/**
* gzip-compressed HTTP response.
*
* @throws Exception
*/
public void testCompressedHttpRecord() throws Exception {
String payload = "hogehogehogehogehoge";
String ctype = "text/plain";
WARCRecordInfo recinfo =
new TestWARCRecordInfo(
TestWARCRecordInfo.buildCompressedHttpResponseBlock(ctype, payload.getBytes()));
recinfo.setMimetype(ctype);
TestARCReader ar = new TestARCReader(recinfo);
ARCRecord rec = ar.get(0);
ArcResource res = new ArcResource(rec, ar);
res.parseHeaders();
assertEquals("statusCode", 200, res.getStatusCode());
assertEquals("content-type", ctype, res.getHeader("Content-Type"));
Resource zres = TextReplayRenderer.decodeResource(res);
assertTrue("wrapped with GzipDecodingResource", (zres instanceof GzipDecodingResource));
byte[] buf = new byte[payload.getBytes().length + 1];
int n = zres.read(buf);
assertEquals("content length", buf.length - 1, n);
res.close();
}
/**
* Read first {@code sniffLength} bytes of {@code resource}'s payload, decoding {@code
* Content-Encoding} if any. Reset {@code resource}'s read position back to zero.
*
* @param resource Resource to load bytes from
* @return bytes, zero-padded if payload is shorter.
* @throws IOException
*/
protected byte[] peekContent(Resource resource) throws IOException {
byte[] bbuffer = new byte[Math.max(sniffLength, MINIMUM_SNIFF_BUFFER_SIZE)];
String encoding = resource.getHeader("content-encoding");
if ("gzip".equalsIgnoreCase(encoding) || "x-gzip".equalsIgnoreCase(encoding)) {
// use larger readlimit, because gzip-ed data can be larger than the original
// at low compression level.
resource.mark(sniffLength + 100);
@SuppressWarnings("resource")
Resource z = new GzipDecodingResource(resource);
z.read(bbuffer, 0, sniffLength);
resource.reset();
} else {
resource.mark(sniffLength);
resource.read(bbuffer, 0, sniffLength);
resource.reset();
}
return bbuffer;
}
/**
* @param charSet
* @throws IOException
*/
public void readFully(String charSet) throws IOException {
this.charSet = charSet;
int recordLength = (int) resource.getRecordLength();
// convert bytes to characters for charset:
InputStreamReader isr = new InputStreamReader(resource, charSet);
char[] cbuffer = new char[C_BUFFER_SIZE];
// slurp the whole thing into RAM:
sb = new StringBuilder(recordLength);
// Skip the UTF-8 BOM 0xFEFF
int firstChar = isr.read();
if ((firstChar != '\uFEFF') && (firstChar != -1)) {
sb.append((char) firstChar);
}
for (int r = -1; (r = isr.read(cbuffer, 0, C_BUFFER_SIZE)) != -1; ) {
sb.append(cbuffer, 0, r);
}
}
