CommonDataFormat.jl

Attribute Descriptor Record (ADR)

Contains a description of an attribute in a CDF. There will be one ADR per attribute. The ADRhead field of the ADR contains the file offset of the first ADR.

ADR(buf, offset, RecordSizeType)

Load an Attribute Descriptor Record from the buffer at the specified position.

AEDR

Attribute g/r Entry Descriptor Record. Describes a global entry (for global attributes) or rVariable entry (for variable attributes).

CDFDataset(filename)

Load a CDF file and return a CDFDataset object.

Example

cdf = CDFDataset("data.cdf")

CDF Descriptor Record (CDR) - the main file header record Contains version, encoding, format information, and pointer to GDR

CDR(buffer, pos, FieldSizeT)

Load a CDF Descriptor Record from the IO stream at the specified offset. This follows the CDF specification for CDR record structure.

Epoch

Milliseconds since Year 0 (01-Jan-0000 00:00:00.000) Represented as a single double.

Epoch16

Picoseconds since Year 0 (01-Jan-0000 00:00:00.000.000.000.000) Represented as two doubles (seconds and picoseconds).

Global Descriptor Record (GDR) - contains global information about the CDF file Points to variable and attribute descriptor records

GDR(buffer::Vector{UInt8}, pos, FieldSizeT)

Load a Global Descriptor Record from the buffer at the specified offset.

CDF Record header structure - common to all record types

TT2000

Nanoseconds since J2000 (01-Jan-2000 12:00:00.000.000.000) with leap seconds, represented as an 8-byte integer.

Variable Descriptor Record (VDR) - describes a single variable

VDR(io::IO, FieldSizeT)

Load a Variable Descriptor Record from the IO stream at the specified offset.

Variable Value Record (VVR) - contains actual variable data

Variable Index Record (VXR) - contains pointers to variable data records

VXR(source, offset, RecordSizeType)

Load a Variable Index Record from the source at the specified offset.

rVDR(io::IO, FieldSizeT)

Load a Variable Descriptor Record from the IO stream at the specified offset.

attrib(cdf::CDFDataset, attribute_name::String)

Retrieve all entries for a named attribute from the CDF file.

attrib(cdf::CDFDataset)

Load all attributes from the CDF file.

attribnames(cdf::CDFDataset)

Return a list of attribute names in the CDF file.

decode_cdr_flags(flags::UInt32)

Decode the CDR flags field into individual boolean flags.

CDF Flags (from CDF specification):

• Bit 0: Majority (1=row-major, 0=column-major)
• Bit 1: File format (1=single-file, 0=multi-file)
• Bit 2: Checksum used (1=checksum present, 0=no checksum)
• Bit 3: MD5 checksum method (requires bit 2=1)

is_big_endian_encoding(encoding)

Determine if a CDF encoding uses big-endian byte order based on CDF specification encoding values.

is_cdf_v3(magic_bytes)

Determine if this is a CDF v3 file based on the magic number.

Whether or not the variable is a non-record variable

load_attribute_entries(buffer::Vector{UInt8}, adr, RecordSizeType, cdf_encoding) -> Vector{AttributeEntry}

Load all attribute entries for a given attribute from its AEDRs.

read_uint32_be(io::IO)

Read a 32-bit unsigned integer in big-endian byte order. CDF format uses big-endian for record fields.

vattrib(cdf, varnum, name)

Optimized version that loads only the requested attribute for the given variable number. Much faster than loading all attributes when only one is needed.

vattrib(cdf::CDFDataset, varnum::Integer)

Get all variable attributes for a specific variable number.

@read_be_fields buffer pos T1 T2 ...

Unrolls sequential big-endian reads starting at pos within buffer. Returns a tuple of the parsed values and the updated position, mirroring read_be_i but without the runtime ntuple/offset bookkeeping.

Example

values, next = @read_be_fields buf pos UInt32 Int16