pjs/xpcom/io/nsFastLoadService.cpp

541 строка
15 KiB
C++
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/* ***** BEGIN LICENSE BLOCK *****
* Version: NPL 1.1/GPL 2.0/LGPL 2.1
*
* The contents of this file are subject to the Netscape Public License
* Version 1.1 (the "License"); you may not use this file except in
* compliance with the License. You may obtain a copy of the License at
* http://www.mozilla.org/NPL/
*
* Software distributed under the License is distributed on an "AS IS" basis,
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
* for the specific language governing rights and limitations under the
* License.
*
* The Original Code is Mozilla FastLoad code.
*
* The Initial Developer of the Original Code is
* Netscape Communications Corporation.
* Portions created by the Initial Developer are Copyright (C) 2001
* the Initial Developer. All Rights Reserved.
*
* Contributor(s):
* Brendan Eich <brendan@mozilla.org> (original author)
*
* Alternatively, the contents of this file may be used under the terms of
* either the GNU General Public License Version 2 or later (the "GPL"), or
* the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
* in which case the provisions of the GPL or the LGPL are applicable instead
* of those above. If you wish to allow use of your version of this file only
* under the terms of either the GPL or the LGPL, and not to allow others to
* use your version of this file under the terms of the NPL, indicate your
* decision by deleting the provisions above and replace them with the notice
* and other provisions required by the GPL or the LGPL. If you do not delete
* the provisions above, a recipient may use your version of this file under
* the terms of any one of the NPL, the GPL or the LGPL.
*
* ***** END LICENSE BLOCK ***** */
#include "prtypes.h"
#include "prio.h"
#include "prtime.h"
#include "pldhash.h"
#include "nsAppDirectoryServiceDefs.h"
#include "nsAutoLock.h"
#include "nsCOMPtr.h"
#include "nsFastLoadFile.h"
#include "nsFastLoadPtr.h"
#include "nsFastLoadService.h"
#include "nsString.h"
#include "nsIComponentManager.h"
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
#include "nsIEnumerator.h"
#include "nsIFastLoadFileControl.h"
#include "nsIFile.h"
#include "nsIObjectInputStream.h"
#include "nsIObjectOutputStream.h"
#include "nsISeekableStream.h"
#include "nsISupports.h"
PR_IMPLEMENT_DATA(nsIFastLoadService*) gFastLoadService_ = nsnull;
NS_IMPL_THREADSAFE_ISUPPORTS1(nsFastLoadService, nsIFastLoadService)
nsFastLoadService::nsFastLoadService()
: mLock(nsnull),
mFastLoadPtrMap(nsnull),
mDirection(0)
{
NS_INIT_REFCNT();
NS_ASSERTION(gFastLoadService_ == nsnull, "double FastLoadService init?");
gFastLoadService_ = this;
}
nsFastLoadService::~nsFastLoadService()
{
gFastLoadService_ = nsnull;
if (mInputStream)
mInputStream->Close();
if (mOutputStream)
mOutputStream->Close();
if (mFastLoadPtrMap)
PL_DHashTableDestroy(mFastLoadPtrMap);
if (mLock)
PR_DestroyLock(mLock);
}
NS_IMETHODIMP
nsFastLoadService::Create(nsISupports *aOuter, REFNSIID aIID, void **aResult)
{
*aResult = nsnull;
if (aOuter)
return NS_ERROR_NO_AGGREGATION;
nsFastLoadService* fastLoadService = new nsFastLoadService();
if (!fastLoadService)
return NS_ERROR_OUT_OF_MEMORY;
fastLoadService->mLock = PR_NewLock();
if (!fastLoadService->mLock) {
delete fastLoadService;
return NS_ERROR_OUT_OF_MEMORY;
}
NS_ADDREF(fastLoadService);
nsresult rv = fastLoadService->QueryInterface(aIID, aResult);
NS_RELEASE(fastLoadService);
return rv;
}
#if defined XP_MAC
// Mac format: "<Basename> FastLoad File" with <basename> capitalized.
# include "nsCRT.h"
# define MASSAGE_BASENAME(bn) (bn.SetCharAt(nsCRT::ToUpper(bn.CharAt(0)), 0))
# define PLATFORM_FASL_SUFFIX " FastLoad File"
#elif defined(XP_UNIX) || defined(XP_BEOS)
// Unix format: "<basename>.mfasl".
# define MASSAGE_BASENAME(bn) /* nothing */
# define PLATFORM_FASL_SUFFIX ".mfasl"
#elif defined(XP_WIN) || defined(XP_OS2)
// Windows format: "<basename>.mfl".
# define MASSAGE_BASENAME(bn) /* nothing */
# define PLATFORM_FASL_SUFFIX ".mfl"
#endif
nsresult
nsFastLoadService::NewFastLoadFile(const char* aBaseName, nsIFile* *aResult)
{
nsresult rv;
nsCOMPtr<nsIFile> file;
rv = NS_GetSpecialDirectory(NS_APP_USER_PROFILE_50_DIR,
getter_AddRefs(file));
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
nsCAutoString name(aBaseName);
MASSAGE_BASENAME(name);
name += PLATFORM_FASL_SUFFIX;
rv = file->Append(name.get());
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
*aResult = file;
NS_ADDREF(*aResult);
return NS_OK;
}
NS_IMETHODIMP
nsFastLoadService::NewInputStream(nsIInputStream* aSrcStream,
nsIObjectInputStream* *aResult)
{
nsAutoLock lock(mLock);
nsCOMPtr<nsIObjectInputStream> stream;
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsresult rv = NS_NewFastLoadFileReader(getter_AddRefs(stream), aSrcStream);
if (NS_FAILED(rv))
return rv;
*aResult = stream;
NS_ADDREF(*aResult);
return NS_OK;
}
NS_IMETHODIMP
nsFastLoadService::NewOutputStream(nsIOutputStream* aDestStream,
nsIObjectOutputStream* *aResult)
{
nsAutoLock lock(mLock);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
return NS_NewFastLoadFileWriter(aResult, aDestStream, mFileIO);
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::GetInputStream(nsIObjectInputStream* *aResult)
{
NS_IF_ADDREF(*aResult = mInputStream);
return NS_OK;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::SetInputStream(nsIObjectInputStream* aStream)
{
nsAutoLock lock(mLock);
mInputStream = aStream;
return NS_OK;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::GetOutputStream(nsIObjectOutputStream* *aResult)
{
NS_IF_ADDREF(*aResult = mOutputStream);
return NS_OK;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::SetOutputStream(nsIObjectOutputStream* aStream)
{
nsAutoLock lock(mLock);
mOutputStream = aStream;
return NS_OK;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::GetFileIO(nsIFastLoadFileIO* *aResult)
{
NS_IF_ADDREF(*aResult = mFileIO);
return NS_OK;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::SetFileIO(nsIFastLoadFileIO* aFileIO)
{
nsAutoLock lock(mLock);
mFileIO = aFileIO;
return NS_OK;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::GetDirection(PRInt32 *aResult)
{
*aResult = mDirection;
return NS_OK;
}
NS_IMETHODIMP
nsFastLoadService::StartMuxedDocument(nsISupports* aURI, const char* aURISpec,
PRInt32 aDirectionFlags)
{
nsresult rv = NS_ERROR_NOT_AVAILABLE;
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsCOMPtr<nsIFastLoadFileControl> control;
nsAutoLock lock(mLock);
// Try for an input stream first, in case aURISpec's data is multiplexed
// in the current FastLoad file.
if ((aDirectionFlags & NS_FASTLOAD_READ) && mInputStream) {
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
control = do_QueryInterface(mInputStream);
if (control) {
// If aURISpec is not in the multiplex, control->StartMuxedDocument
// will return NS_ERROR_NOT_AVAILABLE.
rv = control->StartMuxedDocument(aURI, aURISpec);
if (NS_SUCCEEDED(rv) || rv != NS_ERROR_NOT_AVAILABLE)
return rv;
// Ok, aURISpec is not in the existing mux. If we have no output
// stream yet, wrap the reader with a FastLoad file updater.
if (!mOutputStream && mFileIO) {
nsCOMPtr<nsIOutputStream> output;
rv = mFileIO->GetOutputStream(getter_AddRefs(output));
if (NS_FAILED(rv))
return rv;
// NB: mInputStream must be an nsFastLoadFileReader!
rv = NS_NewFastLoadFileUpdater(getter_AddRefs(mOutputStream),
output,
mInputStream);
if (NS_FAILED(rv))
return rv;
}
if (aDirectionFlags == NS_FASTLOAD_READ) {
// Tell our caller to re-start multiplexing, rather than attempt
// to select and deserialize now.
return NS_ERROR_NOT_AVAILABLE;
}
}
- Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Writer or Updater must close after the Reader, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-17 13:52:55 +04:00
}
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if ((aDirectionFlags & NS_FASTLOAD_WRITE) && mOutputStream) {
control = do_QueryInterface(mOutputStream);
if (control)
rv = control->StartMuxedDocument(aURI, aURISpec);
}
return rv;
}
NS_IMETHODIMP
nsFastLoadService::SelectMuxedDocument(nsISupports* aURI)
{
nsresult rv = NS_ERROR_NOT_AVAILABLE;
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsCOMPtr<nsIFastLoadFileControl> control;
nsAutoLock lock(mLock);
// Try to select the reader, if any; then only if the URI was not in the
// file already, select the writer/updater.
if (mInputStream) {
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
control = do_QueryInterface(mInputStream);
if (control) {
rv = control->SelectMuxedDocument(aURI);
if (NS_SUCCEEDED(rv))
mDirection = NS_FASTLOAD_READ;
}
}
if (rv == NS_ERROR_NOT_AVAILABLE && mOutputStream) {
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
control = do_QueryInterface(mOutputStream);
if (control) {
rv = control->SelectMuxedDocument(aURI);
if (NS_SUCCEEDED(rv))
mDirection = NS_FASTLOAD_WRITE;
}
}
return rv;
}
NS_IMETHODIMP
nsFastLoadService::EndMuxedDocument(nsISupports* aURI)
{
nsresult rv = NS_ERROR_NOT_AVAILABLE;
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsCOMPtr<nsIFastLoadFileControl> control;
nsAutoLock lock(mLock);
// Try to end the document identified by aURI in the reader, if any; then
// only if the URI was not in the file already, end the writer/updater.
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (mInputStream) {
control = do_QueryInterface(mInputStream);
if (control)
rv = control->EndMuxedDocument(aURI);
}
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (rv == NS_ERROR_NOT_AVAILABLE && mOutputStream) {
control = do_QueryInterface(mOutputStream);
if (control)
rv = control->EndMuxedDocument(aURI);
}
mDirection = 0;
return rv;
}
NS_IMETHODIMP
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsFastLoadService::AddDependency(nsIFile* aFile)
{
nsAutoLock lock(mLock);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
nsCOMPtr<nsIFastLoadWriteControl> control(do_QueryInterface(mOutputStream));
if (!control)
return NS_ERROR_NOT_AVAILABLE;
return control->AddDependency(aFile);
}
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
NS_IMETHODIMP
nsFastLoadService::ComputeChecksum(nsIFile* aFile,
nsIFastLoadReadControl* aControl,
PRUint32 *aChecksum)
{
nsXPIDLCString path;
nsresult rv = aFile->GetPath(getter_Copies(path));
if (NS_FAILED(rv))
return rv;
nsCStringKey key(path);
PRUint32 checksum = NS_PTR_TO_INT32(mChecksumTable.Get(&key));
if (checksum) {
*aChecksum = checksum;
return NS_OK;
}
- Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Writer or Updater must close after the Reader, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-17 13:52:55 +04:00
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
rv = aControl->ComputeChecksum(&checksum);
if (NS_FAILED(rv))
return rv;
mChecksumTable.Put(&key, NS_INT32_TO_PTR(checksum));
*aChecksum = checksum;
return NS_OK;
}
NS_IMETHODIMP
nsFastLoadService::CacheChecksum(nsIFile* aFile, nsIObjectOutputStream *aStream)
{
nsCOMPtr<nsIFastLoadFileControl> control(do_QueryInterface(aStream));
if (!control)
return NS_ERROR_FAILURE;
PRUint32 checksum;
nsresult rv = control->GetChecksum(&checksum);
if (NS_FAILED(rv))
return rv;
nsXPIDLCString path;
rv = aFile->GetPath(getter_Copies(path));
if (NS_FAILED(rv))
return rv;
nsCStringKey key(path);
mChecksumTable.Put(&key, NS_INT32_TO_PTR(checksum));
return NS_OK;
}
struct nsFastLoadPtrEntry : public PLDHashEntryHdr {
nsISupports** mPtrAddr; // key, must come first for PL_DHashGetStubOps
PRUint32 mOffset;
};
NS_IMETHODIMP
nsFastLoadService::GetFastLoadReferent(nsISupports* *aPtrAddr)
{
NS_ASSERTION(*aPtrAddr == nsnull,
"aPtrAddr doesn't point to null nsFastLoadPtr<T>::mRawAddr?");
nsAutoLock lock(mLock);
if (!mFastLoadPtrMap || !mInputStream)
return NS_OK;
nsFastLoadPtrEntry* entry =
NS_STATIC_CAST(nsFastLoadPtrEntry*,
PL_DHashTableOperate(mFastLoadPtrMap, aPtrAddr,
PL_DHASH_LOOKUP));
if (PL_DHASH_ENTRY_IS_FREE(entry))
return NS_OK;
nsresult rv;
nsCOMPtr<nsISeekableStream> seekable(do_QueryInterface(mInputStream));
rv = seekable->Seek(nsISeekableStream::NS_SEEK_SET, entry->mOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = mInputStream->ReadObject(PR_TRUE, aPtrAddr);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
// Shrink the table if half the entries are removed sentinels.
PRUint32 size = PL_DHASH_TABLE_SIZE(mFastLoadPtrMap);
if (mFastLoadPtrMap->removedCount >= (size >> 2))
PL_DHashTableOperate(mFastLoadPtrMap, entry, PL_DHASH_REMOVE);
else
PL_DHashTableRawRemove(mFastLoadPtrMap, entry);
return NS_OK;
}
NS_IMETHODIMP
nsFastLoadService::ReadFastLoadPtr(nsIObjectInputStream* aInputStream,
nsISupports* *aPtrAddr)
{
// nsFastLoadPtrs self-construct to null, so if we have a non-null value
// in our inout parameter, we must have been read already, alright!
if (*aPtrAddr)
return NS_OK;
nsresult rv;
PRUint32 nextOffset;
nsAutoLock lock(mLock);
rv = aInputStream->Read32(&nextOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
nsCOMPtr<nsISeekableStream> seekable(do_QueryInterface(aInputStream));
if (!seekable)
return NS_ERROR_FAILURE;
PRUint32 thisOffset;
rv = seekable->Tell(&thisOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = seekable->Seek(nsISeekableStream::NS_SEEK_SET, nextOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
if (!mFastLoadPtrMap) {
mFastLoadPtrMap = PL_NewDHashTable(PL_DHashGetStubOps(), this,
sizeof(nsFastLoadPtrEntry),
PL_DHASH_MIN_SIZE);
if (!mFastLoadPtrMap)
return NS_ERROR_OUT_OF_MEMORY;
}
nsFastLoadPtrEntry* entry =
NS_STATIC_CAST(nsFastLoadPtrEntry*,
PL_DHashTableOperate(mFastLoadPtrMap, aPtrAddr,
PL_DHASH_ADD));
NS_ASSERTION(entry->mPtrAddr == nsnull, "duplicate nsFastLoadPtr?!");
entry->mPtrAddr = aPtrAddr;
entry->mOffset = thisOffset;
return NS_OK;
}
NS_IMETHODIMP
nsFastLoadService::WriteFastLoadPtr(nsIObjectOutputStream* aOutputStream,
nsISupports* aObject)
{
NS_ASSERTION(aObject != nsnull, "writing an unread nsFastLoadPtr?!");
if (!aObject)
return NS_ERROR_UNEXPECTED;
nsresult rv;
nsAutoLock lock(mLock); // serialize writes to aOutputStream
nsCOMPtr<nsISeekableStream> seekable(do_QueryInterface(aOutputStream));
if (!seekable)
return NS_ERROR_FAILURE;
PRUint32 saveOffset;
rv = seekable->Tell(&saveOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = aOutputStream->Write32(0); // nextOffset placeholder
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = aOutputStream->WriteObject(aObject, PR_TRUE);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
PRUint32 nextOffset;
rv = seekable->Tell(&nextOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = seekable->Seek(nsISeekableStream::NS_SEEK_SET, saveOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = aOutputStream->Write32(nextOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
rv = seekable->Seek(nsISeekableStream::NS_SEEK_SET, nextOffset);
Bit checkin for bug 68045, r/sr=waterson&shaver, second attempt. It all works for me on optimized and debug gcc2.96, rh7.1. - Better failure codes from nsXULPrototypeScript::Deserialize. - Call nsXULDocument::AbortFastLoads after nsXULPrototypeScript::Serialize failure, instead of just nulling the FastLoad service's output stream. - Expose nsXULDocument::AbortFastLoads via nsIXULPrototypeCache, for use from nsChromeProtocolHandler.cpp. AbortFastLoads flushes the XUL cache now, for good measure. - The needless "Current" adjective in nsIFastLoadService attribute and method names is no more. - Add a do_GetFastLoadService() helper, to use CID instead of contractid, and to let the compiler consolidate the static inline CID. - Add "nglayout.debug.checksum_xul_fastload_file" pref so people can do without the checksum verification step when reading a FastLoad file. - Verify the FastLoad file checksum, by default. Also, cache it in the FastLoad service so we don't recompute it when re-opening the FastLoad file as mailnews and other top-levels start up. Fill the checksum cache in EndFastLoad, when the last pseudo-concurrent top-level finishes loading. My hope to compute the checksum while writing the FastLoad file ran afoul of misordered writes. The old code to checksum the in-memory nsFastLoadHeader also was broken on little endian platforms. Now all checksumming is done via a separate read pass over the complete file, save for the header's checksum field, which is summed as if it contained zero. - Track and check FastLoad file dependencies. This required groveling with a bunch of Necko interfaces in nsChromeProtocolHandler::NewChannel -- read it and weep. Dependency checking, as well as checksum access and computation, use better-factored nsIFastLoad{File,Read,Write}Control interfaces. - nsBufferedStream::Seek wasn't flushing the buffer when seeking backward within the buffer, but it must, because mCursor bounds the amount to write if the buffer contains the end of file. - Add an unbufferedStream readonly attribute to nsIStreamBufferAccess, so we don't have to screw around with the bufferying layer when checksumming. Also implement nsIStreamBufferAccess in nsBufferedOutputStream. - nsISeekableOutputStream was bogus, based on a bad state I had put the nsBufferedOutputStream code in on its way from being completely broken when you seek backwards outside of the buffer. Removing this interface required using nsIFastLoadFileIO in nsFastLoadFileWriter, and it also required careful ordering of Close calls (the Reader must close after the Writer or Updater, so that the Reader's underlying, unbuffered input stream can be read by nsFastLoadFileWriter::Close to compute the checksum. - Miscellaneous tab/indentation, comment typo, bracing, if( => if ( style, nsnull vs. 0, useless variable elimination, tortured control flow, AutoString instead of String, and gratuitous ; after nsISupportsUtils.h macro call cleanups.
2001-08-22 00:51:34 +04:00
if (NS_FAILED(rv))
return rv;
return NS_OK;
}