ciderpress/app/Actions.cpp

2368 lines
80 KiB
C++
Raw Normal View History

2007-03-27 17:47:10 +00:00
/*
* CiderPress
* Copyright (C) 2007 by faddenSoft, LLC. All Rights Reserved.
* See the file LICENSE for distribution terms.
*/
/*
* File actions. These are actually part of MainWindow, but for readability
* these are split into their own file.
*/
#include "stdafx.h"
#include "Main.h"
#include "ViewFilesDialog.h"
#include "ChooseDirDialog.h"
#include "AddFilesDialog.h"
#include "CreateSubdirDialog.h"
#include "ExtractOptionsDialog.h"
#include "UseSelectionDialog.h"
#include "RecompressOptionsDialog.h"
#include "ConvDiskOptionsDialog.h"
#include "ConvFileOptionsDialog.h"
#include "EditCommentDialog.h"
#include "EditPropsDialog.h"
#include "RenameVolumeDialog.h"
#include "ConfirmOverwriteDialog.h"
#include "ImageFormatDialog.h"
#include "FileNameConv.h"
#include "GenericArchive.h"
#include "NufxArchive.h"
#include "DiskArchive.h"
#include "ChooseAddTargetDialog.h"
#include "CassetteDialog.h"
#include "BasicImport.h"
#include "../diskimg/TwoImg.h"
#include <errno.h>
/*
* ==========================================================================
* View
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsView(void)
2007-03-27 17:47:10 +00:00
{
HandleView();
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsView(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL &&
fpContentList->GetSelectedCount() > 0);
2007-03-27 17:47:10 +00:00
}
void MainWindow::HandleView(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
SelectionSet selSet;
int threadMask = GenericEntry::kAnyThread | GenericEntry::kAllowDamaged |
GenericEntry::kAllowDirectory | GenericEntry::kAllowVolumeDir;
selSet.CreateFromSelection(fpContentList, threadMask);
selSet.Dump();
if (selSet.GetNumEntries() == 0) {
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
MessageBox(L"Nothing viewable found.",
L"No match", MB_OK | MB_ICONEXCLAMATION);
return;
}
ViewFilesDialog vfd(this);
vfd.SetSelectionSet(&selSet);
vfd.SetTextTypeFace(fPreferences.GetPrefString(kPrViewTextTypeFace));
vfd.SetTextPointSize(fPreferences.GetPrefLong(kPrViewTextPointSize));
vfd.SetNoWrapText(fPreferences.GetPrefBool(kPrNoWrapText));
vfd.DoModal();
// remember which font they used (sticky pref, not in registry)
fPreferences.SetPrefString(kPrViewTextTypeFace, vfd.GetTextTypeFace());
fPreferences.SetPrefLong(kPrViewTextPointSize, vfd.GetTextPointSize());
LOGI("Preferences: saving view font %d-point '%ls'",
fPreferences.GetPrefLong(kPrViewTextPointSize),
fPreferences.GetPrefString(kPrViewTextTypeFace));
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Open as disk image
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsOpenAsDisk(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpContentList->GetSelectedCount() == 1);
GenericEntry* pEntry = GetSelectedItem(fpContentList);
if (pEntry->GetHasDiskImage())
TmpExtractAndOpen(pEntry, GenericEntry::kDiskImageThread, kModeDiskImage);
else
TmpExtractAndOpen(pEntry, GenericEntry::kDataThread, kModeDiskImage);
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsOpenAsDisk(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
const int kMinLen = 512 * 7;
bool allow = false;
2014-11-18 05:13:13 +00:00
if (fpContentList != NULL && fpContentList->GetSelectedCount() == 1) {
GenericEntry* pEntry = GetSelectedItem(fpContentList);
2014-11-18 05:13:13 +00:00
if (pEntry != NULL) {
if ((pEntry->GetHasDataFork() || pEntry->GetHasDiskImage()) &&
pEntry->GetUncompressedLen() > kMinLen)
{
allow = true;
}
}
}
pCmdUI->Enable(allow);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Add Files
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsAddFiles(void)
2007-03-27 17:47:10 +00:00
{
LOGI("Add files!");
AddFilesDialog addFiles(this);
2014-11-18 05:13:13 +00:00
DiskImgLib::A2File* pTargetSubdir = NULL;
/*
* Special handling for adding files to disk images.
*/
if (fpOpenArchive->GetArchiveKind() == GenericArchive::kArchiveDiskImage)
{
if (!ChooseAddTarget(&pTargetSubdir, &addFiles.fpTargetDiskFS))
return;
}
addFiles.fStoragePrefix = "";
addFiles.fIncludeSubfolders =
fPreferences.GetPrefBool(kPrAddIncludeSubFolders);
addFiles.fStripFolderNames =
fPreferences.GetPrefBool(kPrAddStripFolderNames);
addFiles.fOverwriteExisting =
fPreferences.GetPrefBool(kPrAddOverwriteExisting);
addFiles.fTypePreservation =
fPreferences.GetPrefLong(kPrAddTypePreservation);
addFiles.fConvEOL =
fPreferences.GetPrefLong(kPrAddConvEOL);
/* if they can't convert EOL when adding files, disable the option */
if (!fpOpenArchive->GetCapability(GenericArchive::kCapCanConvEOLOnAdd)) {
addFiles.fConvEOL = AddFilesDialog::kConvEOLNone;
addFiles.fConvEOLEnable = false;
}
/*
* Disable editing of the storage prefix field. Force pathname
* stripping to be on for non-hierarchical filesystems (i.e. everything
* but ProDOS and HFS).
*/
2014-11-18 05:13:13 +00:00
if (addFiles.fpTargetDiskFS != NULL) {
DiskImg::FSFormat format;
format = addFiles.fpTargetDiskFS->GetDiskImg()->GetFSFormat();
2014-11-18 05:13:13 +00:00
if (pTargetSubdir != NULL) {
ASSERT(!pTargetSubdir->IsVolumeDirectory());
addFiles.fStoragePrefix = pTargetSubdir->GetPathName();
}
addFiles.fStripFolderNamesEnable = false;
addFiles.fStoragePrefixEnable = false;
switch (format) {
case DiskImg::kFormatProDOS:
case DiskImg::kFormatMacHFS:
addFiles.fStripFolderNamesEnable = true;
break;
default:
break;
}
}
if (!addFiles.fStripFolderNamesEnable) {
/* if we disabled it, we did so because it's mandatory */
addFiles.fStripFolderNames = true;
}
addFiles.m_ofn.lpstrInitialDir = fPreferences.GetPrefString(kPrAddFileFolder);
LRESULT addResult = addFiles.DoModal();
if (addResult == IDOK) {
fPreferences.SetPrefBool(kPrAddIncludeSubFolders,
addFiles.fIncludeSubfolders != 0);
if (addFiles.fStripFolderNamesEnable) {
// only update the pref if they had the ability to change it
fPreferences.SetPrefBool(kPrAddStripFolderNames,
addFiles.fStripFolderNames != 0);
}
fPreferences.SetPrefBool(kPrAddOverwriteExisting,
addFiles.fOverwriteExisting != 0);
fPreferences.SetPrefLong(kPrAddTypePreservation,
addFiles.fTypePreservation);
if (addFiles.fConvEOLEnable)
fPreferences.SetPrefLong(kPrAddConvEOL,
addFiles.fConvEOL);
CString saveFolder = addFiles.GetDirectory();
fPreferences.SetPrefString(kPrAddFileFolder, saveFolder);
/*
* Set up a progress dialog and kick things off.
*/
bool result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionAdd, this);
//fpContentList->Invalidate(); // don't allow redraws until done
result = fpOpenArchive->BulkAdd(fpActionProgress, &addFiles);
fpContentList->Reload();
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
if (result)
SuccessBeep();
} else {
LOGI("SFD bailed with Cancel");
}
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsAddFiles(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly());
2007-03-27 17:47:10 +00:00
}
bool MainWindow::ChooseAddTarget(DiskImgLib::A2File** ppTargetSubdir,
DiskImgLib::DiskFS** ppTargetDiskFS)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(ppTargetSubdir != NULL);
ASSERT(ppTargetDiskFS != NULL);
2014-11-18 05:13:13 +00:00
*ppTargetSubdir = NULL;
*ppTargetDiskFS = NULL;
GenericEntry* pEntry = GetSelectedItem(fpContentList);
2014-11-18 05:13:13 +00:00
if (pEntry != NULL &&
(pEntry->GetRecordKind() == GenericEntry::kRecordKindDirectory ||
pEntry->GetRecordKind() == GenericEntry::kRecordKindVolumeDir))
{
/*
* They've selected a single subdirectory. Add the files there.
*/
DiskEntry* pDiskEntry = (DiskEntry*) pEntry;
*ppTargetSubdir = pDiskEntry->GetA2File();
*ppTargetDiskFS = (*ppTargetSubdir)->GetDiskFS();
} else {
/*
* Nothing selected, non-subdir selected, or multiple files
* selected. Whatever the case, pop up the choose target dialog.
*
* This works on DOS 3.3 and Pascal disks, because the absence
* of subdirectories means there's only one possible place to
* put the files. We could short-circuit this code for anything
* but ProDOS and HFS, but we have to be careful about embedded
* sub-volumes.
*/
DiskArchive* pDiskArchive = (DiskArchive*) fpOpenArchive;
LOGD("Trying ChooseAddTarget");
ChooseAddTargetDialog targetDialog(this);
targetDialog.fpDiskFS = pDiskArchive->GetDiskFS();
if (targetDialog.DoModal() != IDOK)
return false;
*ppTargetSubdir = targetDialog.fpChosenSubdir;
*ppTargetDiskFS = targetDialog.fpChosenDiskFS;
/* make sure the subdir is part of the diskfs */
2014-11-18 05:13:13 +00:00
ASSERT(*ppTargetSubdir == NULL ||
(*ppTargetSubdir)->GetDiskFS() == *ppTargetDiskFS);
}
2014-11-18 05:13:13 +00:00
if (*ppTargetSubdir != NULL && (*ppTargetSubdir)->IsVolumeDirectory())
*ppTargetSubdir = NULL;
return true;
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Add Disks
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsAddDisks(void)
2007-03-27 17:47:10 +00:00
{
DIError dierr;
DiskImg img;
CString failed, errMsg;
CString openFilters, saveFolder;
CStringA pathNameA;
AddFilesDialog addOpts;
LOGI("Add disks!");
CheckedLoadString(&failed, IDS_FAILED);
openFilters = kOpenDiskImage;
openFilters += kOpenAll;
openFilters += kOpenEnd;
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
CFileDialog dlg(TRUE, L"dsk", NULL, OFN_FILEMUSTEXIST, openFilters, this);
dlg.m_ofn.lpstrTitle = L"Add Disk Image";
/* file is always opened read-only */
dlg.m_ofn.Flags |= OFN_HIDEREADONLY;
dlg.m_ofn.lpstrInitialDir = fPreferences.GetPrefString(kPrAddFileFolder);
if (dlg.DoModal() != IDOK)
goto bail;
saveFolder = dlg.m_ofn.lpstrFile;
saveFolder = saveFolder.Left(dlg.m_ofn.nFileOffset);
fPreferences.SetPrefString(kPrAddFileFolder, saveFolder);
/* open the image file and analyze it */
pathNameA = dlg.GetPathName();
dierr = img.OpenImage(pathNameA, PathProposal::kLocalFssep, true);
if (dierr != kDIErrNone) {
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
errMsg.Format(L"Unable to open disk image: %hs.",
DiskImgLib::DIStrError(dierr));
MessageBox(errMsg, failed, MB_OK|MB_ICONSTOP);
goto bail;
}
if (img.AnalyzeImage() != kDIErrNone) {
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
errMsg.Format(L"The file '%ls' doesn't seem to hold a valid disk image.",
(LPCWSTR) dlg.GetPathName());
MessageBox(errMsg, failed, MB_OK|MB_ICONSTOP);
goto bail;
}
/* if requested (or necessary), verify the format */
if (/*img.GetFSFormat() == DiskImg::kFormatUnknown ||*/
img.GetSectorOrder() == DiskImg::kSectorOrderUnknown ||
fPreferences.GetPrefBool(kPrQueryImageFormat))
{
ImageFormatDialog imf;
imf.InitializeValues(&img);
imf.fFileSource = dlg.GetPathName();
imf.SetQueryDisplayFormat(false);
LOGI(" On entry, sectord=%d format=%d",
imf.fSectorOrder, imf.fFSFormat);
if (imf.fFSFormat == DiskImg::kFormatUnknown)
imf.fFSFormat = DiskImg::kFormatGenericProDOSOrd;
if (imf.DoModal() != IDOK) {
LOGI("User bailed on IMF dialog");
goto bail;
}
LOGI(" On exit, sectord=%d format=%d",
imf.fSectorOrder, imf.fFSFormat);
if (imf.fSectorOrder != img.GetSectorOrder() ||
imf.fFSFormat != img.GetFSFormat())
{
LOGI("Initial values overridden, forcing img format");
dierr = img.OverrideFormat(img.GetPhysicalFormat(), imf.fFSFormat,
imf.fSectorOrder);
if (dierr != kDIErrNone) {
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
errMsg.Format(L"Unable to access disk image using selected"
L" parameters. Error: %hs.",
DiskImgLib::DIStrError(dierr));
MessageBox(errMsg, failed, MB_OK | MB_ICONSTOP);
goto bail;
}
}
}
/*
* We want to read from the image the way that a ProDOS application
* would, which means forcing the FSFormat to generic ProDOS ordering.
* This way, ProDOS disks load serially, and DOS 3.3 disks load with
* their sectors swapped around.
*/
dierr = img.OverrideFormat(img.GetPhysicalFormat(),
DiskImg::kFormatGenericProDOSOrd, img.GetSectorOrder());
if (dierr != kDIErrNone) {
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
errMsg.Format(L"Internal error: couldn't switch to generic ProDOS: %hs.",
DiskImgLib::DIStrError(dierr));
MessageBox(errMsg, failed, MB_OK | MB_ICONSTOP);
goto bail;
}
/*
* Set up an AddFilesDialog, but don't actually use it as a dialog.
* Instead, we just configure the various options appropriately.
*/
ASSERT(dlg.m_ofn.nFileOffset > 0);
{
CString directory(dlg.m_ofn.lpstrFile, dlg.m_ofn.nFileOffset - 1);
CString file(dlg.m_ofn.lpstrFile + dlg.m_ofn.nFileOffset);
LOGD("Stuffing '%ls' '%ls'", (LPCWSTR) directory, (LPCWSTR) file);
addOpts.StuffSingleFilename(directory, file);
}
addOpts.fStoragePrefix = "";
addOpts.fIncludeSubfolders = false;
addOpts.fStripFolderNames = false;
addOpts.fOverwriteExisting = false;
addOpts.fTypePreservation = AddFilesDialog::kPreserveTypes;
addOpts.fConvEOL = AddFilesDialog::kConvEOLNone;
addOpts.fConvEOLEnable = false; // no EOL conversion on disk images!
addOpts.fpDiskImg = &img;
bool result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionAddDisk, this);
//fpContentList->Invalidate(); // don't allow updates until done
result = fpOpenArchive->AddDisk(fpActionProgress, &addOpts);
fpContentList->Reload();
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
if (result)
SuccessBeep();
2007-03-27 17:47:10 +00:00
bail:
return;
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsAddDisks(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly() &&
fpOpenArchive->GetCapability(GenericArchive::kCapCanAddDisk));
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Create Subdirectory
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsCreateSubdir(void)
2007-03-27 17:47:10 +00:00
{
CreateSubdirDialog csDialog;
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
ASSERT(!fpOpenArchive->IsReadOnly());
GenericEntry* pEntry = GetSelectedItem(fpContentList);
2014-11-18 05:13:13 +00:00
if (pEntry == NULL) {
// can happen for no selection or multi-selection; should not be here
ASSERT(false);
return;
}
if (pEntry->GetRecordKind() != GenericEntry::kRecordKindDirectory &&
pEntry->GetRecordKind() != GenericEntry::kRecordKindVolumeDir)
{
CString errMsg;
errMsg = "Please select a subdirectory.";
ShowFailureMsg(this, errMsg, IDS_MB_APP_NAME);
return;
}
LOGI("Creating subdir in '%ls'", (LPCWSTR) pEntry->GetPathNameUNI());
csDialog.fBasePath = pEntry->GetPathNameUNI();
csDialog.fpArchive = fpOpenArchive;
csDialog.fpParentEntry = pEntry;
csDialog.fNewName = "New.Subdir";
if (csDialog.DoModal() != IDOK)
return;
LOGI("Creating '%ls'", (LPCWSTR) csDialog.fNewName);
fpOpenArchive->CreateSubdir(this, pEntry, csDialog.fNewName);
fpContentList->Reload();
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsCreateSubdir(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
bool enable = fpContentList != NULL && !fpOpenArchive->IsReadOnly() &&
fpContentList->GetSelectedCount() == 1 &&
fpOpenArchive->GetCapability(GenericArchive::kCapCanCreateSubdir);
if (enable) {
/* second-level check: make sure it's a subdir */
GenericEntry* pEntry = GetSelectedItem(fpContentList);
2014-11-18 05:13:13 +00:00
if (pEntry == NULL) {
ASSERT(false);
return;
}
if (pEntry->GetRecordKind() != GenericEntry::kRecordKindDirectory &&
pEntry->GetRecordKind() != GenericEntry::kRecordKindVolumeDir)
{
enable = false;
}
}
pCmdUI->Enable(enable);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Extract
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsExtract(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
/*
* Ask the user about various options.
*/
ExtractOptionsDialog extOpts(fpContentList->GetSelectedCount(), this);
extOpts.fExtractPath = fPreferences.GetPrefString(kPrExtractFileFolder);
extOpts.fConvEOL = fPreferences.GetPrefLong(kPrExtractConvEOL);
extOpts.fConvHighASCII = fPreferences.GetPrefBool(kPrExtractConvHighASCII);
extOpts.fIncludeDataForks = fPreferences.GetPrefBool(kPrExtractIncludeData);
extOpts.fIncludeRsrcForks = fPreferences.GetPrefBool(kPrExtractIncludeRsrc);
extOpts.fIncludeDiskImages = fPreferences.GetPrefBool(kPrExtractIncludeDisk);
extOpts.fEnableReformat = fPreferences.GetPrefBool(kPrExtractEnableReformat);
extOpts.fDiskTo2MG = fPreferences.GetPrefBool(kPrExtractDiskTo2MG);
extOpts.fAddTypePreservation = fPreferences.GetPrefBool(kPrExtractAddTypePreservation);
extOpts.fAddExtension = fPreferences.GetPrefBool(kPrExtractAddExtension);
extOpts.fStripFolderNames = fPreferences.GetPrefBool(kPrExtractStripFolderNames);
extOpts.fOverwriteExisting = fPreferences.GetPrefBool(kPrExtractOverwriteExisting);
if (fpContentList->GetSelectedCount() > 0)
extOpts.fFilesToExtract = ExtractOptionsDialog::kExtractSelection;
else
extOpts.fFilesToExtract = ExtractOptionsDialog::kExtractAll;
if (extOpts.DoModal() != IDOK)
return;
if (extOpts.fExtractPath.Right(1) != "\\")
extOpts.fExtractPath += "\\";
/* push preferences back out */
fPreferences.SetPrefString(kPrExtractFileFolder, extOpts.fExtractPath);
fPreferences.SetPrefLong(kPrExtractConvEOL, extOpts.fConvEOL);
fPreferences.SetPrefBool(kPrExtractConvHighASCII, extOpts.fConvHighASCII != 0);
fPreferences.SetPrefBool(kPrExtractIncludeData, extOpts.fIncludeDataForks != 0);
fPreferences.SetPrefBool(kPrExtractIncludeRsrc, extOpts.fIncludeRsrcForks != 0);
fPreferences.SetPrefBool(kPrExtractIncludeDisk, extOpts.fIncludeDiskImages != 0);
fPreferences.SetPrefBool(kPrExtractEnableReformat, extOpts.fEnableReformat != 0);
fPreferences.SetPrefBool(kPrExtractDiskTo2MG, extOpts.fDiskTo2MG != 0);
fPreferences.SetPrefBool(kPrExtractAddTypePreservation, extOpts.fAddTypePreservation != 0);
fPreferences.SetPrefBool(kPrExtractAddExtension, extOpts.fAddExtension != 0);
fPreferences.SetPrefBool(kPrExtractStripFolderNames, extOpts.fStripFolderNames != 0);
fPreferences.SetPrefBool(kPrExtractOverwriteExisting, extOpts.fOverwriteExisting != 0);
LOGI("Requested extract path is '%ls'", (LPCWSTR) extOpts.fExtractPath);
/*
* Create a "selection set" of things to display.
*/
SelectionSet selSet;
int threadMask = 0;
if (extOpts.fIncludeDataForks)
threadMask |= GenericEntry::kDataThread;
if (extOpts.fIncludeRsrcForks)
threadMask |= GenericEntry::kRsrcThread;
if (extOpts.fIncludeDiskImages)
threadMask |= GenericEntry::kDiskImageThread;
if (extOpts.fFilesToExtract == ExtractOptionsDialog::kExtractSelection) {
selSet.CreateFromSelection(fpContentList, threadMask);
} else {
selSet.CreateFromAll(fpContentList, threadMask);
}
//selSet.Dump();
if (selSet.GetNumEntries() == 0) {
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
MessageBox(L"No files matched the selection criteria.",
L"No match", MB_OK | MB_ICONEXCLAMATION);
return;
}
/*
* Set up the progress dialog then do the extraction.
*/
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionExtract, this);
DoBulkExtract(&selSet, &extOpts);
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsExtract(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL);
2007-03-27 17:47:10 +00:00
}
void MainWindow::DoBulkExtract(SelectionSet* pSelSet,
const ExtractOptionsDialog* pExtOpts)
2007-03-27 17:47:10 +00:00
{
/*
* IMPORTANT: since the pActionProgress dialog has the foreground, it's
* vital that any MessageBox calls go through that. Otherwise the
* progress dialog message handler won't get disabled by MessageBox and
* we can end up permanently hiding the dialog. (Could also use
* ::MessageBox or ::AfxMessageBox instead.)
*/
2014-11-18 05:13:13 +00:00
ReformatHolder* pHolder = NULL;
bool overwriteExisting, ovwrForAll;
2014-11-18 05:13:13 +00:00
ASSERT(pSelSet != NULL);
ASSERT(fpActionProgress != NULL);
pSelSet->IterReset();
/* set up our "overwrite existing files" logic */
overwriteExisting = ovwrForAll = (pExtOpts->fOverwriteExisting != FALSE);
while (true) {
SelectionEntry* pSelEntry;
bool result;
/* make sure we service events */
PeekAndPump();
pSelEntry = pSelSet->IterNext();
2014-11-18 05:13:13 +00:00
if (pSelEntry == NULL) {
SuccessBeep();
break; // out of while (all done!)
}
GenericEntry* pEntry = pSelEntry->GetEntry();
if (pEntry->GetDamaged()) {
LOGI("Skipping '%ls' due to damage", (LPCWSTR) pEntry->GetPathNameUNI());
continue;
}
/*
* Extract all parts of the file -- including those we don't actually
* intend to extract to disk -- and hold them in the ReformatHolder.
* Some formats, e.g. GWP, need the resource fork to do formatting,
* so it's important to have the resource fork available even if
* we're just going to throw it away.
*
* The selection set should have screened out anything totally
* inappropriate, e.g. files with nothing but a resource fork don't
* make it into the set if we're not extracting resource forks.
*
* We only want to reformat files, not disk images, directories,
* volume dirs, etc. We have a reformatter for ProDOS directories,
* but (a) we don't explicitly extract subdirs, and (b) we'd really
* like directories to be directories so we can extract files into
* them.
*/
if (pExtOpts->ShouldTryReformat() &&
(pEntry->GetRecordKind() == GenericEntry::kRecordKindFile ||
pEntry->GetRecordKind() == GenericEntry::kRecordKindForkedFile))
{
fpActionProgress->SetArcName(pEntry->GetDisplayName());
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
fpActionProgress->SetFileName(L"-");
SET_PROGRESS_BEGIN();
if (GetFileParts(pEntry, &pHolder) == 0) {
/*
* Use the prefs, but disable generic text conversion, so
* that we default to "raw". That way we will use the text
* conversion that the user has specified in the "extract"
* dialog.
*
* We might want to just disable any "always"-level
* reformatter, but that would require tweaking the reformat
* code to return "raw" when nothing applies.
*/
ConfigureReformatFromPreferences(pHolder);
pHolder->SetReformatAllowed(ReformatHolder::kReformatTextEOL_HA,
false);
pHolder->SetSourceAttributes(
pEntry->GetFileType(),
pEntry->GetAuxType(),
ReformatterSourceFormat(pEntry->GetSourceFS()),
pEntry->GetFileNameExtensionMOR());
pHolder->TestApplicability();
}
}
if (pExtOpts->fIncludeDataForks && pEntry->GetHasDataFork()) {
result = ExtractEntry(pEntry, GenericEntry::kDataThread,
pHolder, pExtOpts, &overwriteExisting, &ovwrForAll);
if (!result)
break;
}
if (pExtOpts->fIncludeRsrcForks && pEntry->GetHasRsrcFork()) {
result = ExtractEntry(pEntry, GenericEntry::kRsrcThread,
pHolder, pExtOpts, &overwriteExisting, &ovwrForAll);
if (!result)
break;
}
if (pExtOpts->fIncludeDiskImages && pEntry->GetHasDiskImage()) {
result = ExtractEntry(pEntry, GenericEntry::kDiskImageThread,
pHolder, pExtOpts, &overwriteExisting, &ovwrForAll);
if (!result)
break;
}
delete pHolder;
2014-11-18 05:13:13 +00:00
pHolder = NULL;
}
// if they cancelled, delete the "stray"
delete pHolder;
2007-03-27 17:47:10 +00:00
}
bool MainWindow::ExtractEntry(GenericEntry* pEntry, int thread,
ReformatHolder* pHolder, const ExtractOptionsDialog* pExtOpts,
bool* pOverwriteExisting, bool* pOvwrForAll)
2007-03-27 17:47:10 +00:00
{
/*
* This first bit of setup is the same for every file. However, it's
* pretty quick, and it's easier to pass "pExtOpts" in than all of
* this stuff, so we just do it every time.
*/
GenericEntry::ConvertEOL convEOL;
GenericEntry::ConvertHighASCII convHA;
bool convTextByType = false;
/* translate the EOL conversion mode into GenericEntry terms */
switch (pExtOpts->fConvEOL) {
case ExtractOptionsDialog::kConvEOLNone:
convEOL = GenericEntry::kConvertEOLOff;
break;
case ExtractOptionsDialog::kConvEOLType:
convEOL = GenericEntry::kConvertEOLOff;
convTextByType = true;
break;
case ExtractOptionsDialog::kConvEOLAuto:
convEOL = GenericEntry::kConvertEOLAuto;
break;
case ExtractOptionsDialog::kConvEOLAll:
convEOL = GenericEntry::kConvertEOLOn;
break;
default:
ASSERT(false);
convEOL = GenericEntry::kConvertEOLOff;
break;
}
if (pExtOpts->fConvHighASCII)
convHA = GenericEntry::kConvertHAAuto;
else
convHA = GenericEntry::kConvertHAOff;
//LOGI(" DBE initial text conversion: eol=%d ha=%d",
// convEOL, convHA);
ReformatHolder holder;
CString outputPath;
CString failed, errMsg;
CheckedLoadString(&failed, IDS_FAILED);
bool writeFailed = false;
bool extractAs2MG = false;
2014-11-18 05:13:13 +00:00
char* reformatText = NULL;
MyDIBitmap* reformatDib = NULL;
2014-11-18 05:13:13 +00:00
ASSERT(pEntry != NULL);
/*
* If we're interested in extracting disk images as 2MG files,
* see if we want to handle this one that way.
*
* If they said "don't extract disk images", the images should
* have been culled from the selection set earlier.
*/
if (pEntry->GetRecordKind() == GenericEntry::kRecordKindDisk) {
ASSERT(pExtOpts->fIncludeDiskImages);
if (pExtOpts->fDiskTo2MG) {
if (pEntry->GetUncompressedLen() > 0 &&
(pEntry->GetUncompressedLen() % TwoImgHeader::kBlockSize) == 0)
{
extractAs2MG = true;
} else {
LOGI("Not extracting funky image '%ls' as 2MG (len=%I64d)",
(LPCWSTR) pEntry->GetPathNameUNI(),
pEntry->GetUncompressedLen());
}
}
}
/*
* Convert the archived pathname to something suitable for the
* local machine (i.e. Win32).
*/
PathProposal pathProp;
CString convName, convNameExtd, convNameExtdPlus;
pathProp.Init(pEntry);
pathProp.fThreadKind = thread;
if (pExtOpts->fStripFolderNames)
pathProp.fJunkPaths = true;
pathProp.ArchiveToLocal();
convName = pathProp.fLocalPathName;
/* run it through again, this time with optional type preservation */
if (pExtOpts->fAddTypePreservation)
pathProp.fPreservation = true;
pathProp.ArchiveToLocal();
convNameExtd = pathProp.fLocalPathName;
/* and a 3rd time, also taking additional extensions into account */
if (pExtOpts->fAddExtension)
pathProp.fAddExtension = true;
pathProp.ArchiveToLocal();
convNameExtdPlus = pathProp.fLocalPathName;
/*
* Prepend the extraction dir to the local pathname. We also add
* the sub-volume name (if any), which should already be a valid
* Win32 directory name. We can't add it earlier because the fssep
* char might be '\0'.
*/
ASSERT(pExtOpts->fExtractPath.Right(1) == "\\");
CString adjustedExtractPath(pExtOpts->fExtractPath);
if (!pExtOpts->fStripFolderNames && !pEntry->GetSubVolName().IsEmpty()) {
adjustedExtractPath += pEntry->GetSubVolName();
adjustedExtractPath += "\\";
}
outputPath = adjustedExtractPath;
outputPath += convNameExtdPlus;
2014-11-18 05:13:13 +00:00
ReformatOutput* pOutput = NULL;
/*
* If requested, try to reformat this file.
*/
2014-11-18 05:13:13 +00:00
if (pHolder != NULL) {
ReformatHolder::ReformatPart part = ReformatHolder::kPartUnknown;
ReformatHolder::ReformatID id;
CString title;
//int result;
switch (thread) {
case GenericEntry::kDataThread:
part = ReformatHolder::kPartData;
break;
case GenericEntry::kRsrcThread:
part = ReformatHolder::kPartRsrc;
break;
case GenericEntry::kDiskImageThread:
part = ReformatHolder::kPartData;
break;
case GenericEntry::kCommentThread:
default:
assert(false);
return false;
}
fpActionProgress->SetFileName(_T("(reformatting)"));
id = pHolder->FindBest(part);
{
CWaitCursor waitc;
pOutput = pHolder->Apply(part, id);
}
2014-11-18 05:13:13 +00:00
if (pOutput != NULL) {
/* use output pathname without preservation */
CString tmpPath;
bool goodReformat = true;
bool noChangePath = false;
tmpPath = adjustedExtractPath;
tmpPath += convName;
CString lastFour = tmpPath.Right(4);
/*
* Tack on a file extension identifying the reformatted
* contents. If the filename already has the correct
* extension, don't tack it on again.
*/
switch (pOutput->GetOutputKind()) {
case ReformatOutput::kOutputText:
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
if (lastFour.CompareNoCase(L".txt") != 0)
tmpPath += L".txt";
break;
case ReformatOutput::kOutputRTF:
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
if (lastFour.CompareNoCase(L".rtf") != 0)
tmpPath += L".rtf";
break;
case ReformatOutput::kOutputCSV:
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
if (lastFour.CompareNoCase(L".csv") != 0)
tmpPath += L".csv";
break;
case ReformatOutput::kOutputBitmap:
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
if (lastFour.CompareNoCase(L".bmp") != 0)
tmpPath += L".bmp";
break;
case ReformatOutput::kOutputRaw:
noChangePath = true;
break;
default:
// kOutputErrorMsg, kOutputUnknown
goodReformat = false;
break;
}
if (goodReformat) {
if (!noChangePath)
outputPath = tmpPath;
} else {
delete pOutput;
2014-11-18 05:13:13 +00:00
pOutput = NULL;
}
}
}
if (extractAs2MG) {
/*
* Reduce to base name and add 2IMG suffix. Would be nice to keep
* the non-extended file type preservation stuff, but right now we
* only expect unadorned sectors for that (and so does NuLib2).
*/
outputPath = adjustedExtractPath;
outputPath += convName;
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
outputPath += L".2mg";
}
/* update the display in case we renamed it */
if (outputPath != fpActionProgress->GetFileName()) {
LOGI(" Renamed our output, from '%ls' to '%ls'",
(LPCWSTR) fpActionProgress->GetFileName(), (LPCWSTR) outputPath);
fpActionProgress->SetFileName(outputPath);
}
/*
* Update the progress meter output filename, and reset the thermometer.
*/
fpActionProgress->SetArcName(pathProp.fStoredPathName);
fpActionProgress->SetFileName(outputPath);
LOGI("Extracting from '%ls' to '%ls'",
(LPCWSTR) pathProp.fStoredPathName, (LPCWSTR) outputPath);
SET_PROGRESS_BEGIN();
/*
* Open the output file.
*
* Returns IDCANCEL on failures as well as user cancellation.
*/
2014-11-18 05:13:13 +00:00
FILE* fp = NULL;
int result;
result = OpenOutputFile(&outputPath, pathProp, pEntry->GetModWhen(),
pOverwriteExisting, pOvwrForAll, &fp);
if (result == IDCANCEL) {
// no messagebox for this one
delete pOutput;
return false;
}
/* update the display in case they renamed the file */
if (outputPath != fpActionProgress->GetFileName()) {
LOGI(" Detected rename, from '%ls' to '%ls'",
(LPCWSTR) fpActionProgress->GetFileName(), (LPCWSTR) outputPath);
fpActionProgress->SetFileName(outputPath);
}
2014-11-18 05:13:13 +00:00
if (fp == NULL) {
/* looks like they elected to skip extraction of this file */
delete pOutput;
return true;
}
//EventPause(500); // DEBUG DEBUG
/*
* Handle "extract as 2MG" by writing a 2MG header to the start
* of the file before we hand off to the extraction function.
*
* NOTE: we're currently assuming that we're extracting an image
* in ProDOS sector order. This is a valid assumption so long as
* we're only pulling disk images out of ShrinkIt archives.
*
* We don't currently use the WriteFooter call here, because we're
* not adding comments.
*/
if (extractAs2MG) {
TwoImgHeader header;
header.InitHeader(TwoImgHeader::kImageFormatProDOS,
(long) pEntry->GetUncompressedLen(),
(long) (pEntry->GetUncompressedLen() / TwoImgHeader::kBlockSize));
int err;
ASSERT(ftell(fp) == 0);
err = header.WriteHeader(fp);
if (err != 0) {
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
errMsg.Format(L"Unable to save 2MG file '%ls': %hs\n",
(LPCWSTR) outputPath, strerror(err));
fpActionProgress->MessageBox(errMsg, failed,
MB_OK | MB_ICONERROR);
goto open_file_fail;
}
ASSERT(ftell(fp) == 64); // size of 2MG header
}
/*
* In some cases we want to override the automatic text detection.
*
* If we're in "auto" mode, force conversion on for DOS/RDOS text files.
* This is important when "convHA" is off, because in high-ASCII mode
* we might not recognize text files for what they are. We also
* consider 0x00 to be binary, which screws up random-access text.
*
* We don't want to do text conversion on disk images or resource
* forks, ever. Turn them off here.
*/
GenericEntry::ConvertEOL thisConv;
thisConv = convEOL;
if (thisConv == GenericEntry::kConvertEOLAuto) {
if (DiskImg::UsesDOSFileStructure(pEntry->GetSourceFS()) &&
pEntry->GetFileType() == kFileTypeTXT)
{
LOGI("Switching EOLAuto to EOLOn for DOS text file");
thisConv = GenericEntry::kConvertEOLOn;
}
} else if (convTextByType) {
/* force it on or off when in conv-by-type mode */
if (pEntry->GetFileType() == kFileTypeTXT ||
pEntry->GetFileType() == kFileTypeSRC)
{
LOGI("Enabling EOL conv for text file");
thisConv = GenericEntry::kConvertEOLOn;
} else {
ASSERT(thisConv == GenericEntry::kConvertEOLOff);
}
}
if (thisConv != GenericEntry::kConvertEOLOff &&
(thread == GenericEntry::kRsrcThread ||
thread == GenericEntry::kDiskImageThread))
{
LOGI("Disabling EOL conv for resource fork or disk image");
thisConv = GenericEntry::kConvertEOLOff;
}
/*
* Extract the contents to the file.
*
* In some cases, notably when the file size exceeds the limit of
* the reformatter, we will be trying to reformat but won't have
* loaded the original data. In such cases we fall through to the
* normal extraction mode, because we threw out pOutput above when
* the result was kOutputErrorMsg.
*
* (Could also be due to extraction failure, e.g. bad CRC.)
*/
2014-11-18 05:13:13 +00:00
if (pOutput != NULL) {
/*
* We have the data in our buffer. Write it out. No need
* to tweak the progress updater, which already shows 100%.
*
* There are four possibilities:
* - Valid text/rtf/csv converted text. Write reformatted.
* - Valid bitmap converted. Write bitmap.
* - No reformatter found, type is "raw". Write raw. (Note
* this may be zero bytes long for an empty file.)
* - Error message encoded in result. Should not be here!
*/
if (pOutput->GetOutputKind() == ReformatOutput::kOutputText ||
pOutput->GetOutputKind() == ReformatOutput::kOutputRTF ||
pOutput->GetOutputKind() == ReformatOutput::kOutputCSV)
{
LOGI(" Writing text, RTF, CSV, or raw");
2014-11-18 05:13:13 +00:00
ASSERT(pOutput->GetTextBuf() != NULL);
int err = 0;
if (fwrite(pOutput->GetTextBuf(),
pOutput->GetTextLen(), 1, fp) != 1)
err = errno;
if (err != 0) {
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
errMsg.Format(L"Unable to save reformatted file '%ls': %hs\n",
(LPCWSTR) outputPath, strerror(err));
fpActionProgress->MessageBox(errMsg, failed,
MB_OK | MB_ICONERROR);
writeFailed = true;
} else {
SET_PROGRESS_UPDATE(100);
}
} else if (pOutput->GetOutputKind() == ReformatOutput::kOutputBitmap) {
LOGI(" Writing bitmap");
2014-11-18 05:13:13 +00:00
ASSERT(pOutput->GetDIB() != NULL);
int err = pOutput->GetDIB()->WriteToFile(fp);
if (err != 0) {
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
errMsg.Format(L"Unable to save bitmap '%ls': %hs\n",
(LPCWSTR) outputPath, strerror(err));
fpActionProgress->MessageBox(errMsg, failed,
MB_OK | MB_ICONERROR);
writeFailed = true;
} else {
SET_PROGRESS_UPDATE(100);
}
} else if (pOutput->GetOutputKind() == ReformatOutput::kOutputRaw) {
/*
* Send raw data through the text conversion configured in the
* extract files dialog. Any file for which no dedicated
* reformatter could be found ends up here.
*
* We could just send it through to the generic non-reformatter
* case, but that would require reading the file twice.
*/
LOGI(" Writing un-reformatted data (%ld bytes)",
pOutput->GetTextLen());
2014-11-18 05:13:13 +00:00
ASSERT(pOutput->GetTextBuf() != NULL);
bool lastCR = false;
GenericEntry::ConvertHighASCII thisConvHA = convHA;
int err;
err = GenericEntry::WriteConvert(fp, pOutput->GetTextBuf(),
pOutput->GetTextLen(), &thisConv, &thisConvHA, &lastCR);
if (err != 0) {
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
errMsg.Format(L"Unable to write file '%ls': %hs\n",
(LPCWSTR) outputPath, strerror(err));
fpActionProgress->MessageBox(errMsg, failed,
MB_OK | MB_ICONERROR);
writeFailed = true;
} else {
SET_PROGRESS_UPDATE(100);
}
} else {
/* something failed, and we don't have the file */
LOGI("How'd we get here?");
ASSERT(false);
}
} else {
/*
* We don't have the data, probably because we aren't using file
* reformatters. Use the GenericEntry extraction routine to copy
* the data directly into the file.
*
* We also get here if the file has a length of zero.
*/
CString msg;
int result;
2014-11-18 05:13:13 +00:00
ASSERT(fpActionProgress != NULL);
LOGI("Extracting '%ls', requesting thisConv=%d, convHA=%d",
(LPCWSTR) outputPath, thisConv, convHA);
result = pEntry->ExtractThreadToFile(thread, fp,
thisConv, convHA, &msg);
if (result != IDOK) {
if (result == IDCANCEL) {
CString msg;
CheckedLoadString(&msg, IDS_OPERATION_CANCELLED);
fpActionProgress->MessageBox(msg,
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
L"CiderPress", MB_OK | MB_ICONEXCLAMATION);
} else {
LOGI(" FAILED on '%ls': %ls",
(LPCWSTR) outputPath, (LPCWSTR) msg);
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
errMsg.Format(L"Unable to extract file '%ls': %ls\n",
(LPCWSTR) outputPath, (LPCWSTR) msg);
fpActionProgress->MessageBox(errMsg, failed,
MB_OK | MB_ICONERROR);
}
writeFailed = true;
}
}
2007-03-27 17:47:10 +00:00
open_file_fail:
delete pOutput;
fclose(fp);
if (writeFailed) {
// clean up
::DeleteFile(outputPath);
return false;
}
/*
* Fix the modification date.
*/
PathName datePath(outputPath);
datePath.SetModWhen(pEntry->GetModWhen());
// datePath.SetAccess(pEntry->GetAccess());
return true;
2007-03-27 17:47:10 +00:00
}
int MainWindow::OpenOutputFile(CString* pOutputPath, const PathProposal& pathProp,
time_t arcFileModWhen, bool* pOverwriteExisting, bool* pOvwrForAll,
FILE** pFp)
2007-03-27 17:47:10 +00:00
{
const int kUserCancel = -2; // must not conflict with errno values
CString failed;
CString msg;
int err = 0;
2007-03-27 17:47:10 +00:00
CheckedLoadString(&failed, IDS_FAILED);
2007-03-27 17:47:10 +00:00
2014-11-18 05:13:13 +00:00
*pFp = NULL;
2007-03-27 17:47:10 +00:00
did_rename:
PathName path(*pOutputPath);
if (path.Exists()) {
if (*pOverwriteExisting) {
2007-03-27 17:47:10 +00:00
do_overwrite:
/* delete existing */
LOGI(" Deleting existing '%ls'", (LPCWSTR) *pOutputPath);
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
if (::_wunlink(*pOutputPath) != 0) {
err = errno;
LOGI(" Failed deleting '%ls', err=%d",
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
(LPCWSTR)*pOutputPath, err);
if (err == ENOENT) {
/* user might have removed it while dialog was up */
err = 0;
} else {
/* unable to delete, we'd better bail out */
goto bail;
}
}
} else if (*pOvwrForAll) {
/* never overwrite */
LOGI(" Skipping '%ls'", (LPCWSTR) *pOutputPath);
goto bail;
} else {
/* no firm policy, ask the user */
ConfirmOverwriteDialog confOvwr;
PathName path(*pOutputPath);
confOvwr.fExistingFile = *pOutputPath;
confOvwr.fExistingFileModWhen = path.GetModWhen();
confOvwr.fNewFileSource = pathProp.fStoredPathName;
confOvwr.fNewFileModWhen = arcFileModWhen;
if (confOvwr.DoModal() == IDCANCEL) {
err = kUserCancel;
goto bail;
}
if (confOvwr.fResultRename) {
*pOutputPath = confOvwr.fExistingFile;
goto did_rename;
}
if (confOvwr.fResultApplyToAll) {
*pOvwrForAll = confOvwr.fResultApplyToAll;
*pOverwriteExisting = confOvwr.fResultOverwrite;
}
if (confOvwr.fResultOverwrite)
goto do_overwrite;
else
goto bail;
}
}
/* create the subdirectories, if necessary */
err = path.CreatePathIFN();
if (err != 0)
goto bail;
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
*pFp = _wfopen(*pOutputPath, L"wb");
2014-11-18 05:13:13 +00:00
if (*pFp == NULL)
err = errno ? errno : -1;
/* fall through with error */
2007-03-27 17:47:10 +00:00
bail:
/* if we failed, tell the user why */
if (err == ENOTDIR) {
/* part of the output path exists, but isn't a directory */
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
msg.Format(L"Unable to create folders for '%ls': part of the path "
L"already exists but is not a folder.\n",
(LPCWSTR) *pOutputPath);
fpActionProgress->MessageBox(msg, failed, MB_OK | MB_ICONERROR);
return IDCANCEL;
} else if (err == EINVAL) {
/* invalid argument; assume it's an invalid filename */
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
msg.Format(L"Unable to create file '%ls': invalid filename.\n",
(LPCWSTR) *pOutputPath);
fpActionProgress->MessageBox(msg, failed, MB_OK | MB_ICONERROR);
return IDCANCEL;
} else if (err == kUserCancel) {
/* user elected to cancel */
LOGI("Cancelling due to user request");
return IDCANCEL;
} else if (err != 0) {
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
msg.Format(L"Unable to create file '%ls': %hs\n",
(LPCWSTR) *pOutputPath, strerror(err));
fpActionProgress->MessageBox(msg, failed, MB_OK | MB_ICONERROR);
return IDCANCEL;
}
return IDOK;
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Test
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsTest(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
/*
* Ask the user about various options.
*/
UseSelectionDialog selOpts(fpContentList->GetSelectedCount(), this);
selOpts.Setup(IDS_TEST_TITLE, IDS_TEST_OK, IDS_TEST_SELECTED_COUNT,
IDS_TEST_SELECTED_COUNTS_FMT, IDS_TEST_ALL_FILES);
if (fpContentList->GetSelectedCount() > 0)
selOpts.fFilesToAction = UseSelectionDialog::kActionSelection;
else
selOpts.fFilesToAction = UseSelectionDialog::kActionAll;
if (selOpts.DoModal() != IDOK) {
LOGI("Test cancelled");
return;
}
/*
* Create a "selection set" of things to test.
*
* We don't currently test directories, because we don't currently
* allow testing anything that has a directory (NuFX doesn't store
* them explicitly). We could probably add them to the threadMask.
*/
SelectionSet selSet;
int threadMask = GenericEntry::kAnyThread;
if (selOpts.fFilesToAction == UseSelectionDialog::kActionSelection) {
selSet.CreateFromSelection(fpContentList, threadMask);
} else {
selSet.CreateFromAll(fpContentList, threadMask);
}
//selSet.Dump();
if (selSet.GetNumEntries() == 0) {
/* should be impossible */
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
MessageBox(L"No files matched the selection criteria.",
L"No match", MB_OK|MB_ICONEXCLAMATION);
return;
}
/*
* Set up the progress window.
*/
bool result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionTest, this);
result = fpOpenArchive->TestSelection(fpActionProgress, &selSet);
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
//if (result)
// SuccessBeep();
2007-03-27 17:47:10 +00:00
}
void
MainWindow::OnUpdateActionsTest(CCmdUI* pCmdUI)
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && fpContentList->GetItemCount() > 0
&& fpOpenArchive->GetCapability(GenericArchive::kCapCanTest));
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Delete
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsDelete(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
ASSERT(!fpOpenArchive->IsReadOnly());
/*
* We handle deletions specially. If they have selected any
* subdirectories, we recursively select the files in those subdirs
* as well. We want to do it early so that the "#of files to delete"
* display accurately reflects what we're about to do.
*/
fpContentList->SelectSubdirContents();
2007-03-27 17:47:10 +00:00
#if 0
/*
* Ask the user about various options.
*/
UseSelectionDialog delOpts(fpContentList->GetSelectedCount(), this);
delOpts.Setup(IDS_DEL_TITLE, IDS_DEL_OK, IDS_DEL_SELECTED_COUNT,
IDS_DEL_SELECTED_COUNTS_FMT, IDS_DEL_ALL_FILES);
if (fpContentList->GetSelectedCount() > 0)
delOpts.fFilesToAction = UseSelectionDialog::kActionSelection;
else
delOpts.fFilesToAction = UseSelectionDialog::kActionAll;
if (delOpts.DoModal() != IDOK) {
LOGI("Delete cancelled");
return;
}
2007-03-27 17:47:10 +00:00
#endif
/*
* Create a "selection set" of things to delete.
*
* We can't delete volume directories, so they're not included.
*/
SelectionSet selSet;
int threadMask = GenericEntry::kAnyThread |
GenericEntry::kAllowDirectory /*| GenericEntry::kAllowVolumeDir*/;
2007-03-27 17:47:10 +00:00
#if 0
if (delOpts.fFilesToAction == UseSelectionDialog::kActionSelection) {
selSet.CreateFromSelection(fpContentList, threadMask);
} else {
CString appName;
UINT response;
appName.LoadString(IDS_MB_APP_NAME);
response = MessageBox("Are you sure you want to delete everything?",
appName, MB_OKCANCEL | MB_ICONEXCLAMATION);
if (response == IDCANCEL)
return;
selSet.CreateFromAll(fpContentList, threadMask);
}
//selSet.Dump();
2007-03-27 17:47:10 +00:00
#endif
selSet.CreateFromSelection(fpContentList, threadMask);
if (selSet.GetNumEntries() == 0) {
/* can happen if they selected volume dir only */
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
MessageBox(L"Nothing to delete.",
L"No match", MB_OK | MB_ICONEXCLAMATION);
return;
}
2007-03-27 17:47:10 +00:00
CString appName, msg;
2007-03-27 17:47:10 +00:00
CheckedLoadString(&appName, IDS_MB_APP_NAME);
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
msg.Format(L"Delete %d file%ls?", selSet.GetNumEntries(),
selSet.GetNumEntries() == 1 ? L"" : L"s");
if (MessageBox(msg, appName, MB_OKCANCEL | MB_ICONQUESTION) != IDOK)
return;
2007-03-27 17:47:10 +00:00
bool result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionDelete, this);
2007-03-27 17:47:10 +00:00
//fpContentList->Invalidate(); // don't allow updates until done
result = fpOpenArchive->DeleteSelection(fpActionProgress, &selSet);
fpContentList->Reload();
2007-03-27 17:47:10 +00:00
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
2007-03-27 17:47:10 +00:00
if (result)
SuccessBeep();
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsDelete(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly()
&& fpContentList->GetSelectedCount() > 0);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Rename
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsRename(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
ASSERT(!fpOpenArchive->IsReadOnly());
/*
* Create a "selection set" of entries to rename. We always go by
* the selection, so there's no need to present a "all or some?" dialog.
*
* Renaming the volume dir is not done from here, so we don't include
* it in the set. We could theoretically allow renaming of "damaged"
* files, since most of the time the damage is in the file structure
* not the directory, but the disk will be read-only anyway so there's
* no point.
*/
SelectionSet selSet;
int threadMask = GenericEntry::kAnyThread | GenericEntry::kAllowDirectory;
selSet.CreateFromSelection(fpContentList, threadMask);
//selSet.Dump();
if (selSet.GetNumEntries() == 0) {
/* should be impossible */
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
MessageBox(L"No files matched the selection criteria.",
L"No match", MB_OK | MB_ICONEXCLAMATION);
return;
}
//fpContentList->Invalidate(); // this might be unnecessary
fpOpenArchive->RenameSelection(this, &selSet);
fpContentList->Reload();
// user interaction on each step, so skip the SuccessBeep
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsRename(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly()
&& fpContentList->GetSelectedCount() > 0);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Edit Comment
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsEditComment(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
ASSERT(!fpOpenArchive->IsReadOnly());
EditCommentDialog editDlg(this);
CString oldComment;
GenericEntry* pEntry = GetSelectedItem(fpContentList);
2014-11-18 05:13:13 +00:00
if (pEntry == NULL) {
ASSERT(false);
return;
}
if (!pEntry->GetHasComment()) {
CString question, title;
int result;
CheckedLoadString(&question, IDS_NO_COMMENT_ADD);
CheckedLoadString(&title, IDS_EDIT_COMMENT);
result = MessageBox(question, title, MB_OKCANCEL | MB_ICONQUESTION);
if (result == IDCANCEL)
return;
editDlg.fComment = "";
editDlg.fNewComment = true;
} else {
fpOpenArchive->GetComment(this, pEntry, &editDlg.fComment);
}
int result;
result = editDlg.DoModal();
if (result == IDOK) {
//fpContentList->Invalidate(); // probably unnecessary
fpOpenArchive->SetComment(this, pEntry, editDlg.fComment);
fpContentList->Reload();
} else if (result == EditCommentDialog::kDeleteCommentID) {
//fpContentList->Invalidate(); // possibly unnecessary
fpOpenArchive->DeleteComment(this, pEntry);
fpContentList->Reload();
}
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsEditComment(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly() &&
fpContentList->GetSelectedCount() == 1 &&
fpOpenArchive->GetCapability(GenericArchive::kCapCanEditComment));
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Edit Properties
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsEditProps(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
EditPropsDialog propsDlg(this);
CString oldComment;
GenericEntry* pEntry = GetSelectedItem(fpContentList);
2014-11-18 05:13:13 +00:00
if (pEntry == NULL) {
ASSERT(false);
return;
}
propsDlg.InitProps(pEntry);
if (fpOpenArchive->IsReadOnly())
propsDlg.fReadOnly = true;
else if (pEntry->GetRecordKind() == GenericEntry::kRecordKindVolumeDir)
propsDlg.fReadOnly = true;
int result;
result = propsDlg.DoModal();
if (result == IDOK && !propsDlg.fReadOnly) {
(void) fpOpenArchive->SetProps(this, pEntry, &propsDlg.fProps);
// only needed if underlying archive reloads
fpContentList->Reload(true);
}
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsEditProps(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
// allow it in read-only mode, so we can view the props
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL &&
fpContentList->GetSelectedCount() == 1);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Rename Volume
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsRenameVolume(void)
2007-03-27 17:47:10 +00:00
{
RenameVolumeDialog rvDialog;
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
ASSERT(!fpOpenArchive->IsReadOnly());
/* only know how to deal with disk images */
if (fpOpenArchive->GetArchiveKind() != GenericArchive::kArchiveDiskImage) {
ASSERT(false);
return;
}
DiskImgLib::DiskFS* pDiskFS;
pDiskFS = ((DiskArchive*) fpOpenArchive)->GetDiskFS();
2014-11-18 05:13:13 +00:00
ASSERT(pDiskFS != NULL);
rvDialog.fpArchive = (DiskArchive*) fpOpenArchive;
if (rvDialog.DoModal() != IDOK)
return;
//LOGI("Creating '%s'", rvDialog.fNewName);
/* rename the chosen disk to the specified name */
bool result;
result = fpOpenArchive->RenameVolume(this, rvDialog.fpChosenDiskFS,
rvDialog.fNewName);
if (!result) {
LOGW("RenameVolume FAILED");
/* keep going -- reload just in case something partially happened */
}
/*
* We need to do two things: reload the content list, because the
* underlying DiskArchive got reloaded, and update the title bar. We
* put the "volume ID" in the title, and we most likely just changed it.
*
* SetCPTitle invokes fpOpenArchive->GetDescription(), which pulls the
* volume ID out of the primary DiskFS.
*/
fpContentList->Reload();
SetCPTitle(fOpenArchivePathName, fpOpenArchive);
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsRenameVolume(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly() &&
fpOpenArchive->GetCapability(GenericArchive::kCapCanRenameVolume));
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Recompress
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsRecompress(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
/*
* Ask the user about various options.
*/
RecompressOptionsDialog selOpts(fpContentList->GetSelectedCount(), this);
selOpts.Setup(IDS_RECOMP_TITLE, IDS_RECOMP_OK, IDS_RECOMP_SELECTED_COUNT,
IDS_RECOMP_SELECTED_COUNTS_FMT, IDS_RECOMP_ALL_FILES);
if (fpContentList->GetSelectedCount() > 0)
selOpts.fFilesToAction = UseSelectionDialog::kActionSelection;
else
selOpts.fFilesToAction = UseSelectionDialog::kActionAll;
selOpts.fCompressionType = fPreferences.GetPrefLong(kPrCompressionType);
if (selOpts.DoModal() != IDOK) {
LOGI("Recompress cancelled");
return;
}
/*
* Create a "selection set" of data forks, resource forks, and disk
* images. If an entry has nothing but a comment, ignore it.
*/
SelectionSet selSet;
int threadMask = GenericEntry::kDataThread | GenericEntry::kRsrcThread |
GenericEntry::kDiskImageThread;
if (selOpts.fFilesToAction == UseSelectionDialog::kActionSelection) {
selSet.CreateFromSelection(fpContentList, threadMask);
} else {
selSet.CreateFromAll(fpContentList, threadMask);
}
//selSet.Dump();
if (selSet.GetNumEntries() == 0) {
/* should be impossible */
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
MessageBox(L"No files matched the selection criteria.",
L"No match", MB_OK|MB_ICONEXCLAMATION);
return;
}
LONGLONG beforeUncomp, beforeComp;
LONGLONG afterUncomp, afterComp;
CalcTotalSize(&beforeUncomp, &beforeComp);
/*
* Set up the progress window.
*/
int result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionRecompress, this);
//fpContentList->Invalidate(); // possibly unnecessary
result = fpOpenArchive->RecompressSelection(fpActionProgress, &selSet,
&selOpts);
fpContentList->Reload();
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
if (result) {
CString msg, appName;
CalcTotalSize(&afterUncomp, &afterComp);
ASSERT(beforeUncomp == afterUncomp);
CheckedLoadString(&appName, IDS_MB_APP_NAME);
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
msg.Format(L"Total uncompressed size of all files:\t%.1fK\r\n"
L"Total size before recompress:\t\t%.1fK\r\n"
L"Total size after recompress:\t\t%.1fK\r\n"
L"Overall reduction:\t\t\t%.1fK",
beforeUncomp / 1024.0, beforeComp / 1024.0, afterComp / 1024.0,
(beforeComp - afterComp) / 1024.0);
MessageBox(msg, appName, MB_OK|MB_ICONINFORMATION);
}
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsRecompress(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly() &&
fpContentList->GetItemCount() > 0 &&
fpOpenArchive->GetCapability(GenericArchive::kCapCanRecompress));
2007-03-27 17:47:10 +00:00
}
void MainWindow::CalcTotalSize(LONGLONG* pUncomp, LONGLONG* pComp) const
2007-03-27 17:47:10 +00:00
{
GenericEntry* pEntry = fpOpenArchive->GetEntries();
LONGLONG uncomp = 0, comp = 0;
2007-03-27 17:47:10 +00:00
2014-11-18 05:13:13 +00:00
while (pEntry != NULL) {
uncomp += pEntry->GetUncompressedLen();
comp += pEntry->GetCompressedLen();
pEntry = pEntry->GetNext();
}
2007-03-27 17:47:10 +00:00
*pUncomp = uncomp;
*pComp = comp;
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Convert to disk archive
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsConvDisk(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
/*
* Ask the user about various options.
*/
ConvDiskOptionsDialog selOpts(fpContentList->GetSelectedCount(), this);
selOpts.Setup(IDS_CONVDISK_TITLE, IDS_CONVDISK_OK, IDS_CONVDISK_SELECTED_COUNT,
IDS_CONVDISK_SELECTED_COUNTS_FMT, IDS_CONVDISK_ALL_FILES);
if (fpContentList->GetSelectedCount() > 0)
selOpts.fFilesToAction = UseSelectionDialog::kActionSelection;
else
selOpts.fFilesToAction = UseSelectionDialog::kActionAll;
//selOpts.fAllowLower =
// fPreferences.GetPrefBool(kPrConvDiskAllowLower);
//selOpts.fSparseAlloc =
// fPreferences.GetPrefBool(kPrConvDiskAllocSparse);
if (selOpts.DoModal() != IDOK) {
LOGI("ConvDisk cancelled");
return;
}
ASSERT(selOpts.fNumBlocks > 0);
//fPreferences.SetPrefBool(kPrConvDiskAllowLower,
// selOpts.fAllowLower != 0);
//fPreferences.SetPrefBool(kPrConvDiskAllocSparse,
// selOpts.fSparseAlloc != 0);
/*
* Create a "selection set" of data forks, resource forks, and
* disk images. We don't want comment threads, but we can ignore
* them later.
*/
SelectionSet selSet;
int threadMask = GenericEntry::kAnyThread;
if (selOpts.fFilesToAction == UseSelectionDialog::kActionSelection) {
selSet.CreateFromSelection(fpContentList, threadMask);
} else {
selSet.CreateFromAll(fpContentList, threadMask);
}
//selSet.Dump();
if (selSet.GetNumEntries() == 0) {
/* should be impossible */
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
MessageBox(L"No files matched the selection criteria.",
L"No match", MB_OK|MB_ICONEXCLAMATION);
return;
}
XferFileOptions xferOpts;
//xferOpts.fAllowLowerCase =
// fPreferences.GetPrefBool(kPrProDOSAllowLower) != 0;
//xferOpts.fUseSparseBlocks =
// fPreferences.GetPrefBool(kPrProDOSUseSparse) != 0;
LOGI("New volume name will be '%ls'", (LPCWSTR) selOpts.fVolName);
/*
* Create a new disk image.
*/
CString filename, saveFolder, errStr;
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
CFileDialog dlg(FALSE, L"po", NULL,
OFN_OVERWRITEPROMPT|OFN_NOREADONLYRETURN|OFN_HIDEREADONLY,
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
L"Disk Images (*.po)|*.po||", this);
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
dlg.m_ofn.lpstrTitle = L"New Disk Image (.PO)";
dlg.m_ofn.lpstrInitialDir = fPreferences.GetPrefString(kPrOpenArchiveFolder);
if (dlg.DoModal() != IDOK) {
LOGI(" User cancelled xfer from image create dialog");
return;
}
saveFolder = dlg.m_ofn.lpstrFile;
saveFolder = saveFolder.Left(dlg.m_ofn.nFileOffset);
fPreferences.SetPrefString(kPrOpenArchiveFolder, saveFolder);
filename = dlg.GetPathName();
LOGI(" Will xfer to file '%ls'", (LPCWSTR) filename);
/* remove file if it already exists */
CString errMsg;
errMsg = RemoveFile(filename);
if (!errMsg.IsEmpty()) {
ShowFailureMsg(this, errMsg, IDS_FAILED);
return;
}
DiskArchive::NewOptions options;
memset(&options, 0, sizeof(options));
options.base.format = DiskImg::kFormatProDOS;
options.base.sectorOrder = DiskImg::kSectorOrderProDOS;
options.prodos.volName = selOpts.fVolName;
options.prodos.numBlocks = selOpts.fNumBlocks;
xferOpts.fTarget = new DiskArchive;
errStr = xferOpts.fTarget->New(filename, &options);
if (!errStr.IsEmpty()) {
ShowFailureMsg(this, errStr, IDS_FAILED);
delete xferOpts.fTarget;
return;
}
/*
* Set up the progress window.
*/
GenericArchive::XferStatus result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionConvDisk, this);
result = fpOpenArchive->XferSelection(fpActionProgress, &selSet,
fpActionProgress, &xferOpts);
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
if (result == GenericArchive::kXferOK)
SuccessBeep();
/* clean up */
delete xferOpts.fTarget;
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsConvDisk(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
/* right now, only NufxArchive has the Xfer stuff implemented */
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL &&
fpContentList->GetItemCount() > 0 &&
fpOpenArchive->GetArchiveKind() == GenericArchive::kArchiveNuFX);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Convert disk image to NuFX file archive
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsConvFile(void)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
ASSERT(fpContentList != NULL);
ASSERT(fpOpenArchive != NULL);
/*
* Ask the user about various options.
*/
ConvFileOptionsDialog selOpts(fpContentList->GetSelectedCount(), this);
selOpts.Setup(IDS_CONVFILE_TITLE, IDS_CONVFILE_OK, IDS_CONVFILE_SELECTED_COUNT,
IDS_CONVFILE_SELECTED_COUNTS_FMT, IDS_CONVFILE_ALL_FILES);
if (fpContentList->GetSelectedCount() > 0)
selOpts.fFilesToAction = UseSelectionDialog::kActionSelection;
else
selOpts.fFilesToAction = UseSelectionDialog::kActionAll;
//selOpts.fConvDOSText =
// fPreferences.GetPrefBool(kPrConvFileConvDOSText);
//selOpts.fConvPascalText =
// fPreferences.GetPrefBool(kPrConvFileConvPascalText);
selOpts.fPreserveEmptyFolders =
fPreferences.GetPrefBool(kPrConvFileEmptyFolders);
if (selOpts.DoModal() != IDOK) {
LOGI("ConvFile cancelled");
return;
}
//fPreferences.SetPrefBool(kPrConvFileConvDOSText,
// selOpts.fConvDOSText != 0);
//fPreferences.SetPrefBool(kPrConvFileConvPascalText,
// selOpts.fConvPascalText != 0);
fPreferences.SetPrefBool(kPrConvFileEmptyFolders,
selOpts.fPreserveEmptyFolders != 0);
/*
* Create a "selection set" of data forks, resource forks, and
* directories. There are no comments or disk images on a disk image,
* so we just request "any" thread.
*
* We only need to explicitly include directories if "preserve
* empty folders" is set.
*/
SelectionSet selSet;
int threadMask = GenericEntry::kAnyThread;
if (selOpts.fPreserveEmptyFolders)
threadMask |= GenericEntry::kAllowDirectory;
if (selOpts.fFilesToAction == UseSelectionDialog::kActionSelection) {
selSet.CreateFromSelection(fpContentList, threadMask);
} else {
selSet.CreateFromAll(fpContentList, threadMask);
}
//selSet.Dump();
if (selSet.GetNumEntries() == 0) {
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
MessageBox(L"No files matched the selection criteria.",
L"No match", MB_OK|MB_ICONEXCLAMATION);
return;
}
XferFileOptions xferOpts;
//xferOpts.fConvDOSText = (selOpts.fConvDOSText != 0);
//xferOpts.fConvPascalText = (selOpts.fConvPascalText != 0);
xferOpts.fPreserveEmptyFolders = (selOpts.fPreserveEmptyFolders != 0);
/*
* Create a new NuFX archive.
*/
CString filename, saveFolder, errStr;
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
CFileDialog dlg(FALSE, L"shk", NULL,
OFN_OVERWRITEPROMPT|OFN_NOREADONLYRETURN|OFN_HIDEREADONLY,
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
L"ShrinkIt Archives (*.shk)|*.shk||", this);
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
dlg.m_ofn.lpstrTitle = L"New Archive";
dlg.m_ofn.lpstrInitialDir = fPreferences.GetPrefString(kPrOpenArchiveFolder);
if (dlg.DoModal() != IDOK) {
LOGI(" User cancelled xfer from archive create dialog");
return;
}
saveFolder = dlg.m_ofn.lpstrFile;
saveFolder = saveFolder.Left(dlg.m_ofn.nFileOffset);
fPreferences.SetPrefString(kPrOpenArchiveFolder, saveFolder);
filename = dlg.GetPathName();
LOGD(" Will xfer to file '%ls'", (LPCWSTR) filename);
/* remove file if it already exists */
CString errMsg;
errMsg = RemoveFile(filename);
if (!errMsg.IsEmpty()) {
ShowFailureMsg(this, errMsg, IDS_FAILED);
return;
}
xferOpts.fTarget = new NufxArchive;
2014-11-18 05:13:13 +00:00
errStr = xferOpts.fTarget->New(filename, NULL);
if (!errStr.IsEmpty()) {
ShowFailureMsg(this, errStr, IDS_FAILED);
delete xferOpts.fTarget;
return;
}
/*
* Set up the progress window.
*/
GenericArchive::XferStatus result;
fpActionProgress = new ActionProgressDialog;
fpActionProgress->Create(ActionProgressDialog::kActionConvFile, this);
result = fpOpenArchive->XferSelection(fpActionProgress, &selSet,
fpActionProgress, &xferOpts);
fpActionProgress->Cleanup(this);
2014-11-18 05:13:13 +00:00
fpActionProgress = NULL;
if (result == GenericArchive::kXferOK)
SuccessBeep();
/* clean up */
delete xferOpts.fTarget;
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsConvFile(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL &&
fpContentList->GetItemCount() > 0 &&
fpOpenArchive->GetArchiveKind() == GenericArchive::kArchiveDiskImage);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Cassette WAV conversions
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsConvToWav(void)
2007-03-27 17:47:10 +00:00
{
// do this someday
LOGI("Convert TO wav");
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsConvToWav(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
BOOL enable = false;
2014-11-18 05:13:13 +00:00
if (fpContentList != NULL && fpContentList->GetSelectedCount() == 1) {
/* only BAS, INT, and BIN shorter than 64K */
GenericEntry* pEntry = GetSelectedItem(fpContentList);
if ((pEntry->GetFileType() == kFileTypeBAS ||
pEntry->GetFileType() == kFileTypeINT ||
pEntry->GetFileType() == kFileTypeBIN) &&
pEntry->GetDataForkLen() < 65536 &&
pEntry->GetRecordKind() == GenericEntry::kRecordKindFile)
{
enable = true;
}
}
pCmdUI->Enable(enable);
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnActionsConvFromWav(void)
2007-03-27 17:47:10 +00:00
{
CassetteDialog dlg;
CString fileName, saveFolder;
2007-03-27 17:47:10 +00:00
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
CFileDialog fileDlg(TRUE, L"wav", NULL, OFN_FILEMUSTEXIST|OFN_HIDEREADONLY,
L"Sound Files (*.wav)|*.wav||", this);
fileDlg.m_ofn.lpstrTitle = L"Open Sound File";
fileDlg.m_ofn.lpstrInitialDir = fPreferences.GetPrefString(kPrOpenWAVFolder);
2007-03-27 17:47:10 +00:00
if (fileDlg.DoModal() != IDOK)
goto bail;
2007-03-27 17:47:10 +00:00
saveFolder = fileDlg.m_ofn.lpstrFile;
saveFolder = saveFolder.Left(fileDlg.m_ofn.nFileOffset);
fPreferences.SetPrefString(kPrOpenWAVFolder, saveFolder);
2007-03-27 17:47:10 +00:00
fileName = fileDlg.GetPathName();
LOGI("Opening WAV file '%ls'", (LPCWSTR) fileName);
2007-03-27 17:47:10 +00:00
dlg.fFileName = fileName;
// pass in fpOpenArchive?
2007-03-27 17:47:10 +00:00
dlg.DoModal();
if (dlg.IsDirty()) {
2014-11-18 05:13:13 +00:00
assert(fpContentList != NULL);
fpContentList->Reload();
}
2007-03-27 17:47:10 +00:00
bail:
return;
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsConvFromWav(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly());
2007-03-27 17:47:10 +00:00
}
/*static*/ bool MainWindow::SaveToArchive(GenericArchive::LocalFileDetails* pDetails,
const uint8_t* dataBufIn, long dataLen,
const uint8_t* rsrcBufIn, long rsrcLen,
CString* pErrMsg, CWnd* pDialog)
2007-03-27 17:47:10 +00:00
{
MainWindow* pMain = GET_MAIN_WINDOW();
GenericArchive* pArchive = pMain->GetOpenArchive();
2014-11-18 05:13:13 +00:00
DiskImgLib::A2File* pTargetSubdir = NULL;
XferFileOptions xferOpts;
CString storagePrefix;
uint8_t* dataBuf = NULL;
uint8_t* rsrcBuf = NULL;
2014-11-18 05:13:13 +00:00
ASSERT(pArchive != NULL);
ASSERT(pErrMsg->IsEmpty());
/*
* Make a copy of the data for XferFile.
*/
if (dataLen >= 0) {
if (dataLen == 0)
dataBuf = new unsigned char[1];
else
dataBuf = new unsigned char[dataLen];
2014-11-18 05:13:13 +00:00
if (dataBuf == NULL) {
pErrMsg->Format(L"Unable to allocate %ld bytes", dataLen);
goto bail;
}
memcpy(dataBuf, dataBufIn, dataLen);
}
if (rsrcLen >= 0) {
assert(false);
}
/*
* Figure out where we want to put the files. For a disk archive
* this can be complicated.
*
* The target DiskFS (which could be a sub-volume) gets tucked into
* the xferOpts.
*/
if (pArchive->GetArchiveKind() == GenericArchive::kArchiveDiskImage) {
if (!pMain->ChooseAddTarget(&pTargetSubdir, &xferOpts.fpTargetFS))
goto bail;
} else if (pArchive->GetArchiveKind() == GenericArchive::kArchiveNuFX) {
// Always use ':' separator for SHK; this is a matter of
// convenience, so they can specify a full path.
//details.storageName.Replace(':', '_');
pDetails->SetFssep(':');
}
2014-11-18 05:13:13 +00:00
if (pTargetSubdir != NULL) {
storagePrefix = pTargetSubdir->GetPathName();
LOGD(" using storagePrefix '%ls'", (LPCWSTR) storagePrefix);
}
if (!storagePrefix.IsEmpty()) {
CString tmpStr, tmpFileName;
tmpFileName = pDetails->GetStrippedLocalPathName();
tmpFileName.Replace(':', '_'); // strip any ':'s in the name
pDetails->SetFssep(':');
tmpStr = storagePrefix;
tmpStr += ':';
tmpStr += tmpFileName;
pDetails->SetStrippedLocalPathName(tmpStr);
}
/*
* Handle the transfer.
*
* On success, XferFile will null out our dataBuf and rsrcBuf pointers.
*/
pArchive->XferPrepare(&xferOpts);
*pErrMsg = pArchive->XferFile(pDetails, &dataBuf, dataLen,
&rsrcBuf, rsrcLen);
delete[] dataBuf;
delete[] rsrcBuf;
if (pErrMsg->IsEmpty())
pArchive->XferFinish(pDialog);
else
pArchive->XferAbort(pDialog);
2007-03-27 17:47:10 +00:00
bail:
return (pErrMsg->IsEmpty() != 0);
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Import BASIC programs from a text file
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
void MainWindow::OnActionsImportBAS(void)
2007-03-27 17:47:10 +00:00
{
ImportBASDialog dlg;
CString fileName, saveFolder;
2007-03-27 17:47:10 +00:00
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
CFileDialog fileDlg(TRUE, L"txt", NULL, OFN_FILEMUSTEXIST | OFN_HIDEREADONLY,
L"Text files (*.txt)|*.txt||", this);
fileDlg.m_ofn.lpstrTitle = L"Open Text File";
fileDlg.m_ofn.lpstrInitialDir = fPreferences.GetPrefString(kPrAddFileFolder);
2007-03-27 17:47:10 +00:00
if (fileDlg.DoModal() != IDOK)
goto bail;
2007-03-27 17:47:10 +00:00
saveFolder = fileDlg.m_ofn.lpstrFile;
saveFolder = saveFolder.Left(fileDlg.m_ofn.nFileOffset);
fPreferences.SetPrefString(kPrAddFileFolder, saveFolder);
2007-03-27 17:47:10 +00:00
fileName = fileDlg.GetPathName();
LOGI("Opening TXT file '%ls'", (LPCWSTR) fileName);
2007-03-27 17:47:10 +00:00
dlg.SetFileName(fileName);
// pass in fpOpenArchive?
2007-03-27 17:47:10 +00:00
dlg.DoModal();
if (dlg.IsDirty()) {
2014-11-18 05:13:13 +00:00
assert(fpContentList != NULL);
fpContentList->Reload();
}
2007-03-27 17:47:10 +00:00
bail:
return;
2007-03-27 17:47:10 +00:00
}
void MainWindow::OnUpdateActionsImportBAS(CCmdUI* pCmdUI)
2007-03-27 17:47:10 +00:00
{
2014-11-18 05:13:13 +00:00
pCmdUI->Enable(fpContentList != NULL && !fpOpenArchive->IsReadOnly());
2007-03-27 17:47:10 +00:00
}
/*
* ==========================================================================
* Multiple file handling
2007-03-27 17:47:10 +00:00
* ==========================================================================
*/
int MainWindow::GetFileParts(const GenericEntry* pEntry,
ReformatHolder** ppHolder) const
2007-03-27 17:47:10 +00:00
{
ReformatHolder* pHolder = new ReformatHolder;
CString errMsg;
2007-03-27 17:47:10 +00:00
2014-11-18 05:13:13 +00:00
if (pHolder == NULL)
return -1;
2007-03-27 17:47:10 +00:00
if (pEntry->GetHasDataFork())
GetFilePart(pEntry, GenericEntry::kDataThread, pHolder);
if (pEntry->GetHasRsrcFork())
GetFilePart(pEntry, GenericEntry::kRsrcThread, pHolder);
if (pEntry->GetHasComment())
GetFilePart(pEntry, GenericEntry::kCommentThread, pHolder);
if (pEntry->GetHasDiskImage())
GetFilePart(pEntry, GenericEntry::kDiskImageThread, pHolder);
2007-03-27 17:47:10 +00:00
*ppHolder = pHolder;
2007-03-27 17:47:10 +00:00
return 0;
2007-03-27 17:47:10 +00:00
}
void MainWindow::GetFilePart(const GenericEntry* pEntry, int whichThread,
ReformatHolder* pHolder) const
2007-03-27 17:47:10 +00:00
{
CString errMsg;
ReformatHolder::ReformatPart part;
2014-11-18 05:13:13 +00:00
char* buf = NULL;
long len = 0;
di_off_t threadLen;
int result;
switch (whichThread) {
case GenericEntry::kDataThread:
part = ReformatHolder::kPartData;
threadLen = pEntry->GetDataForkLen();
break;
case GenericEntry::kRsrcThread:
part = ReformatHolder::kPartRsrc;
threadLen = pEntry->GetRsrcForkLen();
break;
case GenericEntry::kCommentThread:
part = ReformatHolder::kPartCmmt;
threadLen = -1; // no comment len getter; assume it's small
break;
case GenericEntry::kDiskImageThread:
part = ReformatHolder::kPartData; // put disks into data thread
threadLen = pEntry->GetDataForkLen();
break;
default:
ASSERT(false);
return;
}
if (threadLen > fPreferences.GetPrefLong(kPrMaxViewFileSize)) {
errMsg.Format(
L"[File size (%I64d KBytes) exceeds file viewer maximum (%ld KBytes). "
L"The limit can be adjusted in the file viewer preferences.]\n",
((LONGLONG) threadLen + 1023) / 1024,
(fPreferences.GetPrefLong(kPrMaxViewFileSize) + 1023) / 1024);
pHolder->SetErrorMsg(part, errMsg);
goto bail;
}
result = pEntry->ExtractThreadToBuffer(whichThread, &buf, &len, &errMsg);
if (result == IDOK) {
/* on success, ETTB guarantees a buffer, even for zero-len file */
2014-11-18 05:13:13 +00:00
ASSERT(buf != NULL);
pHolder->SetSourceBuf(part, (unsigned char*) buf, len);
} else if (result == IDCANCEL) {
/* not expected */
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
errMsg = L"Cancelled!";
pHolder->SetErrorMsg(part, errMsg);
2014-11-18 05:13:13 +00:00
ASSERT(buf == NULL);
} else {
/* transfer error message to ReformatHolder buffer */
LOGI("Got error message from ExtractThread: '%ls'",
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
(LPCWSTR) errMsg);
pHolder->SetErrorMsg(part, errMsg);
2014-11-18 05:13:13 +00:00
ASSERT(buf == NULL);
}
2007-03-27 17:47:10 +00:00
bail:
return;
2007-03-27 17:47:10 +00:00
}