How to read local files using browser (no-build) #43
Comments
|
If you're loading files in the browser, you have to do a little bit more - (you have to load the file data to an Arraybuffer, then into a Uint8Array and write that to the virtual emscripten file system), If you're using the no-build browser example, every step in that section of the readme is needed in order to load a file from a URL! I suppose it might be nice to consider adding a helper function async h5wasm.fromURL(url: string): Promise<h5wasm.File>that awaits a fetch, then writes to a random local file, and returns an h5wasm.File object but I don't have anything like that in the library at the moment. |
|
Thanks. Hopefully, you will have something soon. I did the following and it can access the keys. 🎉 ✅ import h5wasm from "https://cdn.jsdelivr.net/npm/h5wasm@0.4.0/dist/esm/hdf5_hl.js";
class Read_HDF5{
constructor(h5_file){
this.h5_file = h5_file;
this.random_name = "random.h5"
}
async getKeys(){
// the WASM loads asychronously, and you can get the module like this:
const Module = await h5wasm.ready;
// then you can get the FileSystem object from the Module:
const { FS } = Module;
console.log(this.h5_file)
let response = await fetch(this.h5_file)
let ab = await response.arrayBuffer();
FS.writeFile(this.random_name, new Uint8Array(ab));
let f = new h5wasm.File(this.random_name, "r");
console.log(f.keys())
let data = f.get("12_rd_p");
console.log(data)
console.log(data.value)
}
}
export {Read_HDF5};As it's write the data to the virtual emscripten file system, is there any limit for the input files. Will ~250 MB HDF5 file create any issue for the performance? |
|
You shouldn't have any issues with files that size. Files > 2GB will have issues with the max size of an ArrayBuffer in javascript. There are ways to read even larger files using HTTP range requests, or web workers and special emscripten filesystems (see discussion in #40 for instance), etc. but it gets more complicated. You can use direct filesystem random access with nodejs instead of a browser if you want to make a native tool/program without limits on file size. |
Hi,
I am trying to read a HDF5 file stored locally.
The file is available at "http://localhost:8080/examples/h5_files/test_data/test.h5". So providing the relative path as input for the file like
/examples/h5_files/test_data/test.h5Test code.
file: HDF5_Reader.js
but getting error something like this.
The text was updated successfully, but these errors were encountered: