Struct BraidzArchive 

pub struct BraidzArchive<R: Read + Seek> {
    pub metadata: BraidMetadata,
    pub expected_fps: f64,
    pub calibration_info: Option<CalibrationInfo>,
    pub kalman_estimates_info: Option<KalmanEstimatesInfo>,
    pub kalman_estimates_table: Option<Vec<KalmanEstimatesRow>>,
    pub data_association: Option<Vec<DataAssocRow>>,
    pub reconstruction_latency_hlog: Option<HistogramLog>,
    pub reprojection_distance_hlog: Option<HistogramLog>,
    pub cam_info: CamInfo,
    pub data2d_distorted: Option<D2DInfo>,
    pub image_sizes: Option<BTreeMap<String, (usize, usize)>>,
    /* private fields */
}

The entire file contents, loaded to memory.

Currently, the implementation does not load everything to memory, but it should and will do so in the future. To load only a summary, use the BraidzSummary type. Currently, a summary can only be made by loading the entire archive first, but more efficient path can be made later.

Fields

metadata: BraidMetadata

expected_fps: f64

calibration_info: Option<CalibrationInfo>

kalman_estimates_info: Option<KalmanEstimatesInfo>

kalman_estimates_table: Option<Vec<KalmanEstimatesRow>>

data_association: Option<Vec<DataAssocRow>>

reconstruction_latency_hlog: Option<HistogramLog>

reprojection_distance_hlog: Option<HistogramLog>

cam_info: CamInfo

data2d_distorted: Option<D2DInfo>

image_sizes: Option<BTreeMap<String, (usize, usize)>>

A mapping from camera name to (width, height).

Implementations

impl<R: Read + Seek> BraidzArchive<R>

pub fn into_inner(self) -> ZipDirArchive<R>

Consume and return the raw storage archive.

pub fn display(&self) -> Display<'_>

Display the path to the archive.

pub fn path(&self) -> &Path

Get the path to the archive.

impl<'a, R: Read + Seek> BraidzArchive<R>

pub fn iter_data2d_distorted( &'a mut self, ) -> Result<impl Iterator<Item = Result<Data2dDistortedRow, Error>> + 'a, Error>

Iterate over the rows of the data2d_distorted table.

This takes a mutable reference because the read location in the archive is changed during operation.

Note that, as described in the “Details about how data are processed online and saved for later analysis” section in the “3D Tracking in Braid” chapter of the User’s Guide, these will not, in general, be returned in a monotonically increasing order.

pub fn iter_grouped_data2d_distorted( &'a mut self, include_nan_data: bool, bufsize: usize, ) -> Result<impl Iterator<Item = Result<GroupedRows<i64, Data2dDistortedRow>, Error>> + 'a, Error>

Iterate over synchronized frames in data2d_distorted table.

This sorts the data by looking ahead up to bufsize rows. Furthermore, it can exclude nan data if include_nan_data is set to false. No guarantees are made about whether frames may be skipped and indeed they often may be.