/*
 * jcphuff.c
 *
 * Copyright (C) 1995-1997, Thomas G. Lane.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
 * This file contains Huffman entropy encoding routines for progressive JPEG.
 *
 * We do not support output suspension in this module, since the library
 * currently does not allow multiple-scan files to be written with output
 * suspension.
 */

#define JPEG_INTERNALS
#include "jinclude.h"
#include "jpeglib.h"
#include "jchuff.h"		/* Declarations shared with jchuff.c */

#ifdef C_PROGRESSIVE_SUPPORTED

/* Expanded entropy encoder object for progressive Huffman encoding. */

typedef struct {
  struct jpeg_entropy_encoder pub; /* public fields */

  /* Mode flag: TRUE for optimization, FALSE for actual data output */
  boolean gather_statistics;

  /* Bit-level coding status.
   * next_output_byte/free_in_buffer are local copies of cinfo->dest fields.
   */
  JOCTET * next_output_byte;	/* => next byte to write in buffer */
  size_t free_in_buffer;	/* # of byte spaces remaining in buffer */
  INT32 put_buffer;		/* current bit-accumulation buffer */
  int put_bits;			/* # of bits now in it */
  j_compress_ptr cinfo;		/* link to cinfo (needed for dump_buffer) */

  /* Coding status for DC components */
  int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */

  /* Coding status for AC components */
  int ac_tbl_no;		/* the table number of the single component */
  unsigned int EOBRUN;		/* run length of EOBs */
  unsigned int BE;		/* # of buffered correction bits before MCU */
  char * bit_buffer;		/* buffer for correction bits (1 per char) */
  /* packing correction bits tightly would save some space but cost time... */

  unsigned int restarts_to_go;	/* MCUs left in this restart interval */
  int next_restart_num;		/* next restart number to write (0-7) */

  /* Pointers to derived tables (these workspaces have image lifespan).
   * Since any one scan codes only DC or only AC, we only need one set
   * of tables, not one for DC and one for AC.
   */
  c_derived_tbl * derived_tbls[NUM_HUFF_TBLS];

  /* Statistics tables for optimization; again, one set is enough */
  long * count_ptrs[NUM_HUFF_TBLS];
} phuff_entropy_encoder;

typedef phuff_entropy_encoder * phuff_entropy_ptr;

/* MAX_CORR_BITS is the number of bits the AC refinement correction-bit
 * buffer can hold.  Larger sizes may slightly improve compression, but
 * 1000 is already well into the realm of overkill.
 * The minimum safe size is 64 bits.
 */

#define MAX_CORR_BITS  1000	/* Max # of correction bits I can buffer */

/* IRIGHT_SHIFT is like RIGHT_SHIFT, but works on int rather than INT32.
 * We assume that int right shift is unsigned if INT32 right shift is,
 * which should be safe.
 */

#ifdef RIGHT_SHIFT_IS_UNSIGNED
#define ISHIFT_TEMPS	int ishift_temp;
#define IRIGHT_SHIFT(x,shft)  \
	((ishift_temp = (x)) < 0 ? \
	 (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \
	 (ishift_temp >> (shft)))
#else
#define ISHIFT_TEMPS
#define IRIGHT_SHIFT(x,shft)	((x) >> (shft))
#endif

/* Forward declarations */
METHODDEF(boolean) encode_mcu_DC_first JPP((j_compress_ptr cinfo,
					    JBLOCKROW *MCU_data));
METHODDEF(boolean) encode_mcu_AC_first JPP((j_compress_ptr cinfo,
					    JBLOCKROW *MCU_data));
METHODDEF(boolean) encode_mcu_DC_refine JPP((j_compress_ptr cinfo,
					     JBLOCKROW *MCU_data));
METHODDEF(boolean) encode_mcu_AC_refine JPP((j_compress_ptr cinfo,
					     JBLOCKROW *MCU_data));
METHODDEF(void) finish_pass_phuff JPP((j_compress_ptr cinfo));
METHODDEF(void) finish_pass_gather_phuff JPP((j_compress_ptr cinfo));


/*
 * Initialize for a Huffman-compressed scan using progressive JPEG.
 */

METHODDEF(void)
start_pass_phuff (j_compress_ptr cinfo, boolean gather_statistics)
{  
  phuff_entropy_ptr entropy = (phuff_entropy_ptr) cinfo->entropy;
  boolean is_DC_band;
  int ci, tbl;
  jpeg_component_info * compptr;

  entropy->cinfo = cinfo;
  entropy->gather_statistics = gather_statistics;

  is_DC_band = (cinfo->Ss == 0);

  /* We assume jcmaster.c already validated the scan parameters. */

  /* Select execution routines */
  if (cinfo->Ah == 0) {
    if (is_DC_band)
      entropy->pub.encode_mcu = encode_mcu_DC_first;
    else
      entropy->pub.encode_mcu = encode_mcu_AC_first;
  } else {
    if (is_DC_band)
      entropy->pub.encode_mcu = encode_mcu_DC_refine;
    else {
      entropy->pub.encode_mcu = encode_mcu_AC_refine;
      /* AC refinement needs a correction bit buffer */
      if (entropy->bit_buffer == NULL)
	entropy->bit_buffer = (char *)
	  (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
				      MAX_CORR_BITS * SIZEOF(char));
    }
  }
  if (gather_statistics)
    entropy->pub.finish_pass = finish_pass_gather_phuff;
  else
    entropy->pub.finish_pass = finish_pass_phuff;

  /* Only DC coefficients may be interleaved, so cinfo->comps_in_scan = 1
   * for AC coefficients.
   */
  for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
    compptr = cinfo->cur_comp_info[ci];
    /* Initialize DC predictions to 0 */
    entropy->last_dc_val[ci] = 0;
    /* Get table index */
    if (is_DC_band) {
      if (cinfo->Ah != 0)	/* DC refinement needs no table */
	continue;
      tbl = compptr->dc_tbl_no;
    } else {
      entropy->ac_tbl_no = tbl = compptr->ac_tbl_no;
    }
    if (gather_statistics) {
      /* Check for invalid table index */
      /* (make_c_derived_tbl does this in the other path) */
      if (tbl < 0 || tbl >= NUM_HUFF_TBLS)
        ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl);
      /* Allocate and zero the statistics tables */
      /* Note that jpeg_gen_optimal_table expects 257 entries in each table! */
      if (entropy->count_ptrs[tbl] == NULL)
	entropy->count_ptrs[tbl] = (long *)
	  (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
				      257 * SIZEOF(long));
      MEMZERO(entropy->count_ptrs[tbl], 257 * SIZEOF(long));
    } else {
      /* Compute derived values for Huffman table */
      /* We may do this more than once for a table, but it's not expensive */
      jpeg_make_c_derived_tbl(cinfo, is_DC_band, tbl,
			      & entropy->derived_tbls[tbl]);
    }
  }

  /* Initialize AC stuff */
  entropy->EOBRUN = 0;
  entropy->BE = 0;

  /* Initialize bit buffer to empty */
  entropy->put_buffer = 0;
  entropy->put_bits = 0;

  /* Initialize restart stuff */
  entropy->restarts_to_go = cinfo->restart_interval;
  entropy->next_restart_num = 0;
}


/* Outputting bytes to the file.
 * NB: these must be called only when actually outputting,
 * that is, entropy->gather_statistics == FALSE.
 */

/* Emit a byte */
#define emit_byte(entropy,val)  \
	{ *(entropy)->next_output_byte++ = (JOCTET) (val);  \
	  if (--(entropy)->free_in_buffer == 0)  \
	    dump_buffer(entropy); }


LOCAL(void)
dump_buffer (phuff_entropy_ptr entropy)
/* Empty the output buffer; we do not support suspension in this module. */
{
  struct jpeg_destination_mgr * dest = entropy->cinfo->dest;

  if (! (*dest->empty_output_buffer) (entropy->cinfo))
    ERREXIT(entropy->cinfo, JERR_CANT_SUSPEND);
  /* After a successful buffer dump, must reset buffer pointers */
  entropy->next_output_byte = dest->next_output_byte;
  entropy->free_in_buffer = dest->free_in_buffer;
}


/* Outputting bits to the file */

/* Only the right 24 bits of put_buffer are used; the valid bits are
 * left-justified in this part.  At most 16 bits can be passed to emit_bits
 * in one call, and we never retain more than 7 bits in put_buffer
 * between calls, so 24 bits are sufficient.
 */

INLINE
LOCAL(void)
emit_bits (phuff_entropy_ptr entropy, unsigned int code, int size)
/* Emit some bits, unless we are in gather mode */
{
  /* This routine is heavily used, so it's worth coding tightly. */
  register INT32 put_buffer = (INT32) code;
  register int put_bits = entropy->put_bits;

  /* if size is 0, caller used an invalid Huffman table entry */
  if (size == 0)
    ERREXIT(entropy->cinfo, JERR_HUFF_MISSING_CODE);

  if (entropy->gather_statistics)
    return;			/* do nothing if we're only getting stats */

  put_buffer &= (((INT32) 1)<<size) - 1; /* mask off any extra bits in code */
  
  put_bits += size;		/* new number of bits in buffer */
  
  put_buffer <<= 24 - put_bits; /* align incoming bits */

  put_buffer |= entropy->put_buffer; /* and merge with old buffer contents */

  while (put_bits >= 8) {
    int c = (int) ((put_buffer >> 16) & 0xFF);
    
    emit_byte(entropy, c);
    if (c == 0xFF) {		/* need to stuff a zero byte? */
      emit_byte(entropy, 0);
    }
    put_buffer <<= 8;
    put_bits -= 8;
  }

  entropy->put_buffer = put_buffer; /* update variables */
  entropy->put_bits = put_bits;
}


LOCAL(void)
flush_bits (phuff_entropy_ptr entropy)
{
  emit_bits(entropy, 0x7F, 7); /* fill any partial byte with ones */
  entropy->put_buffer = 0;     /* and reset bit-buffer to empty */
  entropy->put_bits = 0;
}


/*
 * Emit (or just count) a Huffman symbol.
 */

INLINE
LOCAL(void)
emit_symbol (phuff_entropy_ptr entropy, int tbl_no, int symbol)
{
  if (entropy->gather_statistics)
    entropy->count_ptrs[tbl_no][symbol]++;
  else {
    c_derived_tbl * tbl = entropy->derived_tbls[tbl_no];
    emit_bits(entropy, tbl->ehufco[symbol], tbl->ehufsi[symbol]);
  }
}


/*
 * Emit bits from a correction bit buffer.
 */

LOCAL(void)
emit_buffered_bits (phuff_entropy_ptr entropy, char * bufstart,
		    unsigned int nbits)
{
  if (entropy->gather_statistics)
    return;			/* no real work */

  while (nbits > 0) {
    emit_bits(entropy, (unsigned int) (*bufstart), 1);
    bufstart++;
    nbits--;
  }
}


/*
 * Emit any pending EOBRUN symbol.
 */

LOCAL(void)
emit_eobrun (phuff_entropy_ptr entropy)
{
  register int temp, nbits;

  if (entropy->EOBRUN > 0) {	/* if there is any pending EOBRUN */
    temp = entropy->EOBRUN;
    nbits = 0;
    while ((temp >>= 1))
      nbits++;
    /* safety check: shouldn't happen given limited correction-bit buffer */
    if (nbits > 14)
      ERREXIT(entropy->cinfo, JERR_HUFF_MISSING_CODE);

    emit_symbol(entropy, entropy->ac_tbl_no, nbits << 4);
    if (nbits)
      emit_bits(entropy, entropy->EOBRUN, nbits);

    entropy->EOBRUN = 0;

    /* Emit any buffered correction bits */
    emit_buffered_bits(entropy, entropy->bit_buffer, entropy->BE);
    entropy->BE = 0;
  }
}


/*
 * Emit a restart marker & resynchronize predictions.
 */

LOCAL(void)
emit_restart (phuff_entropy_ptr entropy, int restart_num)
{
  int ci;

  emit_eobrun(entropy);

  if (! entropy->gather_statistics) {
    flush_bits(entropy);
    emit_byte(entropy, 0xFF);
    emit_byte(entropy, JPEG_RST0 + restart_num);
  }

  if (entropy->cinfo->Ss == 0) {
    /* Re-initialize DC predictions to 0 */
    for (ci = 0; ci < entropy->cinfo->comps_in_scan; ci++)
      entropy->last_dc_val[ci] = 0;
  } else {
    /* Re-initialize all AC-related fields to 0 */
    entropy->EOBRUN = 0;
    entropy->BE = 0;
  }
}


/*
 * MCU encoding for DC initial scan (either spectral selection,
 * or first pass of successive approximation).
 */

METHODDEF(boolean)
encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
{
  phuff_entropy_ptr entropy = (phuff_entropy_ptr) cinfo->entropy;
  register int temp, temp2;
  register int nbits;
  int blkn, ci;
  int Al = cinfo->Al;
  JBLOCKROW block;
  jpeg_component_info * compptr;
  ISHIFT_TEMPS

  entropy->next_output_byte = cinfo->dest->next_output_byte;
  entropy->free_in_buffer = cinfo->dest->free_in_buffer;

  /* Emit restart marker if needed */
  if (cinfo->restart_interval)
    if (entropy->restarts_to_go == 0)
      emit_restart(entropy, entropy->next_restart_num);

  /* Encode the MCU data blocks */
  for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) {
    block = MCU_data[blkn];
    ci = cinfo->MCU_membership[blkn];
    compptr = cinfo->cur_comp_info[ci];

    /* Compute the DC value after the required point transform by Al.
     * This is simply an arithmetic right shift.
     */
    temp2 = IRIGHT_SHIFT((int) ((*block)[0]), Al);

    /* DC differences are figured on the point-transformed values. */
    temp = temp2 - entropy->last_dc_val[ci];
    entropy->last_dc_val[ci] = temp2;

    /* Encode the DC coefficient difference per section G.1.2.1 */
    temp2 = temp;
    if (temp < 0) {
      temp = -temp;		/* temp is abs value of input */
      /* For a negative input, want temp2 = bitwise complement of abs(input) */
      /* This code assumes we are on a two's complement machine */
      temp2--;