Unfortunately I am dealing with another library that wants to fetch from URLs itself. I was hoping to wedge a "cjs://" protocol handler in but to do that id need raw byte access to JCS.
Its okay though because I found a leverage point higher up in the other stack where I can take control of the actual loading process. On Wed, Aug 26, 2009 at 12:55 PM, Stuart Maclean <stu...@apl.washington.edu>wrote: > Hi Jeffrey, I am doing a similar thing to you. I get Image objects from > remote URLs, then convert to int[] and store the int[] in JCS. I really did > this since Image is not serializable. > > You could read all bytes from your handler until read returns -1, store all > into a ByteArrayOutputStream, then call toByteArray on that and had the > byte[] to JCS? > > Stuart > > > On Aug 25, 2009, at 6:06 PM, Jeffrey Kesselman wrote: > > Hi Guys, >> >> Im trying to mate JCS at the moment with another system that expects to >> access data as bytes vi a URL. As sun Im writing a jcs protocol handler. >> >> The problem is that, as alluded to above, a protocl handler expcts to hand >> back its data as a byte stream. >> >> Is there a way to push thinsg into and take thinsg otu of the JCS cache as >> blocks of bytes? >> -- >> ~~ Microsoft help desk says: reply hazy, ask again later. ~~ >> > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: jcs-users-unsubscr...@jakarta.apache.org > For additional commands, e-mail: jcs-users-h...@jakarta.apache.org > > -- ~~ Microsoft help desk says: reply hazy, ask again later. ~~
