15 File Formats

15.1 File descriptions

15.1.1 _DBconfig.json

The _DBconfig.json file contains the configuration options of the database. People familiar with the legacy EMU system will recognize this as the replacement file for the legacy template (.tpl) file. By convention, variables or strings written entirely in capital letters indicate a constant variable that usually has a special meaning. This is also the case with strings like this found in the _DBconfig.json ("STRING", "ITEM" ,"SEGMENT", "EVENT", "OSCI", … ).

The _DBconfig.json file contains the following fields:

• "name" specifies the name of the database

• "UUID" a unique ID given to each database

• "mediafileExtension" the main mediafileExtension (currently only uncompressed .wav files are supported in every component of the EMU system. This is also the recommended audio format for the emu-sdms.)

• "ssffTrackDefinitions" an array of definitions defining the SSFF tracks of the database. Each ssffTrackDefinition consists of:

  • "name" the name of the ssffTrackDefinition

  • "columnName" the name of the column of the matching SSFF file. For more information on the columns the various functions of the wrassp produce, see the track fields of the wrasspOutputInfos object that is part of the wrassp package. Further, although the SSFF is a binary file format, it has a plain text header. This means that if you open a SSFF file in the text editor of your choice, you will be able to see the columns contained within it. Another way of accessing column information about a specific SSFF file is to use the wrassp function res = read.AsspDataObj(/path/2/SSFF/file) to read the file from the file system. names(res) will then give you the names of the columns present in this file. In the context of the SSFF, we use the term “column,” while in the context of the EMU system we use either “track” or “SSFF track.” Both refer to the same data.

  • "fileExtention" the file extension of the matching SSFF file. See also ?wrasspOutputInfos for the default extensions produced by the wrassp functions.

• "levelDefinitions" array of definitions defining the levels of the database. Each "levelDefinitions" consists of:

  • "name" The name of the levelDefinition.

  • "type" Specifies the type of level (either "ITEM" | "EVENT" | "SEGMENT").

  • "attributeDefinitions" an array of definitions defining the attributes of the level. Each attributeDefinition consists of:

    • "name" The name of the "attributeDefinition".

    • "type" Specifies the type of the attribute (currently only "STRING" permitted)

    • "labelGroups" An (optional) array containing label group definitions. These can be used as a shorthand notation for querying certain groups of labels and comprise the following:

      • "name" The name of the label group. This will be the value used in a query to refer to this group.

      • "values" An array of strings representing the labels.

    • "legalLabels" An (optional) array of strings specifying which labels are valid or legal for this attribute definition.

  • "anagestConfig" If specified (optional), this will convert the level into a special type of level for labeling articulatory data. This will also serve as a marker for the EMU-webApp to treat this level differently. This optional field may only be set for levels of the type "EVENT".

    • "verticalPosSsffTrackName" The name of the SSFF track containing the vertical position data.

    • "velocitySsffTrackName" The name of the SSFF track containing the velocity data.

    • "autoLinkLevelName" The name of the level to which created events will be linked.

    • "multiplicationFactor" The factor to multiply with (either -1 1).

    • "threshold" A value between 0 and 1 defining the absolute threshold.

    • "gestureOnOffsetLabels" An array containing two strings that specify the on- and offset labels.

    • "maxVelocityOnOffsetLabels" An array containing two strings that specify the on- and offset labels.

    • "constrictionPlateauBeginEndLabels" An array containing two strings that specify the begin- and end labels.

    • "maxConstrictionLabel" A string specifying the maximum constriction label.

• "linkDefinitions" An array of the definitions defining the links between levels of the database. The combination of all link definitions specifies the hierarchy of the database. Each linkDefinition consists of:

  • "type" Specifies the type of link (either "ONE_TO_MANY" "MANY_TO_MANY" "ONE_TO_ONE").

  • "superlevelName" Specifies the name of the super-level.

  • "sublevelName" Specifies the name of the sub-level.

• "labelGroups" An (optional) array containing label group definitions. These can be used as a shorthand notation for querying certain groups of labels. Compared to the "labelGroups", which can be defined within an attributeDefinition, the labelGroups defined here are globally defined for the entire database as follows:

  • "name" The name of the label group.

  • "values" An array of strings containing labels.

• "EMUwebAppConfig" Specifies the configuration options intended for the
  EMU-webApp (i.e., how the database is to be displayed). This field can contain all the configurations options that are specified in the EMU-webApp’s configuration schema (see https://github.com/IPS-LMU/EMU-webApp/tree/master/dist/schemaFiles). The "EMUwebAppConfig" contains the following fields:

  • "main" Main behavior options:

    • "autoConnect": Auto connect to the "serverUrl" on the initial load of the webApp to automatically load a database (mainly used for development).

    • "serverUrl": The default server URL that is displayed in the connect modal window (and used if "autoConnect" is set to true). The default: "ws://localhost:17890" points to the server started by the serve() function of the emuR package.

    • "serverTimeoutInterval": The maximum amount of time the EMU-webApp waits (in milliseconds) for the server to respond.

    • "comMode": Specifies the communication mode the EMU-webApp is in. Currently the only option that is available is "WS" (websocket).

    • "catchMouseForKeyBinding": Check if mouse has to be in labeler for key bindings to work.

  • "keyMappings" Keyboard shortcut definitions. For the sake of brevity, not every key-code is shown (see schema: https://github.com/IPS-LMU/EMU-webApp/blob/master/dist/schemaFiles/emuwebappConfigSchema.json for extensive list).

    • "toggleSideBarLeft" integer value that represents the key-code that toggles the left side bar (== bundleList side bar)

    • "toggleSideBarRight" integer value that represents the key-code that toggles the right side bar (== perspective side bar)

    • …

  • "spectrogramSettings" Specifies the default settings of the spectrogram. The possible settings are:

    • "windowSizeInSecs" Specifies the window size in seconds.

    • "rangeFrom" Specifies the lowest frequency (in Hz) that will be displayed by the spectrogram.

    • "rangeTo" Specifies the highest frequency (in Hz) that will be displayed by the spectrogram.

    • "dynamicRange" Specifies the dynamic rang for maximum decibel dynamic range.

    • "window" Specifies the window type ("BARTLETT" "BARTLETTHANN" "BLACKMAN" "COSINE" "GAUSS" "HAMMING" "HANN" "LANCZOS" "RECTANGULAR" "TRIANGULAR").

    • "preEmphasisFilterFactor" Specifies the preemphasis factor (in formula: s’(k) = s(k) - preEmphasisFilterFactor * s(k-1) ).

    • "transparency" Specifies the transparency of the spectrogram (integer from 0 to 255).

    • "drawHeatMapColors" (optional) Defines whether the spectrogram should be drawn using heat-map colors (either true or false)

    • "heatMapColorAnchors" (optional) Specifies the heat-map color anchors (array of the form [[255, 0, 0], [0, 255, 0], [0, 0, 255]])

  • "perspectives" An array containing perspective configurations. Each "perspective" consists of:

    • "name" Name of the perspective.

    • "signalCanvases" Configuration options for the signalCanvases.

      • "order" An array specifying the order in which the ssff tracks are to be displayed. Note that the ssff track names “OSCI” and “SPEC” are always available in addition to the ssff track defined in the database.

      • "assign" An array of configuration options that assign one ssff track to another, effectively creating a visual overlay of one track over another. Each array element consists of:

      • "signalCanvasName" The name of the signal specified in the "order" array. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "ssffTrackName" The name of the SSFF track to overlay onto "signalCanvasName". (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "minMaxValLims" An array of configuration options to limit the y-axis range that is displayed for a specified SSFF track.

      • "ssffTrackName": A name specifying which ssffTrack should be limited. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "minVal": The minimum value which defines the lower y-axis limit. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "horizontalLines": Add horizontal lines to ssffTrack

      • "ssffTrackName": The name of the SSFF track to draw horizontal lines on. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "yValues": array of number specifying the y values of the horizontal lines (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "contourLims" An array containing contour limit values that specify an index range that is to be displayed. As a track or column can contain multi-dimensional data (e.g. four formant values per time stamp, 256 DFT values per time stamp, etc.) it is possible to specify an index range that specifies which values should be displayed (e.g., display formant 2 through 4).

      • "ssffTrackName" A name specifying which ssffTrack should be limited. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "minContourIdx" The minimum contour index to display (starts at index 0). (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "maxContourIdx" The maximum contour index to display. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "contourColors" An array to specify colors of individual contours. This overrides the default of automatically calculating distinct colors for each contour.

      • "ssffTrackName" The name of the ssffTrackName for which colors are defined. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "colors" An array of RGB strings (e.g. ["rgb(238,130,238)", "rgb(127,255,212)"]) that specify the color of the contour (first value = first contour color and so on). (NOTE: indentation level reduced by 1 to avoid max indentation problems)

    • "levelCanvases" Configuration options for the levelCanvases:

      • "order" An array specifying the order in which the levels are to be displayed. Note that only levels of the type EVENT or SEGMENT can be displayed as levelCanvases. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

    • "twoDimCanvases" Configuration options for the 2D canvas.

      • "order" An array specifying the order in which the levels are to be displayed. Note that currently only a single
        twoDimDrawingDefinition can be displayed so this array can currently only contain a single element.

      • "twoDimDrawingDefinitions" An array containing two dimensional drawing definitions. Each two dimensional drawing definition consists of:

      • "dots" An array containing dot definitions. Each dot definition consist of: (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "name" The name of the dot.(NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "xSsffTrack" The ssffTrackName of the track that contains the x axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "xContourNr" The contour number of the track that contains the x-axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "ySsffTrack" The ssffTrackName of the track that contains the y-axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "yContourNr" The contour number of the track that contains the y-axis values.(NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "color" The RGB color string specifying the color given to dot. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "connectLines" An array specifying which of the dots specified in the "dots" definition array should be connected by a line. (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "fromDot" The dot from which the line should start. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "toDot" The dot at which the line should end. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "color" The RGB string defining the color of the line. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "staticDots" An array containing static dot definitions:(NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "name" The name of the static dots. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "xNameCoordinate" An x-coordinate specifying the location at which name should be drawn. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "yNameCoordinate" y-coordinate specifying the location at which name should be drawn. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "xCoordinates" An array of x-coordinates (e.g. [300, 300, 900, 900, 300]). (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "yCoordinates" An array of y-coordinates (e.g. [880, 2540, 2540, 880, 880]). (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "connect" A boolean value that specifies whether or not to connect the static dots with lines. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "color" An RGB string specifying the color of static dots. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "staticContours" An array containing static contour definitions: (NOTE: indentation level reduced by 1 to avoid max indentation problems)

      • "name" The name of static contour. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "xSsffTrack" The ssffTrackName of the track that contains the x-axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "xContourNr" The contour number of the track that contains the x-axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "ySsffTrack" The ssffTrackName of the track that contains the y-axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "yContourNr" The contour number of the track that contains the y-axis values. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "connect" A boolean value that specifies whether or not to connect the static dots with lines. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

      • "color" An RGB string specifying color of the static contour. (NOTE: indentation level reduced by 2 to avoid max indentation problems)

  • "labelCanvasConfig" Configuration options for the label canvases:

    • "addTimeMode" The mode in which time to boundaries is added and subtracted ("absolute" or "relative").

    • "addTimeValue": The amount of samples added to or subtracted from boundaries.

    • "newSegmentName" The value given to the default label if a new SEGMENT is added (default is "" == empty string).

    • "newEventName" The value given to the default label if a new EVENT is added (default is "" == empty string).

  • "restrictions":

    • "playback" A boolean value specifying whether to allow audio playback.

    • "correctionTool" A boolean value specifying whether correction tools are available.

    • "editItemSize" A boolean value specifying whether to allow the size of a SEGMENT or EVENT to be changed (i.e., move boundaries).

    • "editItemName" A boolean value specifying whether to allow the label of an ITEM to be changed.

    • "deleteItemBoundary" A boolean value specifying whether to allow the deletion of boundaries.

    • "deleteItem" A boolean value specifying whether to allow the deletion of entire ITEMs

    • "deleteLevel" A boolean value specifying whether to allow the deletion of entire levels.

    • "addItem" A boolean value specifying whether to allow new ITEMs to be added.

    • "drawCrossHairs" A boolean value specifying whether to draw the cross hairs on signal canvases.

    • "drawSampleNrs" A boolean value specifying whether to draw the sample numbers in the OSCI canvas if zoomed in close enough to see samples (mainly for debugging and development purposes).

    • "drawZeroLine" A boolean value specifying whether to draw the zero value line in OSCI canvas.

    • "bundleComments" A boolean value specifying whether to allow the annotator to add comments to bundles she or he has annotated. A bundle comment field will show up in the bundle list side bar for each bundle if this is set to true. Note that the server has to support saving these comments, which the serve() function of the emuR package does not.

    • "bundleFinishedEditing" A boolean value specifying whether to allow the annotator to mark when she or he has finished annotating a bundle. A finished editing toggle button will show up in the bundle list side bar for each bundle if this is set to true. Note that the server has to support saving these comments which the serve() function of the emuR package does not.

    • "showPerspectivesSidebar" A boolean value specifying whether to show the perspectives side bar.

  • "activeButtons" Specifies which top- or bottom-menu buttons should be displayed by the EMU-webApp.

    • "addLevelSeg" A boolean value specifying whether to show the add SEGMENT level button in the top menu bar.

    • "addLevelEvent" A boolean value specifying whether to show the add EVENT level button in the top menu bar.

    • "renameSelLevel" A boolean value specifying whether to allow the user to rename the currently selected level.

    • "downloadTextGrid" A boolean value specifying whether to allow the user to download the current annotation as a .TextGrid file by displaying a download TextGrid button in the top menu bar.

    • "downloadAnnotation" A boolean value specifying whether to allow the user to download the current annotation as an _annot.json file by displaying a download annotJSON button in the top menu bar.

    • "specSettings" A boolean value specifying whether to display the spec. settings button in the top menu bar.

    • "connect" A boolean value specifying whether to display the connect button in the top menu bar.

    • "clear" A boolean value specifying whether to display the clear button in the top menu bar.

    • "deleteSingleLevel" A boolean value specifying whether to allow the user to delete a level containing time information.

    • "resizeSingleLevel" A boolean value specifying whether to allow the user to resize a level.

    • "saveSingleLevel" A boolean value specifying whether to allow the user to download a single level in the ESPS/waves+ format.

    • "resizeSignalCanvas" A boolean value specifying whether to allow the user to resize the signalCanvases ("OSCI", "SPEC", …).

    • "openDemoDB" A boolean value specifying whether to show the open demoDB button.

    • "saveBundle" A boolean value specifying whether to show the save button in bundle list side bar for each bundle.

    • "openMenu" A boolean value specifying whether open bundle list side bar button is displayed.

    • "showHierarchy" A boolean value specifying whether to display the show hierarchy button.

• "demoDBs" An array of strings specifying which demoDBs to display in the open demo drop-down menu. Currently available demo databases are ["ae", "ema", "epgdorsal"].

15.1.2 _annot.json

The _annot.json files contain the actual annotation information as well as the hierarchical linking information. Legacy EMU users should note that all the information that used to be split into several ESPS/waves+ label files and a .hlb file is now contained in this single file.

The _annot.json file contains the following fields:

• "name" Specifies the name of the annotation file (has to be equal to the bundle directory prefix as well as the _annot.json prefix).

• "annotates" Specifies the (relative) media file path that this _annot.json file annotates.

• "sampleRate" Specifies the sample rate of the annotation (should be the same as the sample rate of the file listed in "annotates").

• "levels" Contains an array of level annotation informations. Each element consists of:

  • "name" Specifies the name of the level.

  • "items" An array containing the annotation items of the level.

    • "id" The unique ID of the item (only unique within an _annot.json file or bundle, not globally for the emuDB).

    • "sampleStart" Contains start sample value of SEGMENT item.

    • "sampleDur" Contains sample duration value of SEGMENT item. Note that the EMU-webApp does not support overlapping SEGMENTs or SEGMENT sequences containing gaps. This implies that each sample is explicitly and unambiguously associated with a single SEGMENT. This means that the sampleStart value of a following SEGMENT has to be sampleStart + sampleDur + 1 of the previous SEGMENT. When converting the sample values to time values, the start time value is calculated with the formula \(start = \frac{sampleStart}{sampleRate} - \frac{0.5}{sampleRate}\) and the end time value with the formula \(end = \frac{sampleStart + sampleDur}{sampleRate} + \frac{0.5}{ sampleRate}\). This is done to have gapless time values for successive SEGMENTs. To avoid a negative time value when dealing with the first sample of an audio file (sampleStart value of \(0\)), the start time value is simply set to \(0\) in this case. One implication of this design is that there can be no zero-length segments. If for example a SEGMENT has sampleStart: 1420 and sampleDur: 0, it will be considered to start at the start time of sample \(1420 - \frac{0.5}{sampleRate}\) and end at the end time of sample \(1420 + \frac{0.5}{sampleRate}\). The actual duration in seconds is therefore expressed as \(\frac{sampleDur + 1}{sampleRate}\). The start and end time value calculation is performed by both the query engine of emuR if the calcTimes parameter is set to TRUE and the EMU-webApp to display the time information in the signal canvases.

    • "samplePoint" Contains sample point values of EVENT items. When calculating the start time values for EVENTs the following formula is used: \(start = \frac{samplePoint}{sampleRate}\)

    • "labels" An array containing labels that belong to this item. Each element consists of:

      • "name" Specifies the attributeDefinition that this label is for.

      • "value" Specifies the label value.

• "links" An array containing links between two items. These links have to adhere to the links specified in linkDefinitions of the corresponding emuDB. Each link consists of:

  • "fromID" The ID value of the item to link from (i.e., item in super-level).

  • "toID" The ID value of item to link to (i.e., item in sub-level).

15.1.3 The SSFF file format

The SSFF file format is a binary file format which has a plain text header. This means that the header is human-readable and can be viewed with any text editor including common UNIX command line tools such as less or cat. Within R it is possible to view the header by using R’s readLines() function as displayed in the R code snippet below.

# load the emuR and wrassp packages
library(emuR, warn.conflicts = FALSE)
library(wrassp)

## Loading required package: tibble

# create demo data in directory 
# provided by tempdir()
create_emuRdemoData(dir = tempdir())

# create path to demo database
path2ae = file.path(tempdir(), "emuR_demoData", "ae_emuDB")

# create path to bundle in database
path2bndl = file.path(path2ae, "0000_ses", "msajc003_bndl")

# create path to .fms file 
path2fmsFile = file.path(path2bndl, 
                        paste0("msajc003.", 
                               wrasspOutputInfos$forest$ext))

# read in first 8 lines of .fms file.
# Note that the header length may vary in other SSFF files.
readLines(path2fmsFile, n = 8)

## [1] "SSFF -- (c) SHLRC"            "Machine IBM-PC"              
## [3] "Record_Freq 200.0"            "Start_Time 0.0025"           
## [5] "Column fm SHORT 4"            "Column bw SHORT 4"           
## [7] "Original_Freq DOUBLE 20000.0" "-----------------"

The general line item structure of the plain text head of an SSFF file can be described as follows:

• SSFF – (c) SHLRC (required line): File type marker.

• Machine IBM-PC (required line): System architecture of the machine that generated the file. This is mainly used to specify the endianness of the data block (see below). Machine IBM-PC indicates little-endian and Machine SPARC indicates big-endian. To date, we have not encountered other machine types.

• Record_Freq SR (required line): Sample rate of current file in Hz. If, for example, SR is 200.0 (see the above R code snippet) then the sample rate is 200 Hz.

• Start_Time ST (required line): Time of first sample block in data block in seconds. This often deviates from 0.0 as wrassp’s windowed signal processing functions start with the first window centered around windowShift / 2. If the windowShift parameter’s default value is 5 ms, the start time ST of the first sample block will be 0.0025 sec (see above R code snippet).

• Column CN CDT CDL (required line(s)): A Column line entry contains four space-separated values, where Column is the initial key word value. The second value, CN (fm in above R code snippet), specifies the name for the column; the third, CDT (SHORT in the above R code snippet), indicates the column’s data type; and the fourth, CDL (4 in the above R code snippet), is the column’s data length in bytes. As can be seen in the R code snippet above, it is quite common for SSFF files to have multiple column entries. The sequence of these entries is relevant, as it specifies the sequence of the data in the binary data block (see below).

• NAME DT DV (optional line(s)): Optional single value definitions that have a NAME, a data type DT and a data value DV (see Original_Freq DOUBLE 20000.0 in the R code snippet above specifying the original sample rate of the audio file the .fms file was generated from).

• Comment CHAR string of variable length (optional line(s)): The Comment CHAR allows for comment strings to be added to the header.

• —————– (required line): marks the end of the plain text header.

The binary data block of the SSFF file format stores its data in a so-called interleaved fashion. This means it does not store the binary data belonging to every column in a separate data block. Rather, it interleaves the columns to form sample blocks that occur at the same point in time. Figure 15.1 displays a sequence of short integer values where the subscript text indicates the index in the sequence. This sequence represents a schematic representation of the data block of the .fms file of R Example 15.1. The first four INT161-4 (green) blocks represent the first four INT16 formant values that belong to the fm column and the next four INT165-8 (orange) represent the first four bandwidth values belonging to the bm column. Therefore, the dashed square marks the first sample block (i.e., the first eight F1, F2, F3 and F4; and F1_bandwidth, F2_bandwidth, F3_bandwidth and F4_bandwidth values) that occur at the time specified by the Start_Time 0.0025 header entry. The time of all subsequent sample blocks of eight INT16 values (e.g., INT169-16) can be calculated as follows: 0.0025 (== Start_Time) + 1 / 200.0 (== Record_Freq) * sample block index.

Figure 15.1: Schematic representation of the data block of the msajc003.fms file of the above R code snippet.

15.2 Example files

15.2.1 _bundleList.json

Compared to the _DBconfig.json and _annot.json files, the _bndl.json format is not part of the emuDB database specification. Rather, it is part of the EMU-webApp-websocket-protocol and is used as a standardized format to transport information about all the available bundles to the EMU-webApp. It is not meant as an on-disk file format but rather should be generated on-demand by the server implementing the EMU-webApp-websocket-protocol. A schematic example of a _bndl.json file is displayed in the listing below.

[
  {
  "name": "msajc003",
  "session": "0000",
  "finishedEditing": false,
  "comment": "",
  "timeAnchors": [
    {
      "sample_start": 1000,
      "sample_end": 2000
    }, ...
  ]
  },
  {
  "name": "msajc010",
  "session": "0000",
  "finishedEditing": false,
  "comment": ""
  }
]

15.2.2 _bndl.json

Compared to the _DBconfig.json and _annot.json files, the _bndl.json format is not part of the emuDB database specification. Rather, it is part of the EMU-webApp-websocket-protocol and is used as a standardized format to transport all the data belonging to a single bundle to the EMU-webApp. It is not meant as an on-disk file format by rather should generated on-demand by the server implementing the EMU-webApp-websocket-protocol. A schematic example of a _bndl.json file is displayed in the listing below.

{
 "ssffFiles": [
  {
   "fileExtension": "fms",
   "encoding": "BASE64",
   "data": "U1N..."
  }
 ],
 "mediaFile": {
  "encoding": "BASE64",
  "data": "Ukl..."
 },
 "annotation": contentOfAnnot.json
}

contentOfAnnot.json in the listing above refers to the content of a _annot.json file.