view dump.cpp @ 408:f0a69693eaef api-inversion

The lesser of two evils, part 1.

Most of the body of audiodb_insert_datum() will apply to
"LARGE_ADB"-type insertions: checking for the right flags, checking for
enough space free, synchronizing the header.  Wouldn't it be nice if we
could reuse all that code (or at least the bits that apply) without one
horrible almost-identical cut-and-paste job (see
batchinsert_large_adb(), or if that's not compelling enough, the four
almost-identical query loops from before the Great Refactoring).

Well, yes, it would.  Sadly C makes it mildly difficult, because its
functions are explicitly typed (so we can't pass arbitrary arguments of
other types, even if they're ABI-compatible), while its macros are
textual (which makes writing and maintaining them horrible).  The
thought of a union argument was briefly entertained and then discarded
as being just Too Weird.

So, instead, (ab)use the oldest trick in the book: void *.  Define an
adb_datum_internal_t which has void * instead of double *; the intention
is that this internal data type can be constructed both from an
adb_datum_t and some notional adb_reference_t (which looks very much
like an adb_insert_t at the time of writing, with char * structure
entries representing filenames).  This adb_datum_internal_t structure is
very much an internals-only thing, so put its definition in the
internals header.

Call what was previously audiodb_insert_datum() a new function
audiodb_insert_datum_internal(), made static so that really no-one is
tempted to call it other than ourselves.  audiodb_insert_datum() is then
trivial in terms of this new function, if stupidly tedious.  (If we were
playing dangerously, we could just perform a cast, but relying on the
fact that sizeof(double *) = sizeof(void *) would almost certainly end
up biting when we least expect.

Incidental inclusion in this patch, since I noticed it at the time:
actually check for the O2_FLAG_L2NORM before scribbling all over the
l2norm table.  Somewhat unsurprisingly, there are as yet no tests to
defend against this (harmless, as it turns out) erroneous behaviour.

#include "audioDB.h"
extern "C" {
#include "audioDB_API.h"
#include "audioDB-internals.h"
}

int audiodb_dump(adb_t *adb, const char *output) {
  char *fileTable = 0; /* key_table */
  unsigned *trackTable = 0; /* track_size_table */
  double *timesTable = 0; /* timestamps_table */
  double *powerTable = 0; /* power_table */

  size_t fileTableLength = 0;
  size_t trackTableLength = 0;
  size_t timesTableLength = 0;
  size_t powerTableLength = 0;

  char *featureFileNameTable = 0;
  char *powerFileNameTable = 0;
  char *timesFileNameTable = 0;
 
  char cwd[PATH_MAX];
  int directory_changed = 0;

  int fLfd = 0, tLfd = 0, pLfd = 0, kLfd = 0;
  FILE *fLFile = 0, *tLFile = 0, *pLFile = 0, *kLFile = 0;

  int times, power;

  char fName[256];
  int ffd, pfd;
  FILE *tFile;
  unsigned pos = 0;
  double *data_buffer;
  size_t data_buffer_size;
  FILE *scriptFile = 0;

  unsigned nfiles = adb->header->numFiles;

  if(adb->header->length > 0) {
    fileTableLength = ALIGN_PAGE_UP(nfiles * O2_FILETABLE_ENTRY_SIZE);
    trackTableLength = ALIGN_PAGE_UP(nfiles * O2_TRACKTABLE_ENTRY_SIZE);
    if(!(adb->header->flags & O2_FLAG_LARGE_ADB)) {
      off_t length = adb->header->length;
      unsigned dim = adb->header->dim;
      timesTableLength = ALIGN_PAGE_UP(2*length/dim);
      powerTableLength = ALIGN_PAGE_UP(length/dim);
    }

    mmap_or_goto_error(char *, fileTable, adb->header->fileTableOffset, fileTableLength);
    mmap_or_goto_error(unsigned *, trackTable, adb->header->trackTableOffset, trackTableLength);
    if (adb->header->flags & O2_FLAG_LARGE_ADB) {
      mmap_or_goto_error(char *, featureFileNameTable, adb->header->dataOffset, fileTableLength);
      mmap_or_goto_error(char *, powerFileNameTable, adb->header->powerTableOffset, fileTableLength);
      mmap_or_goto_error(char *, timesFileNameTable, adb->header->timesTableOffset, fileTableLength);
    } else {
      mmap_or_goto_error(double *, powerTable, adb->header->powerTableOffset, powerTableLength);
      mmap_or_goto_error(double *, timesTable, adb->header->timesTableOffset, timesTableLength);
    }
  }

  if((mkdir(output, S_IRWXU|S_IRWXG|S_IRWXO)) < 0) {
    goto error;
  }

  if ((getcwd(cwd, PATH_MAX)) == 0) {
    goto error;
  }

  /* FIXME: Hrm.  How does chdir(2) interact with threads?  Does each
   * thread have its own working directory? */
  if((chdir(output)) < 0) {
    goto error;
  }
  directory_changed = 1;

  if ((fLfd = open("featureList.txt", O_CREAT|O_RDWR|O_EXCL, S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH)) < 0) {
    goto error;
  }

  times = adb->header->flags & O2_FLAG_TIMES;
  if (times) {
    if ((tLfd = open("timesList.txt", O_CREAT|O_RDWR|O_EXCL, S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH)) < 0) {
      goto error;
    }
  }

  power = adb->header->flags & O2_FLAG_POWER;
  if (power) {
    if ((pLfd = open("powerList.txt", O_CREAT|O_RDWR|O_EXCL, S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH)) < 0) {
      goto error;
    }
  }

  if ((kLfd = open("keyList.txt", O_CREAT|O_RDWR|O_EXCL, S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH)) < 0) {
    goto error;
  }
  
  /* can these fail?  I sincerely hope not. */
  fLFile = fdopen(fLfd, "w");
  if (times) {
    tLFile = fdopen(tLfd, "w");
  }
  if (power) {
    pLFile = fdopen(pLfd, "w");
  }
  kLFile = fdopen(kLfd, "w");

  lseek(adb->fd, adb->header->dataOffset, SEEK_SET);

  for(unsigned k = 0; k < nfiles; k++) {
    fprintf(kLFile, "%s\n", fileTable + k*O2_FILETABLE_ENTRY_SIZE);
    if(adb->header->flags & O2_FLAG_LARGE_ADB) {
      char *featureFileName = featureFileNameTable+k*O2_FILETABLE_ENTRY_SIZE;
      if(*featureFileName != '/') {
        goto error;
      }
      fprintf(fLFile, "%s\n", featureFileName);
      if(times) {
	char *timesFileName = timesFileNameTable + k*O2_FILETABLE_ENTRY_SIZE;
	if(*timesFileName != '/') {
          goto error;
	}
	fprintf(tLFile, "%s\n", timesFileName);
      }
      if(power) {
	char *powerFileName = powerFileNameTable + k*O2_FILETABLE_ENTRY_SIZE;
	if(*powerFileName != '/') {
          goto error;
	}
	fprintf(pLFile, "%s\n", powerFileName);
      }
    } else {
      snprintf(fName, 256, "%05d.features", k);
      if ((ffd = open(fName, O_CREAT|O_RDWR|O_EXCL, S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH)) < 0) {
        goto error;
      }
      if ((write(ffd, &(adb->header->dim), sizeof(uint32_t))) < 0) {
        goto error;
      }
      
      /* FIXME: this repeated malloc()/free() of data buffers is
	 inefficient. */
      data_buffer_size = trackTable[k] * adb->header->dim * sizeof(double);
      
      {
	void *tmp = malloc(data_buffer_size);
	if (tmp == NULL) {
          goto error;
	}
	data_buffer = (double *) tmp;
      }
      
      if ((read(adb->fd, data_buffer, data_buffer_size)) != (ssize_t) data_buffer_size) {
        goto error;
      }
      
      if ((write(ffd, data_buffer, data_buffer_size)) < 0) {
        goto error;
      }
      
      free(data_buffer);
      
      fprintf(fLFile, "%s\n", fName);
      close(ffd);
      ffd = 0;

      if (times) {
	snprintf(fName, 256, "%05d.times", k);
	tFile = fopen(fName, "w");
	for(unsigned i = 0; i < trackTable[k]; i++) {
	  // KLUDGE: specifying 16 digits of precision after the decimal
	  // point is (but check this!) sufficient to uniquely identify
	  // doubles; however, that will cause ugliness, as that's
	  // vastly too many for most values of interest.  Moving to %a
	  // here and scanf() in the timesFile reading might fix this.
	  // -- CSR, 2007-10-19
	  fprintf(tFile, "%.16e\n", *(timesTable + 2*pos + 2*i));
	}
	fprintf(tFile, "%.16e\n", *(timesTable + 2*pos + 2*trackTable[k]-1));
        fclose(tFile);
	
	fprintf(tLFile, "%s\n", fName);
      }
      
      if (power) {
	uint32_t one = 1;
	snprintf(fName, 256, "%05d.power", k);
	if ((pfd = open(fName, O_CREAT|O_RDWR|O_EXCL, S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH)) < 0) {
          goto error;
	}
	if ((write(pfd, &one, sizeof(uint32_t))) < 0) {
          goto error;
	}
	if ((write(pfd, powerTable + pos, trackTable[k] * sizeof(double))) < 0) {
          goto error;
	}
	fprintf(pLFile, "%s\n", fName);
	close(pfd);
        pfd = 0;
      } 
      
      pos += trackTable[k];
      std::cout << fileTable+k*O2_FILETABLE_ENTRY_SIZE << " " << trackTable[k] << std::endl;
    }
  }

  scriptFile = fopen("restore.sh", "w");
  fprintf(scriptFile, "\
#! /bin/sh\n\
#\n\
# usage: AUDIODB=/path/to/audioDB sh ./restore.sh <newdb>\n\
\n\
if [ -z \"${AUDIODB}\" ]; then echo set AUDIODB variable; exit 1; fi\n\
if [ -z \"$1\" ]; then echo usage: $0 newdb; exit 1; fi\n\n\
\"${AUDIODB}\" -d \"$1\" -N --datasize=%d --ntracks=%d --datadim=%d\n",
          (int) ((adb->header->timesTableOffset - adb->header->dataOffset) / (1024*1024)),
          // fileTable entries (char[256]) are bigger than trackTable
          // (int), so the granularity of page aligning is finer.
          (int) ((adb->header->trackTableOffset - adb->header->fileTableOffset) / O2_FILETABLE_ENTRY_SIZE),
          (int) ceil(((double) (adb->header->timesTableOffset - adb->header->dataOffset)) / ((double) (adb->header->dbSize - adb->header->l2normTableOffset))));
  if(adb->header->flags & O2_FLAG_L2NORM) {
    fprintf(scriptFile, "\"${AUDIODB}\" -d \"$1\" -L\n");
  }
  if(power) {
    fprintf(scriptFile, "\"${AUDIODB}\" -d \"$1\" -P\n");
  }
  fprintf(scriptFile, "\"${AUDIODB}\" -d \"$1\" -B -F featureList.txt -K keyList.txt");
  if(times) {
    fprintf(scriptFile, " -T timesList.txt");
  }
  if(power) {
    fprintf(scriptFile, " -W powerList.txt");
  }
  fprintf(scriptFile, "\n");
  fclose(scriptFile);

  fclose(fLFile);
  if(times) {
    fclose(tLFile);
  }
  if(power) {
    fclose(pLFile);
  }
  fclose(kLFile);
    
  maybe_munmap(fileTable, fileTableLength);
  maybe_munmap(trackTable, trackTableLength);
  maybe_munmap(timesTable, timesTableLength);
  maybe_munmap(powerTable, powerTableLength);
  maybe_munmap(featureFileNameTable, fileTableLength);
  maybe_munmap(timesFileNameTable, fileTableLength);
  maybe_munmap(powerFileNameTable, fileTableLength);

  if((chdir(cwd)) < 0) {
    /* don't goto error because the error handling will try to
     * chdir() */
    return 1;
  }

  return 0;

 error:
  if(fLFile) {
    fclose(fLFile);
  } else if(fLfd) {
    close(fLfd);
  }
  if(tLFile) {
    fclose(tLFile);
  } else if(tLfd) {
    close(fLfd);
  }
  if(pLFile) {
    fclose(pLFile);
  } else if(pLfd) {
    close(pLfd);
  }
  if(kLFile) {
    fclose(kLFile);
  } else if(kLfd) {
    close(kLfd);
  }
  if(scriptFile) {
    fclose(scriptFile);
  }

  maybe_munmap(fileTable, fileTableLength);
  maybe_munmap(trackTable, trackTableLength);
  maybe_munmap(timesTable, timesTableLength);
  maybe_munmap(powerTable, powerTableLength);
  maybe_munmap(featureFileNameTable, fileTableLength);
  maybe_munmap(timesFileNameTable, fileTableLength);
  maybe_munmap(powerFileNameTable, fileTableLength);

  if(directory_changed) {
    chdir(cwd);
  }
  return 1;
}