pjs/modules/libjar/nsJARChannel.cpp

709 строки
18 KiB
C++
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*-
*
* The contents of this file are subject to the Netscape Public License
* Version 1.0 (the "NPL"); you may not use this file except in
* compliance with the NPL. You may obtain a copy of the NPL at
* http://www.mozilla.org/NPL/
*
* Software distributed under the NPL is distributed on an "AS IS" basis,
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the NPL
* for the specific language governing rights and limitations under the
* NPL.
*
* The Initial Developer of this code under the NPL is Netscape
* Communications Corporation. Portions created by Netscape are
* Copyright (C) 1998,2000 Netscape Communications Corporation. All Rights
* Reserved.
*
*/
#include "nsNetUtil.h"
#include "nsIComponentManager.h"
#include "nsIServiceManager.h"
#include "nsJARChannel.h"
#include "nsCRT.h"
#include "nsIFileTransportService.h"
#include "nsIURI.h"
#include "nsCExternalHandlerService.h"
#include "nsIMIMEService.h"
#include "nsAutoLock.h"
2000-01-25 00:28:28 +03:00
#include "nsIFileStreams.h"
#include "nsMimeTypes.h"
#include "nsScriptSecurityManager.h"
#include "nsIAggregatePrincipal.h"
#include "nsIProgressEventSink.h"
#include "nsXPIDLString.h"
#include "nsIJAR.h"
#include "nsIResProtocolHandler.h"
static NS_DEFINE_CID(kFileTransportServiceCID, NS_FILETRANSPORTSERVICE_CID);
static NS_DEFINE_CID(kZipReaderCID, NS_ZIPREADER_CID);
static NS_DEFINE_CID(kScriptSecurityManagerCID, NS_SCRIPTSECURITYMANAGER_CID);
#if defined(PR_LOGGING)
//
2001-08-02 05:35:28 +04:00
// Log module for JarChannel logging...
//
// To enable logging (see prlog.h for full details):
//
// set NSPR_LOG_MODULES=nsJarProtocol:5
// set NSPR_LOG_FILE=nspr.log
//
// this enables PR_LOG_DEBUG level information and places all output in
// the file nspr.log
//
PRLogModuleInfo* gJarProtocolLog = nsnull;
#endif /* PR_LOGGING */
////////////////////////////////////////////////////////////////////////////////
#define NS_DEFAULT_JAR_BUFFER_SEGMENT_SIZE (16*1024)
#define NS_DEFAULT_JAR_BUFFER_MAX_SIZE (256*1024)
nsJARChannel::nsJARChannel()
: mLoadFlags(LOAD_NORMAL),
mContentType(nsnull),
mContentLength(-1),
mJAREntry(nsnull),
mStatus(NS_OK),
mMonitor(nsnull)
{
NS_INIT_REFCNT();
#if defined(PR_LOGGING)
//
// Initialize the global PRLogModule for socket transport logging
// if necessary...
//
if (nsnull == gJarProtocolLog) {
gJarProtocolLog = PR_NewLogModule("nsJarProtocol");
}
#endif /* PR_LOGGING */
}
nsJARChannel::~nsJARChannel()
{
if (mContentType)
nsCRT::free(mContentType);
if (mJAREntry)
nsCRT::free(mJAREntry);
if (mMonitor)
PR_DestroyMonitor(mMonitor);
}
NS_IMPL_THREADSAFE_ISUPPORTS7(nsJARChannel,
nsIJARChannel,
nsIChannel,
nsIRequest,
nsIRequestObserver,
nsIStreamListener,
nsIStreamIO,
nsIDownloadObserver)
NS_METHOD
nsJARChannel::Create(nsISupports *aOuter, REFNSIID aIID, void **aResult)
{
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;
if (aOuter)
return NS_ERROR_NO_AGGREGATION;
nsJARChannel* jarChannel = new nsJARChannel();
if (jarChannel == nsnull)
return NS_ERROR_OUT_OF_MEMORY;
NS_ADDREF(jarChannel);
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 = jarChannel->QueryInterface(aIID, aResult);
NS_RELEASE(jarChannel);
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 rv;
}
nsresult
nsJARChannel::Init(nsIJARProtocolHandler* aHandler, nsIURI* uri)
{
nsresult rv;
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
mURI = do_QueryInterface(uri, &rv);
if (NS_FAILED(rv)) return rv;
mMonitor = PR_NewMonitor();
if (mMonitor == nsnull)
return NS_ERROR_OUT_OF_MEMORY;
mJARProtocolHandler = aHandler;
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_OK;
}
////////////////////////////////////////////////////////////////////////////////
// nsIRequest methods
NS_IMETHODIMP
nsJARChannel::GetName(PRUnichar* *result)
{
nsresult rv;
nsXPIDLCString urlStr;
rv = mURI->GetSpec(getter_Copies(urlStr));
if (NS_FAILED(rv)) return rv;
nsString name;
name.AppendWithConversion(urlStr);
*result = name.ToNewUnicode();
return *result ? NS_OK : NS_ERROR_OUT_OF_MEMORY;
}
NS_IMETHODIMP
nsJARChannel::IsPending(PRBool* result)
{
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_NOTREACHED("nsJARChannel::IsPending");
return NS_ERROR_NOT_IMPLEMENTED;
}
NS_IMETHODIMP
nsJARChannel::GetStatus(nsresult *status)
{
*status = mStatus;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::Cancel(nsresult status)
{
NS_ASSERTION(NS_FAILED(status), "shouldn't cancel with a success code");
nsresult rv = NS_OK;
nsAutoMonitor monitor(mMonitor);
if (mJarExtractionTransport) {
rv = mJarExtractionTransport->Cancel(status);
mJarExtractionTransport = nsnull;
}
mStatus = status;
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 rv;
}
NS_IMETHODIMP
nsJARChannel::Suspend()
{
nsresult rv = NS_OK;
nsAutoMonitor monitor(mMonitor);
if (mJarExtractionTransport) {
rv = mJarExtractionTransport->Suspend();
}
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 rv;
}
NS_IMETHODIMP
nsJARChannel::Resume()
{
nsresult rv = NS_OK;
nsAutoMonitor monitor(mMonitor);
if (mJarExtractionTransport) {
rv = mJarExtractionTransport->Resume();
}
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 rv;
}
////////////////////////////////////////////////////////////////////////////////
// nsIChannel methods
NS_IMETHODIMP
nsJARChannel::GetOriginalURI(nsIURI* *aOriginalURI)
{
if (mOriginalURI)
*aOriginalURI = mOriginalURI;
else
*aOriginalURI = NS_STATIC_CAST(nsIURI*, mURI);
NS_IF_ADDREF(*aOriginalURI);
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::SetOriginalURI(nsIURI* aOriginalURI)
{
mOriginalURI = aOriginalURI;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::GetURI(nsIURI* *aURI)
{
*aURI = mURI;
NS_ADDREF(*aURI);
return NS_OK;
}
nsresult
nsJARChannel::OpenJARElement()
{
nsresult rv;
nsAutoCMonitor mon(this);
rv = Open(nsnull, nsnull);
if (NS_SUCCEEDED(rv))
rv = GetInputStream(getter_AddRefs(mSynchronousInputStream));
mon.Notify(); // wake up nsIChannel::Open
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 rv;
}
NS_IMETHODIMP
nsJARChannel::Open(nsIInputStream* *result)
{
nsAutoCMonitor mon(this);
nsresult rv;
mSynchronousRead = PR_TRUE;
rv = EnsureJARFileAvailable();
if (NS_FAILED(rv)) return rv;
if (mSynchronousInputStream == nsnull)
mon.Wait();
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 (!mSynchronousInputStream)
return NS_ERROR_FAILURE;
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
*result = mSynchronousInputStream; // Result of GetInputStream called on transport thread
NS_ADDREF(*result);
mSynchronousInputStream = 0;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::AsyncOpen(nsIStreamListener* listener, nsISupports* ctxt)
{
nsresult rv;
mUserContext = ctxt;
mUserListener = listener;
if (mLoadGroup) {
rv = mLoadGroup->AddRequest(this, nsnull);
if (NS_FAILED(rv)) return rv;
}
mSynchronousRead = PR_FALSE;
return EnsureJARFileAvailable();
}
nsresult
nsJARChannel::EnsureJARFileAvailable()
{
nsresult rv;
nsCOMPtr<nsIFileURL> fileURL;
#ifdef PR_LOGGING
nsXPIDLCString jarURLStr;
mURI->GetSpec(getter_Copies(jarURLStr));
PR_LOG(gJarProtocolLog, PR_LOG_DEBUG,
("nsJarProtocol: EnsureJARFileAvailable %s", (const char*)jarURLStr));
#endif
rv = mURI->GetJARFile(getter_AddRefs(mJARBaseURI));
if (NS_FAILED(rv)) goto error;
rv = mURI->GetJAREntry(&mJAREntry);
if (NS_FAILED(rv)) goto error;
// try to get a nsIFile directly from the url, which will often succeed.
fileURL = do_QueryInterface(mJARBaseURI);
if (fileURL)
fileURL->GetFile(getter_AddRefs(mDownloadedJARFile));
if (mDownloadedJARFile) {
// after successfully downloading the jar file to the cache,
// start the extraction process:
if (mSynchronousRead)
rv = OpenJARElement();
else
rv = AsyncReadJARElement();
}
else {
rv = NS_NewDownloader(getter_AddRefs(mDownloader),
mJARBaseURI, this, nsnull, mSynchronousRead,
mLoadGroup, mCallbacks, mLoadFlags);
// if DownloadComplete() was called early, need to release the reference.
if (mSynchronousRead && mSynchronousInputStream)
mDownloader = 0;
}
error:
if (NS_FAILED(rv) && mLoadGroup) {
nsresult rv2 = mLoadGroup->RemoveRequest(this, nsnull, NS_OK);
NS_ASSERTION(NS_SUCCEEDED(rv2), "RemoveChannel failed");
}
return rv;
}
nsresult
nsJARChannel::AsyncReadJARElement()
{
nsresult rv;
nsAutoMonitor monitor(mMonitor);
nsCOMPtr<nsIFileTransportService> fts =
do_GetService(kFileTransportServiceCID, &rv);
if (NS_FAILED(rv)) return rv;
nsCOMPtr<nsITransport> jarTransport;
rv = fts->CreateTransportFromStreamIO(this, getter_AddRefs(jarTransport));
if (NS_FAILED(rv)) return rv;
if (mCallbacks) {
nsCOMPtr<nsIProgressEventSink> sink = do_GetInterface(mCallbacks);
if (sink) {
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
// XXX don't think that this is needed anymore
// jarTransport->SetProgressEventSink(sink);
}
}
#ifdef PR_LOGGING
nsXPIDLCString jarURLStr;
mURI->GetSpec(getter_Copies(jarURLStr));
PR_LOG(gJarProtocolLog, PR_LOG_DEBUG,
("nsJarProtocol: AsyncRead jar entry %s", (const char*)jarURLStr));
#endif
rv = jarTransport->AsyncRead(this, nsnull, 0, PRUint32(-1), 0,
getter_AddRefs(mJarExtractionTransport));
mJarExtractionTransport = 0;
jarTransport = 0;
return rv;
}
NS_IMETHODIMP
nsJARChannel::GetLoadFlags(PRUint32* aLoadFlags)
{
*aLoadFlags = mLoadFlags;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::SetLoadFlags(PRUint32 aLoadFlags)
{
mLoadFlags = aLoadFlags;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::GetContentType(char* *aContentType)
{
nsresult rv = NS_OK;
if (mContentType == nsnull) {
if (!mJAREntry)
return NS_ERROR_FAILURE;
char* fileName = nsCRT::strdup(mJAREntry);
if (fileName != nsnull) {
PRInt32 len = nsCRT::strlen(fileName);
const char* ext = nsnull;
for (PRInt32 i = len-1; i >= 0; i--) {
if (fileName[i] == '.') {
ext = &fileName[i + 1];
break;
}
}
if (ext) {
nsCOMPtr<nsIMIMEService> mimeServ (do_GetService(NS_MIMESERVICE_CONTRACTID, &rv));
if (NS_SUCCEEDED(rv)) {
rv = mimeServ->GetTypeFromExtension(ext, &mContentType);
}
}
else
rv = NS_ERROR_OUT_OF_MEMORY;
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
nsCRT::free(fileName);
}
else {
2000-09-12 09:11:11 +04:00
rv = NS_ERROR_OUT_OF_MEMORY;
}
if (NS_FAILED(rv)) {
mContentType = nsCRT::strdup(UNKNOWN_CONTENT_TYPE);
if (mContentType == nsnull)
rv = NS_ERROR_OUT_OF_MEMORY;
else
rv = NS_OK;
}
}
if (NS_SUCCEEDED(rv)) {
*aContentType = nsCRT::strdup(mContentType);
if (*aContentType == nsnull)
rv = NS_ERROR_OUT_OF_MEMORY;
}
return rv;
}
NS_IMETHODIMP
nsJARChannel::SetContentType(const char *aContentType)
{
if (mContentType) {
nsCRT::free(mContentType);
}
mContentType = nsCRT::strdup(aContentType);
if (!mContentType) return NS_ERROR_OUT_OF_MEMORY;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::GetContentLength(PRInt32* aContentLength)
{
if (mContentLength == -1)
return NS_ERROR_FAILURE;
*aContentLength = mContentLength;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::SetContentLength(PRInt32 aContentLength)
{
NS_NOTREACHED("nsJARChannel::SetContentLength");
return NS_ERROR_NOT_IMPLEMENTED;
}
NS_IMETHODIMP
nsJARChannel::GetLoadGroup(nsILoadGroup* *aLoadGroup)
{
*aLoadGroup = mLoadGroup;
NS_IF_ADDREF(*aLoadGroup);
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::SetLoadGroup(nsILoadGroup* aLoadGroup)
{
mLoadGroup = aLoadGroup;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::GetOwner(nsISupports* *aOwner)
{
nsresult rv;
if (mOwner == nsnull) {
//-- Verify signature, if one is present, and set owner accordingly
rv = EnsureZipReader();
if (NS_FAILED(rv)) return rv;
nsCOMPtr<nsIJAR> jar = do_QueryInterface(mJAR, &rv);
NS_ASSERTION(NS_SUCCEEDED(rv), "Zip reader is not an nsIJAR");
nsCOMPtr<nsIPrincipal> certificate;
rv = jar->GetCertificatePrincipal(mJAREntry,
getter_AddRefs(certificate));
if (NS_FAILED(rv)) return rv;
if (certificate)
{ // Get the codebase principal
nsCOMPtr<nsIScriptSecurityManager> secMan =
do_GetService(kScriptSecurityManagerCID, &rv);
if (NS_FAILED(rv)) return NS_ERROR_FAILURE;
nsCOMPtr<nsIPrincipal> codebase;
rv = secMan->GetCodebasePrincipal(mJARBaseURI,
getter_AddRefs(codebase));
if (NS_FAILED(rv)) return rv;
// Join the certificate and the codebase
nsCOMPtr<nsIAggregatePrincipal> agg;
agg = do_QueryInterface(certificate, &rv);
rv = agg->SetCodebase(codebase);
if (NS_FAILED(rv)) return rv;
mOwner = do_QueryInterface(agg, &rv);
if (NS_FAILED(rv)) return rv;
}
}
*aOwner = mOwner;
NS_IF_ADDREF(*aOwner);
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::SetOwner(nsISupports* aOwner)
{
mOwner = aOwner;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::GetNotificationCallbacks(nsIInterfaceRequestor* *aNotificationCallbacks)
{
*aNotificationCallbacks = mCallbacks.get();
NS_IF_ADDREF(*aNotificationCallbacks);
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::SetNotificationCallbacks(nsIInterfaceRequestor* aNotificationCallbacks)
{
mCallbacks = aNotificationCallbacks;
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::GetSecurityInfo(nsISupports * *aSecurityInfo)
{
*aSecurityInfo = nsnull;
return NS_OK;
}
////////////////////////////////////////////////////////////////////////////////
// nsIDownloadObserver methods:
NS_IMETHODIMP
nsJARChannel::OnDownloadComplete(nsIDownloader* aDownloader, nsISupports* aClosure,
nsresult aStatus, nsIFile* aFile)
{
nsresult rv=aStatus;
if(NS_SUCCEEDED(aStatus)) {
NS_ASSERTION(!mDownloader ||(aDownloader == mDownloader.get()), "wrong downloader");
mDownloadedJARFile = aFile;
// after successfully downloading the jar file to the cache,
// start the extraction process:
if (mSynchronousRead)
rv = OpenJARElement();
else
rv = AsyncReadJARElement();
}
mDownloader = 0;
return rv;
}
////////////////////////////////////////////////////////////////////////////////
// nsIRequestObserver methods:
NS_IMETHODIMP
nsJARChannel::OnStartRequest(nsIRequest* jarExtractionTransport,
nsISupports* context)
{
return mUserListener->OnStartRequest(this, mUserContext);
}
NS_IMETHODIMP
nsJARChannel::OnStopRequest(nsIRequest* jarExtractionTransport, nsISupports* context,
nsresult aStatus)
{
nsresult rv;
#ifdef PR_LOGGING
nsCOMPtr<nsIURI> jarURI;
nsXPIDLCString jarURLStr;
rv = mURI->GetSpec(getter_Copies(jarURLStr));
if (NS_SUCCEEDED(rv)) {
PR_LOG(gJarProtocolLog, PR_LOG_DEBUG,
("nsJarProtocol: jar extraction complete %s status=%x",
(const char*)jarURLStr, aStatus));
}
#endif
rv = mUserListener->OnStopRequest(this, mUserContext, aStatus);
NS_ASSERTION(NS_SUCCEEDED(rv), "OnStopRequest failed");
if (mLoadGroup)
mLoadGroup->RemoveRequest(this, context, aStatus);
mUserListener = nsnull;
mUserContext = nsnull;
mJarExtractionTransport = nsnull;
return rv;
}
////////////////////////////////////////////////////////////////////////////////
// nsIStreamListener methods:
NS_IMETHODIMP
nsJARChannel::OnDataAvailable(nsIRequest* jarCacheTransport,
nsISupports* context,
nsIInputStream *inStr,
PRUint32 sourceOffset,
PRUint32 count)
{
return mUserListener->OnDataAvailable(this, mUserContext,
inStr, sourceOffset, count);
}
////////////////////////////////////////////////////////////////////////////////
// nsIStreamIO methods:
nsresult
nsJARChannel::EnsureZipReader()
{
if (mJAR == nsnull) {
nsresult rv;
if (mDownloadedJARFile == nsnull)
return NS_ERROR_FAILURE;
nsCOMPtr<nsIZipReaderCache> jarCache;
rv = mJARProtocolHandler->GetJARCache(getter_AddRefs(jarCache));
if (NS_FAILED(rv)) return rv;
rv = jarCache->GetZip(mDownloadedJARFile, getter_AddRefs(mJAR));
if (NS_FAILED(rv)) return rv;
}
return NS_OK;
}
NS_IMETHODIMP
nsJARChannel::Open(char* *contentType, PRInt32 *contentLength)
{
nsresult rv;
rv = EnsureZipReader();
if (NS_FAILED(rv)) return rv;
nsCOMPtr<nsIZipEntry> entry;
rv = mJAR->GetEntry(mJAREntry, getter_AddRefs(entry));
if (NS_FAILED(rv)) return rv;
if (contentLength) {
rv = entry->GetRealSize((PRUint32*)contentLength);
if (NS_FAILED(rv)) return rv;
}
if (contentType) {
rv = GetContentType(contentType);
if (NS_FAILED(rv)) return rv;
}
return rv;
}
NS_IMETHODIMP
nsJARChannel::Close(nsresult status)
{
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_OK;
}
NS_IMETHODIMP
nsJARChannel::GetInputStream(nsIInputStream* *aInputStream)
{
#ifdef PR_LOGGING
nsXPIDLCString jarURLStr;
mURI->GetSpec(getter_Copies(jarURLStr));
PR_LOG(gJarProtocolLog, PR_LOG_DEBUG,
("nsJarProtocol: GetInputStream jar entry %s", (const char*)jarURLStr));
#endif
NS_ENSURE_TRUE(mJAR, NS_ERROR_NULL_POINTER);
return mJAR->GetInputStream(mJAREntry, aInputStream);
}
NS_IMETHODIMP
nsJARChannel::GetOutputStream(nsIOutputStream* *aOutputStream)
{
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_NOTREACHED("nsJARChannel::GetOutputStream");
return NS_ERROR_NOT_IMPLEMENTED;
}
NS_IMETHODIMP
nsJARChannel::GetName(char* *aName)
{
return mURI->GetSpec(aName);
}
////////////////////////////////////////////////////////////////////////////////
// nsIJARChannel methods:
NS_IMETHODIMP
nsJARChannel::EnumerateEntries(const char *aRoot, nsISimpleEnumerator **_retval)
{
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_NOTREACHED("nsJARChannel::EnumerateEntries");
return NS_ERROR_NOT_IMPLEMENTED;
}
////////////////////////////////////////////////////////////////////////////////