On Mon, May 18, 2020 at 10:30 AM J Decker <[email protected]> wrote: > > > On Mon, May 18, 2020 at 6:02 AM Eric Mandel <[email protected]> > wrote: > >> Hi Talha, >> >> Our browser-based astronomical image display (https://js9.si.edu) is >> transformed into a full desktop app using Electron and a few hundred lines >> of wrapper code. Having the same code base for both environments is a huge >> benefit. We also very cautiously allow use of Electron's node capabilities >> to make the underlying file system available on the desktop app.The main >> drawback for us (compared to a pure desktop app using Aqua or X11) is lack >> of mmap() and the limit on the amount of available memory: our files are >> large (> 100Mb) or even huge (> 1Gb) and special techniques are required to >> cut down on memory use. >> > > re no mmap... > https://www.npmjs.com/package/sack.vfs has a mapFile() function which > isn't very complex itself, that just returns an ArrayBuffer... > https://github.com/d3x0r/sack.vfs/blob/master/src/vfs_module.cc#L921-L960 > >
Sadly even if node supports mmap, using methods such as this, there is no way (today) to make that usable by a WebAssebmly module. > > Also, `--max-old-space-size=8192 ` should also work as an electron option > to increase the JS heap size. (which apparently needs to be wrapped in an > option https://github.com/electron/electron/issues/2056 > --js-flags='--max-old-space-size=10192' > ) > > > >> We use Emscripten to compile cpu-intensive astronomical i/o, >> reprojection, and world coordinate system libraries to wasm. Although the >> results run at near-native speed thanks to Emscripten, our data processing >> takes much more time than an ordinary web app. As a result, we probably >> don't notice Electron-specific inefficiencies that others might. >> >> Regards, >> >> Eric >> >> -- >> You received this message because you are subscribed to the Google Groups >> "emscripten-discuss" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/emscripten-discuss/7017d53b-c824-4ea3-a5f8-1e6c941f48b6%40googlegroups.com >> <https://groups.google.com/d/msgid/emscripten-discuss/7017d53b-c824-4ea3-a5f8-1e6c941f48b6%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > -- > You received this message because you are subscribed to the Google Groups > "emscripten-discuss" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/emscripten-discuss/CAA2GJqU24CyweeeR7hpwzm5MsGNtijY9nsRJm4V2O8xDUDQ_uw%40mail.gmail.com > <https://groups.google.com/d/msgid/emscripten-discuss/CAA2GJqU24CyweeeR7hpwzm5MsGNtijY9nsRJm4V2O8xDUDQ_uw%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/emscripten-discuss/CAL_va2-xE4Rv%2B3FPsP0F7y82a1sKZHsm5Zn706_51K-_-Yo2PQ%40mail.gmail.com.
