X-Git-Url: https://wimlib.net/git/?p=wimlib;a=blobdiff_plain;f=src%2Fwim.c;h=11e2f9048aeb345995b8ddea83a3250c138129be;hp=20afd389aa535f8bf604ad32bb4df9ea3b00d885;hb=d92446abf69f8a9eff9aed1123f37122647cc7e0;hpb=465a630d28e93b09e55ca07b1a6cae8def3b42f5

diff --git a/src/wim.c b/src/wim.c
index 20afd389..11e2f904 100644
--- a/src/wim.c
+++ b/src/wim.c
@@ -38,6 +38,8 @@
 #include "wimlib/security.h"
 #include "wimlib/wim.h"
 #include "wimlib/xml.h"
+#include "wimlib/compressor_ops.h"
+#include "wimlib/decompressor_ops.h"
 
 #ifdef __WIN32__
 #  include "wimlib/win32.h" /* for realpath() replacement */
@@ -62,19 +64,116 @@ image_print_metadata(WIMStruct *wim)
 				  wim->lookup_table);
 }
 
+static int
+wim_default_pack_compression_type(void)
+{
+	return WIMLIB_COMPRESSION_TYPE_LZMS;
+}
+
+static u32
+wim_default_pack_chunk_size(int ctype) {
+	switch (ctype) {
+	case WIMLIB_COMPRESSION_TYPE_LZMS:
+		/* Note: WIMGAPI uses 1 << 26, but lower sizes are compatible.
+		 * */
+		return 1U << 25; /* 33554432  */
+	default:
+		return 1U << 15; /* 32768     */
+	}
+}
 
 static WIMStruct *
 new_wim_struct(void)
 {
 	WIMStruct *wim = CALLOC(1, sizeof(WIMStruct));
 	if (wim) {
-		wim->in_fd.fd = -1;
-		wim->out_fd.fd = -1;
+		filedes_invalidate(&wim->in_fd);
+		filedes_invalidate(&wim->out_fd);
+		wim->out_pack_compression_type = wim_default_pack_compression_type();
+		wim->out_pack_chunk_size = wim_default_pack_chunk_size(
+						wim->out_pack_compression_type);
+		INIT_LIST_HEAD(&wim->subwims);
 	}
-	INIT_LIST_HEAD(&wim->subwims);
 	return wim;
 }
 
+/* Determine if the chunk size is valid for the specified compression type.  */
+static bool
+wim_chunk_size_valid(u32 chunk_size, int ctype)
+{
+	u32 order;
+
+	/* Chunk size is meaningless for uncompressed WIMs --- any value is
+	 * okay.  */
+	if (ctype == WIMLIB_COMPRESSION_TYPE_NONE)
+		return true;
+
+	/* Chunk size must be power of 2.  */
+	if (chunk_size == 0)
+		return false;
+	order = bsr32(chunk_size);
+	if (chunk_size != 1U << order)
+		return false;
+
+	/* Order	Size
+	 * =====	====
+	 * 15		32768
+	 * 16		65536
+	 * 17		131072
+	 * 18		262144
+	 * 19		524288
+	 * 20		1048576
+	 * 21		2097152
+	 * 22		4194304
+	 * 23		8388608
+	 * 24		16777216
+	 * 25		33554432
+	 * 26		67108864
+	 */
+	switch (ctype) {
+	case WIMLIB_COMPRESSION_TYPE_LZX:
+		/* For LZX compression, the chunk size corresponds to the LZX
+		 * window size, which according the LZX specification can be any
+		 * power of 2 between 2^15 and 2^21, inclusively.  All these are
+		 * supported by wimlib; however, unfortunately only 2^15 is
+		 * supported by WIMGAPI[1] so this value is used by default.
+		 *
+		 * [1] WIMGAPI (Windows 7) attempts to decompress LZX chunk
+		 * sizes > 2^15 but seems to have bug(s) that cause it to fail
+		 * or crash.  (I tried several tweaks to the LZX data but none
+		 * resulted in successful decompression.)  WIMGAPI (Windows 8)
+		 * appears to refuse to open WIMs with chunk size > 2^15
+		 * entirely.  */
+		return order >= 15 && order <= 21;
+
+	case WIMLIB_COMPRESSION_TYPE_XPRESS:
+		/* WIMGAPI (Windows 7, Windows 8) doesn't seem to support XPRESS
+		 * chunk size below 32768 bytes, but larger power-of-two sizes,
+		 * up ta 67108864 bytes, appear to work.  (Note, however, that
+		 * the offsets of XPRESS matches are still limited to 65535
+		 * bytes even when a much larger chunk size is used!)  */
+		return order >= 15 && order <= 26;
+
+	case WIMLIB_COMPRESSION_TYPE_LZMS:
+		return order >= 15 && order <= 26;
+	}
+	return false;
+}
+
+/* Return the default chunk size to use for the specified compression type.
+ *
+ * See notes above in wim_chunk_size_valid().  */
+static u32
+wim_default_chunk_size(int ctype)
+{
+	switch (ctype) {
+	case WIMLIB_COMPRESSION_TYPE_LZMS:
+		return 1U << 17; /* 131072  */
+	default:
+		return 1U << 15; /* 32768   */
+	}
+}
+
 /*
  * Calls a function on images in the WIM.  If @image is WIMLIB_ALL_IMAGES, @visitor
  * is called on the WIM once for each image, with each image selected as the
@@ -124,21 +223,24 @@ wimlib_create_new_wim(int ctype, WIMStruct **wim_ret)
 
 	/* Allocate the WIMStruct. */
 	wim = new_wim_struct();
-	if (!wim)
+	if (wim == NULL)
 		return WIMLIB_ERR_NOMEM;
 
-	ret = init_wim_header(&wim->hdr, ctype);
-	if (ret != 0)
+	ret = init_wim_header(&wim->hdr, ctype, wim_default_chunk_size(ctype));
+	if (ret)
 		goto out_free;
 
 	table = new_lookup_table(9001);
-	if (!table) {
+	if (table == NULL) {
 		ret = WIMLIB_ERR_NOMEM;
 		goto out_free;
 	}
 	wim->lookup_table = table;
 	wim->refcnts_ok = 1;
 	wim->compression_type = ctype;
+	wim->out_compression_type = ctype;
+	wim->chunk_size = wim->hdr.chunk_size;
+	wim->out_chunk_size = wim->hdr.chunk_size;
 	*wim_ret = wim;
 	return 0;
 out_free:
@@ -214,6 +316,8 @@ wimlib_get_compression_type_string(int ctype)
 			return T("LZX");
 		case WIMLIB_COMPRESSION_TYPE_XPRESS:
 			return T("XPRESS");
+		case WIMLIB_COMPRESSION_TYPE_LZMS:
+			return T("LZMS");
 		default:
 			return T("Invalid");
 	}
@@ -292,8 +396,8 @@ wimlib_get_wim_info(WIMStruct *wim, struct wimlib_wim_info *info)
 	memcpy(info->guid, wim->hdr.guid, WIMLIB_GUID_LEN);
 	info->image_count = wim->hdr.image_count;
 	info->boot_index = wim->hdr.boot_idx;
-	info->wim_version = WIM_VERSION;
-	info->chunk_size = WIM_CHUNK_SIZE;
+	info->wim_version = wim->hdr.wim_version;
+	info->chunk_size = wim->chunk_size;
 	info->part_number = wim->hdr.part_number;
 	info->total_parts = wim->hdr.total_parts;
 	info->compression_type = wim->compression_type;
@@ -356,6 +460,95 @@ wimlib_set_wim_info(WIMStruct *wim, const struct wimlib_wim_info *info, int whic
 	return 0;
 }
 
+static int
+set_out_ctype(int ctype, u8 *out_ctype_p)
+{
+	switch (ctype) {
+	case WIMLIB_COMPRESSION_TYPE_INVALID:
+		break;
+	case WIMLIB_COMPRESSION_TYPE_NONE:
+	case WIMLIB_COMPRESSION_TYPE_LZX:
+	case WIMLIB_COMPRESSION_TYPE_XPRESS:
+	case WIMLIB_COMPRESSION_TYPE_LZMS:
+		*out_ctype_p = ctype;
+		return 0;
+	}
+	return WIMLIB_ERR_INVALID_COMPRESSION_TYPE;
+}
+
+/* API function documented in wimlib.h  */
+WIMLIBAPI int
+wimlib_set_output_compression_type(WIMStruct *wim, int ctype)
+{
+	int ret = set_out_ctype(ctype, &wim->out_compression_type);
+	if (ret)
+		return ret;
+
+	/* Reset the chunk size if it's no longer valid.  */
+	if (!wim_chunk_size_valid(ctype, wim->out_chunk_size))
+		wim->out_chunk_size = wim_default_chunk_size(ctype);
+	return 0;
+}
+
+/* API function documented in wimlib.h  */
+WIMLIBAPI int
+wimlib_set_output_pack_compression_type(WIMStruct *wim, int ctype)
+{
+	int ret = set_out_ctype(ctype, &wim->out_pack_compression_type);
+	if (ret)
+		return ret;
+
+	/* Reset the chunk size if it's no longer valid.  */
+	if (!wim_chunk_size_valid(ctype, wim->out_pack_chunk_size))
+		wim->out_pack_chunk_size = wim_default_pack_chunk_size(ctype);
+	return 0;
+}
+
+static int
+set_out_chunk_size(u32 chunk_size, int ctype, u32 *out_chunk_size_p)
+{
+	if (!wim_chunk_size_valid(chunk_size, ctype)) {
+		ERROR("Invalid chunk size (%"PRIu32" bytes) "
+		      "for compression type %"TS"!",
+		      chunk_size,
+		      wimlib_get_compression_type_string(ctype));
+		return WIMLIB_ERR_INVALID_CHUNK_SIZE;
+	}
+
+	*out_chunk_size_p = chunk_size;
+	return 0;
+}
+
+/* API function documented in wimlib.h  */
+WIMLIBAPI int
+wimlib_set_output_chunk_size(WIMStruct *wim, uint32_t chunk_size)
+{
+	if (chunk_size == 0) {
+		wim->out_chunk_size =
+			wim_default_chunk_size(wim->out_compression_type);
+		return 0;
+	}
+
+	return set_out_chunk_size(chunk_size,
+				  wim->out_compression_type,
+				  &wim->out_chunk_size);
+}
+
+/* API function documented in wimlib.h  */
+WIMLIBAPI int
+wimlib_set_output_pack_chunk_size(WIMStruct *wim, uint32_t chunk_size)
+{
+	if (chunk_size == 0) {
+		wim->out_pack_chunk_size =
+			wim_default_pack_chunk_size(wim->out_pack_compression_type);
+		return 0;
+	}
+
+	return set_out_chunk_size(chunk_size,
+				  wim->out_pack_compression_type,
+				  &wim->out_pack_chunk_size);
+}
+
 static int
 do_open_wim(const tchar *filename, struct filedes *fd_ret)
 {
@@ -405,7 +598,6 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 		wim->in_fd.is_pipe = 1;
 	} else {
 		wimfile = wim_filename_or_fd;
-		DEBUG("Reading the WIM file `%"TS"'", wimfile);
 		ret = do_open_wim(wimfile, &wim->in_fd);
 		if (ret)
 			return ret;
@@ -420,10 +612,10 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 		 * intervening time...
 		 *
 		 * Warning: in Windows native builds, realpath() calls the
-		 * replacement function in win32.c.
+		 * replacement function in win32_replacements.c.
 		 */
 		wim->filename = realpath(wimfile, NULL);
-		if (!wim->filename) {
+		if (wim->filename == NULL) {
 			ERROR_WITH_ERRNO("Failed to resolve WIM filename");
 			if (errno == ENOMEM)
 				return WIMLIB_ERR_NOMEM;
@@ -432,14 +624,15 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 		}
 	}
 
-	ret = read_wim_header(wim->filename, &wim->in_fd, &wim->hdr);
+	ret = read_wim_header(wim, &wim->hdr);
 	if (ret)
 		return ret;
 
 	if (wim->hdr.flags & WIM_HDR_FLAG_WRITE_IN_PROGRESS) {
-		WARNING("The WIM_HDR_FLAG_WRITE_IN_PROGRESS is set in the header of \"%"TS"\".\n"
-			"          It may be being changed by another process, or a process\n"
-			"          may have crashed while writing the WIM.", wimfile);
+		WARNING("The WIM_HDR_FLAG_WRITE_IN_PROGRESS flag is set in the header of\n"
+			"          \"%"TS"\".  It may be being changed by another process,\n"
+			"          or a process may have crashed while writing the WIM.",
+			wimfile);
 	}
 
 	if (open_flags & WIMLIB_OPEN_FLAG_WRITE_ACCESS) {
@@ -448,11 +641,9 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 			return ret;
 	}
 
-	if (wim->hdr.total_parts != 1 && !(open_flags & WIMLIB_OPEN_FLAG_SPLIT_OK)) {
-		ERROR("\"%"TS"\": This WIM is part %u of a %u-part WIM",
-		      wimfile, wim->hdr.part_number, wim->hdr.total_parts);
-		return WIMLIB_ERR_SPLIT_UNSUPPORTED;
-	}
+	if ((open_flags & WIMLIB_OPEN_FLAG_ERROR_IF_SPLIT) &&
+	    (wim->hdr.total_parts != 1))
+		return WIMLIB_ERR_IS_SPLIT_WIM;
 
 	DEBUG("According to header, WIM contains %u images", wim->hdr.image_count);
 
@@ -467,23 +658,32 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 	/* Check and cache the compression type */
 	if (wim->hdr.flags & WIM_HDR_FLAG_COMPRESSION) {
 		if (wim->hdr.flags & WIM_HDR_FLAG_COMPRESS_LZX) {
-			if (wim->hdr.flags & WIM_HDR_FLAG_COMPRESS_XPRESS) {
-				ERROR("Multiple compression flags are set in \"%"TS"\"",
-				      wimfile);
-				return WIMLIB_ERR_INVALID_COMPRESSION_TYPE;
-			}
 			wim->compression_type = WIMLIB_COMPRESSION_TYPE_LZX;
 		} else if (wim->hdr.flags & WIM_HDR_FLAG_COMPRESS_XPRESS) {
 			wim->compression_type = WIMLIB_COMPRESSION_TYPE_XPRESS;
+		} else if (wim->hdr.flags & WIM_HDR_FLAG_COMPRESS_LZMS) {
+			wim->compression_type = WIMLIB_COMPRESSION_TYPE_LZMS;
 		} else {
 			ERROR("The compression flag is set on \"%"TS"\", but "
-			      "neither the XPRESS nor LZX flag is set",
+			      "a flag for a recognized format is not",
 			      wimfile);
 			return WIMLIB_ERR_INVALID_COMPRESSION_TYPE;
 		}
 	} else {
 		wim->compression_type = WIMLIB_COMPRESSION_TYPE_NONE;
 	}
+	wim->out_compression_type = wim->compression_type;
+
+	/* Check and cache the chunk size.  */
+	wim->chunk_size = wim->hdr.chunk_size;
+	wim->out_chunk_size = wim->chunk_size;
+	if (!wim_chunk_size_valid(wim->chunk_size, wim->compression_type)) {
+		ERROR("Invalid chunk size (%"PRIu32" bytes) "
+		      "for compression type %"TS"!",
+		      wim->chunk_size,
+		      wimlib_get_compression_type_string(wim->compression_type));
+		return WIMLIB_ERR_INVALID_CHUNK_SIZE;
+	}
 
 	if (open_flags & WIMLIB_OPEN_FLAG_CHECK_INTEGRITY) {
 		ret = check_wim_integrity(wim, progress_func);
@@ -500,18 +700,15 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 
 	if (wim->hdr.image_count != 0 && wim->hdr.part_number == 1) {
 		wim->image_metadata = new_image_metadata_array(wim->hdr.image_count);
-		if (!wim->image_metadata)
+		if (wim->image_metadata == NULL)
 			return WIMLIB_ERR_NOMEM;
 	}
 
 	if (open_flags & WIMLIB_OPEN_FLAG_FROM_PIPE) {
 		wim->lookup_table = new_lookup_table(9001);
-		if (!wim->lookup_table)
+		if (wim->lookup_table == NULL)
 			return WIMLIB_ERR_NOMEM;
 	} else {
-		ret = read_wim_lookup_table(wim);
-		if (ret)
-			return ret;
 
 		ret = read_wim_xml_data(wim);
 		if (ret)
@@ -525,6 +722,11 @@ begin_read(WIMStruct *wim, const void *wim_filename_or_fd,
 			      "<IMAGE> element per image.", wim->hdr.image_count);
 			return WIMLIB_ERR_IMAGE_COUNT;
 		}
+
+		ret = read_wim_lookup_table(wim);
+		if (ret)
+			return ret;
+
 		DEBUG("Done beginning read of WIM file `%"TS"'.", wimfile);
 	}
 	return 0;
@@ -537,28 +739,29 @@ open_wim_as_WIMStruct(const void *wim_filename_or_fd, int open_flags,
 	WIMStruct *wim;
 	int ret;
 
+	if (open_flags & WIMLIB_OPEN_FLAG_FROM_PIPE)
+		DEBUG("Opening pipable WIM from file descriptor %d.", *(const int*)wim_filename_or_fd);
+	else
+		DEBUG("Opening WIM file \"%"TS"\"", (const tchar*)wim_filename_or_fd);
+
 	wimlib_global_init(WIMLIB_INIT_FLAG_ASSUME_UTF8);
 
-	ret = WIMLIB_ERR_INVALID_PARAM;
-	if (!wim_ret)
-		goto out;
+	if (wim_ret == NULL)
+		return WIMLIB_ERR_INVALID_PARAM;
 
-	ret = WIMLIB_ERR_NOMEM;
 	wim = new_wim_struct();
-	if (!wim)
-		goto out;
+	if (wim == NULL)
+		return WIMLIB_ERR_NOMEM;
 
 	ret = begin_read(wim, wim_filename_or_fd, open_flags, progress_func);
-	if (ret)
-		goto out_wimlib_free;
+	if (ret) {
+		wimlib_free(wim);
+		return ret;
+	}
 
-	ret = 0;
+	DEBUG("Successfully opened WIM and created WIMStruct.");
 	*wim_ret = wim;
-	goto out;
-out_wimlib_free:
-	wimlib_free(wim);
-out:
-	return ret;
+	return 0;
 }
 
 /* API function documented in wimlib.h  */
@@ -585,7 +788,7 @@ destroy_image_metadata(struct wim_image_metadata *imd,
 		free_lookup_table_entry(imd->metadata_lte);
 		imd->metadata_lte = NULL;
 	}
-	if (!table) {
+	if (table == NULL) {
 		struct wim_lookup_table_entry *lte, *tmp;
 		list_for_each_entry_safe(lte, tmp, &imd->unhashed_streams, unhashed_list)
 			free_lookup_table_entry(lte);
@@ -622,7 +825,7 @@ append_image_metadata(WIMStruct *wim, struct wim_image_metadata *imd)
 	imd_array = REALLOC(wim->image_metadata,
 			    sizeof(wim->image_metadata[0]) * (wim->hdr.image_count + 1));
 
-	if (!imd_array)
+	if (imd_array == NULL)
 		return WIMLIB_ERR_NOMEM;
 	wim->image_metadata = imd_array;
 	imd_array[wim->hdr.image_count++] = imd;
@@ -657,14 +860,14 @@ new_image_metadata_array(unsigned num_images)
 
 	imd_array = CALLOC(num_images, sizeof(imd_array[0]));
 
-	if (!imd_array) {
+	if (imd_array == NULL) {
 		ERROR("Failed to allocate memory for %u image metadata structures",
 		      num_images);
 		return NULL;
 	}
 	for (unsigned i = 0; i < num_images; i++) {
 		imd_array[i] = new_image_metadata();
-		if (!imd_array[i]) {
+		if (imd_array[i] == NULL) {
 			for (unsigned j = 0; j < i; j++)
 				put_image_metadata(imd_array[j], NULL);
 			FREE(imd_array);
@@ -682,13 +885,19 @@ int
 wim_checksum_unhashed_streams(WIMStruct *wim)
 {
 	int ret;
+
+	if (!wim_has_metadata(wim))
+		return 0;
 	for (int i = 0; i < wim->hdr.image_count; i++) {
 		struct wim_lookup_table_entry *lte, *tmp;
 		struct wim_image_metadata *imd = wim->image_metadata[i];
 		image_for_each_unhashed_stream_safe(lte, tmp, imd) {
-			ret = hash_unhashed_stream(lte, wim->lookup_table, NULL);
+			struct wim_lookup_table_entry *new_lte;
+			ret = hash_unhashed_stream(lte, wim->lookup_table, &new_lte);
 			if (ret)
 				return ret;
+			if (new_lte != lte)
+				free_lookup_table_entry(lte);
 		}
 	}
 	return 0;
@@ -714,7 +923,7 @@ can_modify_wim(WIMStruct *wim)
 		}
 	}
 	if (wim->hdr.total_parts != 1) {
-		ERROR("Cannot modify \"%"TS"\": is part of a spanned set",
+		ERROR("Cannot modify \"%"TS"\": is part of a split WIM",
 		      wim->filename);
 		return WIMLIB_ERR_WIM_IS_READONLY;
 	}
@@ -745,8 +954,11 @@ can_delete_from_wim(WIMStruct *wim)
 	ret = can_modify_wim(wim);
 	if (ret)
 		return ret;
-	if (!wim->refcnts_ok)
-		wim_recalculate_refcnts(wim);
+	if (!wim->refcnts_ok) {
+		ret = wim_recalculate_refcnts(wim);
+		if (ret)
+			return ret;
+	}
 	return 0;
 }
 
@@ -754,7 +966,7 @@ can_delete_from_wim(WIMStruct *wim)
 WIMLIBAPI void
 wimlib_free(WIMStruct *wim)
 {
-	if (!wim)
+	if (wim == NULL)
 		return;
 
 	DEBUG("Freeing WIMStruct (filename=\"%"TS"\", image_count=%u)",
@@ -774,9 +986,10 @@ wimlib_free(WIMStruct *wim)
 	if (filedes_valid(&wim->out_fd))
 		filedes_close(&wim->out_fd);
 
-
 	free_lookup_table(wim->lookup_table);
 
+	wimlib_free_decompressor(wim->decompressor);
+
 	FREE(wim->filename);
 	free_wim_info(wim->wim_info);
 	if (wim->image_metadata) {
@@ -807,7 +1020,6 @@ WIMLIBAPI int
 wimlib_global_init(int init_flags)
 {
 	static bool already_inited = false;
-	int ret;
 
 	if (already_inited)
 		return 0;
@@ -820,14 +1032,19 @@ wimlib_global_init(int init_flags)
 	#endif
 	}
 #ifdef __WIN32__
-	ret = win32_global_init(init_flags);
-	if (ret)
-		return ret;
-#else
-	ret = 0;
+	{
+		int ret = win32_global_init(init_flags);
+		if (ret)
+			return ret;
+	}
 #endif
+	init_upcase();
+	if (init_flags & WIMLIB_INIT_FLAG_DEFAULT_CASE_SENSITIVE)
+		default_ignore_case = false;
+	else if (init_flags & WIMLIB_INIT_FLAG_DEFAULT_CASE_INSENSITIVE)
+		default_ignore_case = true;
 	already_inited = true;
-	return ret;
+	return 0;
 }
 
 /* API function documented in wimlib.h  */
@@ -839,4 +1056,6 @@ wimlib_global_cleanup(void)
 #ifdef __WIN32__
 	win32_global_cleanup();
 #endif
+	cleanup_decompressor_params();
+	cleanup_compressor_params();
 }