Bytearray uncompress filestream wirtebytes error

Question

Hi,
I'm developing an app for iPad in which I'm trying to load a zip file from a remote url, save that zip file on device and then trying to unzip it.
It works fine both on laptop(while testing as AIR app) and also on the device(iPad) if zip file size is around 20 - 25MB, with it's internal file being around 140 - 170MB. But if the zip file size is larger (beyond 50MB) then it crashes while uncompressing the zip file on the device. It still works fine on laptop though.
I'm compiling the code with Flex 4.6 overlayed on AIR 3.6
iPad iOS version is 6.0.1
Using following code :
private function unzipFile():void
            progress = progress + fileName +' unzipFile started\n';
            saveLog();
            var zStream:FileStream = new FileStream();
            var bytes:ByteArray = new ByteArray();
            var fileName:String = new String();
            var flNameLength:uint;
            var xfldLength:uint;
            var offset:uint;
            var compSize:uint;
            var uncompSize:uint;
            var compMethod:int;
            var signature:int;
            progress = progress + fileName +'before zStream.open\n';
            saveLog();
            zStream.open(fileLocal, FileMode.READ);
            progress = progress + fileName +'after zStream.open\n';
            saveLog();
            bytes.endian = Endian.LITTLE_ENDIAN; 
            while (zStream.position < fileLocal.size)
            {  progress = progress + fileName +'before zStream.readBytesbytes, 0, 30\n';
                saveLog();
                zStream.readBytes(bytes, 0, 30);
                progress = progress + fileName +'after zStream.readBytesbytes, 0, 30\n';
                saveLog();
                bytes.position = 0;
                signature = bytes.readInt();
                if (signature != 0x04034b50)
                    break;
                bytes.position = 8;
                compMethod = bytes.readByte();
                offset = 0;
                bytes.position = 26;
                flNameLength = bytes.readShort();   
                offset += flNameLength;    
                bytes.position = 28;    
                xfldLength = bytes.readShort();
                offset += xfldLength;   
                zStream.readBytes(bytes, 30, offset);
                bytes.position = 30;
                fileName = bytes.readUTFBytes(flNameLength); 
                bytes.position = 18;
                compSize = bytes.readUnsignedInt(); 
                bytes.position = 22;  
                uncompSize = bytes.readUnsignedInt();
                progress = progress + fileName +'before zStream.readBytes(bytes, 0, compSize)\n';
                saveLog();
                zStream.readBytes(bytes, 0, compSize);
                progress = progress + fileName +'after zStream.readBytes(bytes, 0, compSize)\n';
                saveLog();
                if (compMethod == 8)
                    try
                        progress = progress + fileName +' before bytes.uncompress\n';
                        saveLog();
                        bytes.uncompress(CompressionAlgorithm.DEFLATE);
                        //bytes.uncompress(CompressionAlgorithm.LZMA);
                        progress = progress + fileName +' after bytes.uncompress\n';
                        saveLog();
                        //outFile(fileName, bytes); 
                    catch(error:Error)
                        progress = progress + fileName +' bytes.uncompress catch\n';
                        saveLog();
              //write bytes to a file locally here
It fails on this line:
bytes.uncompress(CompressionAlgorithm.DEFLATE);
and gets inside catch block.
To avoid this problem, I also tried to load the file which was there in the zip file, directly using remote url(file size and download time will be more), but in this case, after loading the file, reading the bytearray data of it, when I try to write this bytearray to a filestream, it crashes again!
Just wanted to know if there is any file size limit on mobile\iOS devices, while unzipping a file and while writing bytearray data to a filesystem or am I doing something wrong here?
Kindly help, as I'm stuck with this and cannot really proceed on this project using Flex AIR if this doesn't work.
-Deepak

Answer 1

Hi Brent,
Yes, for unzipping, we really have to split the zip files, as uncompress doesn't work for larger zip files(it crashes on device).
Since my file size would range from 200-400MB, for now, I'm planning to load the raw file directly(running out of time to deliver it :| ). This too was failing, when i tried to readByes\writeBytes, after loading the file completely(since it runs out of memory). But I came across a solution right here:
http://stackoverflow.com/questions/14583247/air-as3-download-large-file-trough-ipad-applic ation
It basically writes data to the disk, as and when data gets downloaded in chunks. I felt it was a great idea! Apparantly, there won't be any memory issues too with that approach
And yes, as you have mentioned future plan would be to load and unzip zip files.
Thanks Brent, that helped too