transitive)

• - Removedprocess-nextick-args@2.0.1(transitive)

• - Removedreadable-stream@2.3.8(transitive)

• - Removedsafe-buffer@5.1.2(transitive)

• - Removedstring_decoder@1.1.1(transitive)

• - Removedthrough2@2.0.5(transitive)

• - Removedutil-deprecate@1.0.2(transitive)

• - Removedxtend@4.0.2(transitive)