Spaces:
Sleeping
Sleeping
File size: 14,188 Bytes
6ee917b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 | import Foundation
import OmFileFormat
import Vapor
import SwiftNetCDF
/**
Small helper tool to convert between `om` versions and between om file format and NetCDF for debugging
Examples:
- Convert to NetCDF: openmeteo-api convert-om data.om --format netcdf -o output.nc --domain ecmwf_ifs025
- Convert between OM versions: openmeteo-api convert-om data.om --format om3 -o data.om3 --domain ecmwf_ifs025
*/
struct ConvertOmCommand: Command {
var help: String {
return "Convert between om file format version or convert to NetCDF"
}
struct Signature: CommandSignature {
@Argument(name: "infile", help: "Input file")
var infile: String
@Option(name: "format", help: "Conversion target format: 'netcdf' or 'om3'")
var format: String?
@Option(name: "output", short: "o", help: "Output file name. Default: [infile].nc or [infile].om3")
var outfile: String?
@Flag(name: "transpose", help: "Transpose data to fast space")
var transpose: Bool
@Option(name: "domain", help: "Domain used for grid definition")
var domain: String?
}
func run(using context: CommandContext, signature: Signature) throws {
let logger = context.application.logger
logger.info("Processing file: \(signature.infile)")
let format = signature.format?.lowercased() ?? "netcdf"
if format == "om3" {
// Handle conversion to OM3
guard let domain = signature.domain else {
throw ConvertOmError("Domain parameter is required for OM3 conversion")
}
let domainObj = try DomainRegistry.load(rawValue: domain)
let outfile = signature.outfile ?? signature.infile.withoutOmSuffix + ".om3"
if signature.transpose {
logger.warning("Transpose flag is currently not supported for OM3 conversion")
}
logger.info("Converting OM file to v3 with domain: \(domain). Outfile will be: \(outfile)")
try convertOmv3(src: signature.infile, dest: outfile, grid: domainObj.getDomain().grid)
return
} else if format == "netcdf" {
// Handle conversion to NetCDF
guard let om = try OmFileReader(file: signature.infile).asArray(of: Float.self) else {
throw ConvertOmError("Not a float array")
}
let dimensions = om.getDimensions().map { $0 }
let chunks = om.getChunkDimensions().map { $0 }
logger.info("File dimensions: \(dimensions), chunks: \(chunks)")
let data = try om.read()
let outfile = signature.outfile ?? signature.infile.withoutOmSuffix + ".nc"
logger.info("Converting to NetCDF: \(outfile)")
try convertToNetCDF(data: data, dimensions: dimensions, outfile: outfile, transpose: signature.transpose, domain: signature.domain, logger: logger)
return
} else {
throw ConvertOmError("Unsupported conversion target: \(format)")
}
}
/// Convert data to NetCDF format
private func convertToNetCDF(data: [Float], dimensions: [UInt64], outfile: String, transpose: Bool, domain: String?, logger: Logger) throws {
let ncFile = try NetCDF.create(path: outfile, overwriteExisting: true)
try ncFile.setAttribute("TITLE", "open-meteo data")
switch dimensions.count {
case 2:
try convertToNetCDF2D(data: data, dimensions: dimensions, ncFile: ncFile, transpose: transpose, domain: domain, logger: logger)
case 3:
try convertToNetCDF3D(data: data, dimensions: dimensions, ncFile: ncFile, transpose: transpose)
default:
logger.error("Unsupported number of dimensions for netcdf conversion: \(dimensions.count)")
throw ConvertOmError("Unsupported number of dimensions: \(dimensions.count)")
}
logger.info("NetCDF conversion completed successfully")
}
/// Handle 2D data conversion to NetCDF
private func convertToNetCDF2D(data: [Float], dimensions: [UInt64], ncFile: Group, transpose: Bool, domain: String?, logger: Logger) throws {
if let domain = domain {
let grid = try DomainRegistry.load(rawValue: domain).getDomain().grid
let ny = grid.ny
let nx = grid.nx
let nt = Int(dimensions[1])
guard dimensions[0] == nx * ny, ny > 1, nx > 1 else {
throw ConvertOmError("Wrong grid! Expected \(nx * ny) locations, got \(dimensions[0])")
}
if transpose {
// Fast time dimension (locations, time) -> (time, locations)
var ncVariable = try ncFile.createVariable(name: "data", type: Float.self, dimensions: [
try ncFile.createDimension(name: "time", length: nt),
try ncFile.createDimension(name: "LAT", length: ny),
try ncFile.createDimension(name: "LON", length: nx)
])
let transposedData = Array2DFastTime(data: data, nLocations: nx * ny, nTime: nt).transpose()
try ncVariable.write(transposedData.data)
} else {
// Default layout
var ncVariable = try ncFile.createVariable(name: "data", type: Float.self, dimensions: [
try ncFile.createDimension(name: "LAT", length: ny),
try ncFile.createDimension(name: "LON", length: nx),
try ncFile.createDimension(name: "time", length: nt)
])
try ncVariable.write(data)
}
} else {
logger.warning("No domain provided, converting to LAT and LON dimensions, which might not be what you want for weather domains!")
logger.warning("If you want to convert to a proper 3-dimensional NetCDF file, please provide a domain (for grid dimensions).")
// Default layout
var ncVariable = try ncFile.createVariable(name: "data", type: Float.self, dimensions: [
try ncFile.createDimension(name: "LAT", length: Int(dimensions[0])),
try ncFile.createDimension(name: "LON", length: Int(dimensions[1]))
])
try ncVariable.write(data)
}
}
/// Handle 3D data conversion to NetCDF
private func convertToNetCDF3D(data: [Float], dimensions: [UInt64], ncFile: Group, transpose: Bool) throws {
let ny = Int(dimensions[0])
let nx = Int(dimensions[1])
let nt = Int(dimensions[2])
if transpose {
// Transpose to fast space (lat, lon, time) -> (time, lat, lon)
var ncVariable = try ncFile.createVariable(name: "data", type: Float.self, dimensions: [
try ncFile.createDimension(name: "time", length: nt),
try ncFile.createDimension(name: "LAT", length: ny),
try ncFile.createDimension(name: "LON", length: nx)
])
let transposed = Array3D(data: data, dim0: ny, dim1: nx, dim2: nt).transpose()
try ncVariable.write(transposed.data)
} else {
var ncVariable = try ncFile.createVariable(name: "data", type: Float.self, dimensions: [
try ncFile.createDimension(name: "LAT", length: ny),
try ncFile.createDimension(name: "LON", length: nx),
try ncFile.createDimension(name: "time", length: nt)
])
try ncVariable.write(data)
}
}
/// Read om file and write it as version 3 and reshape data to proper 3d files
func convertOmv3(src: String, dest: String, grid: Gridable) throws {
// Read data from the input OM file
guard let readfile = try? OmFileReader(fn: try MmapFile(fn: FileHandle.openFileReading(file: src))),
let reader = readfile.asArray(of: Float.self) else {
throw ConvertOmError("Failed to open file: \(src)")
}
let dimensions = Array(reader.getDimensions())
let chunks = Array(reader.getChunkDimensions())
print("compression: \(reader.compression)")
print("dimensions: \(dimensions)")
print("chunks: \(chunks)")
print("scaleFactor: \(reader.scaleFactor)")
let ny = UInt64(grid.ny)
let nx = UInt64(grid.nx)
let nt = dimensions[1]
guard dimensions.count == 2, nx * ny == dimensions[0], ny > 1, nx > 1 else {
throw ConvertOmError("Wrong grid! Expected \(nx * ny) locations, got \(dimensions[0])")
}
let dimensionsOut = [ny, nx, nt]
let chunksOut = [1,chunks[0],chunks[1]]
// TODO somehow 5x5 is larger than 1x25....
/*let dataRaw = try reader.read(range: [0..<ny*nx, 0..<nt])
print("data read")
if false {
let ncFile = try NetCDF.create(path: "\(dest).nc", overwriteExisting: true)
try ncFile.setAttribute("TITLE", "open-meteo file convert")
// to fast space
var ncVariable = try ncFile.createVariable(name: "data", type: Float.self, dimensions: [
try ncFile.createDimension(name: "time", length: Int(nt)),
try ncFile.createDimension(name: "LAT", length: Int(ny)),
try ncFile.createDimension(name: "LON", length: Int(nx))
])
let data2 = Array2DFastTime(data: dataRaw, nLocations: Int(nx*ny), nTime: Int(nt)).transpose()
print("transpose done")
try ncVariable.write(data2.data)
print("nc wwrite done")
return
}*/
try FileManager.default.removeItemIfExists(at: dest)
let fileHandle = try FileHandle.createNewFile(file: dest)
// Write the compressed data to the output OM file
let fileWriter = OmFileWriter(fn: fileHandle, initialCapacity: 1024 * 1024 * 10) // Initial capacity of 10MB
print("created writer")
let writer = try fileWriter.prepareArray(
type: Float.self,
dimensions: dimensionsOut,
chunkDimensions: chunksOut,
compression: reader.compression,
scale_factor: reader.scaleFactor,
add_offset: reader.addOffset
)
print("prepared array")
/// Reshape data from flated 2D to 3D context
for yStart in stride(from: 0, to: ny, by: UInt64.Stride(chunksOut[0])) {
for xStart in stride(from: 0, to: nx, by: UInt64.Stride(chunksOut[1])) {
for tStart in stride(from: 0, to: nt, by: UInt64.Stride(chunksOut[2])) {
let yRange = yStart ..< min(yStart + chunksOut[0], ny)
let xRange = xStart ..< min(xStart + chunksOut[1], nx)
let tRange = tStart ..< min(tStart + chunksOut[2], nt)
//print("chunk y=\(yRange) x=\(xRange) t=\(tRange)")
var chunk = [Float](repeating: .nan, count: yRange.count * xRange.count * tRange.count)
for (row, y) in yRange.enumerated() {
try reader.read(
into: &chunk,
range: [y * nx + xRange.startIndex ..< y * nx + xRange.endIndex, tRange],
intoCubeOffset: [UInt64(row * xRange.count), 0],
intoCubeDimension: [UInt64(yRange.count * xRange.count), UInt64(tRange.count)]
)
}
try writer.writeData(
array: chunk,
arrayDimensions: [UInt64(yRange.count), UInt64(xRange.count), UInt64(tRange.count)],
arrayOffset: nil,
arrayCount: nil
)
}
}
}
let variableMeta = try writer.finalise()
print("Finalized Array")
let variable = try fileWriter.write(array: variableMeta, name: "data", children: [])
try fileWriter.writeTrailer(rootVariable: variable)
print("Finished writing")
/*// Verify the output
guard let verificationFile = try? OmFileReader(fn: try MmapFile(fn: FileHandle.openFileReading(file: dest))),
let verificationReader = verificationFile.asArray(of: Float.self) else {
fatalError("Failed to open file: \(dest)")
}
let dataVerify = try verificationReader.read(range: [0..<ny, 0..<nx, 0..<nt])
guard dataVerify == dataRaw else {
for i in 0..<min(dataVerify.count, 1000) {
if dataVerify[i] != dataRaw[i] {//}&& !dataRaw[i].isNaN && !dataVerify[i].isNaN {
print(i, dataVerify[i], dataRaw[i])
}
}
print("verify failed")
fatalError()
}*/
}
}
struct ConvertOmError: Error {
let message: String
init(_ message: String) {
self.message = message
}
}
extension String {
var withoutOmSuffix: String {
if hasSuffix(".om") {
return String(dropLast(3))
}
return self
}
}
extension Array3D {
/// Transpose the array to swap dimensions: (dim0, dim1, dim2) -> (dim2, dim0, dim1)
/// This effectively changes from (lat, lon, time) to (time, lat, lon)
func transpose() -> Array3D {
precondition(data.count == dim0 * dim1 * dim2)
return data.withUnsafeBufferPointer { data in
let out = [Float](unsafeUninitializedCapacity: data.count) { buffer, initializedCount in
for d0 in 0..<dim0 {
for d1 in 0..<dim1 {
for d2 in 0..<dim2 {
// From (d0, d1, d2) to (d2, d0, d1)
let srcIdx = d0 * dim1 * dim2 + d1 * dim2 + d2
let dstIdx = d2 * dim0 * dim1 + d0 * dim1 + d1
buffer[dstIdx] = data[srcIdx]
}
}
}
initializedCount = data.count
}
return Array3D(data: out, dim0: dim2, dim1: dim0, dim2: dim1)
}
}
}
|