ciderpress/nufxlib/Deflate.c

296 lines
8.5 KiB
C
Raw Normal View History

Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
/*
* NuFX archive manipulation library
* Copyright (C) 2000-2007 by Andy McFadden, All Rights Reserved.
* This is free software; you can redistribute it and/or modify it under the
* terms of the BSD License, see the file COPYING-LIB.
*
* Support for the "deflate" algorithm, via the "zlib" library.
*
* This compression format is totally unsupported on the Apple II. This
* is provided primarily for the benefit of Apple II emulators that want
* a better storage format for disk images than SHK+LZW or a ZIP file.
*
* This code was developed and tested with ZLIB_VERSION "1.1.3". It is
* expected to work with any version >= 1.1.3 and < 2.x. Please visit
* http://www.zlib.org/ for more information.
*/
#include "NufxLibPriv.h"
#ifdef ENABLE_DEFLATE
#include "zlib.h"
#define kNuDeflateLevel 9 /* use maximum compression */
/*
* Alloc and free functions provided to zlib.
*/
static voidpf Nu_zalloc(voidpf opaque, uInt items, uInt size)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
{
return Nu_Malloc(opaque, items * size);
}
static void Nu_zfree(voidpf opaque, voidpf address)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
{
Nu_Free(opaque, address);
}
/*
* ===========================================================================
* Compression
* ===========================================================================
*/
/*
* Compress "srcLen" bytes from "pStraw" to "fp".
*/
NuError Nu_CompressDeflate(NuArchive* pArchive, NuStraw* pStraw, FILE* fp,
uint32_t srcLen, uint32_t* pDstLen, uint16_t* pCrc)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
{
NuError err = kNuErrNone;
z_stream zstream;
int zerr;
Bytef* outbuf = NULL;
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
Assert(pArchive != NULL);
Assert(pStraw != NULL);
Assert(fp != NULL);
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
Assert(srcLen > 0);
Assert(pDstLen != NULL);
Assert(pCrc != NULL);
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
err = Nu_AllocCompressionBufferIFN(pArchive);
if (err != kNuErrNone)
return err;
/* allocate a similarly-sized buffer for the output */
outbuf = Nu_Malloc(pArchive, kNuGenCompBufSize);
BailAlloc(outbuf);
/*
* Initialize the zlib stream.
*/
zstream.zalloc = Nu_zalloc;
zstream.zfree = Nu_zfree;
zstream.opaque = pArchive;
zstream.next_in = NULL;
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
zstream.avail_in = 0;
zstream.next_out = outbuf;
zstream.avail_out = kNuGenCompBufSize;
zstream.data_type = Z_UNKNOWN;
zerr = deflateInit(&zstream, kNuDeflateLevel);
if (zerr != Z_OK) {
err = kNuErrInternal;
if (zerr == Z_VERSION_ERROR) {
Nu_ReportError(NU_BLOB, err,
"installed zlib is not compatible with linked version (%s)",
ZLIB_VERSION);
} else {
Nu_ReportError(NU_BLOB, err,
"call to deflateInit failed (zerr=%d)", zerr);
}
goto bail;
}
/*
* Loop while we have data.
*/
do {
uint32_t getSize;
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
int flush;
/* should be able to read a full buffer every time */
if (zstream.avail_in == 0 && srcLen) {
getSize = (srcLen > kNuGenCompBufSize) ? kNuGenCompBufSize : srcLen;
DBUG(("+++ reading %ld bytes\n", getSize));
err = Nu_StrawRead(pArchive, pStraw, pArchive->compBuf, getSize);
if (err != kNuErrNone) {
Nu_ReportError(NU_BLOB, err, "deflate read failed");
goto z_bail;
}
srcLen -= getSize;
*pCrc = Nu_CalcCRC16(*pCrc, pArchive->compBuf, getSize);
zstream.next_in = pArchive->compBuf;
zstream.avail_in = getSize;
}
if (srcLen == 0)
flush = Z_FINISH; /* tell zlib that we're done */
else
flush = Z_NO_FLUSH; /* more to come! */
zerr = deflate(&zstream, flush);
if (zerr != Z_OK && zerr != Z_STREAM_END) {
err = kNuErrInternal;
Nu_ReportError(NU_BLOB, err, "zlib deflate call failed (zerr=%d)",
zerr);
goto z_bail;
}
/* write when we're full or when we're done */
if (zstream.avail_out == 0 ||
(zerr == Z_STREAM_END && zstream.avail_out != kNuGenCompBufSize))
{
DBUG(("+++ writing %d bytes\n", zstream.next_out - outbuf));
err = Nu_FWrite(fp, outbuf, zstream.next_out - outbuf);
if (err != kNuErrNone) {
Nu_ReportError(NU_BLOB, err, "fwrite failed in deflate");
goto z_bail;
}
zstream.next_out = outbuf;
zstream.avail_out = kNuGenCompBufSize;
}
} while (zerr == Z_OK);
Assert(zerr == Z_STREAM_END); /* other errors should've been caught */
*pDstLen = zstream.total_out;
z_bail:
deflateEnd(&zstream); /* free up any allocated structures */
bail:
if (outbuf != NULL)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
free(outbuf);
return err;
}
/*
* ===========================================================================
* Expansion
* ===========================================================================
*/
/*
* Expand from "infp" to "pFunnel".
*/
NuError Nu_ExpandDeflate(NuArchive* pArchive, const NuRecord* pRecord,
const NuThread* pThread, FILE* infp, NuFunnel* pFunnel, uint16_t* pCrc)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
{
NuError err = kNuErrNone;
z_stream zstream;
int zerr;
uint32_t compRemaining;
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
Bytef* outbuf;
Assert(pArchive != NULL);
Assert(pThread != NULL);
Assert(infp != NULL);
Assert(pFunnel != NULL);
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
err = Nu_AllocCompressionBufferIFN(pArchive);
if (err != kNuErrNone)
return err;
/* allocate a similarly-sized buffer for the output */
outbuf = Nu_Malloc(pArchive, kNuGenCompBufSize);
BailAlloc(outbuf);
compRemaining = pThread->thCompThreadEOF;
/*
* Initialize the zlib stream.
*/
zstream.zalloc = Nu_zalloc;
zstream.zfree = Nu_zfree;
zstream.opaque = pArchive;
zstream.next_in = NULL;
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
zstream.avail_in = 0;
zstream.next_out = outbuf;
zstream.avail_out = kNuGenCompBufSize;
zstream.data_type = Z_UNKNOWN;
zerr = inflateInit(&zstream);
if (zerr != Z_OK) {
err = kNuErrInternal;
if (zerr == Z_VERSION_ERROR) {
Nu_ReportError(NU_BLOB, err,
"installed zlib is not compatible with linked version (%s)",
ZLIB_VERSION);
} else {
Nu_ReportError(NU_BLOB, err,
"call to inflateInit failed (zerr=%d)", zerr);
}
goto bail;
}
/*
* Loop while we have data.
*/
do {
uint32_t getSize;
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
/* read as much as we can */
if (zstream.avail_in == 0) {
getSize = (compRemaining > kNuGenCompBufSize) ?
kNuGenCompBufSize : compRemaining;
DBUG(("+++ reading %ld bytes (%ld left)\n", getSize,
compRemaining));
err = Nu_FRead(infp, pArchive->compBuf, getSize);
if (err != kNuErrNone) {
Nu_ReportError(NU_BLOB, err, "inflate read failed");
goto z_bail;
}
compRemaining -= getSize;
zstream.next_in = pArchive->compBuf;
zstream.avail_in = getSize;
}
/* uncompress the data */
zerr = inflate(&zstream, Z_NO_FLUSH);
if (zerr != Z_OK && zerr != Z_STREAM_END) {
err = kNuErrInternal;
Nu_ReportError(NU_BLOB, err, "zlib inflate call failed (zerr=%d)",
zerr);
goto z_bail;
}
/* write every time there's anything (buffer will usually be full) */
if (zstream.avail_out != kNuGenCompBufSize) {
DBUG(("+++ writing %d bytes\n", zstream.next_out - outbuf));
err = Nu_FunnelWrite(pArchive, pFunnel, outbuf,
zstream.next_out - outbuf);
if (err != kNuErrNone) {
Nu_ReportError(NU_BLOB, err, "write failed in inflate");
goto z_bail;
}
if (pCrc != NULL)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
*pCrc = Nu_CalcCRC16(*pCrc, outbuf, zstream.next_out - outbuf);
zstream.next_out = outbuf;
zstream.avail_out = kNuGenCompBufSize;
}
} while (zerr == Z_OK);
Assert(zerr == Z_STREAM_END); /* other errors should've been caught */
if (zstream.total_out != pThread->actualThreadEOF) {
err = kNuErrBadData;
Nu_ReportError(NU_BLOB, err,
"size mismatch on inflated file (%ld vs %u)",
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
zstream.total_out, pThread->actualThreadEOF);
goto z_bail;
}
z_bail:
inflateEnd(&zstream); /* free up any allocated structures */
bail:
if (outbuf != NULL)
Large set of changes to restore CiderPress build. CiderPress and MDC now compile, and execute far enough to open their respective "about" boxes, but I doubt they'll do much more than that. * Switch from MBCS to UNICODE APIs Microsoft switched to UTF-16 (by way of UCS-2) a long time ago, and the support for MBCS seems to be getting phased out. So it's time to switch to wide strings. This is a bit awkward for CiderPress because it works with disk and file archives with 8-bit filenames, and I want NufxLib and DiskImgLib to continue to work on Linux (which has largely taken the UTF-8 approach to Unicode). The libraries will continue to work with 8-bit filenames, with CiderPress/MDC doing the conversion at the appropriate point. There were a couple of places where strings from a structure handed back by one of the libraries were used directly in the UI, or vice-versa, which is a problem because we have nowhere to store the result of the conversion. These currently have fixed place-holder "xyzzy" strings. All UI strings are now wide. Various format strings now use "%ls" and "%hs" to explicitly specify wide and narrow. This doesn't play well with gcc, so only the Windows-specific parts use those. * Various updates to vcxproj files The project-file conversion had some cruft that is now largely gone. The build now has a common output directory for the EXEs and libraries, avoiding the old post-build copy steps. * Added zlib 1.2.8 and nufxlib 2.2.2 source snapshots The old "prebuilts" directory is now gone. The libraries are now built as part of building the apps. I added a minimal set of files for zlib, and a full set for nufxlib. The Linux-specific nufxlib goodies are included for the benefit of the Linux utilities, which are currently broken (don't build). * Replace symbols used for include guards Symbols with a leading "__" are reserved.
2014-11-10 23:32:55 +00:00
free(outbuf);
return err;
}
#endif /*ENABLE_DEFLATE*/