freebsd-dev/lib/libarchive/archive_private.h

233 lines
8.3 KiB
C
Raw Normal View History

/*-
* Copyright (c) 2003-2004 Tim Kientzle
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer
* in this position and unchanged.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* $FreeBSD$
*/
#ifndef ARCHIVE_PRIVATE_H_INCLUDED
#define ARCHIVE_PRIVATE_H_INCLUDED
#include "archive.h"
#include "archive_string.h"
#define ARCHIVE_WRITE_MAGIC (0xb0c5c0deU)
#define ARCHIVE_READ_MAGIC (0xdeb0c5U)
/*
* This is used by archive_extract to keep track of non-writable
* directories so that they can be initially restored writable, then
* fixed up at end. This also handles mtime/atime fixups.
*/
struct archive_extract_dir_entry;
struct archive {
/*
* The magic/state values are used to sanity-check the
* client's usage. If an API function is called at a
* rediculous time, or the client passes us an invalid
* pointer, these values allow me to catch that.
*/
unsigned magic;
unsigned state;
struct archive_entry *entry;
/* Utility: Pointer to a block of nulls. */
const char *nulls;
size_t null_length;
/*
* Used to limit reads of entry data. Eventually, each reader
* will be able to register it's own read_data routine and these
* will move into the per-format data for the formats that use them.
*/
Many fixes: * Disabled shared-library building, as some API breakage is still likely. (I didn't realize it was turned on by default.) If you have an existing /usr/lib/libarchive.so.2, I recommend deleting it. * Pax interchange format now correctly stores and reads UTF8 for extended attributes. In particular, pax format can portably handle arbitrarily long pathnames containing arbitrary characters. * Library compiles cleanly at -O2, -O3, and WARNS=6 on all FreeBSD-CURRENT platforms. * Minor portability improvements inspired by Juergen Lock and Greg Lewis. (Less reliance on stdint.h, isolating of various portability-challenged constructs.) * archive_entry transparently converts multi-byte <-> wide character strings, allowing clients and format handlers to deal with either one, as appropriate. * Support for reading 'L' and 'K' entries in standard tar archives for star compatibility. * Recognize (but don't yet handle) ACL entries from Solaris tar. * Pushed format-specific data for format readers down into format-specific storage and out of library-global storage. This should make it easier to maintain individual formats without mucking with the core library management. * Documentation updates to track the above changes. * Updates to tar.5 to correct a few mistakes and add some additional information about GNU tar and Solaris tar formats. Notes: * The basic 'tar' reader is getting more general; there's not much point in keeping the 'gnutar' reader separate. Merging the two would lose a bunch of duplicate code. * The libc ACL support is looking increasingly inadequate for my needs here. I might need to assemble some fairly significant code for parsing and building ACLs. <sigh>
2004-03-19 22:37:06 +00:00
off_t entry_bytes_remaining;
off_t entry_padding; /* Skip this much after entry data. */
uid_t user_uid; /* UID of current user. */
/* Callbacks to open/read/write/close archive stream. */
archive_open_callback *client_opener;
archive_read_callback *client_reader;
archive_write_callback *client_writer;
archive_close_callback *client_closer;
void *client_data;
/*
* Blocking information. Note that bytes_in_last_block is
* misleadingly named; I should find a better name. These
* control the final output from all compressors, including
* compression_none.
*/
int bytes_per_block;
int bytes_in_last_block;
/*
* These control whether data within a gzip/bzip2 compressed
* stream gets padded or not. If pad_uncompressed is set,
* the data will be padded to a full block before being
* compressed. The pad_uncompressed_byte determines the value
* that will be used for padding. Note that these have no
* effect on compression "none."
*/
int pad_uncompressed;
int pad_uncompressed_byte; /* TODO: Support this. */
/* Position in UNCOMPRESSED data stream. */
Many fixes: * Disabled shared-library building, as some API breakage is still likely. (I didn't realize it was turned on by default.) If you have an existing /usr/lib/libarchive.so.2, I recommend deleting it. * Pax interchange format now correctly stores and reads UTF8 for extended attributes. In particular, pax format can portably handle arbitrarily long pathnames containing arbitrary characters. * Library compiles cleanly at -O2, -O3, and WARNS=6 on all FreeBSD-CURRENT platforms. * Minor portability improvements inspired by Juergen Lock and Greg Lewis. (Less reliance on stdint.h, isolating of various portability-challenged constructs.) * archive_entry transparently converts multi-byte <-> wide character strings, allowing clients and format handlers to deal with either one, as appropriate. * Support for reading 'L' and 'K' entries in standard tar archives for star compatibility. * Recognize (but don't yet handle) ACL entries from Solaris tar. * Pushed format-specific data for format readers down into format-specific storage and out of library-global storage. This should make it easier to maintain individual formats without mucking with the core library management. * Documentation updates to track the above changes. * Updates to tar.5 to correct a few mistakes and add some additional information about GNU tar and Solaris tar formats. Notes: * The basic 'tar' reader is getting more general; there's not much point in keeping the 'gnutar' reader separate. Merging the two would lose a bunch of duplicate code. * The libc ACL support is looking increasingly inadequate for my needs here. I might need to assemble some fairly significant code for parsing and building ACLs. <sigh>
2004-03-19 22:37:06 +00:00
off_t file_position;
/* File offset of beginning of most recently-read header. */
Many fixes: * Disabled shared-library building, as some API breakage is still likely. (I didn't realize it was turned on by default.) If you have an existing /usr/lib/libarchive.so.2, I recommend deleting it. * Pax interchange format now correctly stores and reads UTF8 for extended attributes. In particular, pax format can portably handle arbitrarily long pathnames containing arbitrary characters. * Library compiles cleanly at -O2, -O3, and WARNS=6 on all FreeBSD-CURRENT platforms. * Minor portability improvements inspired by Juergen Lock and Greg Lewis. (Less reliance on stdint.h, isolating of various portability-challenged constructs.) * archive_entry transparently converts multi-byte <-> wide character strings, allowing clients and format handlers to deal with either one, as appropriate. * Support for reading 'L' and 'K' entries in standard tar archives for star compatibility. * Recognize (but don't yet handle) ACL entries from Solaris tar. * Pushed format-specific data for format readers down into format-specific storage and out of library-global storage. This should make it easier to maintain individual formats without mucking with the core library management. * Documentation updates to track the above changes. * Updates to tar.5 to correct a few mistakes and add some additional information about GNU tar and Solaris tar formats. Notes: * The basic 'tar' reader is getting more general; there's not much point in keeping the 'gnutar' reader separate. Merging the two would lose a bunch of duplicate code. * The libc ACL support is looking increasingly inadequate for my needs here. I might need to assemble some fairly significant code for parsing and building ACLs. <sigh>
2004-03-19 22:37:06 +00:00
off_t header_position;
/*
* Detection functions for decompression: bid functions are
* given a block of data from the beginning of the stream and
* can bid on whether or not they support the data stream.
* General guideline: bid the number of bits that you actually
* test, e.g., 16 if you test a 2-byte magic value. The
* highest bidder will have their init function invoked, which
* can set up pointers to specific handlers.
*
* On write, the client just invokes an archive_write_set function
* which sets up the data here directly.
*/
int compression_code; /* Currently active compression. */
const char *compression_name;
struct {
int (*bid)(const void *buff, size_t);
int (*init)(struct archive *, const void *buff, size_t);
} decompressors[4];
/* Read/write data stream (with compression). */
void *compression_data; /* Data for (de)compressor. */
int (*compression_init)(struct archive *); /* Initialize. */
int (*compression_finish)(struct archive *);
ssize_t (*compression_write)(struct archive *, const void *, size_t);
/*
* Read uses a peek/consume I/O model: the decompression code
* returns a pointer to the requested block and advances the
* file position only when requested by a consume call. This
* reduces copying and also simplifies look-ahead for format
* detection.
*/
ssize_t (*compression_read_ahead)(struct archive *,
const void **, size_t request);
ssize_t (*compression_read_consume)(struct archive *, size_t);
/*
* Format detection is mostly the same as compression
* detection, with two significant differences: The bidders
* use the read_ahead calls above to examine the stream rather
* than having the supervisor hand them a block of data to
* examine, and the auction is repeated for every header.
* Winning bidders should set the archive_format and
* archive_format_name appropriately. Bid routines should
* check archive_format and decline to bid if the format of
* the last header was incompatible.
*
* Again, write support is considerably simpler because there's
* no need for an auction.
*/
int archive_format;
const char *archive_format_name;
struct archive_format_descriptor {
int (*bid)(struct archive *);
int (*read_header)(struct archive *, struct archive_entry *);
int (*cleanup)(struct archive *);
void *format_data; /* Format-specific data for readers. */
} formats[4];
struct archive_format_descriptor *format; /* Active format. */
/*
* Storage for format-specific data. Note that there can be
* multiple format readers active at one time, so we need to
* allow for multiple format readers to have their data
* available. The pformat_data slot here is the solution: on
Many fixes: * Disabled shared-library building, as some API breakage is still likely. (I didn't realize it was turned on by default.) If you have an existing /usr/lib/libarchive.so.2, I recommend deleting it. * Pax interchange format now correctly stores and reads UTF8 for extended attributes. In particular, pax format can portably handle arbitrarily long pathnames containing arbitrary characters. * Library compiles cleanly at -O2, -O3, and WARNS=6 on all FreeBSD-CURRENT platforms. * Minor portability improvements inspired by Juergen Lock and Greg Lewis. (Less reliance on stdint.h, isolating of various portability-challenged constructs.) * archive_entry transparently converts multi-byte <-> wide character strings, allowing clients and format handlers to deal with either one, as appropriate. * Support for reading 'L' and 'K' entries in standard tar archives for star compatibility. * Recognize (but don't yet handle) ACL entries from Solaris tar. * Pushed format-specific data for format readers down into format-specific storage and out of library-global storage. This should make it easier to maintain individual formats without mucking with the core library management. * Documentation updates to track the above changes. * Updates to tar.5 to correct a few mistakes and add some additional information about GNU tar and Solaris tar formats. Notes: * The basic 'tar' reader is getting more general; there's not much point in keeping the 'gnutar' reader separate. Merging the two would lose a bunch of duplicate code. * The libc ACL support is looking increasingly inadequate for my needs here. I might need to assemble some fairly significant code for parsing and building ACLs. <sigh>
2004-03-19 22:37:06 +00:00
* read, it is gauranteed to always point to a void* variable
* that the format can use.
*/
void **pformat_data; /* Pointer to current format_data. */
void *format_data; /* Used by writers. */
/*
* Pointers to format-specific functions. On read, these are
* initialized in the bid process. On write, they're initialized by
* archive_write_set_format_XXX() calls.
*/
int (*format_init)(struct archive *); /* Only used on write. */
int (*format_finish)(struct archive *);
int (*format_finish_entry)(struct archive *);
int (*format_write_header)(struct archive *,
struct archive_entry *);
int (*format_write_data)(struct archive *,
const void *buff, size_t);
/*
* Various information needed by archive_extract.
*/
struct archive_string extract_mkdirpath;
struct archive_extract_dir_entry *archive_extract_dir_list;
void (*cleanup_archive_extract)(struct archive *);
int archive_error_number;
const char *error;
struct archive_string error_string;
};
/* Utility function to format a USTAR header into a buffer. */
int
__archive_write_format_header_ustar(struct archive *, char buff[512],
struct archive_entry *);
#define ARCHIVE_STATE_ANY 0xFFFFU
#define ARCHIVE_STATE_NEW 1U
#define ARCHIVE_STATE_HEADER 2U
#define ARCHIVE_STATE_DATA 4U
#define ARCHIVE_STATE_EOF 8U
#define ARCHIVE_STATE_CLOSED 0x10U
#define ARCHIVE_STATE_FATAL 0x8000U
/* Check magic value and state; exit if it isn't valid. */
void
__archive_check_magic(struct archive *, unsigned magic,
unsigned state, const char *func);
#define archive_check_magic(a,m,s) \
__archive_check_magic((a), (m), (s), __func__)
int __archive_read_register_format(struct archive *a,
void *format_data,
int (*bid)(struct archive *),
int (*read_header)(struct archive *, struct archive_entry *),
int (*cleanup)(struct archive *));
int __archive_read_register_compression(struct archive *a,
int (*bid)(const void *, size_t),
int (*init)(struct archive *, const void *, size_t));
Many fixes: * Disabled shared-library building, as some API breakage is still likely. (I didn't realize it was turned on by default.) If you have an existing /usr/lib/libarchive.so.2, I recommend deleting it. * Pax interchange format now correctly stores and reads UTF8 for extended attributes. In particular, pax format can portably handle arbitrarily long pathnames containing arbitrary characters. * Library compiles cleanly at -O2, -O3, and WARNS=6 on all FreeBSD-CURRENT platforms. * Minor portability improvements inspired by Juergen Lock and Greg Lewis. (Less reliance on stdint.h, isolating of various portability-challenged constructs.) * archive_entry transparently converts multi-byte <-> wide character strings, allowing clients and format handlers to deal with either one, as appropriate. * Support for reading 'L' and 'K' entries in standard tar archives for star compatibility. * Recognize (but don't yet handle) ACL entries from Solaris tar. * Pushed format-specific data for format readers down into format-specific storage and out of library-global storage. This should make it easier to maintain individual formats without mucking with the core library management. * Documentation updates to track the above changes. * Updates to tar.5 to correct a few mistakes and add some additional information about GNU tar and Solaris tar formats. Notes: * The basic 'tar' reader is getting more general; there's not much point in keeping the 'gnutar' reader separate. Merging the two would lose a bunch of duplicate code. * The libc ACL support is looking increasingly inadequate for my needs here. I might need to assemble some fairly significant code for parsing and building ACLs. <sigh>
2004-03-19 22:37:06 +00:00
#define err_combine(a,b) ((a) < (b) ? (a) : (b))
#endif