[wimlib] / src / resource.c

/*
 * resource.c
 *
 * Read uncompressed and compressed metadata and file resources.
 */

/*
 * Copyright (C) 2010 Carl Thijssen
 * Copyright (C) 2012 Eric Biggers
 *
 * This file is part of wimlib, a library for working with WIM files.
 *
 * wimlib is free software; you can redistribute it and/or modify it under the
 * terms of the GNU Lesser General Public License as published by the Free
 * Software Foundation; either version 2.1 of the License, or (at your option)
 * any later version.
 *
 * wimlib is distributed in the hope that it will be useful, but WITHOUT ANY
 * WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
 * A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
 * details.
 *
 * You should have received a copy of the GNU Lesser General Public License
 * along with wimlib; if not, see http://www.gnu.org/licenses/.
 */

#include "wimlib_internal.h"
#include "lookup_table.h"
#include "io.h"
#include "lzx.h"
#include "xpress.h"
#include "sha1.h"
#include "dentry.h"
#include "config.h"
#include <unistd.h>
#include <errno.h>
#include <alloca.h>

#ifdef WITH_NTFS_3G
#include <ntfs-3g/attrib.h>
#include <ntfs-3g/inode.h>
#include <ntfs-3g/dir.h>
#endif

/* 
 * Reads all or part of a compressed resource into an in-memory buffer.
 *
 * @fp:                 The FILE* for the WIM file.
 * @resource_compressed_size:    The compressed size of the resource.  
 * @resource_uncompressed_size:  The uncompressed size of the resource.
 * @resource_offset:             The offset of the start of the resource from
 *                                      the start of the stream @fp.
 * @resource_ctype:     The compression type of the resource. 
 * @len:                The number of bytes of uncompressed data to read from
 *                              the resource.
 * @offset:             The offset of the bytes to read within the uncompressed
 *                              resource.
 * @contents_len:       An array into which the uncompressed data is written.
 *                              It must be at least @len bytes long.
 *
 * Returns zero on success, nonzero on failure.
 */
static int read_compressed_resource(FILE *fp, u64 resource_compressed_size, 
                                    u64 resource_uncompressed_size, 
                                    u64 resource_offset, int resource_ctype, 
                                    u64 len, u64 offset, u8  contents_ret[])
{

        DEBUG2("comp size = %"PRIu64", uncomp size = %"PRIu64", "
               "res offset = %"PRIu64"",
               resource_compressed_size,
               resource_uncompressed_size,
               resource_offset);
        DEBUG2("resource_ctype = %s, len = %"PRIu64", offset = %"PRIu64"",
               wimlib_get_compression_type_string(resource_ctype), len, offset);
        /* Trivial case */
        if (len == 0)
                return 0;

        int (*decompress)(const void *, uint, void *, uint);
        /* Set the appropriate decompress function. */
        if (resource_ctype == WIM_COMPRESSION_TYPE_LZX)
                decompress = lzx_decompress;
        else
                decompress = xpress_decompress;

        /* The structure of a compressed resource consists of a table of chunk
         * offsets followed by the chunks themselves.  Each chunk consists of
         * compressed data, and there is one chunk for each WIM_CHUNK_SIZE =
         * 32768 bytes of the uncompressed file, with the last chunk having any
         * remaining bytes.
         *
         * The chunk offsets are measured relative to the end of the chunk
         * table.  The first chunk is omitted from the table in the WIM file
         * because its offset is implicitly given by the fact that it directly
         * follows the chunk table and therefore must have an offset of 0. 
         */

        /* Calculate how many chunks the resource conists of in its entirety. */
        u64 num_chunks = (resource_uncompressed_size + WIM_CHUNK_SIZE - 1) /
                                                                WIM_CHUNK_SIZE;
        /* As mentioned, the first chunk has no entry in the chunk table. */
        u64 num_chunk_entries = num_chunks - 1;


        /* The index of the chunk that the read starts at. */
        u64 start_chunk = offset / WIM_CHUNK_SIZE;
        /* The byte offset at which the read starts, within the start chunk. */
        u64 start_chunk_offset = offset % WIM_CHUNK_SIZE;

        /* The index of the chunk that contains the last byte of the read. */
        u64 end_chunk   = (offset + len - 1) / WIM_CHUNK_SIZE;
        /* The byte offset of the last byte of the read, within the end chunk */
        u64 end_chunk_offset = (offset + len - 1) % WIM_CHUNK_SIZE;

        /* Number of chunks that are actually needed to read the requested part
         * of the file. */
        u64 num_needed_chunks = end_chunk - start_chunk + 1;

        /* If the end chunk is not the last chunk, an extra chunk entry is
         * needed because we need to know the offset of the chunk after the last
         * chunk read to figure out the size of the last read chunk. */
        if (end_chunk != num_chunks - 1)
                num_needed_chunks++;

        /* Declare the chunk table.  It will only contain offsets for the chunks
         * that are actually needed for this read. */
        u64 chunk_offsets[num_needed_chunks];

        /* Set the implicit offset of the first chunk if it is included in the
         * needed chunks.
         *
         * Note: M$'s documentation includes a picture that shows the first
         * chunk starting right after the chunk entry table, labeled as offset
         * 0x10.  However, in the actual file format, the offset is measured
         * from the end of the chunk entry table, so the first chunk has an
         * offset of 0. */
        if (start_chunk == 0)
                chunk_offsets[0] = 0;

        /* According to M$'s documentation, if the uncompressed size of
         * the file is greater than 4 GB, the chunk entries are 8-byte
         * integers.  Otherwise, they are 4-byte integers. */
        u64 chunk_entry_size = (resource_uncompressed_size >= (u64)1 << 32) ? 
                                                                        8 : 4;

        /* Size of the full chunk table in the WIM file. */
        u64 chunk_table_size = chunk_entry_size * num_chunk_entries;

        /* Read the needed chunk offsets from the table in the WIM file. */

        /* Index, in the WIM file, of the first needed entry in the
         * chunk table. */
        u64 start_table_idx = (start_chunk == 0) ? 0 : start_chunk - 1;

        /* Number of entries we need to actually read from the chunk
         * table (excludes the implicit first chunk). */
        u64 num_needed_chunk_entries = (start_chunk == 0) ? 
                                num_needed_chunks - 1 : num_needed_chunks;

        /* Skip over unneeded chunk table entries. */
        u64 file_offset_of_needed_chunk_entries = resource_offset + 
                                start_table_idx * chunk_entry_size;
        if (fseeko(fp, file_offset_of_needed_chunk_entries, SEEK_SET) != 0) {
                ERROR_WITH_ERRNO("Failed to seek to byte %"PRIu64" to read "
                                 "chunk table of compressed resource",
                                 file_offset_of_needed_chunk_entries);
                return WIMLIB_ERR_READ;
        }

        /* Number of bytes we need to read from the chunk table. */
        size_t size = num_needed_chunk_entries * chunk_entry_size;

        u8 chunk_tab_buf[size];

        if (fread(chunk_tab_buf, 1, size, fp) != size)
                goto err;

        /* Now fill in chunk_offsets from the entries we have read in
         * chunk_tab_buf. */

        u64 *chunk_tab_p = chunk_offsets;
        if (start_chunk == 0)
                chunk_tab_p++;

        if (chunk_entry_size == 4) {
                u32 *entries = (u32*)chunk_tab_buf;
                while (num_needed_chunk_entries--)
                        *chunk_tab_p++ = to_le32(*entries++);
        } else {
                u64 *entries = (u64*)chunk_tab_buf;
                while (num_needed_chunk_entries--)
                        *chunk_tab_p++ = to_le64(*entries++);
        }

        /* Done with the chunk table now.  We must now seek to the first chunk
         * that is needed for the read. */

        u64 file_offset_of_first_needed_chunk = resource_offset + 
                                chunk_table_size + chunk_offsets[0];
        if (fseeko(fp, file_offset_of_first_needed_chunk, SEEK_SET) != 0) {
                ERROR_WITH_ERRNO("Failed to seek to byte %"PRIu64" to read "
                                 "first chunk of compressed resource",
                                 file_offset_of_first_needed_chunk);
                return WIMLIB_ERR_READ;
        }

        /* Pointer to current position in the output buffer for uncompressed
         * data. */
        u8 *out_p = (u8*)contents_ret;

        /* Buffer for compressed data.  While most compressed chunks will have a
         * size much less than WIM_CHUNK_SIZE, WIM_CHUNK_SIZE - 1 is the maximum
         * size in the worst-case.  This assumption is valid only if chunks that
         * happen to compress to more than the uncompressed size (i.e. a
         * sequence of random bytes) are always stored uncompressed. But this seems
         * to be the case in M$'s WIM files, even though it is undocumented. */
        u8 compressed_buf[WIM_CHUNK_SIZE - 1];


        /* Decompress all the chunks. */
        for (u64 i = start_chunk; i <= end_chunk; i++) {

                DEBUG2("Chunk %"PRIu64" (start %"PRIu64", end %"PRIu64").",
                       i, start_chunk, end_chunk);

                /* Calculate the sizes of the compressed chunk and of the
                 * uncompressed chunk. */
                uint compressed_chunk_size, uncompressed_chunk_size;
                if (i != num_chunks - 1) {
                        /* All the chunks except the last one in the resource
                         * expand to WIM_CHUNK_SIZE uncompressed, and the amount
                         * of compressed data for the chunk is given by the
                         * difference of offsets in the chunk offset table. */
                        compressed_chunk_size = chunk_offsets[i + 1 - start_chunk] - 
                                                chunk_offsets[i - start_chunk];
                        uncompressed_chunk_size = WIM_CHUNK_SIZE;
                } else {
                        /* The last compressed chunk consists of the remaining
                         * bytes in the file resource, and the last uncompressed
                         * chunk has size equal to however many bytes are left-
                         * that is, the remainder of the uncompressed size when
                         * divided by WIM_CHUNK_SIZE. 
                         *
                         * Note that the resource_compressed_size includes the
                         * chunk table, so the size of it must be subtracted. */
                        compressed_chunk_size = resource_compressed_size - 
                                                chunk_table_size -
                                                chunk_offsets[i - start_chunk];

                        uncompressed_chunk_size = resource_uncompressed_size % 
                                                                WIM_CHUNK_SIZE;

                        /* If the remainder is 0, the last chunk actually
                         * uncompresses to a full WIM_CHUNK_SIZE bytes. */
                        if (uncompressed_chunk_size == 0)
                                uncompressed_chunk_size = WIM_CHUNK_SIZE;
                }

                DEBUG2("compressed_chunk_size = %u, "
                       "uncompressed_chunk_size = %u",
                       compressed_chunk_size, uncompressed_chunk_size);


                /* Figure out how much of this chunk we actually need to read */
                u64 start_offset;
                if (i == start_chunk)
                        start_offset = start_chunk_offset;
                else
                        start_offset = 0;
                u64 end_offset;
                if (i == end_chunk)
                        end_offset = end_chunk_offset;
                else
                        end_offset = WIM_CHUNK_SIZE - 1;

                u64 partial_chunk_size = end_offset + 1 - start_offset;
                bool is_partial_chunk = (partial_chunk_size != 
                                                uncompressed_chunk_size);

                DEBUG2("start_offset = %u, end_offset = %u", start_offset,
                                        end_offset);
                DEBUG2("partial_chunk_size = %u", partial_chunk_size);

                /* This is undocumented, but chunks can be uncompressed.  This
                 * appears to always be the case when the compressed chunk size
                 * is equal to the uncompressed chunk size. */
                if (compressed_chunk_size == uncompressed_chunk_size) {
                        /* Probably an uncompressed chunk */

                        if (start_offset != 0) {
                                if (fseeko(fp, start_offset, SEEK_CUR) != 0) {
                                        ERROR_WITH_ERRNO("Uncompressed partial "
                                                         "chunk fseek() error");
                                        return WIMLIB_ERR_READ;
                                }
                        }
                        if (fread(out_p, 1, partial_chunk_size, fp) != 
                                        partial_chunk_size)
                                goto err;
                } else {
                        /* Compressed chunk */
                        int ret;

                        /* Read the compressed data into compressed_buf. */
                        if (fread(compressed_buf, 1, compressed_chunk_size, 
                                                fp) != compressed_chunk_size)
                                goto err;

                        /* For partial chunks we must buffer the uncompressed
                         * data because we don't need all of it. */
                        if (is_partial_chunk) {
                                u8 uncompressed_buf[uncompressed_chunk_size];

                                ret = decompress(compressed_buf,
                                                compressed_chunk_size,
                                                uncompressed_buf, 
                                                uncompressed_chunk_size);
                                if (ret != 0)
                                        return WIMLIB_ERR_DECOMPRESSION;
                                memcpy(out_p, uncompressed_buf + start_offset,
                                                partial_chunk_size);
                        } else {
                                ret = decompress(compressed_buf,
                                                compressed_chunk_size,
                                                out_p,
                                                uncompressed_chunk_size);
                                if (ret != 0)
                                        return WIMLIB_ERR_DECOMPRESSION;
                        }
                }

                /* Advance the pointer into the uncompressed output data by the
                 * number of uncompressed bytes that were written.  */
                out_p += partial_chunk_size;
        }

        return 0;

err:
        if (feof(fp))
                ERROR("Unexpected EOF in compressed file resource");
        else
                ERROR_WITH_ERRNO("Error reading compressed file resource");
        return WIMLIB_ERR_READ;
}

/* 
 * Reads uncompressed data from an open file stream.
 */
int read_uncompressed_resource(FILE *fp, u64 offset, u64 len,
                               u8 contents_ret[])
{
        if (fseeko(fp, offset, SEEK_SET) != 0) {
                ERROR("Failed to seek to byte %"PRIu64" of input file "
                      "to read uncompressed resource (len = %"PRIu64")",
                      offset, len);
                return WIMLIB_ERR_READ;
        }
        if (fread(contents_ret, 1, len, fp) != len) {
                if (feof(fp)) {
                        ERROR("Unexpected EOF in uncompressed file resource");
                } else {
                        ERROR("Failed to read %"PRIu64" bytes from "
                              "uncompressed resource at offset %"PRIu64,
                              len, offset);
                }
                return WIMLIB_ERR_READ;
        }
        return 0;
}




/* Reads the contents of a struct resource_entry, as represented in the on-disk
 * format, from the memory pointed to by @p, and fills in the fields of @entry.
 * A pointer to the byte after the memory read at @p is returned. */
const u8 *get_resource_entry(const u8 *p, struct resource_entry *entry)
{
        u64 size;
        u8 flags;

        p = get_u56(p, &size);
        p = get_u8(p, &flags);
        entry->size = size;
        entry->flags = flags;
        p = get_u64(p, &entry->offset);
        p = get_u64(p, &entry->original_size);
        return p;
}

/* Copies the struct resource_entry @entry to the memory pointed to by @p in the
 * on-disk format.  A pointer to the byte after the memory written at @p is
 * returned. */
u8 *put_resource_entry(u8 *p, const struct resource_entry *entry)
{
        p = put_u56(p, entry->size);
        p = put_u8(p, entry->flags);
        p = put_u64(p, entry->offset);
        p = put_u64(p, entry->original_size);
        return p;
}

/*
 * Reads some data from the resource corresponding to a WIM lookup table entry.
 *
 * @lte:        The WIM lookup table entry for the resource.
 * @buf:        Buffer into which to write the data.
 * @size:       Number of bytes to read.
 * @offset:     Offset at which to start reading the resource.
 * @raw:        If %true, compressed data is read literally rather than being
 *                      decompressed first.
 *
 * Returns zero on success, nonzero on failure.
 */
int read_wim_resource(const struct lookup_table_entry *lte, u8 buf[],
                      size_t size, u64 offset, bool raw)
{
        /* We shouldn't be allowing read over-runs in any part of the library.
         * */
        if (raw)
                wimlib_assert(offset + size <= lte->resource_entry.size);
        else
                wimlib_assert(offset + size <= lte->resource_entry.original_size);

        int ctype;
        int ret;
        FILE *fp;
        switch (lte->resource_location) {
        case RESOURCE_IN_WIM:
                /* The resource is in a WIM file, and its WIMStruct is given by
                 * the lte->wim member.  The resource may be either compressed
                 * or uncompressed. */
                wimlib_assert(lte->wim);
                wimlib_assert(lte->wim->fp);
                ctype = wim_resource_compression_type(lte);

                wimlib_assert(ctype != WIM_COMPRESSION_TYPE_NONE ||
                              (lte->resource_entry.original_size ==
                               lte->resource_entry.size));

                if (raw || ctype == WIM_COMPRESSION_TYPE_NONE)
                        return read_uncompressed_resource(lte->wim->fp,
                                                          lte->resource_entry.offset + offset,
                                                          size, buf);
                else
                        return read_compressed_resource(lte->wim->fp,
                                                        lte->resource_entry.size,
                                                        lte->resource_entry.original_size,
                                                        lte->resource_entry.offset,
                                                        ctype, size, offset, buf);
                break;
        case RESOURCE_IN_STAGING_FILE:
        case RESOURCE_IN_FILE_ON_DISK:
                /* The resource is in some file on the external filesystem and
                 * needs to be read uncompressed */
                wimlib_assert(lte->file_on_disk);
                wimlib_assert(&lte->file_on_disk == &lte->staging_file_name);
                /* Use existing file pointer if available; otherwise open one
                 * temporarily */
                if (lte->file_on_disk_fp) {
                        fp = lte->file_on_disk_fp;
                } else {
                        fp = fopen(lte->file_on_disk, "rb");
                        if (!fp) {
                                ERROR_WITH_ERRNO("Failed to open the file "
                                                 "`%s'", lte->file_on_disk);
                                return WIMLIB_ERR_OPEN;
                        }
                }
                ret = read_uncompressed_resource(fp, offset, size, buf);
                if (fp != lte->file_on_disk_fp)
                        fclose(fp);
                return ret;
                break;
        case RESOURCE_IN_ATTACHED_BUFFER:
                /* The resource is directly attached uncompressed in an
                 * in-memory buffer. */
                wimlib_assert(lte->attached_buffer);
                memcpy(buf, lte->attached_buffer + offset, size);
                return 0;
                break;
#ifdef WITH_NTFS_3G
        case RESOURCE_IN_NTFS_VOLUME:
                if (lte->attr) {
                        if (ntfs_attr_pread(lte->attr, offset, size, buf) == size) {
                                return 0;
                        } else {
                                ERROR_WITH_ERRNO("Error reading NTFS attribute "
                                                 "at `%s'",
                                                 lte->ntfs_loc->path_utf8);
                                return WIMLIB_ERR_NTFS_3G;
                        }
                } else {
                        wimlib_assert(0);
                }
                break;
#endif
        default:
                assert(0);
        }
}

/* 
 * Reads all the data from the resource corresponding to a WIM lookup table
 * entry.
 *
 * @lte:        The WIM lookup table entry for the resource.
 * @buf:        Buffer into which to write the data.  It must be at least
 *              wim_resource_size(lte) bytes long.
 *
 * Returns 0 on success; nonzero on failure.
 */
int read_full_wim_resource(const struct lookup_table_entry *lte, u8 buf[])
{
        return read_wim_resource(lte, buf, wim_resource_size(lte), 0, false);
}

/* Chunk table that's located at the beginning of each compressed resource in
 * the WIM.  (This is not the on-disk format; the on-disk format just has an
 * array of offsets.) */
struct chunk_table {
        off_t file_offset;
        u64 num_chunks;
        u64 original_resource_size;
        u64 bytes_per_chunk_entry;
        u64 table_disk_size;
        u64 cur_offset;
        u64 *cur_offset_p;
        u64 offsets[0];
};

/* 
 * Allocates and initializes a chunk table, and reserves space for it in the
 * output file.
 */
static int
begin_wim_resource_chunk_tab(const struct lookup_table_entry *lte,
                             FILE *out_fp,
                             off_t file_offset,
                             struct chunk_table **chunk_tab_ret)
{
        u64 size = wim_resource_size(lte);
        u64 num_chunks = (size + WIM_CHUNK_SIZE - 1) / WIM_CHUNK_SIZE;
        struct chunk_table *chunk_tab = MALLOC(sizeof(struct chunk_table) +
                                               num_chunks * sizeof(u64));
        int ret = 0;

        wimlib_assert(size != 0);

        if (!chunk_tab) {
                ERROR("Failed to allocate chunk table for %"PRIu64" byte "
                      "resource", size);
                ret = WIMLIB_ERR_NOMEM;
                goto out;
        }
        chunk_tab->file_offset = file_offset;
        chunk_tab->num_chunks = num_chunks;
        chunk_tab->original_resource_size = size;
        chunk_tab->bytes_per_chunk_entry = (size >= (1ULL << 32)) ? 8 : 4;
        chunk_tab->table_disk_size = chunk_tab->bytes_per_chunk_entry *
                                     (num_chunks - 1);
        chunk_tab->cur_offset = 0;
        chunk_tab->cur_offset_p = chunk_tab->offsets;

        if (fwrite(chunk_tab, 1, chunk_tab->table_disk_size, out_fp) !=
                   chunk_tab->table_disk_size) {
                ERROR_WITH_ERRNO("Failed to write chunk table in compressed "
                                 "file resource");
                ret = WIMLIB_ERR_WRITE;
                goto out;
        }

        *chunk_tab_ret = chunk_tab;
out:
        return ret;
}

/* 
 * Compresses a chunk of a WIM resource.
 *
 * @chunk:              Uncompressed data of the chunk.
 * @chunk_size:         Size of the uncompressed chunk in bytes.
 * @compressed_chunk:   Pointer to output buffer of size at least
 *                              (@chunk_size - 1) bytes.
 * @compressed_chunk_len_ret:   Pointer to an unsigned int into which the size
 *                                      of the compressed chunk will be
 *                                      returned.
 * @ctype:      Type of compression to use.  Must be WIM_COMPRESSION_TYPE_LZX
 *              or WIM_COMPRESSION_TYPE_XPRESS.
 *
 * Returns zero if compressed succeeded, and nonzero if the chunk could not be
 * compressed to any smaller than @chunk_size.  This function cannot fail for
 * any other reasons.
 */
static int compress_chunk(const u8 chunk[], unsigned chunk_size,
                          u8 compressed_chunk[],
                          unsigned *compressed_chunk_len_ret,
                          int ctype)
{
        int (*compress)(const void *, unsigned, void *, unsigned *);
        switch (ctype) {
        case WIM_COMPRESSION_TYPE_LZX:
                compress = lzx_compress;
                break;
        case WIM_COMPRESSION_TYPE_XPRESS:
                compress = xpress_compress;
                break;
        default:
                wimlib_assert(0);
                break;
        }
        return (*compress)(chunk, chunk_size, compressed_chunk,
                           compressed_chunk_len_ret);
}

/*
 * Writes a chunk of a WIM resource to an output file.
 *
 * @chunk:        Uncompressed data of the chunk.
 * @chunk_size:   Size of the chunk (<= WIM_CHUNK_SIZE)
 * @out_fp:       FILE * to write tho chunk to.
 * @out_ctype:    Compression type to use when writing the chunk (ignored if no 
 *                      chunk table provided)
 * @chunk_tab:    Pointer to chunk table being created.  It is updated with the
 *                      offset of the chunk we write.
 *
 * Returns 0 on success; nonzero on failure.
 */
static int write_wim_resource_chunk(const u8 chunk[], unsigned chunk_size,
                                    FILE *out_fp, int out_ctype,
                                    struct chunk_table *chunk_tab)
{
        const u8 *out_chunk;
        unsigned out_chunk_size;

        wimlib_assert(chunk_size <= WIM_CHUNK_SIZE);

        if (!chunk_tab) {
                out_chunk = chunk;
                out_chunk_size = chunk_size;
        } else {
                u8 *compressed_chunk = alloca(chunk_size);
                int ret;

                ret = compress_chunk(chunk, chunk_size, compressed_chunk,
                                     &out_chunk_size, out_ctype);
                if (ret == 0) {
                        out_chunk = compressed_chunk;
                } else {
                        out_chunk = chunk;
                        out_chunk_size = chunk_size;
                }
                *chunk_tab->cur_offset_p++ = chunk_tab->cur_offset;
                chunk_tab->cur_offset += out_chunk_size;
        }
        
        if (fwrite(out_chunk, 1, out_chunk_size, out_fp) != out_chunk_size) {
                ERROR_WITH_ERRNO("Failed to write WIM resource chunk");
                return WIMLIB_ERR_WRITE;
        }
        return 0;
}

/* 
 * Finishes a WIM chunk tale and writes it to the output file at the correct
 * offset.
 *
 * The final size of the full compressed resource is returned in the
 * @compressed_size_p.
 */
static int
finish_wim_resource_chunk_tab(struct chunk_table *chunk_tab,
                              FILE *out_fp, u64 *compressed_size_p)
{
        size_t bytes_written;
        if (fseeko(out_fp, chunk_tab->file_offset, SEEK_SET) != 0) {
                ERROR_WITH_ERRNO("Failed to seek to byte %"PRIu64" of output "
                                 "WIM file", chunk_tab->file_offset);
                return WIMLIB_ERR_WRITE;
        }

        if (chunk_tab->bytes_per_chunk_entry == 8) {
                array_to_le64(chunk_tab->offsets, chunk_tab->num_chunks);
        } else {
                for (u64 i = 0; i < chunk_tab->num_chunks; i++)
                        ((u32*)chunk_tab->offsets)[i] =
                                to_le32(chunk_tab->offsets[i]);
        }
        bytes_written = fwrite((u8*)chunk_tab->offsets +
                                        chunk_tab->bytes_per_chunk_entry,
                               1, chunk_tab->table_disk_size, out_fp);
        if (bytes_written != chunk_tab->table_disk_size) {
                ERROR_WITH_ERRNO("Failed to write chunk table in compressed "
                                 "file resource");
                return WIMLIB_ERR_WRITE;
        }
        if (fseeko(out_fp, 0, SEEK_END) != 0) {
                ERROR_WITH_ERRNO("Failed to seek to end of output WIM file");
                return WIMLIB_ERR_WRITE;
        }
        *compressed_size_p = chunk_tab->cur_offset + chunk_tab->table_disk_size;
        return 0;
}

/*
 * Writes a WIM resource to a FILE * opened for writing.  The resource may be
 * written uncompressed or compressed depending on the @out_ctype parameter.
 *
 * If by chance the resource compresses to more than the original size (this may
 * happen with random data or files than are pre-compressed), the resource is
 * instead written uncompressed (and this is reflected in the @out_res_entry by
 * removing the WIM_RESHDR_FLAG_COMPRESSED flag).
 *
 * @lte:        The lookup table entry for the WIM resource.
 * @out_fp:     The FILE * to write the resource to.
 * @out_ctype:  The compression type of the resource to write.  Note: if this is
 *                      the same as the compression type of the WIM resource we
 *                      need to read, we simply copy the data (i.e. we do not
 *                      uncompress it, then compress it again).
 * @out_res_entry:  If non-NULL, a resource entry that is filled in with the 
 *                  offset, original size, compressed size, and compression flag
 *                  of the output resource.
 *
 * Returns 0 on success; nonzero on failure.
 */
static int write_wim_resource(struct lookup_table_entry *lte,
                              FILE *out_fp, int out_ctype,
                              struct resource_entry *out_res_entry)
{
        u64 bytes_remaining;
        u64 original_size;
        u64 old_compressed_size;
        u64 new_compressed_size;
        u64 offset = 0;
        int ret = 0;
        struct chunk_table *chunk_tab = NULL;
        bool raw;
        off_t file_offset;
#ifdef WITH_NTFS_3G
        ntfs_inode *ni;
#endif

        /* Original size of the resource */
        original_size = wim_resource_size(lte);

        /* Compressed size of the resource (as it exists now) */
        old_compressed_size = wim_resource_compressed_size(lte);

        /* Current offset in output file */
        file_offset = ftello(out_fp);
        if (file_offset == -1) {
                ERROR_WITH_ERRNO("Failed to get offset in output "
                                 "stream");
                return WIMLIB_ERR_WRITE;
        }
        
        /* Are the compression types the same?  If so, do a raw copy (copy
         * without decompressing and recompressing the data). */
        raw = (wim_resource_compression_type(lte) == out_ctype
               && out_ctype != WIM_COMPRESSION_TYPE_NONE);
        if (raw)
                bytes_remaining = old_compressed_size;
        else
                bytes_remaining = original_size;

        /* Empty resource; nothing needs to be done, so just return success. */
        if (bytes_remaining == 0)
                return 0;

        /* Buffer for reading chunks for the resource */
        u8 buf[min(WIM_CHUNK_SIZE, bytes_remaining)];

        /* If we are writing a compressed resource and not doing a raw copy, we
         * need to initialize the chunk table */
        if (out_ctype != WIM_COMPRESSION_TYPE_NONE && !raw) {
                ret = begin_wim_resource_chunk_tab(lte, out_fp, file_offset,
                                                   &chunk_tab);
                if (ret != 0)
                        goto out;
        }

        /* If the WIM resource is in an external file, open a FILE * to it so we
         * don't have to open a temporary one in read_wim_resource() for each
         * chunk. */
        if (lte->resource_location == RESOURCE_IN_FILE_ON_DISK
             && !lte->file_on_disk_fp)
        {
                wimlib_assert(lte->file_on_disk);
                lte->file_on_disk_fp = fopen(lte->file_on_disk, "rb");
                if (!lte->file_on_disk_fp) {
                        ERROR_WITH_ERRNO("Failed to open the file `%s' for "
                                         "reading", lte->file_on_disk);
                        ret = WIMLIB_ERR_OPEN;
                        goto out;
                }
        }
#ifdef WITH_NTFS_3G
        else if (lte->resource_location == RESOURCE_IN_NTFS_VOLUME
                   && !lte->attr)
        {
                struct ntfs_location *loc = lte->ntfs_loc;
                wimlib_assert(loc);
                ni = ntfs_pathname_to_inode(*loc->ntfs_vol_p, NULL, loc->path_utf8);
                if (!ni) {
                        ERROR_WITH_ERRNO("Failed to open inode `%s' in NTFS "
                                         "volume", loc->path_utf8);
                }
                lte->attr = ntfs_attr_open(ni,
                                           loc->is_reparse_point ? AT_REPARSE_POINT : AT_DATA,
                                           (ntfschar*)loc->stream_name_utf16,
                                           loc->stream_name_utf16_num_chars);
                if (!lte->attr) {
                        ntfs_inode_close(ni);
                        ERROR_WITH_ERRNO("Failed to open attribute of `%s' in "
                                         "NTFS volume", loc->path_utf8);
                }
        }
#endif

        /* If we aren't doing a raw copy, we will compute the SHA1 message
         * digest of the resource as we read it, and verify it's the same as the
         * hash given in the lookup table entry once we've finished reading the
         * resource. */
        SHA_CTX ctx;
        if (!raw)
                sha1_init(&ctx);

        /* While there are still bytes remaining in the WIM resource, read a
         * chunk of the resource, update SHA1, then write that chunk using the
         * desired compression type. */
        do {
                u64 to_read = min(bytes_remaining, WIM_CHUNK_SIZE);
                ret = read_wim_resource(lte, buf, to_read, offset, raw);
                if (ret != 0)
                        goto out_fclose;
                if (!raw)
                        sha1_update(&ctx, buf, to_read);
                ret = write_wim_resource_chunk(buf, to_read, out_fp,
                                               out_ctype, chunk_tab);
                if (ret != 0)
                        goto out_fclose;
                bytes_remaining -= to_read;
                offset += to_read;
        } while (bytes_remaining);

        /* If writing a compressed resource and not doing a raw copy, write the
         * chunk table, and finish_wim_resource_chunk_tab() will provide the
         * compressed size of the resource we wrote.  Otherwise, the compressed
         * size of the written resource is the same as the compressed size of
         * the existing resource. */
        if (out_ctype != WIM_COMPRESSION_TYPE_NONE && !raw) {
                ret = finish_wim_resource_chunk_tab(chunk_tab, out_fp,
                                                    &new_compressed_size);
                if (ret != 0)
                        goto out_fclose;
        } else {
                new_compressed_size = old_compressed_size;
        }

        /* Verify SHA1 message digest of the resource, unless we are doing a raw
         * write (in which case we never even saw the uncompressed data).  Or,
         * if the hash we had before is all 0's, just re-set it to be the new
         * hash. */
        if (!raw) {
                u8 md[SHA1_HASH_SIZE];
                sha1_final(md, &ctx);
                if (is_zero_hash(lte->hash)) {
                        copy_hash(lte->hash, md);
                } else if (!hashes_equal(md, lte->hash)) {
                        ERROR("WIM resource has incorrect hash!");
                        if (lte->resource_location == RESOURCE_IN_FILE_ON_DISK) {
                                ERROR("We were reading it from `%s'; maybe it changed "
                                      "while we were reading it.",
                                      lte->file_on_disk);
                        }
                        ret = WIMLIB_ERR_INVALID_RESOURCE_HASH;
                        goto out_fclose;
                }
        }

        if (new_compressed_size >= original_size &&
            out_ctype != WIM_COMPRESSION_TYPE_NONE && !raw)
        {
                /* Oops!  We compressed the resource to larger than the original
                 * size.  Write the resource uncompressed instead. */
                if (fseeko(out_fp, file_offset, SEEK_SET) != 0) {
                        ERROR_WITH_ERRNO("Failed to seek to byte %"PRIu64" "
                                         "of output WIM file", file_offset);
                        ret = WIMLIB_ERR_WRITE;
                        goto out_fclose;
                }
                ret = write_wim_resource(lte, out_fp, WIM_COMPRESSION_TYPE_NONE,
                                         out_res_entry);
                if (ret != 0)
                        goto out_fclose;
                if (fflush(out_fp) != 0) {
                        ERROR_WITH_ERRNO("Failed to flush output WIM file");
                        ret = WIMLIB_ERR_WRITE;
                        goto out_fclose;
                }
                if (ftruncate(fileno(out_fp), file_offset + out_res_entry->size) != 0) {
                        ERROR_WITH_ERRNO("Failed to truncate output WIM file");
                        ret = WIMLIB_ERR_WRITE;
                }
                goto out_fclose;
        }
        wimlib_assert(new_compressed_size <= original_size);
        if (out_res_entry) {
                out_res_entry->size          = new_compressed_size;
                out_res_entry->original_size = original_size;
                out_res_entry->offset        = file_offset;
                out_res_entry->flags         = lte->resource_entry.flags
                                                & ~WIM_RESHDR_FLAG_COMPRESSED;
                if (out_ctype != WIM_COMPRESSION_TYPE_NONE)
                        out_res_entry->flags |= WIM_RESHDR_FLAG_COMPRESSED;
        }
out_fclose:
        if (lte->resource_location == RESOURCE_IN_FILE_ON_DISK
             && lte->file_on_disk_fp) {
                fclose(lte->file_on_disk_fp);
                lte->file_on_disk_fp = NULL;
        }
#ifdef WITH_NTFS_3G
        else if (lte->resource_location == RESOURCE_IN_NTFS_VOLUME
                 && lte->attr) {
                ntfs_attr_close(lte->attr);
                ntfs_inode_close(ni);
        }
#endif
out:
        FREE(chunk_tab);
        return ret;
}

/* Like write_wim_resource(), but the resource is specified by a buffer of
 * uncompressed data rather a lookup table entry; also writes the SHA1 hash of
 * the buffer to @hash.  */
static int write_wim_resource_from_buffer(const u8 *buf, u64 buf_size,
                                          FILE *out_fp, int out_ctype,
                                          struct resource_entry *out_res_entry,
                                          u8 hash[SHA1_HASH_SIZE])
{
        /* Set up a temporary lookup table entry that we provide to
         * write_wim_resource(). */
        struct lookup_table_entry lte;
        int ret;
        lte.resource_entry.flags         = 0;
        lte.resource_entry.original_size = buf_size;
        lte.resource_entry.size          = buf_size;
        lte.resource_entry.offset        = 0;
        lte.resource_location            = RESOURCE_IN_ATTACHED_BUFFER;
        lte.attached_buffer              = (u8*)buf;

        zero_hash(lte.hash);
        ret = write_wim_resource(&lte, out_fp, out_ctype, out_res_entry);
        if (ret != 0)
                return ret;
        copy_hash(hash, lte.hash);
        return 0;
}

/* 
 * Extracts the first @size bytes of the WIM resource specified by @lte to the
 * open file descriptor @fd.
 * 
 * Returns 0 on success; nonzero on failure.
 */
int extract_wim_resource_to_fd(const struct lookup_table_entry *lte, int fd,
                               u64 size)
{
        u64 bytes_remaining = size;
        u8 buf[min(WIM_CHUNK_SIZE, bytes_remaining)];
        u64 offset = 0;
        int ret = 0;
        u8 hash[SHA1_HASH_SIZE];

        SHA_CTX ctx;
        sha1_init(&ctx);

        while (bytes_remaining) {
                u64 to_read = min(bytes_remaining, WIM_CHUNK_SIZE);
                ret = read_wim_resource(lte, buf, to_read, offset, false);
                if (ret != 0)
                        break;
                sha1_update(&ctx, buf, to_read);
                if (full_write(fd, buf, to_read) < 0) {
                        ERROR_WITH_ERRNO("Error extracting WIM resource");
                        return WIMLIB_ERR_WRITE;
                }
                bytes_remaining -= to_read;
                offset += to_read;
        }
        sha1_final(hash, &ctx);
        if (!hashes_equal(hash, lte->hash)) {
                ERROR("Invalid checksum on a WIM resource "
                      "(detected when extracting to external file)");
                ERROR("The following WIM resource is invalid:");
                print_lookup_table_entry(lte);
                return WIMLIB_ERR_INVALID_RESOURCE_HASH;
        }
        return 0;
}

/* 
 * Extracts the WIM resource specified by @lte to the open file descriptor @fd.
 * 
 * Returns 0 on success; nonzero on failure.
 */
int extract_full_wim_resource_to_fd(const struct lookup_table_entry *lte, int fd)
{
        return extract_wim_resource_to_fd(lte, fd, wim_resource_size(lte));
}

/* 
 * Copies the file resource specified by the lookup table entry @lte from the
 * input WIM to the output WIM that has its FILE * given by
 * ((WIMStruct*)wim)->out_fp.
 *
 * The output_resource_entry, out_refcnt, and part_number fields of @lte are
 * updated.
 *
 * Metadata resources are not copied (they are handled elsewhere for joining and
 * splitting).
 */
int copy_resource(struct lookup_table_entry *lte, void *wim)
{
        WIMStruct *w = wim;
        int ret;

        if ((lte->resource_entry.flags & WIM_RESHDR_FLAG_METADATA) &&
            !w->write_metadata)
                return 0;

        ret = write_wim_resource(lte, w->out_fp,
                                 wim_resource_compression_type(lte), 
                                 &lte->output_resource_entry);
        if (ret != 0)
                return ret;
        lte->out_refcnt = lte->refcnt;
        lte->part_number = w->hdr.part_number;
        return 0;
}

/* 
 * Writes a dentry's resources, including the main file resource as well as all
 * alternate data streams, to the output file. 
 *
 * @dentry:  The dentry for the file.
 * @wim_p:   A pointer to the WIMStruct containing @dentry.
 *
 * @return zero on success, nonzero on failure. 
 */
int write_dentry_resources(struct dentry *dentry, void *wim_p)
{
        WIMStruct *w = wim_p;
        int ret = 0;
        struct lookup_table_entry *lte;
        int ctype = wimlib_get_compression_type(w);

        if (w->write_flags & WIMLIB_WRITE_FLAG_VERBOSE) {
                wimlib_assert(dentry->full_path_utf8);
                printf("Writing streams for `%s'\n", dentry->full_path_utf8);
        }

        for (unsigned i = 0; i <= dentry->num_ads; i++) {
                lte = dentry_stream_lte(dentry, i, w->lookup_table);
                if (lte && ++lte->out_refcnt == 1) {
                        ret = write_wim_resource(lte, w->out_fp, ctype,
                                                 &lte->output_resource_entry);
                        if (ret != 0)
                                break;
                }
        }
        return ret;
}

/* 
 * Reads the metadata metadata resource from the WIM file.  The metadata
 * resource consists of the security data, followed by the directory entry for
 * the root directory, followed by all the other directory entries in the
 * filesystem.  The subdir_offset field of each directory entry gives the start
 * of its child entries from the beginning of the metadata resource.  An
 * end-of-directory is signaled by a directory entry of length '0', really of
 * length 8, because that's how long the 'length' field is.
 *
 * @fp:         The FILE* for the input WIM file.
 * @wim_ctype:  The compression type of the WIM file.
 * @imd:        Pointer to the image metadata structure.  Its `metadata_lte'
 *              member specifies the lookup table entry for the metadata
 *              resource.  The rest of the image metadata entry will be filled
 *              in by this function.
 *
 * @return:     Zero on success, nonzero on failure.
 */
int read_metadata_resource(WIMStruct *w, struct image_metadata *imd)
{
        u8 *buf;
        u32 dentry_offset;
        int ret;
        struct dentry *dentry;
        struct link_group_table *lgt;
        const struct lookup_table_entry *metadata_lte;
        u64 metadata_len;
        u64 metadata_offset;

        metadata_lte = imd->metadata_lte;
        metadata_len = wim_resource_size(metadata_lte);
        metadata_offset = metadata_lte->resource_entry.offset;

        DEBUG("Reading metadata resource: length = %"PRIu64", "
              "offset = %"PRIu64"", metadata_len, metadata_offset);

        /* There is no way the metadata resource could possibly be less than (8
         * + WIM_DENTRY_DISK_SIZE) bytes, where the 8 is for security data (with
         * no security descriptors) and WIM_DENTRY_DISK_SIZE is for the root
         * dentry. */
        if (metadata_len < 8 + WIM_DENTRY_DISK_SIZE) {
                ERROR("Expected at least %u bytes for the metadata resource",
                      8 + WIM_DENTRY_DISK_SIZE);
                return WIMLIB_ERR_INVALID_RESOURCE_SIZE;
        }

        /* Allocate memory for the uncompressed metadata resource. */
        buf = MALLOC(metadata_len);

        if (!buf) {
                ERROR("Failed to allocate %"PRIu64" bytes for uncompressed "
                      "metadata resource", metadata_len);
                return WIMLIB_ERR_NOMEM;
        }

        /* Read the metadata resource into memory.  (It may be compressed.) */
        ret = read_full_wim_resource(metadata_lte, buf);
        if (ret != 0)
                goto out_free_buf;

        DEBUG("Finished reading metadata resource into memory.");

        /* The root directory entry starts after security data, aligned on an
         * 8-byte boundary within the metadata resource.
         *
         * The security data starts with a 4-byte integer giving its total
         * length, so if we round that up to an 8-byte boundary that gives us
         * the offset of the root dentry.
         *
         * Here we read the security data into a wim_security_data structure,
         * and if successful, go ahead and calculate the offset in the metadata
         * resource of the root dentry. */

        ret = read_security_data(buf, metadata_len, &imd->security_data);
        if (ret != 0)
                goto out_free_buf;

        get_u32(buf, &dentry_offset);
        if (dentry_offset == 0)
                dentry_offset = 8;
        dentry_offset = (dentry_offset + 7) & ~7;

        /* Allocate memory for the root dentry and read it into memory */
        dentry = MALLOC(sizeof(struct dentry));
        if (!dentry) {
                ERROR("Failed to allocate %zu bytes for root dentry",
                      sizeof(struct dentry));
                ret = WIMLIB_ERR_NOMEM;
                goto out_free_security_data;
        }
                
        ret = read_dentry(buf, metadata_len, dentry_offset, dentry);

        /* This is the root dentry, so set its pointers correctly. */
        dentry->parent = dentry;
        dentry->next   = dentry;
        dentry->prev   = dentry;
        if (ret != 0)
                goto out_free_dentry_tree;

        /* Now read the entire directory entry tree into memory. */
        DEBUG("Reading dentry tree");
        ret = read_dentry_tree(buf, metadata_len, dentry);
        if (ret != 0)
                goto out_free_dentry_tree;

        /* Calculate the full paths in the dentry tree. */
        DEBUG("Calculating dentry full paths");
        ret = for_dentry_in_tree(dentry, calculate_dentry_full_path, NULL);
        if (ret != 0)
                goto out_free_dentry_tree;

        /* Build hash table that maps hard link group IDs to dentry sets */
        DEBUG("Building link group table");
        lgt = new_link_group_table(9001);
        if (!lgt)
                goto out_free_dentry_tree;
        ret = for_dentry_in_tree(dentry, link_group_table_insert, lgt);
        if (ret != 0)
                goto out_free_lgt;

        DEBUG("Fixing inconsistencies in the link groups");
        ret = fix_link_groups(lgt);
        if (ret != 0)
                goto out_free_lgt;

        DEBUG("Running miscellaneous verifications on the dentry tree");
        ret = for_dentry_in_tree(dentry, verify_dentry, w);
        if (ret != 0)
                goto out_free_lgt;

        DEBUG("Done reading image metadata");

        imd->lgt           = lgt;
        imd->root_dentry   = dentry;
        goto out_free_buf;
out_free_lgt:
        free_link_group_table(lgt);
out_free_dentry_tree:
        free_dentry_tree(dentry, NULL);
out_free_security_data:
        free_security_data(imd->security_data);
        imd->security_data = NULL;
out_free_buf:
        FREE(buf);
        return ret;
}

/* Write the metadata resource for the current WIM image. */
int write_metadata_resource(WIMStruct *w)
{
        u8 *buf;
        u8 *p;
        int ret;
        u64 subdir_offset;
        struct dentry *root;
        struct lookup_table_entry *lte;
        u64 metadata_original_size;
        const struct wim_security_data *sd;
        const unsigned random_tail_len = 20;

        DEBUG("Writing metadata resource for image %d", w->current_image);

        root = wim_root_dentry(w);
        sd = wim_security_data(w);

        /* We do not allow the security data pointer to be NULL, although it may
         * point to an empty security data with no entries. */
        wimlib_assert(sd);

        /* Offset of first child of the root dentry.  It's equal to:
         * - The total length of the security data, rounded to the next 8-byte
         *   boundary,
         * - plus the total length of the root dentry,
         * - plus 8 bytes for an end-of-directory entry following the root
         *   dentry (shouldn't really be needed, but just in case...)
         */
        subdir_offset = ((sd->total_length + 7) & ~7) +
                        dentry_correct_total_length(root) + 8;

        /* Calculate the subdirectory offsets for the entire dentry tree. */
        calculate_subdir_offsets(root, &subdir_offset);

        /* Total length of the metadata resource (uncompressed) */
        metadata_original_size = subdir_offset + random_tail_len;

        /* Allocate a buffer to contain the uncompressed metadata resource */
        buf = MALLOC(metadata_original_size);
        if (!buf) {
                ERROR("Failed to allocate %"PRIu64" bytes for "
                      "metadata resource", metadata_original_size);
                return WIMLIB_ERR_NOMEM;
        }

        /* Write the security data into the resource buffer */
        p = write_security_data(sd, buf);

        /* Write the dentry tree into the resource buffer */
        DEBUG("Writing dentry tree.");
        p = write_dentry_tree(root, p);

        /* 
         * Append 20 random bytes to the metadata resource so that we don't have
         * identical metadata resources if we happen to append exactly the same
         * image twice without any changes in timestamps.  If this were to
         * happen, it would cause confusion about the number and order of images
         * in the WIM.
         */
        randomize_byte_array(p, random_tail_len);

        /* We MUST have exactly filled the buffer; otherwise we calculated its
         * size incorrectly or wrote the data incorrectly. */
        wimlib_assert(p - buf + random_tail_len == metadata_original_size);

        /* Get the lookup table entry for the metadata resource so we can update
         * it. */
        lte = wim_metadata_lookup_table_entry(w);

        /* Write the metadata resource to the output WIM using the proper
         * compression type.  The lookup table entry for the metadata resource
         * is updated. */
        ret = write_wim_resource_from_buffer(buf, metadata_original_size,
                                             w->out_fp,
                                             wimlib_get_compression_type(w),
                                             &lte->output_resource_entry,
                                             lte->hash);
        if (ret != 0)
                goto out;

        /* It's very likely the SHA1 message digest of the metadata resource, so
         * re-insert the lookup table entry into the lookup table. */
        lookup_table_unlink(w->lookup_table, lte);
        lookup_table_insert(w->lookup_table, lte);

        /* We do not allow a metadata resource to be referenced multiple times,
         * and the 20 random bytes appended to it should make it extremely
         * likely for each metadata resource to be unique, even if the exact
         * same image is captured. */
        wimlib_assert(lte->out_refcnt == 0);
        lte->out_refcnt = 1;

        /* Make sure that the resource entry is written marked with the metadata
         * flag. */
        lte->output_resource_entry.flags |= WIM_RESHDR_FLAG_METADATA;
out:
        /* All the data has been written to the new WIM; no need for the buffer
         * anymore */
        FREE(buf);
        return ret;
}