NodeJS addon for reading and writing the files in the Network Common Data Form (NetCDF) version <= 4, built upon the C-library for netcdf.
netcdf4-js is built with nodejs >= 4.x
Install using npm:
$ npm install netcdf4
Prerequisites:
You will need libnetcdf >= 4.x installed.
- Install NetCDF4 using your package manager, e.g., on Ubuntu/Debian:
$ sudo apt-get install libnetcdf-dev
or download it here
- Make sure your system fulfills all the prerequisites of node-gyp
- Install NetCDF4 from here
- Make sure to select at least "dependencies", "headers" and "libraries" to install in the NetCDF installation wizard
- Install the build tools as described here
- Set the environment variable
NETCDF_DIRto your NetCDF installation, e.g.,
C:\> set NETCDF_DIR=C:\Program Files\netCDF 4.6.1Open files with
var netcdf4 = require("netcdf4");
var file = new netcdf4.File("test/testrh.nc", "r");
File modes are "r" for "reading", "w" for "writing", "c" for
"creation", and "c!" for "overwriting".
Then you can read variables using read or readSlice. The following example reads values at positions 5 to 15:
console.log(file.root.variables['var1'].readSlice(5, 10));
Properties marked (r/w) can be read and will be written to the file when set.
Properties:
root: MainGroup-object in file
Methods:
close(): Close filesync(): Sync (or "flush") file to disk
Properties:
id: ID used by C-libraryname: Namefullname: Full name (path in file)variables: Associative array of variables in groupdimensions: Associative array of dimensions in groupunlimited: Associative array of unlimited Dimensions in groupattribute: Associative array of attributes of groupsubgroups: Associative array of subgroups of group
Methods:
addVariable(name, type, dimensions): Add a new variable in group.typeis one of"byte", "char", "short", "int", "ubyte", "ushort", "uint", "float", "double".dimensionsis an array of ids of dimensions for the new variable. Returns new variable.addDimension(name, length): Add new dimension of lengthlength(can be"unlimited"for unlimited dimension). Returns new dimension.addSubgroup(name): Add subgroup. Returns new subgroup.addAttribute(name, type, value): Add and set new attribute. Returns new attribute.
Properties:
id: ID used by C-libraryname: Name (r/w)length: Length or currently used length for unlimited dimensions
Properties:
id: ID used by C-libraryname: Name (r/w)value: Value (r/w)
Methods:
delete(): Delete attribute
Properties:
id: ID used by C-libraryname: Name (r/w)type: Type of variableattributes: Associative array of attributesdimensions: Array of dimensions used by variableendianness: Endianness:"little","big", or"native"(r/w)checksummode: Checksum mode:"none", or"fletcher32"(r/w)chunkmode: Chunk mode:"contiguous", or"chunked"(r/w)chunksizes: Array of chunk sizes (one size per dimension) (r/w)fillmode: Boolean switch for fill mode (r/w)fillvalue: Fill value (r/w)compressionshuffle: Boolean switch for shuffle (r/w)compressiondeflate: Boolean switch for compression (r/w)compressionlevel: Compression level (1-9) (r/w)
Methods:
read(pos....): Reads and returns a single value at positions given as forwrite.readSlice(pos, size....): Reads and returns an array of values (cf. "Specify a Hyperslab") at positions and sizes given for each dimension,readSlice(pos1, size1, pos2, size2, ...)e.g.readSlice(2, 3, 4, 2)gives an array of the values at position 2 for 3 steps along the first dimension and position 4 for 2 steps along the second one.readStridedSlice(pos, size, stride....): Similar toreadSlice(), but it adds a stride (interval between indices) parameter to each dimension. If stride is 4, the function will take 1 value, discard 3, take 1 again, etc. So for instancereadStridedSlice(2, 3, 2, 4, 2, 1)gives an array of the values at position 2 for 3 steps with stride 2 (i.e. every other value) along the first dimension and position 4 for 2 steps with stride 1 (i.e. with no dropping) along the second dimension.write(pos..., value): Writevalueat positions given, e.g.write(2, 3, "a")writes"a"at position 2 along the first dimension and position 3 along the second one.writeSlice(pos, size..., valuearray): Write values invaluearray(must be a typed array) at positions and sizes given for each dimension, e.g.writeSlice(2, 3, 4, 2, new Int32Array([0, 1, 2, 3, 4, 5]))writes the array at position 2 for 3 steps along the first dimension and position 4 for 2 step along the second one (cf. "Specify a Hyperslab").writeStridedSlice(pos, size, stride..., valuearray): Similar towriteSlice(), but it adds a stride parameter to each dimension. So for instancewriteStridedSlice(2, 3, 2, 4, 2, 1), new Int32Array([0, 1, 2, 3, 4, 5])writes the array at position 2 for 3 steps with stride 2 (i.e. every other value) along the first dimension and position 4 for 2 steps with stride 1 (i.e. with no dropping) along the second dimension.addAttribute(name, type, value): Adds and sets new attribute. Returns new attribute.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent
