summaryrefslogtreecommitdiffstats
path: root/src/h5repart.c
diff options
context:
space:
mode:
authorRobb Matzke <matzke@llnl.gov>1998-05-13 17:58:24 (GMT)
committerRobb Matzke <matzke@llnl.gov>1998-05-13 17:58:24 (GMT)
commit34f5a59b9eb995997536575eeb796f745a2be784 (patch)
treec70515f61d6167c265f43c7d13c7c75d1e3608cb /src/h5repart.c
parentbb86785ce5b597cdc007f8494c6649b01e386455 (diff)
downloadhdf5-34f5a59b9eb995997536575eeb796f745a2be784.zip
hdf5-34f5a59b9eb995997536575eeb796f745a2be784.tar.gz
hdf5-34f5a59b9eb995997536575eeb796f745a2be784.tar.bz2
[svn-r398] Changes since 19980501
---------------------- ./src/H5Bprivate.h ./src/H5F.c ./src/H5Ffamily.c ./src/H5Fprivate.h ./src/H5Gpkg.h ./src/H5MF.c ./src/H5P.c ./src/H5Ppublic.h ./test/big.c ./html/Big.html ./html/Files.html Family members can now be any size >1kB. Got rid of some places where we were reading a property list after it was closed. ./MANIFEST ./src/Makefile.in ./src/h5repart.c A program to repartition file families. The source and/or destination may be files or file families. Examples: testhdf5 -c -o stab h5repart -m 1m tstab2.h5 x%05d.h5 h5repart -m 100k x%05d.h5 y%05d.h5 h5repart y%05d.h5 z.h5 diff tstab2.h5 z.h5 ./src/h5ls.c Added a usage message, replaced assertions with error messages. ./config/linux Made a warning message more obvious. Added better optimization flags for Pentium-Pro's.
Diffstat (limited to 'src/h5repart.c')
-rw-r--r--src/h5repart.c388
1 files changed, 388 insertions, 0 deletions
diff --git a/src/h5repart.c b/src/h5repart.c
new file mode 100644
index 0000000..8faa9ea
--- /dev/null
+++ b/src/h5repart.c
@@ -0,0 +1,388 @@
+/*
+ * Copyright (C) 1998 NCSA
+ * All rights reserved.
+ *
+ * Programmer: Robb Matzke <matzke@llnl.gov>
+ * Wednesday, May 13, 1998
+ *
+ * Purpose: Repartitions a file family. This program can be used to
+ * split a single file into a family of files, join a family of
+ * files into a single file, or copy one family to another while
+ * changing the size of the family members. It can also be used
+ * to copy a single file to a single file with holes.
+ */
+#include <ctype.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <hdf5.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#define FALSE 0
+#define TRUE 1
+#define NAMELEN 4096
+#define GB *1024*1024*1024
+
+#define MIN(X,Y) ((X)<(Y)?(X):(Y))
+#define MIN3(X,Y,Z) MIN(MIN(X,Y),Z)
+
+
+/*-------------------------------------------------------------------------
+ * Function: usage
+ *
+ * Purpose: Prints a usage message.
+ *
+ * Return: void
+ *
+ * Programmer: Robb Matzke
+ * Wednesday, May 13, 1998
+ *
+ * Modifications:
+ *
+ *-------------------------------------------------------------------------
+ */
+static void
+usage (const char *progname)
+{
+ fprintf (stderr, "usage: %s [-[b|m] N[g|m|k]] SRC DST\n", progname);
+ fprintf (stderr, " -b N The I/O block size, defaults to 1kB\n");
+ fprintf (stderr, " -m N The destination member size or 1GB\n");
+ fprintf (stderr, " SRC The name of the source file\n");
+ fprintf (stderr, " DST The name of the destination files\n");
+ fprintf (stderr, "Sizes may be suffixed with `g' for GB, `m' for MB or "
+ "`k' for kB.\n");
+ fprintf (stderr, "File family names include an integer printf "
+ "format such as `%%d'\n");
+ exit (1);
+}
+
+
+/*-------------------------------------------------------------------------
+ * Function: get_size
+ *
+ * Purpose: Reads a size option of the form `-XNS' where `X' is any
+ * letter, `N' is a multi-character positive decimal number, and
+ * `S' is an optional suffix letter in the set [GgMmk]. The
+ * option may also be split among two arguments as: `-X NS'.
+ * The input value of ARGNO is the argument number for the
+ * switch in the ARGV vector and ARGC is the number of entries
+ * in that vector.
+ *
+ * Return: Success: The value N multiplied according to the
+ * suffix S. On return ARGNO will be the number
+ * of the next argument to process.
+ *
+ * Failure: Calls usage() which exits with a non-zero
+ * status.
+ *
+ * Programmer: Robb Matzke
+ * Wednesday, May 13, 1998
+ *
+ * Modifications:
+ *
+ *-------------------------------------------------------------------------
+ */
+static off_t
+get_size (const char *progname, int *argno, int argc, char *argv[])
+{
+ off_t retval;
+ char *suffix;
+
+ if (isdigit (argv[*argno][2])) {
+ retval = strtol (argv[*argno]+2, &suffix, 10);
+ (*argno)++;
+ } else if (argv[*argno][2] || *argno+1>=argc) {
+ usage (progname);
+ } else {
+ retval = strtol (argv[*argno+1], &suffix, 0);
+ if (suffix==argv[*argno+1]) usage (progname);
+ *argno += 2;
+ }
+ if (suffix && suffix[0] && !suffix[1]) {
+ switch (*suffix) {
+ case 'G':
+ case 'g':
+ retval *= 1024 * 1024 * 1024;
+ break;
+ case 'M':
+ case 'm':
+ retval *= 1024 * 1024;
+ break;
+ case 'k':
+ retval *= 1024;
+ break;
+ default:
+ usage (progname);
+ }
+ } else if (suffix && suffix[0]) {
+ usage (progname);
+ }
+ return retval;
+}
+
+
+/*-------------------------------------------------------------------------
+ * Function: main
+ *
+ * Purpose: Split an hdf5 file
+ *
+ * Return: Success:
+ *
+ * Failure:
+ *
+ * Programmer: Robb Matzke
+ * Wednesday, May 13, 1998
+ *
+ * Modifications:
+ *
+ *-------------------------------------------------------------------------
+ */
+int
+main (int argc, char *argv[])
+{
+ const char *prog_name; /*program name */
+ size_t blk_size=1024; /*size of each I/O block */
+ char *buf=NULL; /*I/O block buffer */
+ size_t n, i; /*counters */
+ ssize_t nio; /*I/O return value */
+ int argno=1; /*program argument number */
+ int src, dst=-1; /*source & destination files */
+ int need_seek=FALSE; /*destination needs to seek? */
+ int need_write; /*data needs to be written? */
+ struct stat sb; /*temporary file stat buffer */
+ int verbose=FALSE; /*display file names? */
+ size_t left_overs=0; /*amount of zeros left over */
+
+ const char *src_gen_name; /*general source name */
+ char src_name[NAMELEN]; /*source member name */
+ off_t src_offset=0; /*offset in source member */
+ int src_is_family; /*is source name a family name? */
+ int src_membno=0; /*source member number */
+ off_t src_size; /*source logical member size */
+ off_t src_act_size; /*source actual member size */
+
+ const char *dst_gen_name; /*general destination name */
+ char dst_name[NAMELEN]; /*destination member name */
+ off_t dst_offset=0; /*offset in destination member */
+ int dst_is_family; /*is dst name a family name? */
+ int dst_membno=0; /*destination member number */
+ off_t dst_size=1 GB; /*destination logical memb size */
+
+ /*
+ * Get the program name from argv[0]. Use only the last component.
+ */
+ if ((prog_name=strrchr (argv[0], '/'))) prog_name++;
+ else prog_name = argv[0];
+
+ /*
+ * Parse switches.
+ */
+ while (argno<argc && '-'==argv[argno][0]) {
+ if (!strcmp (argv[argno], "-v")) {
+ verbose = TRUE;
+ argno++;
+ } else if ('b'==argv[argno][1]) {
+ blk_size = get_size (prog_name, &argno, argc, argv);
+ } else if ('m'==argv[argno][1]) {
+ dst_size = get_size (prog_name, &argno, argc, argv);
+ } else {
+ usage (prog_name);
+ }
+ }
+
+ /*
+ * Get the name for the source file and open the first member. The size
+ * of the first member determines the logical size of all the members.
+ */
+ if (argno>=argc) usage (prog_name);
+ src_gen_name = argv[argno++];
+ sprintf (src_name, src_gen_name, src_membno);
+ src_is_family = strcmp (src_name, src_gen_name);
+ if ((src=open (src_name, O_RDONLY))<0) {
+ perror (src_name);
+ exit (1);
+ }
+ if (fstat (src, &sb)<0) {
+ perror ("fstat");
+ exit (1);
+ }
+ src_size = src_act_size = sb.st_size;
+ if (verbose) fprintf (stderr, "< %s\n", src_name);
+
+ /*
+ * Get the name for the destination file and open the first member.
+ */
+ if (argno>=argc) usage (prog_name);
+ dst_gen_name = argv[argno++];
+ sprintf (dst_name, dst_gen_name, dst_membno);
+ dst_is_family = strcmp (dst_name, dst_gen_name);
+ if ((dst=open (dst_name, O_RDWR|O_CREAT|O_TRUNC, 0666))<0) {
+ perror (dst_name);
+ exit (1);
+ }
+ if (verbose) fprintf (stderr, "> %s\n", dst_name);
+
+ /* No more arguments */
+ if (argno<argc) usage (prog_name);
+
+ /* Now the real work, split the file */
+ buf = malloc (blk_size);
+ while (src_offset<src_size) {
+
+ /* Read a block. The amount to read is the minimum of:
+ * 1. The I/O block size
+ * 2. What's left to write in the destination member
+ * 3. Left over zeros or what's left in the source member.
+ */
+ n = blk_size;
+ if (dst_is_family) n = (size_t)MIN((off_t)n, dst_size-dst_offset);
+ if (left_overs) {
+ n = MIN (n, left_overs);
+ left_overs -= n;
+ need_write = FALSE;
+ } else if (src_offset<src_act_size) {
+ n = (size_t)MIN ((off_t)n, src_act_size-src_offset);
+ if ((nio=read (src, buf, n))<0) {
+ perror ("read");
+ exit (1);
+ } else if ((size_t)nio!=n) {
+ fprintf (stderr, "%s: short read\n", src_name);
+ exit (1);
+ }
+ for (i=0; i<n; i++) {
+ if (buf[i]) break;
+ }
+ need_write = (i<n);
+ } else {
+ n = 0;
+ left_overs = src_size - src_act_size;
+ need_write = FALSE;
+ }
+
+ /*
+ * If the block contains non-zero data then write it to the
+ * destination, otherwise just remember that we'll have to do a seek
+ * later in the destination when we finally get non-zero data.
+ */
+ if (need_write) {
+ if (need_seek && lseek (dst, dst_offset, SEEK_SET)<0) {
+ perror ("lseek");
+ exit (1);
+ }
+ if ((nio=write (dst, buf, n))<0) {
+ perror ("write");
+ exit (1);
+ } else if ((size_t)nio!=n) {
+ fprintf (stderr, "%s: short write\n", dst_name);
+ exit (1);
+ }
+ need_seek = FALSE;
+ } else {
+ need_seek = TRUE;
+ }
+
+ /*
+ * Update the source offset and open the next source family member if
+ * necessary. The source stream ends at the first member which
+ * cannot be opened because it doesn't exist. At the end of the
+ * source stream, update the destination offset and break out of the
+ * loop. The destination offset must be updated so we can fix
+ * trailing holes.
+ */
+ src_offset += n;
+ if (src_offset==src_act_size) {
+ close (src);
+ if (!src_is_family) {
+ dst_offset += n;
+ break;
+ }
+ sprintf (src_name, src_gen_name, ++src_membno);
+ if ((src=open (src_name, O_RDONLY))<0 && ENOENT==errno) {
+ dst_offset += n;
+ break;
+ } else if (src<0) {
+ perror (src_name);
+ exit (1);
+ }
+ if (fstat (src, &sb)<0) {
+ perror ("fstat");
+ exit (1);
+ }
+ src_act_size = sb.st_size;
+ if (src_act_size>src_size) {
+ fprintf (stderr, "%s: member truncated to %lu bytes\n",
+ src_name, (unsigned long)src_size);
+ }
+ src_offset = 0;
+ if (verbose) fprintf (stderr, "< %s\n", src_name);
+ }
+
+ /*
+ * Update the destination offset, opening a new member if one will be
+ * needed. The first member is extended to the logical member size
+ * but other members might be smaller if they end with a hole.
+ */
+ dst_offset += n;
+ if (dst_is_family && dst_offset==dst_size) {
+ if (0==dst_membno) {
+ if (lseek (dst, dst_size-1, SEEK_SET)<0) {
+ perror ("lseek");
+ exit (1);
+ }
+ if (read (dst, buf, 1)<0) {
+ perror ("read");
+ exit (1);
+ }
+ if (lseek (dst, dst_size-1, SEEK_SET)<0) {
+ perror ("lseek");
+ exit (1);
+ }
+ if (write (dst, buf, 1)<0) {
+ perror ("write");
+ exit (1);
+ }
+ }
+ close (dst);
+ sprintf (dst_name, dst_gen_name, ++dst_membno);
+ if ((dst=open (dst_name, O_RDWR|O_CREAT|O_TRUNC, 0666))<0) {
+ perror (dst_name);
+ exit (1);
+ }
+ dst_offset = 0;
+ need_seek = FALSE;
+ if (verbose) fprintf (stderr, "> %s\n", dst_name);
+ }
+ }
+
+ /*
+ * Make sure the last family member is the right size and then close it.
+ * The last member can't end with a hole or hdf5 will think that the
+ * family has been truncated.
+ */
+ if (need_seek) {
+ if (lseek (dst, dst_offset-1, SEEK_SET)<0) {
+ perror ("lseek");
+ exit (1);
+ }
+ if (read (dst, buf, 1)<0) {
+ perror ("read");
+ exit (1);
+ }
+ if (lseek (dst, dst_offset-1, SEEK_SET)<0) {
+ perror ("lseek");
+ exit (1);
+ }
+ if (write (dst, buf, 1)<0) {
+ perror ("write");
+ exit (1);
+ }
+ }
+ close (dst);
+
+ /* Free resources and return */
+ free (buf);
+ return 0;
+}