I don't know if my phone is 32 or 64 bit. But J is 32 bit. On Oct 7, 2014 8:32 AM, "Don Guinn" <[email protected]> wrote:
> I took your numbers and did > > 256#.0 0 1 72 224 139 75 119 > 1412516498295 > > on my laptop and phone and got the same answer. But it was an integer on > my laptop and floating on my phone. Might have to flip things around if you > have enian problems. > > On Tue, Oct 7, 2014 at 8:23 AM, Joe Bogner <[email protected]> wrote: > >> Thanks Don. It looks like a long on java is 64 bits, 8 bytes, even on a 32 >> bit architecture. >> >> I'm uncertain of the right path forward. I could change the java app to >> log >> a 4 byte timestamp, or I can figure out a way to convert the 8 byte to an >> integer on J32. >> >> The simplest change would be to have some mechanism to convert it on J32, >> but it would need to be fairly performant as the file may easily have 1M+ >> values. >> >> >> >> On Tue, Oct 7, 2014 at 9:34 AM, Don Guinn <[email protected]> wrote: >> >> > I checked IF64 on my phone and its value was zero. Help says that >> _3(3!:4) >> > only is valid for J64 >> > >> > On Tue, Oct 7, 2014 at 6:00 AM, Joe Bogner <[email protected]> wrote: >> > >> > > I have a android app that is logging sensor data to a text file using >> > > java's DataInputStream.writeLong[1]. It is logging a timestamp each >> > > time it gets a step (pedometer). The timestamp comes from getTime() >> > > [2] >> > > >> > > I can read the file on Windows >> > > >> > > bin=:fread 'c:/joe/j/steps.txt' >> > > times=: |. _3 (3!:4) |. bin >> > > >> > > 0 { times >> > > 1412516498295 >> > > >> > > tsFrom1970 =: (1970 1 1 0 0 0) tsPlus 1000 %~ ] >> > > tsFrom1970 1 { times >> > > 2014 10 5 13 41 38.644 >> > > >> > > >> > > Here is the first number: >> > > >> > > a. i. (8{. bin) >> > > 0 0 1 72 224 139 75 119 >> > > >> > > _3 (3!:4) |. (0 0 1 72 224 139 75 119 { a.) >> > > 1412516498295 >> > > >> > > I'd like to create a JHS app on android to show some metrics. I'm >> > > unable to convert the binary data to a jint on android: >> > > >> > > _3 (3!:4) bin >> > > |domain error >> > > | _3 (3!:4)bin >> > > _3 (3!:4) |. bin >> > > |domain error >> > > | _3 (3!:4)|.bin >> > > >> > > >> > > The results are the same on Windows. I reverse it on Windows because >> > > of big/little endian (I think) >> > > >> > > I could try to manually implement the encoding that's used by >> > > writeLong but figured I'd ask here first to see if I'm missing >> > > something simple >> > > >> > > [1] - >> > > >> > >> http://docs.oracle.com/javase/7/docs/api/java/io/DataOutput.html#writeLong(long) >> > > >> > > [2] - >> > > >> http://docs.oracle.com/javase/7/docs/api/java/util/Date.html#getTime() >> > > ---------------------------------------------------------------------- >> > > For information about J forums see >> http://www.jsoftware.com/forums.htm >> > > >> > ---------------------------------------------------------------------- >> > For information about J forums see http://www.jsoftware.com/forums.htm >> > >> ---------------------------------------------------------------------- >> For information about J forums see http://www.jsoftware.com/forums.htm >> > > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
