pjs/xpcom/io/nsFastLoadFile.h

585 строки
21 KiB
C
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/* ***** BEGIN LICENSE BLOCK *****
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
*
* The contents of this file are subject to the Mozilla 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/MPL/
*
* 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 of 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 MPL, 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 MPL, the GPL or the LGPL.
*
* ***** END LICENSE BLOCK ***** */
#ifndef nsFastLoadFile_h___
#define nsFastLoadFile_h___
/**
* Mozilla FastLoad file format and helper types.
*/
#include "prtypes.h"
#include "pldhash.h"
#include "nsBinaryStream.h"
#include "nsCOMPtr.h"
#include "nsDebug.h"
#include "nsID.h"
#include "nsMemory.h"
#include "nsVoidArray.h"
#include "nsIFastLoadFileControl.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 "nsIFastLoadService.h"
#include "nsISeekableStream.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 "nsISupportsArray.h"
/**
* FastLoad file Object ID (OID) is an identifier for multiply and cyclicly
* connected objects in the serialized graph of all reachable objects.
*
* Holy Mixed Metaphors: JS, after Common Lisp, uses #n= to define a "sharp
* variable" naming an object that's multiply or cyclicly connected, and #n#
* to stand for a connection to an already-defined object. We too call any
* object with multiple references "sharp", and (here it comes) any object
* with only one reference "dull".
*
* Note that only sharp objects require a mapping from OID to FastLoad file
* offset and other information. Dull objects can be serialized _in situ_
* (where they are referenced) and deserialized when their (singular, shared)
* OID is scanned.
*
* We also compress 16-byte XPCOM IDs into 32-bit dense identifiers to save
* space. See nsFastLoadFooter, below, for the mapping data structure used to
* compute an nsID given an NSFastLoadID.
*/
typedef PRUint32 NSFastLoadID; // nsFastLoadFooter::mIDMap index
typedef PRUint32 NSFastLoadOID; // nsFastLoadFooter::mObjectMap index
/**
* A Mozilla FastLoad file is an untagged (in general) stream of objects and
* primitive-type data. Small integers are fairly common, and could easily be
* confused for NSFastLoadIDs and NSFastLoadOIDs. To help catch bugs where
* reader and writer code fail to match, we XOR unlikely 32-bit numbers with
* NSFastLoad*IDs when storing and fetching. The following unlikely values are
* irrational numbers ((sqrt(5)-1)/2, sqrt(2)-1) represented in fixed point.
*
* The reader XORs, converts the ID to an index, and bounds-checks all array
* accesses that use the index. Array access code asserts that the index is in
* bounds, and returns a dummy array element if it isn't.
*/
#define MFL_ID_XOR_KEY 0x9E3779B9 // key XOR'd with ID when serialized
#define MFL_OID_XOR_KEY 0x6A09E667 // key XOR'd with OID when serialized
/**
* An OID can be tagged to introduce the serialized definition of the object,
* or to stand for a strong or weak reference to that object. Thus the high
* 29 bits actually identify the object, and the low three bits tell whether
* the object is being defined or just referenced -- and via what inheritance
* chain or inner object, if necessary.
*
* The MFL_QUERY_INTERFACE_TAG bit helps us cope with aggregation and multiple
* inheritance: object identity follows the XPCOM rule, but a deserializer may
* need to query for an interface not on the primary inheritance chain ending
* in the nsISupports whose address uniquely identifies the XPCOM object being
* referenced or defined.
*/
#define MFL_OBJECT_TAG_BITS 3
#define MFL_OBJECT_TAG_MASK PR_BITMASK(MFL_OBJECT_TAG_BITS)
#define MFL_OBJECT_DEF_TAG 1U // object definition follows this OID
#define MFL_WEAK_REF_TAG 2U // OID weakly refers to a prior object
// NB: do not confuse with nsWeakPtr!
#define MFL_QUERY_INTERFACE_TAG 4U // QI object to the ID follows this OID
// NB: an NSFastLoadID, not an nsIID!
/**
* The dull object identifier introduces the definition of all objects that
* have only one (necessarily strong) ref in the serialization. The definition
* appears at the point of reference.
*/
#define MFL_DULL_OBJECT_OID MFL_OBJECT_DEF_TAG
/**
* Convert an OID to an index into nsFastLoadFooter::mObjectMap.
*/
#define MFL_OID_TO_SHARP_INDEX(oid) (((oid) >> MFL_OBJECT_TAG_BITS) - 1)
#define MFL_SHARP_INDEX_TO_OID(index) (((index) + 1) << MFL_OBJECT_TAG_BITS)
/**
* Magic "number" at start of a FastLoad file. Inspired by the PNG "magic"
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
* string, which inspired XPCOM's typelib (.xpt) file magic. Guaranteed to be
* corrupted by FTP-as-ASCII and other likely errors, meaningful to clued-in
* humans, and ending in ^Z to terminate erroneous text input on Windows.
*/
#define MFL_FILE_MAGIC "XPCOM\nMozFASL\r\n\032"
#define MFL_FILE_MAGIC_SIZE 16
#define MFL_FILE_VERSION_0 0
#define MFL_FILE_VERSION_1 1000
#define MFL_FILE_VERSION 5 // rev'ed to defend against unversioned
// XPCOM JS component fastload files
/**
* Compute Fletcher's 16-bit checksum over aLength bytes starting at aBuffer,
* with the initial accumulators seeded from *aChecksum, and final checksum
* returned in *aChecksum. The return value is the number of unchecked bytes,
* which may be non-zero if aBuffer is misaligned or aLength is odd. Callers
* should copy any remaining bytes to the front of the next buffer.
*
* If aLastBuffer is false, do not check any bytes remaining due to misaligned
* aBuffer or odd aLength, instead returning the remaining byte count. But if
* aLastBuffer is true, treat aBuffer as the last buffer in the file and check
* every byte, returning 0. Here's a read-loop checksumming sketch:
*
* char buf[BUFSIZE];
* PRUint32 len, rem = 0;
* PRUint32 checksum = 0;
*
* while (NS_SUCCEEDED(rv = Read(buf + rem, sizeof buf - rem, &len)) && len) {
* len += rem;
* rem = NS_AccumulateFastLoadChecksum(&checksum,
* NS_REINTERPRET_CAST(PRUint8*, buf),
* len,
* PR_FALSE);
* if (rem)
* memcpy(buf, buf + len - rem, rem);
* }
*
* if (rem) {
* NS_AccumulateFastLoadChecksum(&checksum,
* NS_REINTERPRET_CAST(PRUint8*, buf),
* rem,
* PR_TRUE);
* }
*
* After this, if NS_SUCCEEDED(rv), checksum contains a valid FastLoad sum.
*/
NS_COM PRUint32
NS_AccumulateFastLoadChecksum(PRUint32 *aChecksum,
const PRUint8* aBuffer,
PRUint32 aLength,
PRBool aLastBuffer);
NS_COM PRUint32
NS_AddFastLoadChecksums(PRUint32 sum1, PRUint32 sum2, PRUint32 sum2ByteCount);
/**
* Header at the start of a FastLoad file.
*/
struct nsFastLoadHeader {
char mMagic[MFL_FILE_MAGIC_SIZE];
PRUint32 mChecksum;
PRUint32 mVersion;
PRUint32 mFooterOffset;
PRUint32 mFileSize;
};
/**
* Footer prefix structure (footer header, ugh), after which come arrays of
* structures or strings counted by these members.
*/
struct nsFastLoadFooterPrefix {
PRUint32 mNumIDs;
PRUint32 mNumSharpObjects;
PRUint32 mNumMuxedDocuments;
PRUint32 mNumDependencies;
};
struct nsFastLoadSharpObjectInfo {
PRUint32 mCIDOffset; // offset of object's NSFastLoadID and data
PRUint16 mStrongRefCnt;
PRUint16 mWeakRefCnt; // high bit is singleton flag, see below
};
#define MFL_SINGLETON_FLAG 0x8000
#define MFL_WEAK_REFCNT_MASK 0x7fff
#define MFL_GET_SINGLETON_FLAG(ip) ((ip)->mWeakRefCnt & MFL_SINGLETON_FLAG)
#define MFL_GET_WEAK_REFCNT(ip) ((ip)->mWeakRefCnt & MFL_WEAK_REFCNT_MASK)
#define MFL_SET_SINGLETON_FLAG(ip) \
((ip)->mWeakRefCnt |= MFL_SINGLETON_FLAG)
#define MFL_SET_WEAK_REFCNT(ip,rc) \
((ip)->mWeakRefCnt = (((ip)->mWeakRefCnt & MFL_SINGLETON_FLAG) | (rc)))
#define MFL_BUMP_WEAK_REFCNT(ip) (++(ip)->mWeakRefCnt)
#define MFL_DROP_WEAK_REFCNT(ip) (--(ip)->mWeakRefCnt)
struct nsFastLoadMuxedDocumentInfo {
const char* mURISpec;
PRUint32 mInitialSegmentOffset;
};
// forward declarations of opaque types defined in nsFastLoadFile.cpp
struct nsDocumentMapReadEntry;
struct nsDocumentMapWriteEntry;
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
// So nsFastLoadFileUpdater can verify that its nsIObjectInputStream parameter
// is an nsFastLoadFileReader.
#define NS_FASTLOADFILEREADER_IID \
{0x7d37d1bb,0xcef3,0x4c5f,{0x97,0x68,0x0f,0x89,0x7f,0x1a,0xe1,0x40}}
struct nsIFastLoadFileReader : public nsISupports {
NS_DECLARE_STATIC_IID_ACCESSOR(NS_FASTLOADFILEREADER_IID)
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_DEFINE_STATIC_IID_ACCESSOR(nsIFastLoadFileReader, NS_FASTLOADFILEREADER_IID)
/**
* Inherit from the concrete class nsBinaryInputStream, which inherits from
* abstract nsIObjectInputStream but does not implement its direct methods.
* Though the names are not as clear as I'd like, this seems to be the best
* way to share nsBinaryStream.cpp code.
*/
class nsFastLoadFileReader
: public nsBinaryInputStream,
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
public nsIFastLoadReadControl,
public nsISeekableStream,
public nsIFastLoadFileReader
{
public:
nsFastLoadFileReader(nsIInputStream *aStream)
: mCurrentDocumentMapEntry(nsnull) {
SetInputStream(aStream);
MOZ_COUNT_CTOR(nsFastLoadFileReader);
}
virtual ~nsFastLoadFileReader() {
MOZ_COUNT_DTOR(nsFastLoadFileReader);
}
private:
// nsISupports methods
NS_DECL_ISUPPORTS_INHERITED
// overridden nsIObjectInputStream methods
NS_IMETHOD ReadObject(PRBool aIsStrongRef, nsISupports* *_retval);
NS_IMETHOD ReadID(nsID *aResult);
// nsIFastLoadFileControl methods
NS_DECL_NSIFASTLOADFILECONTROL
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
// nsIFastLoadReadControl methods
NS_DECL_NSIFASTLOADREADCONTROL
// nsISeekableStream methods
NS_DECL_NSISEEKABLESTREAM
// Override Read so we can demultiplex a document interleaved with others.
NS_IMETHOD Read(char* aBuffer, PRUint32 aCount, PRUint32 *aBytesRead);
Reduce seeks outside the FastLoad file's underlying stream buffer (195010, r=ben, sr=bryner). - Pass null scope objects into nsIScriptContext::Compile{EventHandler,Script} when precompiling for brutal sharing. The JS engine does not need a non-null object parameter for static scope. That parameter can be non-null only if it helps the compiler "pre-bind" functions to the same scope object that they'll be parented by when executing, but with brutal sharing, functions are precompiled once and executed against many different scope objects. - A problem in XUL FastLoad was that it would serialize master .xul document out-of-line scripts (those included via script src= from non-overlay, "master" .xul docs) far from the place in the FastLoad file where XUL prototype script info was serialized for the <script src=> tag itself. I fixed that so that, unless the out-of-line script was previously serialized (by a different src= reference from another .xul file), the OOL script data immediately follows the proto-script info in the FastLoad file. This required adding a SerializeOutOfLine method to nsXULPrototypeScript, which restores symmetry by matching the existing DeserializeOutOfLine(Script) (note: I dropped the redundant "Script" from the end of the latter method's name. We need SerializeOutOfLine to handle overlay OOL scripts. They are serialized by nsXULDocument::OnStreamComplete, because that code knows the difference between an overlay and a master doc. This removes all trace of FastLoad writing from nsXULPrototypeScript::Compile -- FastLoad stuff didn't belong there, not only because we now want to write master OOL scripts later, when walking the master XUL prototype doc's element tree, but also for modularity reasons. The caller knows about FastLoad, nsXULPrototypeScript::Compile does just what its name implies. There are 132 seeks with the patch, only 49 of which dump the underlying file stream's buffer, vs. 133 seeks without the patch, 87 of which dump the buffer. - Nit-picked some comments to fit in 80 columns, and made other cosmetic fixes. - Implicated the nsXULDocument::mIsWritingFastLoad flag from the useXULCache "is the XUL cache enabled?" flag, so other places that test mIsWritingFastLoad don't have to query whether the cache is enabled. - Added METERING synchronous meter-dumping to /tmp/bufstats, only ifdef DEBUG_brendan, in netwerk/base/src/nsBufferedStreams.cpp. - Added the deferred seek optimization from the first patch in bug 195010 to nsFastLoadFile.cpp. - Fixed nsFastLoadFileReader so it overrides readSegments as well as read. This catches up with the interface extension made to nsIInputStream to add readSegments. The nsFastLoadFileReader extends nsBinaryInputStream, which is a concrete class, and overrides Read in order to snoop on *all* bytes read. It does this in order to demultiplex documents interleaved when the FastLoad file was written. But since the readSegments move into nsIInputStream.idl, certain primitives in nsBinaryStream.cpp, e.g., nsBinaryInputStream::ReadCString, have used ReadSegments, not Read, to consume parts of the underlying stream (to read the C string's chars, in that example), and the FastLoad file implementation has not accounted for those bytes. - Added a new method to nsIFastLoadFileControl and its impls: hasMuxedDocument. This is needed when serializing master XUL doc OOL scripts, because we must not serialize twice, and any OOL script that other XUL docs can include via script src= could already be in the FastLoad mux. /be
2003-04-12 05:34:11 +04:00
// Override ReadSegments too, as nsBinaryInputStream::ReadSegments does
// not call through our overridden Read method -- it calls directly into
// the underlying input stream.
NS_IMETHODIMP ReadSegments(nsWriteSegmentFun aWriter, void* aClosure,
PRUint32 aCount, PRUint32 *aResult);
Reduce seeks outside the FastLoad file's underlying stream buffer (195010, r=ben, sr=bryner). - Pass null scope objects into nsIScriptContext::Compile{EventHandler,Script} when precompiling for brutal sharing. The JS engine does not need a non-null object parameter for static scope. That parameter can be non-null only if it helps the compiler "pre-bind" functions to the same scope object that they'll be parented by when executing, but with brutal sharing, functions are precompiled once and executed against many different scope objects. - A problem in XUL FastLoad was that it would serialize master .xul document out-of-line scripts (those included via script src= from non-overlay, "master" .xul docs) far from the place in the FastLoad file where XUL prototype script info was serialized for the <script src=> tag itself. I fixed that so that, unless the out-of-line script was previously serialized (by a different src= reference from another .xul file), the OOL script data immediately follows the proto-script info in the FastLoad file. This required adding a SerializeOutOfLine method to nsXULPrototypeScript, which restores symmetry by matching the existing DeserializeOutOfLine(Script) (note: I dropped the redundant "Script" from the end of the latter method's name. We need SerializeOutOfLine to handle overlay OOL scripts. They are serialized by nsXULDocument::OnStreamComplete, because that code knows the difference between an overlay and a master doc. This removes all trace of FastLoad writing from nsXULPrototypeScript::Compile -- FastLoad stuff didn't belong there, not only because we now want to write master OOL scripts later, when walking the master XUL prototype doc's element tree, but also for modularity reasons. The caller knows about FastLoad, nsXULPrototypeScript::Compile does just what its name implies. There are 132 seeks with the patch, only 49 of which dump the underlying file stream's buffer, vs. 133 seeks without the patch, 87 of which dump the buffer. - Nit-picked some comments to fit in 80 columns, and made other cosmetic fixes. - Implicated the nsXULDocument::mIsWritingFastLoad flag from the useXULCache "is the XUL cache enabled?" flag, so other places that test mIsWritingFastLoad don't have to query whether the cache is enabled. - Added METERING synchronous meter-dumping to /tmp/bufstats, only ifdef DEBUG_brendan, in netwerk/base/src/nsBufferedStreams.cpp. - Added the deferred seek optimization from the first patch in bug 195010 to nsFastLoadFile.cpp. - Fixed nsFastLoadFileReader so it overrides readSegments as well as read. This catches up with the interface extension made to nsIInputStream to add readSegments. The nsFastLoadFileReader extends nsBinaryInputStream, which is a concrete class, and overrides Read in order to snoop on *all* bytes read. It does this in order to demultiplex documents interleaved when the FastLoad file was written. But since the readSegments move into nsIInputStream.idl, certain primitives in nsBinaryStream.cpp, e.g., nsBinaryInputStream::ReadCString, have used ReadSegments, not Read, to consume parts of the underlying stream (to read the C string's chars, in that example), and the FastLoad file implementation has not accounted for those bytes. - Added a new method to nsIFastLoadFileControl and its impls: hasMuxedDocument. This is needed when serializing master XUL doc OOL scripts, because we must not serialize twice, and any OOL script that other XUL docs can include via script src= could already be in the FastLoad mux. /be
2003-04-12 05:34:11 +04:00
// Override SetInputStream so we can update mSeekableInput
NS_IMETHOD SetInputStream(nsIInputStream* aInputStream);
nsresult ReadHeader(nsFastLoadHeader *aHeader);
/**
* In-memory representation of an indexed nsFastLoadSharpObjectInfo record.
*/
struct nsObjectMapEntry : public nsFastLoadSharpObjectInfo {
nsCOMPtr<nsISupports> mReadObject;
PRInt64 mSkipOffset;
PRUint16 mSaveStrongRefCnt; // saved for an Update
PRUint16 mSaveWeakRefCnt; // after a Read
};
/**
* In-memory representation of the FastLoad file footer.
*/
struct nsFastLoadFooter : public nsFastLoadFooterPrefix {
nsFastLoadFooter()
: mIDMap(nsnull),
mObjectMap(nsnull) {
mDocumentMap.ops = mURIMap.ops = nsnull;
}
~nsFastLoadFooter() {
delete[] mIDMap;
delete[] mObjectMap;
if (mDocumentMap.ops)
PL_DHashTableFinish(&mDocumentMap);
if (mURIMap.ops)
PL_DHashTableFinish(&mURIMap);
}
// These can't be static within GetID and GetSharpObjectEntry or the
// toolchains on HP-UX 10.20's, RH 7.0, and Mac OS X all barf at link
// time ("common symbols not allowed with MY_DHLIB output format", to
// quote the OS X rev of gcc).
static nsID gDummyID;
static nsObjectMapEntry gDummySharpObjectEntry;
const nsID& GetID(NSFastLoadID aFastId) const {
PRUint32 index = aFastId - 1;
NS_ASSERTION(index < mNumIDs, "aFastId out of range");
if (index >= mNumIDs)
return gDummyID;
return mIDMap[index];
}
nsObjectMapEntry&
GetSharpObjectEntry(NSFastLoadOID aOID) const {
PRUint32 index = MFL_OID_TO_SHARP_INDEX(aOID);
NS_ASSERTION(index < mNumSharpObjects, "aOID out of range");
if (index >= mNumSharpObjects)
return gDummySharpObjectEntry;
return mObjectMap[index];
}
// Map from dense, zero-based, uint32 NSFastLoadID to 16-byte nsID.
nsID* mIDMap;
// Map from dense, zero-based MFL_OID_TO_SHARP_INDEX(oid) to sharp
// object offset and refcnt information.
nsObjectMapEntry* mObjectMap;
// Map from URI spec string to nsDocumentMapReadEntry, which helps us
// demultiplex a document's objects from among the interleaved object
// stream segments in the FastLoad file.
PLDHashTable mDocumentMap;
// Fast mapping from URI object pointer to mDocumentMap entry, valid
// only while the muxed document is loading.
PLDHashTable mURIMap;
// List of source filename dependencies that should trigger regeneration
// of the FastLoad 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
nsCOMPtr<nsISupportsArray> mDependencies;
};
nsresult ReadFooter(nsFastLoadFooter *aFooter);
nsresult ReadFooterPrefix(nsFastLoadFooterPrefix *aFooterPrefix);
nsresult ReadSlowID(nsID *aID);
nsresult ReadFastID(NSFastLoadID *aID);
nsresult ReadSharpObjectInfo(nsFastLoadSharpObjectInfo *aInfo);
nsresult ReadMuxedDocumentInfo(nsFastLoadMuxedDocumentInfo *aInfo);
nsresult DeserializeObject(nsISupports* *aObject);
nsresult Open();
NS_IMETHOD Close();
protected:
// Kept in sync with mInputStream to avoid repeated QI
nsCOMPtr<nsISeekableStream> mSeekableInput;
nsFastLoadHeader mHeader;
nsFastLoadFooter mFooter;
nsDocumentMapReadEntry* mCurrentDocumentMapEntry;
friend class nsFastLoadFileUpdater;
};
NS_COM nsresult
NS_NewFastLoadFileReader(nsIObjectInputStream* *aResult,
nsIInputStream* aSrcStream);
/**
* Inherit from the concrete class nsBinaryInputStream, which inherits from
* abstract nsIObjectInputStream but does not implement its direct methods.
* Though the names are not as clear as I'd like, this seems to be the best
* way to share nsBinaryStream.cpp code.
*/
class nsFastLoadFileWriter
: public nsBinaryOutputStream,
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
public nsIFastLoadWriteControl,
public nsISeekableStream
{
public:
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
nsFastLoadFileWriter(nsIOutputStream *aStream, nsIFastLoadFileIO* aFileIO)
: mCurrentDocumentMapEntry(nsnull),
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
mFileIO(aFileIO)
{
SetOutputStream(aStream);
mHeader.mChecksum = 0;
mIDMap.ops = mObjectMap.ops = mDocumentMap.ops = mURIMap.ops = nsnull;
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
mDependencyMap.ops = nsnull;
MOZ_COUNT_CTOR(nsFastLoadFileWriter);
}
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
virtual ~nsFastLoadFileWriter()
{
if (mIDMap.ops)
PL_DHashTableFinish(&mIDMap);
if (mObjectMap.ops)
PL_DHashTableFinish(&mObjectMap);
if (mDocumentMap.ops)
PL_DHashTableFinish(&mDocumentMap);
if (mURIMap.ops)
PL_DHashTableFinish(&mURIMap);
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 (mDependencyMap.ops)
PL_DHashTableFinish(&mDependencyMap);
MOZ_COUNT_DTOR(nsFastLoadFileWriter);
}
private:
// nsISupports methods
NS_DECL_ISUPPORTS_INHERITED
// overridden nsIObjectOutputStream methods
NS_IMETHOD WriteObject(nsISupports* aObject, PRBool aIsStrongRef);
NS_IMETHOD WriteSingleRefObject(nsISupports* aObject);
NS_IMETHOD WriteCompoundObject(nsISupports* aObject,
const nsIID& aIID,
PRBool aIsStrongRef);
NS_IMETHOD WriteID(const nsID& aID);
// Override SetOutputStream so we can update mSeekableOutput
NS_IMETHOD SetOutputStream(nsIOutputStream* aOutputStream);
// nsIFastLoadFileControl methods
NS_DECL_NSIFASTLOADFILECONTROL
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
// nsIFastLoadWriteControl methods
NS_DECL_NSIFASTLOADWRITECONTROL
// nsISeekableStream methods
NS_DECL_NSISEEKABLESTREAM
nsresult MapID(const nsID& aSlowID, NSFastLoadID *aResult);
nsresult WriteHeader(nsFastLoadHeader *aHeader);
nsresult WriteFooter();
nsresult WriteFooterPrefix(const nsFastLoadFooterPrefix& aFooterPrefix);
nsresult WriteSlowID(const nsID& aID);
nsresult WriteFastID(NSFastLoadID aID);
nsresult WriteSharpObjectInfo(const nsFastLoadSharpObjectInfo& aInfo);
nsresult WriteMuxedDocumentInfo(const nsFastLoadMuxedDocumentInfo& aInfo);
nsresult Init();
nsresult Open();
NS_IMETHOD Close();
nsresult WriteObjectCommon(nsISupports* aObject,
PRBool aIsStrongRef,
PRUint32 aQITag);
static PLDHashOperator PR_CALLBACK
IDMapEnumerate(PLDHashTable *aTable,
PLDHashEntryHdr *aHdr,
PRUint32 aNumber,
void *aData);
static PLDHashOperator PR_CALLBACK
ObjectMapEnumerate(PLDHashTable *aTable,
PLDHashEntryHdr *aHdr,
PRUint32 aNumber,
void *aData);
static PLDHashOperator PR_CALLBACK
DocumentMapEnumerate(PLDHashTable *aTable,
PLDHashEntryHdr *aHdr,
PRUint32 aNumber,
void *aData);
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
static PLDHashOperator PR_CALLBACK
DependencyMapEnumerate(PLDHashTable *aTable,
PLDHashEntryHdr *aHdr,
PRUint32 aNumber,
void *aData);
protected:
// Kept in sync with mOutputStream to avoid repeated QI
nsCOMPtr<nsISeekableStream> mSeekableOutput;
nsFastLoadHeader mHeader;
PLDHashTable mIDMap;
PLDHashTable mObjectMap;
PLDHashTable mDocumentMap;
PLDHashTable mURIMap;
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
PLDHashTable mDependencyMap;
nsDocumentMapWriteEntry* mCurrentDocumentMapEntry;
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<nsIFastLoadFileIO> mFileIO;
};
NS_COM nsresult
NS_NewFastLoadFileWriter(nsIObjectOutputStream* *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
nsIOutputStream* aDestStream,
nsIFastLoadFileIO* aFileIO);
/**
* Subclass of nsFastLoadFileWriter, friend of nsFastLoadFileReader which it
* wraps when a FastLoad file needs to be updated. The wrapped reader can be
* used to demulitplex data for documents already in the FastLoad file, while
* the updater writes new data over the old footer, then writes a new footer
* that maps all data on Close.
*/
class nsFastLoadFileUpdater
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
: public nsFastLoadFileWriter,
nsIFastLoadFileIO
{
public:
nsFastLoadFileUpdater(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
: nsFastLoadFileWriter(aOutputStream, nsnull) {
MOZ_COUNT_CTOR(nsFastLoadFileUpdater);
}
virtual ~nsFastLoadFileUpdater() {
MOZ_COUNT_DTOR(nsFastLoadFileUpdater);
}
private:
// nsISupports methods
NS_DECL_ISUPPORTS_INHERITED
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
// nsIFastLoadFileIO methods
NS_DECL_NSIFASTLOADFILEIO
nsresult Open(nsFastLoadFileReader* aReader);
NS_IMETHOD Close();
static PLDHashOperator PR_CALLBACK
CopyReadDocumentMapEntryToUpdater(PLDHashTable *aTable,
PLDHashEntryHdr *aHdr,
PRUint32 aNumber,
void *aData);
friend class nsFastLoadFileReader;
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
protected:
nsCOMPtr<nsIInputStream> mInputStream;
// Kept in sync with mInputStream to avoid repeated QI
nsCOMPtr<nsISeekableStream> mSeekableInput;
};
NS_COM nsresult
NS_NewFastLoadFileUpdater(nsIObjectOutputStream* *aResult,
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
nsIObjectInputStream* aReaderAsStream);
#endif // nsFastLoadFile_h___