Lines Matching defs:data
27 #include "ordered-data.h"
787 * Heuristic uses systematic sampling to collect data from the input data
791 * @SAMPLING_INTERVAL - range from which the sampled data can be collected
797 * For statistical analysis of the input data we consider bytes that form a
809 * If the data contain the maximum number of symbols, which is 256, we obtain a
812 * For a sample of at most 8KB of data per data range: 16 consecutive bytes
823 /* Partial copy of input data */
1169 * pages_in is an array of pages with compressed data.
1178 * The pages in the bio are for the uncompressed data, and they may not
1196 * a less complex decompression routine. Our compressed data fits in a
1198 * start_byte tells us the offset into the compressed data we're interested in
1231 * Copy uncompressed data from working buffer to pages.
1252 * copying into relative to the start of the compressed data.
1256 /* we haven't yet hit data corresponding to this page */
1261 * the start of the data we care about is offset into
1329 * Pure byte distribution analysis fails to determine compressibility of data.
1331 * needed to encode the sampled data.
1339 * @ENTROPY_LVL_HIGH - data are not compressible with high probability
1450 * Normal radix expects to move data from a temporary array, to
1481 * There are several types of structured binary data that use nearly all byte
1483 * nearly the same (eg. encrypted data). Unlikely to be compressible.
1485 * Other possibility is normal (Gaussian) distribution, where the data could
1491 * @BYTE_CORE_SET_HIGH - data have uniform distribution and with high
1524 * This heuristic can detect textual data (configs, xml, json, html, etc).
1525 * Because in most text-like data byte set is restricted to limited number of
1526 * possible characters, and that restriction in most cases makes data easy to
1529 * @BYTE_SET_THRESHOLD - consider all data within this byte set size:
1548 * the detection technique would fail for this type of data.
1564 const u8 *data = ws->sample;
1566 return memcmp(&data[0], &data[half_of_sample], half_of_sample) == 0;
1578 * Compression handles the input data by chunks of 128KiB
1625 * quickly (compared to direct compression) detect data characteristics
1627 * data.
1630 * - detect mostly zero data
1631 * - detect data with low "byte set" size (text, etc)
1632 * - detect data with low/high "core byte" set
1690 * 1. it is possible to defrag the data later
1692 * 2. the data would turn out to be hardly compressible, eg. 150 byte
1694 * be confused. This can happen when data have some internal repeated