gecko-dev/dom/events/ContentEventHandler.cpp

3081 строка
110 KiB
C++
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
2012-05-21 15:12:37 +04:00
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#include "ContentEventHandler.h"
#include "mozilla/IMEStateManager.h"
#include "mozilla/TextComposition.h"
#include "mozilla/TextEvents.h"
#include "mozilla/dom/Element.h"
#include "mozilla/dom/HTMLUnknownElement.h"
#include "mozilla/dom/Selection.h"
#include "nsCaret.h"
#include "nsCOMPtr.h"
#include "nsContentUtils.h"
#include "nsCopySupport.h"
#include "nsFocusManager.h"
#include "nsFontMetrics.h"
#include "nsFrameSelection.h"
#include "nsIContentIterator.h"
#include "nsIPresShell.h"
#include "nsISelection.h"
#include "nsIFrame.h"
#include "nsIObjectFrame.h"
#include "nsLayoutUtils.h"
#include "nsPresContext.h"
#include "nsQueryObject.h"
#include "nsRange.h"
#include "nsTextFragment.h"
#include "nsTextFrame.h"
#include "nsView.h"
#include <algorithm>
namespace mozilla {
using namespace dom;
using namespace widget;
/******************************************************************/
/* ContentEventHandler */
/******************************************************************/
// NOTE
//
// ContentEventHandler *creates* ranges as following rules:
// 1. Start of range:
// 1.1. Cases: [textNode or text[Node or textNode[
// When text node is start of a range, start node is the text node and
// start offset is any number between 0 and the length of the text.
// 1.2. Case: [<element>:
// When start of an element node is start of a range, start node is
// parent of the element and start offset is the element's index in the
// parent.
// 1.3. Case: <element/>[
// When after an empty element node is start of a range, start node is
// parent of the element and start offset is the element's index in the
// parent + 1.
// 1.4. Case: <element>[
// When start of a non-empty element is start of a range, start node is
// the element and start offset is 0.
// 1.5. Case: <root>[
// When start of a range is 0 and there are no nodes causing text,
// start node is the root node and start offset is 0.
// 1.6. Case: [</root>
// When start of a range is out of bounds, start node is the root node
// and start offset is number of the children.
// 2. End of range:
// 2.1. Cases: ]textNode or text]Node or textNode]
// When a text node is end of a range, end node is the text node and
// end offset is any number between 0 and the length of the text.
// 2.2. Case: ]<element>
// When before an element node (meaning before the open tag of the
// element) is end of a range, end node is previous node causing text.
// Note that this case shouldn't be handled directly. If rule 2.1 and
// 2.3 are handled correctly, the loop with nsContentIterator shouldn't
// reach the element node since the loop should've finished already at
// handling the last node which caused some text.
// 2.3. Case: <element>]
// When a line break is caused before a non-empty element node and it's
// end of a range, end node is the element and end offset is 0.
// (i.e., including open tag of the element)
// 2.4. Cases: <element/>]
// When after an empty element node is end of a range, end node is
// parent of the element node and end offset is the element's index in
// the parent + 1. (i.e., including close tag of the element or empty
// element)
// 2.5. Case: ]</root>
// When end of a range is out of bounds, end node is the root node and
// end offset is number of the children.
//
// ContentEventHandler *treats* ranges as following additional rules:
// 1. When the start node is an element node which doesn't have children,
// it includes a line break caused before itself (i.e., includes its open
// tag). For example, if start position is { <br>, 0 }, the line break
// caused by <br> should be included into the flatten text.
// 2. When the end node is an element node which doesn't have children,
// it includes the end (i.e., includes its close tag except empty element).
// Although, currently, any close tags don't cause line break, this also
// includes its open tag. For example, if end position is { <br>, 0 }, the
// line break caused by the <br> should be included into the flatten text.
ContentEventHandler::ContentEventHandler(nsPresContext* aPresContext)
: mPresContext(aPresContext)
, mPresShell(aPresContext->GetPresShell())
, mSelection(nullptr)
, mFirstSelectedRange(nullptr)
, mRootContent(nullptr)
{
}
nsresult
ContentEventHandler::InitBasic()
{
NS_ENSURE_TRUE(mPresShell, NS_ERROR_NOT_AVAILABLE);
// If text frame which has overflowing selection underline is dirty,
// we need to flush the pending reflow here.
mPresShell->FlushPendingNotifications(Flush_Layout);
// Flushing notifications can cause mPresShell to be destroyed (bug 577963).
NS_ENSURE_TRUE(!mPresShell->IsDestroying(), NS_ERROR_FAILURE);
return NS_OK;
}
nsresult
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
ContentEventHandler::InitRootContent(Selection* aNormalSelection)
{
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
MOZ_ASSERT(aNormalSelection);
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
// Root content should be computed with normal selection because normal
// selection is typically has at least one range but the other selections
// not so. If there is a range, computing its root is easy, but if
// there are no ranges, we need to use ancestor limit instead.
MOZ_ASSERT(aNormalSelection->Type() == SelectionType::eNormal);
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
if (!aNormalSelection->RangeCount()) {
// If there is no selection range, we should compute the selection root
// from ancestor limiter or root content of the document.
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
nsresult rv =
aNormalSelection->GetAncestorLimiter(getter_AddRefs(mRootContent));
if (NS_WARN_IF(NS_FAILED(rv))) {
return NS_ERROR_FAILURE;
}
if (!mRootContent) {
mRootContent = mPresShell->GetDocument()->GetRootElement();
if (NS_WARN_IF(!mRootContent)) {
return NS_ERROR_NOT_AVAILABLE;
}
}
return NS_OK;
}
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
RefPtr<nsRange> range(aNormalSelection->GetRangeAt(0));
if (NS_WARN_IF(!range)) {
return NS_ERROR_UNEXPECTED;
}
// If there is a selection, we should retrieve the selection root from
// the range since when the window is inactivated, the ancestor limiter
// of selection was cleared by blur event handler of EditorBase but the
// selection range still keeps storing the nodes. If the active element of
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
// the deactive window is <input> or <textarea>, we can compute the
// selection root from them.
nsINode* startNode = range->GetStartParent();
nsINode* endNode = range->GetEndParent();
if (NS_WARN_IF(!startNode) || NS_WARN_IF(!endNode)) {
return NS_ERROR_FAILURE;
}
// See bug 537041 comment 5, the range could have removed node.
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
if (NS_WARN_IF(startNode->GetUncomposedDoc() != mPresShell->GetDocument())) {
return NS_ERROR_FAILURE;
}
NS_ASSERTION(startNode->GetUncomposedDoc() == endNode->GetUncomposedDoc(),
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
"firstNormalSelectionRange crosses the document boundary");
mRootContent = startNode->GetSelectionRootContent(mPresShell);
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
if (NS_WARN_IF(!mRootContent)) {
return NS_ERROR_FAILURE;
}
return NS_OK;
}
nsresult
ContentEventHandler::InitCommon(SelectionType aSelectionType)
{
if (mSelection && mSelection->Type() == aSelectionType) {
return NS_OK;
}
mSelection = nullptr;
mFirstSelectedRange = nullptr;
mRootContent = nullptr;
nsresult rv = InitBasic();
NS_ENSURE_SUCCESS(rv, rv);
nsCOMPtr<nsISelectionController> selectionController =
mPresShell->GetSelectionControllerForFocusedContent();
if (NS_WARN_IF(!selectionController)) {
return NS_ERROR_NOT_AVAILABLE;
}
nsCOMPtr<nsISelection> selection;
rv = selectionController->GetSelection(ToRawSelectionType(aSelectionType),
getter_AddRefs(selection));
if (NS_WARN_IF(NS_FAILED(rv))) {
return NS_ERROR_UNEXPECTED;
}
mSelection = static_cast<Selection*>(selection.get());
if (NS_WARN_IF(!mSelection)) {
return NS_ERROR_NOT_AVAILABLE;
}
RefPtr<Selection> normalSelection;
if (mSelection->Type() == SelectionType::eNormal) {
normalSelection = mSelection;
} else {
nsCOMPtr<nsISelection> domSelection;
nsresult rv =
selectionController->GetSelection(
nsISelectionController::SELECTION_NORMAL,
getter_AddRefs(domSelection));
if (NS_WARN_IF(NS_FAILED(rv))) {
return NS_ERROR_UNEXPECTED;
}
if (NS_WARN_IF(!domSelection)) {
return NS_ERROR_NOT_AVAILABLE;
}
normalSelection = domSelection->AsSelection();
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
if (NS_WARN_IF(!normalSelection)) {
return NS_ERROR_NOT_AVAILABLE;
}
}
rv = InitRootContent(normalSelection);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
if (mSelection->RangeCount()) {
mFirstSelectedRange = mSelection->GetRangeAt(0);
if (NS_WARN_IF(!mFirstSelectedRange)) {
return NS_ERROR_UNEXPECTED;
}
return NS_OK;
}
// Even if there are no selection ranges, it's usual case if aSelectionType
// is a special selection.
if (aSelectionType != SelectionType::eNormal) {
MOZ_ASSERT(!mFirstSelectedRange);
return NS_OK;
}
// But otherwise, we need to assume that there is a selection range at the
// beginning of the root content if aSelectionType is eNormal.
rv = nsRange::CreateRange(mRootContent, 0, mRootContent, 0,
getter_AddRefs(mFirstSelectedRange));
if (NS_WARN_IF(NS_FAILED(rv)) || NS_WARN_IF(!mFirstSelectedRange)) {
return NS_ERROR_UNEXPECTED;
}
return NS_OK;
}
nsresult
ContentEventHandler::Init(WidgetQueryContentEvent* aEvent)
{
NS_ASSERTION(aEvent, "aEvent must not be null");
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
MOZ_ASSERT(aEvent->mMessage == eQuerySelectedText ||
aEvent->mInput.mSelectionType == SelectionType::eNormal);
if (NS_WARN_IF(!aEvent->mInput.IsValidOffset()) ||
NS_WARN_IF(!aEvent->mInput.IsValidEventMessage(aEvent->mMessage))) {
return NS_ERROR_FAILURE;
}
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
// Note that we should ignore WidgetQueryContentEvent::Input::mSelectionType
// if the event isn't eQuerySelectedText.
SelectionType selectionType =
aEvent->mMessage == eQuerySelectedText ? aEvent->mInput.mSelectionType :
SelectionType::eNormal;
if (NS_WARN_IF(selectionType == SelectionType::eNone)) {
return NS_ERROR_FAILURE;
}
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
nsresult rv = InitCommon(selectionType);
NS_ENSURE_SUCCESS(rv, rv);
// Be aware, WidgetQueryContentEvent::mInput::mOffset should be made absolute
// offset before sending it to ContentEventHandler because querying selection
// every time may be expensive. So, if the caller caches selection, it
// should initialize the event with the cached value.
if (aEvent->mInput.mRelativeToInsertionPoint) {
MOZ_ASSERT(selectionType == SelectionType::eNormal);
RefPtr<TextComposition> composition =
IMEStateManager::GetTextCompositionFor(aEvent->mWidget);
if (composition) {
uint32_t compositionStart = composition->NativeOffsetOfStartComposition();
if (NS_WARN_IF(!aEvent->mInput.MakeOffsetAbsolute(compositionStart))) {
return NS_ERROR_FAILURE;
}
} else {
LineBreakType lineBreakType = GetLineBreakType(aEvent);
uint32_t selectionStart = 0;
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
rv = GetStartOffset(mFirstSelectedRange, &selectionStart, lineBreakType);
if (NS_WARN_IF(NS_FAILED(rv))) {
return NS_ERROR_FAILURE;
}
if (NS_WARN_IF(!aEvent->mInput.MakeOffsetAbsolute(selectionStart))) {
return NS_ERROR_FAILURE;
}
}
}
aEvent->mSucceeded = false;
aEvent->mReply.mContentsRoot = mRootContent.get();
aEvent->mReply.mHasSelection = !mSelection->IsCollapsed();
nsRect r;
nsIFrame* frame = nsCaret::GetGeometry(mSelection, &r);
if (!frame) {
frame = mRootContent->GetPrimaryFrame();
if (NS_WARN_IF(!frame)) {
return NS_ERROR_FAILURE;
}
}
aEvent->mReply.mFocusedWidget = frame->GetNearestWidget();
return NS_OK;
}
nsresult
ContentEventHandler::Init(WidgetSelectionEvent* aEvent)
{
NS_ASSERTION(aEvent, "aEvent must not be null");
nsresult rv = InitCommon();
NS_ENSURE_SUCCESS(rv, rv);
aEvent->mSucceeded = false;
return NS_OK;
}
nsIContent*
ContentEventHandler::GetFocusedContent()
{
nsIDocument* doc = mPresShell->GetDocument();
if (!doc) {
return nullptr;
}
nsCOMPtr<nsPIDOMWindowOuter> window = doc->GetWindow();
nsCOMPtr<nsPIDOMWindowOuter> focusedWindow;
return nsFocusManager::GetFocusedDescendant(window, true,
getter_AddRefs(focusedWindow));
}
bool
ContentEventHandler::IsPlugin(nsIContent* aContent)
{
return aContent &&
aContent->GetDesiredIMEState().mEnabled == IMEState::PLUGIN;
}
nsresult
ContentEventHandler::QueryContentRect(nsIContent* aContent,
WidgetQueryContentEvent* aEvent)
{
NS_PRECONDITION(aContent, "aContent must not be null");
nsIFrame* frame = aContent->GetPrimaryFrame();
NS_ENSURE_TRUE(frame, NS_ERROR_FAILURE);
// get rect for first frame
nsRect resultRect(nsPoint(0, 0), frame->GetRect().Size());
nsresult rv = ConvertToRootRelativeOffset(frame, resultRect);
NS_ENSURE_SUCCESS(rv, rv);
// account for any additional frames
while ((frame = frame->GetNextContinuation()) != nullptr) {
nsRect frameRect(nsPoint(0, 0), frame->GetRect().Size());
rv = ConvertToRootRelativeOffset(frame, frameRect);
NS_ENSURE_SUCCESS(rv, rv);
resultRect.UnionRect(resultRect, frameRect);
}
aEvent->mReply.mRect = LayoutDeviceIntRect::FromUnknownRect(
resultRect.ToOutsidePixels(mPresContext->AppUnitsPerDevPixel()));
// Returning empty rect may cause native IME confused, let's make sure to
// return non-empty rect.
EnsureNonEmptyRect(aEvent->mReply.mRect);
aEvent->mSucceeded = true;
return NS_OK;
}
// Editor places a bogus BR node under its root content if the editor doesn't
// have any text. This happens even for single line editors.
// When we get text content and when we change the selection,
// we don't want to include the bogus BRs at the end.
static bool IsContentBR(nsIContent* aContent)
{
return aContent->IsHTMLElement(nsGkAtoms::br) &&
!aContent->AttrValueIs(kNameSpaceID_None,
nsGkAtoms::type,
nsGkAtoms::moz,
eIgnoreCase) &&
!aContent->AttrValueIs(kNameSpaceID_None,
nsGkAtoms::mozeditorbogusnode,
nsGkAtoms::_true,
eIgnoreCase);
}
static bool IsMozBR(nsIContent* aContent)
{
return aContent->IsHTMLElement(nsGkAtoms::br) && !IsContentBR(aContent);
}
static void ConvertToNativeNewlines(nsAFlatString& aString)
{
#if defined(XP_WIN)
aString.ReplaceSubstring(NS_LITERAL_STRING("\n"), NS_LITERAL_STRING("\r\n"));
#endif
}
static void AppendString(nsAString& aString, nsIContent* aContent)
{
NS_ASSERTION(aContent->IsNodeOfType(nsINode::eTEXT),
"aContent is not a text node!");
const nsTextFragment* text = aContent->GetText();
if (!text) {
return;
}
text->AppendTo(aString);
}
static void AppendSubString(nsAString& aString, nsIContent* aContent,
uint32_t aXPOffset, uint32_t aXPLength)
{
NS_ASSERTION(aContent->IsNodeOfType(nsINode::eTEXT),
"aContent is not a text node!");
const nsTextFragment* text = aContent->GetText();
if (!text) {
return;
}
text->AppendTo(aString, int32_t(aXPOffset), int32_t(aXPLength));
}
#if defined(XP_WIN)
static uint32_t CountNewlinesInXPLength(nsIContent* aContent,
uint32_t aXPLength)
{
NS_ASSERTION(aContent->IsNodeOfType(nsINode::eTEXT),
"aContent is not a text node!");
const nsTextFragment* text = aContent->GetText();
if (!text) {
return 0;
}
// For automated tests, we should abort on debug build.
MOZ_ASSERT(aXPLength == UINT32_MAX || aXPLength <= text->GetLength(),
"aXPLength is out-of-bounds");
const uint32_t length = std::min(aXPLength, text->GetLength());
uint32_t newlines = 0;
for (uint32_t i = 0; i < length; ++i) {
if (text->CharAt(i) == '\n') {
++newlines;
}
}
return newlines;
}
static uint32_t CountNewlinesInNativeLength(nsIContent* aContent,
uint32_t aNativeLength)
{
NS_ASSERTION(aContent->IsNodeOfType(nsINode::eTEXT),
"aContent is not a text node!");
const nsTextFragment* text = aContent->GetText();
if (!text) {
return 0;
}
// For automated tests, we should abort on debug build.
MOZ_ASSERT(
(aNativeLength == UINT32_MAX || aNativeLength <= text->GetLength() * 2),
"aNativeLength is unexpected value");
const uint32_t xpLength = text->GetLength();
uint32_t newlines = 0;
for (uint32_t i = 0, nativeOffset = 0;
i < xpLength && nativeOffset < aNativeLength;
++i, ++nativeOffset) {
// For automated tests, we should abort on debug build.
MOZ_ASSERT(i < text->GetLength(), "i is out-of-bounds");
if (text->CharAt(i) == '\n') {
++newlines;
++nativeOffset;
}
}
return newlines;
}
#endif
/* static */ uint32_t
ContentEventHandler::GetNativeTextLength(nsIContent* aContent,
uint32_t aStartOffset,
uint32_t aEndOffset)
{
MOZ_ASSERT(aEndOffset >= aStartOffset,
"aEndOffset must be equals or larger than aStartOffset");
if (NS_WARN_IF(!aContent->IsNodeOfType(nsINode::eTEXT))) {
return 0;
}
if (aStartOffset == aEndOffset) {
return 0;
}
return GetTextLength(aContent, LINE_BREAK_TYPE_NATIVE, aEndOffset) -
GetTextLength(aContent, LINE_BREAK_TYPE_NATIVE, aStartOffset);
}
/* static */ uint32_t
ContentEventHandler::GetNativeTextLength(nsIContent* aContent,
uint32_t aMaxLength)
{
if (NS_WARN_IF(!aContent->IsNodeOfType(nsINode::eTEXT))) {
return 0;
}
return GetTextLength(aContent, LINE_BREAK_TYPE_NATIVE, aMaxLength);
}
/* static */ uint32_t
ContentEventHandler::GetNativeTextLengthBefore(nsIContent* aContent,
nsINode* aRootNode)
{
if (NS_WARN_IF(aContent->IsNodeOfType(nsINode::eTEXT))) {
return 0;
}
return ShouldBreakLineBefore(aContent, aRootNode) ?
GetBRLength(LINE_BREAK_TYPE_NATIVE) : 0;
}
/* static inline */ uint32_t
ContentEventHandler::GetBRLength(LineBreakType aLineBreakType)
{
#if defined(XP_WIN)
// Length of \r\n
return (aLineBreakType == LINE_BREAK_TYPE_NATIVE) ? 2 : 1;
#else
return 1;
#endif
}
/* static */ uint32_t
ContentEventHandler::GetTextLength(nsIContent* aContent,
LineBreakType aLineBreakType,
uint32_t aMaxLength)
{
MOZ_ASSERT(aContent->IsNodeOfType(nsINode::eTEXT));
uint32_t textLengthDifference =
#if defined(XP_WIN)
// On Windows, the length of a native newline ("\r\n") is twice the length
// of the XP newline ("\n"), so XP length is equal to the length of the
// native offset plus the number of newlines encountered in the string.
(aLineBreakType == LINE_BREAK_TYPE_NATIVE) ?
CountNewlinesInXPLength(aContent, aMaxLength) : 0;
#else
// On other platforms, the native and XP newlines are the same.
0;
#endif
const nsTextFragment* text = aContent->GetText();
if (!text) {
return 0;
}
uint32_t length = std::min(text->GetLength(), aMaxLength);
return length + textLengthDifference;
}
static uint32_t ConvertToXPOffset(nsIContent* aContent, uint32_t aNativeOffset)
{
#if defined(XP_WIN)
// On Windows, the length of a native newline ("\r\n") is twice the length of
// the XP newline ("\n"), so XP offset is equal to the length of the native
// offset minus the number of newlines encountered in the string.
return aNativeOffset - CountNewlinesInNativeLength(aContent, aNativeOffset);
#else
// On other platforms, the native and XP newlines are the same.
return aNativeOffset;
#endif
}
/* static */ bool
ContentEventHandler::ShouldBreakLineBefore(nsIContent* aContent,
nsINode* aRootNode)
{
// We don't need to append linebreak at the start of the root element.
if (aContent == aRootNode) {
return false;
}
// If it's not an HTML element (including other markup language's elements),
// we shouldn't insert like break before that for now. Becoming this is a
// problem must be edge case. E.g., when ContentEventHandler is used with
// MathML or SVG elements.
if (!aContent->IsHTMLElement()) {
return false;
}
// If the element is <br>, we need to check if the <br> is caused by web
// content. Otherwise, i.e., it's caused by internal reason of Gecko,
// it shouldn't be exposed as a line break to flatten text.
if (aContent->IsHTMLElement(nsGkAtoms::br)) {
return IsContentBR(aContent);
}
// Note that ideally, we should refer the style of the primary frame of
// aContent for deciding if it's an inline. However, it's difficult
// IMEContentObserver to notify IME of text change caused by style change.
// Therefore, currently, we should check only from the tag for now.
if (aContent->IsAnyOfHTMLElements(nsGkAtoms::a,
nsGkAtoms::abbr,
nsGkAtoms::acronym,
nsGkAtoms::b,
nsGkAtoms::bdi,
nsGkAtoms::bdo,
nsGkAtoms::big,
nsGkAtoms::cite,
nsGkAtoms::code,
nsGkAtoms::data,
nsGkAtoms::del,
nsGkAtoms::dfn,
nsGkAtoms::em,
nsGkAtoms::font,
nsGkAtoms::i,
nsGkAtoms::ins,
nsGkAtoms::kbd,
nsGkAtoms::mark,
nsGkAtoms::s,
nsGkAtoms::samp,
nsGkAtoms::small,
nsGkAtoms::span,
nsGkAtoms::strike,
nsGkAtoms::strong,
nsGkAtoms::sub,
nsGkAtoms::sup,
nsGkAtoms::time,
nsGkAtoms::tt,
nsGkAtoms::u,
nsGkAtoms::var)) {
return false;
}
// If the element is unknown element, we shouldn't insert line breaks before
// it since unknown elements should be ignored.
RefPtr<HTMLUnknownElement> unknownHTMLElement = do_QueryObject(aContent);
return !unknownHTMLElement;
}
nsresult
ContentEventHandler::GenerateFlatTextContent(nsIContent* aContent,
nsAFlatString& aString,
LineBreakType aLineBreakType)
{
MOZ_ASSERT(aString.IsEmpty());
RefPtr<nsRange> range = new nsRange(mRootContent);
ErrorResult rv;
range->SelectNodeContents(*aContent, rv);
if (NS_WARN_IF(rv.Failed())) {
return rv.StealNSResult();
}
return GenerateFlatTextContent(range, aString, aLineBreakType);
}
nsresult
ContentEventHandler::GenerateFlatTextContent(nsRange* aRange,
nsAFlatString& aString,
LineBreakType aLineBreakType)
{
MOZ_ASSERT(aString.IsEmpty());
if (aRange->Collapsed()) {
return NS_OK;
}
nsINode* startNode = aRange->GetStartParent();
nsINode* endNode = aRange->GetEndParent();
if (NS_WARN_IF(!startNode) || NS_WARN_IF(!endNode)) {
return NS_ERROR_FAILURE;
}
if (startNode == endNode && startNode->IsNodeOfType(nsINode::eTEXT)) {
nsIContent* content = startNode->AsContent();
AppendSubString(aString, content, aRange->StartOffset(),
aRange->EndOffset() - aRange->StartOffset());
ConvertToNativeNewlines(aString);
return NS_OK;
}
nsCOMPtr<nsIContentIterator> iter = NS_NewPreContentIterator();
nsresult rv = iter->Init(aRange);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
for (; !iter->IsDone(); iter->Next()) {
nsINode* node = iter->GetCurrentNode();
if (NS_WARN_IF(!node)) {
break;
}
if (!node->IsContent()) {
continue;
}
nsIContent* content = node->AsContent();
if (content->IsNodeOfType(nsINode::eTEXT)) {
if (content == startNode) {
AppendSubString(aString, content, aRange->StartOffset(),
content->TextLength() - aRange->StartOffset());
} else if (content == endNode) {
AppendSubString(aString, content, 0, aRange->EndOffset());
} else {
AppendString(aString, content);
}
} else if (ShouldBreakLineBefore(content, mRootContent)) {
aString.Append(char16_t('\n'));
}
}
if (aLineBreakType == LINE_BREAK_TYPE_NATIVE) {
ConvertToNativeNewlines(aString);
}
return NS_OK;
}
static FontRange*
AppendFontRange(nsTArray<FontRange>& aFontRanges, uint32_t aBaseOffset)
{
FontRange* fontRange = aFontRanges.AppendElement();
fontRange->mStartOffset = aBaseOffset;
return fontRange;
}
/* static */ uint32_t
ContentEventHandler::GetTextLengthInRange(nsIContent* aContent,
uint32_t aXPStartOffset,
uint32_t aXPEndOffset,
LineBreakType aLineBreakType)
{
MOZ_ASSERT(aContent->IsNodeOfType(nsINode::eTEXT));
return aLineBreakType == LINE_BREAK_TYPE_NATIVE ?
GetNativeTextLength(aContent, aXPStartOffset, aXPEndOffset) :
aXPEndOffset - aXPStartOffset;
}
/* static */ void
ContentEventHandler::AppendFontRanges(FontRangeArray& aFontRanges,
nsIContent* aContent,
int32_t aBaseOffset,
int32_t aXPStartOffset,
int32_t aXPEndOffset,
LineBreakType aLineBreakType)
{
MOZ_ASSERT(aContent->IsNodeOfType(nsINode::eTEXT));
nsIFrame* frame = aContent->GetPrimaryFrame();
if (!frame) {
// It is a non-rendered content, create an empty range for it.
AppendFontRange(aFontRanges, aBaseOffset);
return;
}
int32_t baseOffset = aBaseOffset;
nsTextFrame* curr = do_QueryFrame(frame);
MOZ_ASSERT(curr, "Not a text frame");
while (curr) {
int32_t frameXPStart = std::max(curr->GetContentOffset(), aXPStartOffset);
int32_t frameXPEnd = std::min(curr->GetContentEnd(), aXPEndOffset);
if (frameXPStart >= frameXPEnd) {
curr = static_cast<nsTextFrame*>(curr->GetNextContinuation());
continue;
}
gfxSkipCharsIterator iter = curr->EnsureTextRun(nsTextFrame::eInflated);
gfxTextRun* textRun = curr->GetTextRun(nsTextFrame::eInflated);
nsTextFrame* next = nullptr;
if (frameXPEnd < aXPEndOffset) {
next = static_cast<nsTextFrame*>(curr->GetNextContinuation());
while (next && next->GetTextRun(nsTextFrame::eInflated) == textRun) {
frameXPEnd = std::min(next->GetContentEnd(), aXPEndOffset);
next = frameXPEnd < aXPEndOffset ?
static_cast<nsTextFrame*>(next->GetNextContinuation()) : nullptr;
}
}
gfxTextRun::Range skipRange(iter.ConvertOriginalToSkipped(frameXPStart),
iter.ConvertOriginalToSkipped(frameXPEnd));
gfxTextRun::GlyphRunIterator runIter(textRun, skipRange);
int32_t lastXPEndOffset = frameXPStart;
while (runIter.NextRun()) {
gfxFont* font = runIter.GetGlyphRun()->mFont.get();
int32_t startXPOffset =
iter.ConvertSkippedToOriginal(runIter.GetStringStart());
// It is possible that the first glyph run has exceeded the frame,
// because the whole frame is filled by skipped chars.
if (startXPOffset >= frameXPEnd) {
break;
}
if (startXPOffset > lastXPEndOffset) {
// Create range for skipped leading chars.
AppendFontRange(aFontRanges, baseOffset);
baseOffset += GetTextLengthInRange(
aContent, lastXPEndOffset, startXPOffset, aLineBreakType);
lastXPEndOffset = startXPOffset;
}
FontRange* fontRange = AppendFontRange(aFontRanges, baseOffset);
fontRange->mFontName = font->GetName();
fontRange->mFontSize = font->GetAdjustedSize();
// The converted original offset may exceed the range,
// hence we need to clamp it.
int32_t endXPOffset =
iter.ConvertSkippedToOriginal(runIter.GetStringEnd());
endXPOffset = std::min(frameXPEnd, endXPOffset);
baseOffset += GetTextLengthInRange(aContent, startXPOffset, endXPOffset,
aLineBreakType);
lastXPEndOffset = endXPOffset;
}
if (lastXPEndOffset < frameXPEnd) {
// Create range for skipped trailing chars. It also handles case
// that the whole frame contains only skipped chars.
AppendFontRange(aFontRanges, baseOffset);
baseOffset += GetTextLengthInRange(
aContent, lastXPEndOffset, frameXPEnd, aLineBreakType);
}
curr = next;
}
}
nsresult
ContentEventHandler::GenerateFlatFontRanges(nsRange* aRange,
FontRangeArray& aFontRanges,
uint32_t& aLength,
LineBreakType aLineBreakType)
{
MOZ_ASSERT(aFontRanges.IsEmpty(), "aRanges must be empty array");
if (aRange->Collapsed()) {
return NS_OK;
}
nsINode* startNode = aRange->GetStartParent();
nsINode* endNode = aRange->GetEndParent();
if (NS_WARN_IF(!startNode) || NS_WARN_IF(!endNode)) {
return NS_ERROR_FAILURE;
}
// baseOffset is the flattened offset of each content node.
int32_t baseOffset = 0;
nsCOMPtr<nsIContentIterator> iter = NS_NewPreContentIterator();
nsresult rv = iter->Init(aRange);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
for (; !iter->IsDone(); iter->Next()) {
nsINode* node = iter->GetCurrentNode();
if (NS_WARN_IF(!node)) {
break;
}
if (!node->IsContent()) {
continue;
}
nsIContent* content = node->AsContent();
if (content->IsNodeOfType(nsINode::eTEXT)) {
int32_t startOffset = content != startNode ? 0 : aRange->StartOffset();
int32_t endOffset = content != endNode ?
content->TextLength() : aRange->EndOffset();
AppendFontRanges(aFontRanges, content, baseOffset,
startOffset, endOffset, aLineBreakType);
baseOffset += GetTextLengthInRange(content, startOffset, endOffset,
aLineBreakType);
} else if (ShouldBreakLineBefore(content, mRootContent)) {
if (aFontRanges.IsEmpty()) {
MOZ_ASSERT(baseOffset == 0);
FontRange* fontRange = AppendFontRange(aFontRanges, baseOffset);
nsIFrame* frame = content->GetPrimaryFrame();
if (frame) {
const nsFont& font = frame->GetParent()->StyleFont()->mFont;
const FontFamilyList& fontList = font.fontlist;
const FontFamilyName& fontName = fontList.IsEmpty() ?
FontFamilyName(fontList.GetDefaultFontType()) :
fontList.GetFontlist()[0];
fontName.AppendToString(fontRange->mFontName, false);
fontRange->mFontSize =
frame->PresContext()->AppUnitsToDevPixels(font.size);
}
}
baseOffset += GetBRLength(aLineBreakType);
}
}
aLength = baseOffset;
return NS_OK;
}
nsresult
ContentEventHandler::ExpandToClusterBoundary(nsIContent* aContent,
bool aForward,
uint32_t* aXPOffset)
{
// XXX This method assumes that the frame boundaries must be cluster
// boundaries. It's false, but no problem now, maybe.
if (!aContent->IsNodeOfType(nsINode::eTEXT) ||
*aXPOffset == 0 || *aXPOffset == aContent->TextLength()) {
return NS_OK;
}
NS_ASSERTION(*aXPOffset <= aContent->TextLength(),
"offset is out of range.");
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<nsFrameSelection> fs = mPresShell->FrameSelection();
int32_t offsetInFrame;
CaretAssociationHint hint =
aForward ? CARET_ASSOCIATE_BEFORE : CARET_ASSOCIATE_AFTER;
nsIFrame* frame = fs->GetFrameForNodeOffset(aContent, int32_t(*aXPOffset),
hint, &offsetInFrame);
if (!frame) {
// This content doesn't have any frames, we only can check surrogate pair...
const nsTextFragment* text = aContent->GetText();
NS_ENSURE_TRUE(text, NS_ERROR_FAILURE);
if (NS_IS_LOW_SURROGATE(text->CharAt(*aXPOffset)) &&
NS_IS_HIGH_SURROGATE(text->CharAt(*aXPOffset - 1))) {
*aXPOffset += aForward ? 1 : -1;
}
return NS_OK;
}
int32_t startOffset, endOffset;
nsresult rv = frame->GetOffsets(startOffset, endOffset);
NS_ENSURE_SUCCESS(rv, rv);
if (*aXPOffset == static_cast<uint32_t>(startOffset) ||
*aXPOffset == static_cast<uint32_t>(endOffset)) {
return NS_OK;
}
if (frame->GetType() != nsGkAtoms::textFrame) {
return NS_ERROR_FAILURE;
}
nsTextFrame* textFrame = static_cast<nsTextFrame*>(frame);
int32_t newOffsetInFrame = *aXPOffset - startOffset;
newOffsetInFrame += aForward ? -1 : 1;
textFrame->PeekOffsetCharacter(aForward, &newOffsetInFrame);
*aXPOffset = startOffset + newOffsetInFrame;
return NS_OK;
}
nsresult
ContentEventHandler::SetRangeFromFlatTextOffset(nsRange* aRange,
uint32_t aOffset,
uint32_t aLength,
LineBreakType aLineBreakType,
bool aExpandToClusterBoundaries,
uint32_t* aNewOffset,
nsIContent** aLastTextNode)
{
if (aNewOffset) {
*aNewOffset = aOffset;
}
if (aLastTextNode) {
*aLastTextNode = nullptr;
}
// Special case like <br contenteditable>
if (!mRootContent->HasChildren()) {
nsresult rv = aRange->SetStart(mRootContent, 0);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
rv = aRange->SetEnd(mRootContent, 0);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
}
nsCOMPtr<nsIContentIterator> iter = NS_NewPreContentIterator();
nsresult rv = iter->Init(mRootContent);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
uint32_t offset = 0;
uint32_t endOffset = aOffset + aLength;
bool startSet = false;
for (; !iter->IsDone(); iter->Next()) {
nsINode* node = iter->GetCurrentNode();
if (NS_WARN_IF(!node)) {
break;
}
// FYI: mRootContent shouldn't cause any text. So, we can skip it simply.
if (node == mRootContent || !node->IsContent()) {
continue;
}
nsIContent* content = node->AsContent();
if (aLastTextNode && content->IsNodeOfType(nsINode::eTEXT)) {
NS_IF_RELEASE(*aLastTextNode);
NS_ADDREF(*aLastTextNode = content);
}
uint32_t textLength =
content->IsNodeOfType(nsINode::eTEXT) ?
GetTextLength(content, aLineBreakType) :
(ShouldBreakLineBefore(content, mRootContent) ?
GetBRLength(aLineBreakType) : 0);
if (!textLength) {
continue;
}
// When the start offset is in between accumulated offset and the last
// offset of the node, the node is the start node of the range.
if (!startSet && aOffset <= offset + textLength) {
nsINode* startNode = nullptr;
int32_t startNodeOffset = -1;
if (content->IsNodeOfType(nsINode::eTEXT)) {
// Rule #1.1: [textNode or text[Node or textNode[
uint32_t xpOffset = aOffset - offset;
if (aLineBreakType == LINE_BREAK_TYPE_NATIVE) {
xpOffset = ConvertToXPOffset(content, xpOffset);
}
if (aExpandToClusterBoundaries) {
uint32_t oldXPOffset = xpOffset;
rv = ExpandToClusterBoundary(content, false, &xpOffset);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
if (aNewOffset) {
// This is correct since a cluster shouldn't include line break.
*aNewOffset -= (oldXPOffset - xpOffset);
}
}
startNode = content;
startNodeOffset = static_cast<int32_t>(xpOffset);
} else if (aOffset < offset + textLength) {
// Rule #1.2 [<element>
startNode = content->GetParent();
if (NS_WARN_IF(!startNode)) {
return NS_ERROR_FAILURE;
}
startNodeOffset = startNode->IndexOf(content);
if (NS_WARN_IF(startNodeOffset == -1)) {
// The content is being removed from the parent!
return NS_ERROR_FAILURE;
}
} else if (!content->HasChildren()) {
// Rule #1.3: <element/>[
startNode = content->GetParent();
if (NS_WARN_IF(!startNode)) {
return NS_ERROR_FAILURE;
}
startNodeOffset = startNode->IndexOf(content) + 1;
if (NS_WARN_IF(startNodeOffset == 0)) {
// The content is being removed from the parent!
return NS_ERROR_FAILURE;
}
} else {
// Rule #1.4: <element>[
startNode = content;
startNodeOffset = 0;
}
NS_ASSERTION(startNode, "startNode must not be nullptr");
NS_ASSERTION(startNodeOffset >= 0,
"startNodeOffset must not be negative");
rv = aRange->SetStart(startNode, startNodeOffset);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
startSet = true;
if (!aLength) {
rv = aRange->SetEnd(startNode, startNodeOffset);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
}
// When the end offset is in the content, the node is the end node of the
// range.
if (endOffset <= offset + textLength) {
MOZ_ASSERT(startSet,
"The start of the range should've been set already");
if (content->IsNodeOfType(nsINode::eTEXT)) {
// Rule #2.1: ]textNode or text]Node or textNode]
uint32_t xpOffset = endOffset - offset;
if (aLineBreakType == LINE_BREAK_TYPE_NATIVE) {
uint32_t xpOffsetCurrent = ConvertToXPOffset(content, xpOffset);
if (xpOffset && GetBRLength(aLineBreakType) > 1) {
MOZ_ASSERT(GetBRLength(aLineBreakType) == 2);
uint32_t xpOffsetPre = ConvertToXPOffset(content, xpOffset - 1);
// If previous character's XP offset is same as current character's,
// it means that the end offset is between \r and \n. So, the
// range end should be after the \n.
if (xpOffsetPre == xpOffsetCurrent) {
xpOffset = xpOffsetCurrent + 1;
} else {
xpOffset = xpOffsetCurrent;
}
}
}
if (aExpandToClusterBoundaries) {
rv = ExpandToClusterBoundary(content, true, &xpOffset);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
}
NS_ASSERTION(xpOffset <= INT32_MAX,
"The end node offset is too large");
rv = aRange->SetEnd(content, static_cast<int32_t>(xpOffset));
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
if (endOffset == offset) {
// Rule #2.2: ]<element>
// NOTE: Please don't crash on release builds because it must be
// overreaction but we shouldn't allow this bug when some
// automated tests find this.
MOZ_ASSERT(false, "This case should've already been handled at "
"the last node which caused some text");
return NS_ERROR_FAILURE;
}
if (content->HasChildren() &&
ShouldBreakLineBefore(content, mRootContent)) {
// Rule #2.3: </element>]
rv = aRange->SetEnd(content, 0);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
// Rule #2.4: <element/>]
nsINode* endNode = content->GetParent();
if (NS_WARN_IF(!endNode)) {
return NS_ERROR_FAILURE;
}
int32_t indexInParent = endNode->IndexOf(content);
if (NS_WARN_IF(indexInParent == -1)) {
// The content is being removed from the parent!
return NS_ERROR_FAILURE;
}
rv = aRange->SetEnd(endNode, indexInParent + 1);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
offset += textLength;
}
if (!startSet) {
MOZ_ASSERT(!mRootContent->IsNodeOfType(nsINode::eTEXT));
if (!offset) {
// Rule #1.5: <root>[</root>
// When there are no nodes causing text, the start of the DOM range
// should be start of the root node since clicking on such editor (e.g.,
// <div contenteditable><span></span></div>) sets caret to the start of
// the editor (i.e., before <span> in the example).
rv = aRange->SetStart(mRootContent, 0);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
if (!aLength) {
rv = aRange->SetEnd(mRootContent, 0);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
} else {
// Rule #1.5: [</root>
rv = aRange->SetStart(mRootContent,
static_cast<int32_t>(mRootContent->GetChildCount()));
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
}
if (aNewOffset) {
*aNewOffset = offset;
}
}
// Rule #2.5: ]</root>
rv = aRange->SetEnd(mRootContent,
static_cast<int32_t>(mRootContent->GetChildCount()));
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
/* static */ LineBreakType
ContentEventHandler::GetLineBreakType(WidgetQueryContentEvent* aEvent)
{
return GetLineBreakType(aEvent->mUseNativeLineBreak);
}
/* static */ LineBreakType
ContentEventHandler::GetLineBreakType(WidgetSelectionEvent* aEvent)
{
return GetLineBreakType(aEvent->mUseNativeLineBreak);
}
/* static */ LineBreakType
ContentEventHandler::GetLineBreakType(bool aUseNativeLineBreak)
{
return aUseNativeLineBreak ?
LINE_BREAK_TYPE_NATIVE : LINE_BREAK_TYPE_XP;
}
nsresult
ContentEventHandler::HandleQueryContentEvent(WidgetQueryContentEvent* aEvent)
{
switch (aEvent->mMessage) {
case eQuerySelectedText:
return OnQuerySelectedText(aEvent);
case eQueryTextContent:
return OnQueryTextContent(aEvent);
case eQueryCaretRect:
return OnQueryCaretRect(aEvent);
case eQueryTextRect:
return OnQueryTextRect(aEvent);
case eQueryTextRectArray:
return OnQueryTextRectArray(aEvent);
case eQueryEditorRect:
return OnQueryEditorRect(aEvent);
case eQueryContentState:
return OnQueryContentState(aEvent);
case eQuerySelectionAsTransferable:
return OnQuerySelectionAsTransferable(aEvent);
case eQueryCharacterAtPoint:
return OnQueryCharacterAtPoint(aEvent);
case eQueryDOMWidgetHittest:
return OnQueryDOMWidgetHittest(aEvent);
default:
return NS_ERROR_NOT_IMPLEMENTED;
}
return NS_OK;
}
// Similar to nsFrameSelection::GetFrameForNodeOffset,
// but this is more flexible for OnQueryTextRect to use
static nsresult GetFrameForTextRect(nsINode* aNode,
int32_t aNodeOffset,
bool aHint,
nsIFrame** aReturnFrame)
{
NS_ENSURE_TRUE(aNode && aNode->IsNodeOfType(nsINode::eCONTENT),
NS_ERROR_UNEXPECTED);
nsIContent* content = static_cast<nsIContent*>(aNode);
nsIFrame* frame = content->GetPrimaryFrame();
NS_ENSURE_TRUE(frame, NS_ERROR_FAILURE);
int32_t childNodeOffset = 0;
return frame->GetChildFrameContainingOffset(aNodeOffset, aHint,
&childNodeOffset, aReturnFrame);
}
nsresult
ContentEventHandler::OnQuerySelectedText(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
Bug 1275914 part.5 Support special selections at handling eQuerySelectedText r=smaug TextComposition needs to query first IME selection. Therefore, we need to add support to query special selection range with eQuerySelectedText. First, WidgetQueryContentEvent::mInput should have mSelectionType which should be initialized with InitForQuerySelectedText() but unfortunately, there is no method for eQuerySelectedText. Therefore, this patch initializes WidgetQueryContentEvent::mInit::mSelectionType with SelectionType::eNormal in its constructor. Next, ContentEventHandler needs to support to handle eQuerySelectedText with special selection types. However, we need to create 2 paths in some cases, one is for normal selection and the other is for special selections because there are no selection ranges may be usual case for special selections but not so for normal selection. Therefore, ContentEventHandler::InitCommon() becomes a little bit more complicated. ContentEventHandler::mSelection and ContentEventHandler::mFirstSelectedRange is initialized with the specified selection type but normal selection type is also necessary to compute the selection root since the selection root is computed from the first selected range which may not be anywhere if its selection type is not normal. Finally, ContentEventHandler::OnQuerySelectedText() returns "there are no selections" as succeeded case in special selection type cases. MozReview-Commit-ID: 9WzUx8b5piw --HG-- extra : rebase_source : fb07b40748b594d36315f1fc21c0a02ff9f668bb
2016-06-21 07:13:11 +03:00
if (!mFirstSelectedRange) {
MOZ_ASSERT(aEvent->mInput.mSelectionType != SelectionType::eNormal);
MOZ_ASSERT(aEvent->mReply.mOffset == WidgetQueryContentEvent::NOT_FOUND);
MOZ_ASSERT(aEvent->mReply.mString.IsEmpty());
MOZ_ASSERT(!aEvent->mReply.mHasSelection);
aEvent->mSucceeded = true;
return NS_OK;
}
nsINode* const startNode = mFirstSelectedRange->GetStartParent();
nsINode* const endNode = mFirstSelectedRange->GetEndParent();
// Make sure the selection is within the root content range.
if (!nsContentUtils::ContentIsDescendantOf(startNode, mRootContent) ||
!nsContentUtils::ContentIsDescendantOf(endNode, mRootContent)) {
return NS_ERROR_NOT_AVAILABLE;
}
NS_ASSERTION(aEvent->mReply.mString.IsEmpty(),
"The reply string must be empty");
LineBreakType lineBreakType = GetLineBreakType(aEvent);
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
rv = GetStartOffset(mFirstSelectedRange,
&aEvent->mReply.mOffset, lineBreakType);
NS_ENSURE_SUCCESS(rv, rv);
nsCOMPtr<nsINode> anchorNode, focusNode;
int32_t anchorOffset = 0, focusOffset = 0;
if (mSelection->RangeCount()) {
// If there is only one selection range, the anchor/focus node and offset
// are the information of the range. Therefore, we have the direction
// information.
if (mSelection->RangeCount() == 1) {
anchorNode = mSelection->GetAnchorNode();
focusNode = mSelection->GetFocusNode();
if (NS_WARN_IF(!anchorNode) || NS_WARN_IF(!focusNode)) {
return NS_ERROR_FAILURE;
}
anchorOffset = static_cast<int32_t>(mSelection->AnchorOffset());
focusOffset = static_cast<int32_t>(mSelection->FocusOffset());
if (NS_WARN_IF(anchorOffset < 0) || NS_WARN_IF(focusOffset < 0)) {
return NS_ERROR_FAILURE;
}
int16_t compare = nsContentUtils::ComparePoints(anchorNode, anchorOffset,
focusNode, focusOffset);
aEvent->mReply.mReversed = compare > 0;
}
// However, if there are 2 or more selection ranges, we have no information
// of that.
else {
aEvent->mReply.mReversed = false;
}
if (!mFirstSelectedRange->Collapsed()) {
rv = GenerateFlatTextContent(mFirstSelectedRange, aEvent->mReply.mString,
lineBreakType);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
} else {
aEvent->mReply.mString.Truncate();
}
} else {
NS_ASSERTION(mFirstSelectedRange->Collapsed(),
"When mSelection doesn't have selection, mFirstSelectedRange must be "
"collapsed");
anchorNode = focusNode = mFirstSelectedRange->GetStartParent();
if (NS_WARN_IF(!anchorNode)) {
return NS_ERROR_FAILURE;
}
anchorOffset = focusOffset =
static_cast<int32_t>(mFirstSelectedRange->StartOffset());
if (NS_WARN_IF(anchorOffset < 0)) {
return NS_ERROR_FAILURE;
}
aEvent->mReply.mReversed = false;
aEvent->mReply.mString.Truncate();
}
nsIFrame* frame = nullptr;
rv = GetFrameForTextRect(focusNode, focusOffset, true, &frame);
if (NS_SUCCEEDED(rv) && frame) {
aEvent->mReply.mWritingMode = frame->GetWritingMode();
} else {
aEvent->mReply.mWritingMode = WritingMode();
}
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryTextContent(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
NS_ASSERTION(aEvent->mReply.mString.IsEmpty(),
"The reply string must be empty");
LineBreakType lineBreakType = GetLineBreakType(aEvent);
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<nsRange> range = new nsRange(mRootContent);
rv = SetRangeFromFlatTextOffset(range, aEvent->mInput.mOffset,
aEvent->mInput.mLength, lineBreakType, false,
&aEvent->mReply.mOffset);
NS_ENSURE_SUCCESS(rv, rv);
rv = GenerateFlatTextContent(range, aEvent->mReply.mString, lineBreakType);
NS_ENSURE_SUCCESS(rv, rv);
if (aEvent->mWithFontRanges) {
uint32_t fontRangeLength;
rv = GenerateFlatFontRanges(range, aEvent->mReply.mFontRanges,
fontRangeLength, lineBreakType);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
MOZ_ASSERT(fontRangeLength == aEvent->mReply.mString.Length(),
"Font ranges doesn't match the string");
}
aEvent->mSucceeded = true;
return NS_OK;
}
void
ContentEventHandler::EnsureNonEmptyRect(nsRect& aRect) const
{
// See the comment in ContentEventHandler.h why this doesn't set them to
// one device pixel.
aRect.height = std::max(1, aRect.height);
aRect.width = std::max(1, aRect.width);
}
void
ContentEventHandler::EnsureNonEmptyRect(LayoutDeviceIntRect& aRect) const
{
aRect.height = std::max(1, aRect.height);
aRect.width = std::max(1, aRect.width);
}
ContentEventHandler::NodePosition
ContentEventHandler::GetNodePositionHavingFlatText(
const NodePosition& aNodePosition)
{
return GetNodePositionHavingFlatText(aNodePosition.mNode,
aNodePosition.mOffset);
}
ContentEventHandler::NodePosition
ContentEventHandler::GetNodePositionHavingFlatText(nsINode* aNode,
int32_t aNodeOffset)
{
if (aNode->IsNodeOfType(nsINode::eTEXT)) {
return NodePosition(aNode, aNodeOffset);
}
int32_t childCount = static_cast<int32_t>(aNode->GetChildCount());
// If it's a empty element node, returns itself.
if (!childCount) {
MOZ_ASSERT(!aNodeOffset || aNodeOffset == 1);
return NodePosition(aNode, aNodeOffset);
}
// If there is a node at given position, return the start of it.
if (aNodeOffset < childCount) {
return NodePosition(aNode->GetChildAt(aNodeOffset), 0);
}
// If the offset represents "after" the node, we need to return the last
// child of it. For example, if a range is |<p>[<br>]</p>|, then, the
// end point is {<p>, 1}. In such case, callers need the <br> node.
if (aNodeOffset == childCount) {
NodePosition result;
result.mNode = aNode->GetChildAt(childCount - 1);
result.mOffset = result.mNode->IsNodeOfType(nsINode::eTEXT) ?
static_cast<int32_t>(result.mNode->AsContent()->TextLength()) : 1;
}
NS_WARNING("aNodeOffset is invalid value");
return NodePosition();
}
ContentEventHandler::FrameAndNodeOffset
ContentEventHandler::GetFirstFrameInRangeForTextRect(nsRange* aRange)
{
NodePosition nodePosition;
nsCOMPtr<nsIContentIterator> iter = NS_NewPreContentIterator();
for (iter->Init(aRange); !iter->IsDone(); iter->Next()) {
nsINode* node = iter->GetCurrentNode();
if (NS_WARN_IF(!node)) {
break;
}
if (!node->IsContent()) {
continue;
}
if (node->IsNodeOfType(nsINode::eTEXT)) {
// If the range starts at the end of a text node, we need to find
// next node which causes text.
int32_t offsetInNode =
node == aRange->GetStartParent() ? aRange->StartOffset() : 0;
if (static_cast<uint32_t>(offsetInNode) < node->Length()) {
nodePosition.mNode = node;
nodePosition.mOffset = offsetInNode;
break;
}
continue;
}
// If the element node causes a line break before it, it's the first
// node causing text.
if (ShouldBreakLineBefore(node->AsContent(), mRootContent) ||
IsMozBR(node->AsContent())) {
nodePosition.mNode = node;
nodePosition.mOffset = 0;
}
}
if (!nodePosition.IsValid()) {
return FrameAndNodeOffset();
}
nsIFrame* firstFrame = nullptr;
GetFrameForTextRect(nodePosition.mNode, nodePosition.mOffset,
true, &firstFrame);
return FrameAndNodeOffset(firstFrame, nodePosition.mOffset);
}
ContentEventHandler::FrameAndNodeOffset
ContentEventHandler::GetLastFrameInRangeForTextRect(nsRange* aRange)
{
NodePosition nodePosition;
nsCOMPtr<nsIContentIterator> iter = NS_NewPreContentIterator();
iter->Init(aRange);
nsINode* endNode = aRange->GetEndParent();
uint32_t endOffset = static_cast<uint32_t>(aRange->EndOffset());
// If the end point is start of a text node or specified by its parent and
// index, the node shouldn't be included into the range. For example,
// with this case, |<p>abc[<br>]def</p>|, the range ends at 3rd children of
// <p> (see the range creation rules, "2.4. Cases: <element/>]"). This causes
// following frames:
// +----+-----+
// | abc|[<br>|
// +----+-----+
// +----+
// |]def|
// +----+
// So, if this method includes the 2nd text frame's rect to its result, the
// caller will return too tall rect which includes 2 lines in this case isn't
// expected by native IME (e.g., popup of IME will be positioned at bottom
// of "d" instead of right-bottom of "c"). Therefore, this method shouldn't
// include the last frame when its content isn't really in aRange.
nsINode* nextNodeOfRangeEnd = nullptr;
if (endNode->IsNodeOfType(nsINode::eTEXT)) {
// Don't set nextNodeOfRangeEnd to the start node of aRange because if
// endNode is same as start node of the range, the text node shouldn't be
// next of range end even if the offset is 0. This could occur with empty
// text node.
if (!endOffset && aRange->GetStartParent() != endNode) {
nextNodeOfRangeEnd = endNode;
}
} else if (endOffset < endNode->GetChildCount()) {
nextNodeOfRangeEnd = endNode->GetChildAt(endOffset);
}
for (iter->Last(); !iter->IsDone(); iter->Prev()) {
nsINode* node = iter->GetCurrentNode();
if (NS_WARN_IF(!node)) {
break;
}
if (!node->IsContent() || node == nextNodeOfRangeEnd) {
continue;
}
if (node->IsNodeOfType(nsINode::eTEXT)) {
nodePosition.mNode = node;
if (node == aRange->GetEndParent()) {
nodePosition.mOffset = aRange->EndOffset();
} else {
nodePosition.mOffset = node->Length();
}
// If the text node is empty or the last node of the range but the index
// is 0, we should store current position but continue looking for
// previous node (If there are no nodes before it, we should use current
// node position for returning its frame).
if (!nodePosition.mOffset) {
continue;
}
break;
}
if (ShouldBreakLineBefore(node->AsContent(), mRootContent) ||
IsMozBR(node->AsContent())) {
nodePosition.mNode = node;
nodePosition.mOffset = 0;
break;
}
}
if (!nodePosition.IsValid()) {
return FrameAndNodeOffset();
}
nsIFrame* lastFrame = nullptr;
GetFrameForTextRect(nodePosition.mNode, nodePosition.mOffset,
true, &lastFrame);
if (!lastFrame) {
return FrameAndNodeOffset();
}
// If the last frame is a text frame, we need to check if the range actually
// includes at least one character in the range. Therefore, if it's not a
// text frame, we need to do nothing anymore.
if (lastFrame->GetType() != nsGkAtoms::textFrame) {
return FrameAndNodeOffset(lastFrame, nodePosition.mOffset);
}
int32_t start, end;
if (NS_WARN_IF(NS_FAILED(lastFrame->GetOffsets(start, end)))) {
return FrameAndNodeOffset();
}
// If the start offset in the node is same as the computed offset in the
// node and it's not 0, the frame shouldn't be added to the text rect. So,
// this should return previous text frame and its last offset if there is
// at least one text frame.
if (nodePosition.mOffset && nodePosition.mOffset == start) {
GetFrameForTextRect(nodePosition.mNode, --nodePosition.mOffset,
true, &lastFrame);
if (NS_WARN_IF(!lastFrame)) {
return FrameAndNodeOffset();
}
}
return FrameAndNodeOffset(lastFrame, nodePosition.mOffset);
}
ContentEventHandler::FrameRelativeRect
ContentEventHandler::GetLineBreakerRectBefore(nsIFrame* aFrame)
{
// Note that this method should be called only with an element's frame whose
// open tag causes a line break or moz-<br> for computing empty last line's
// rect.
MOZ_ASSERT(ShouldBreakLineBefore(aFrame->GetContent(), mRootContent) ||
IsMozBR(aFrame->GetContent()));
nsIFrame* frameForFontMetrics = aFrame;
// If it's not a <br> frame, this method computes the line breaker's rect
// outside the frame. Therefore, we need to compute with parent frame's
// font metrics in such case.
if (aFrame->GetType() != nsGkAtoms::brFrame && aFrame->GetParent()) {
frameForFontMetrics = aFrame->GetParent();
}
// Note that <br> element's rect is decided with line-height but we need
// a rect only with font height. Additionally, <br> frame's width and
// height are 0 in quirks mode if it's not an empty line. So, we cannot
// use frame rect information even if it's a <br> frame.
FrameRelativeRect result(aFrame);
RefPtr<nsFontMetrics> fontMetrics =
nsLayoutUtils::GetInflatedFontMetricsForFrame(frameForFontMetrics);
if (NS_WARN_IF(!fontMetrics)) {
return FrameRelativeRect();
}
const WritingMode kWritingMode = frameForFontMetrics->GetWritingMode();
nscoord baseline = aFrame->GetCaretBaseline();
if (kWritingMode.IsVertical()) {
if (kWritingMode.IsLineInverted()) {
result.mRect.x = baseline - fontMetrics->MaxDescent();
} else {
result.mRect.x = baseline - fontMetrics->MaxAscent();
}
result.mRect.width = fontMetrics->MaxHeight();
} else {
result.mRect.y = baseline - fontMetrics->MaxAscent();
result.mRect.height = fontMetrics->MaxHeight();
}
// If aFrame isn't a <br> frame, caret should be at outside of it because
// the line break is before its open tag. For example, case of
// |<div><p>some text</p></div>|, caret is before <p> element and in <div>
// element, the caret should be left of top-left corner of <p> element like:
//
// +-<div>------------------- <div>'s border box
// | I +-<p>----------------- <p>'s border box
// | I |
// | I |
// | |
// ^- caret
//
// However, this is a hack for unusual scenario. This hack shouldn't be
// used as far as possible.
if (aFrame->GetType() != nsGkAtoms::brFrame) {
if (kWritingMode.IsVertical()) {
if (kWritingMode.IsLineInverted()) {
// above of top-left corner of aFrame.
result.mRect.x = 0;
} else {
// above of top-right corner of aFrame.
result.mRect.x = aFrame->GetRect().XMost() - result.mRect.width;
}
result.mRect.y = -mPresContext->AppUnitsPerDevPixel();
} else {
// left of top-left corner of aFrame.
result.mRect.x = -mPresContext->AppUnitsPerDevPixel();
result.mRect.y = 0;
}
}
return result;
}
ContentEventHandler::FrameRelativeRect
ContentEventHandler::GuessLineBreakerRectAfter(nsIContent* aTextContent)
{
// aTextContent should be a text node.
MOZ_ASSERT(aTextContent->IsNodeOfType(nsINode::eTEXT));
FrameRelativeRect result;
int32_t length = static_cast<int32_t>(aTextContent->Length());
if (NS_WARN_IF(length < 0)) {
return result;
}
// Get the last nsTextFrame which is caused by aTextContent. Note that
// a text node can cause multiple text frames, e.g., the text is too long
// and wrapped by its parent block or the text has line breakers and its
// white-space property respects the line breakers (e.g., |pre|).
nsIFrame* lastTextFrame = nullptr;
nsresult rv = GetFrameForTextRect(aTextContent, length, true, &lastTextFrame);
if (NS_WARN_IF(NS_FAILED(rv)) || NS_WARN_IF(!lastTextFrame)) {
return result;
}
const nsRect kLastTextFrameRect = lastTextFrame->GetRect();
if (lastTextFrame->GetWritingMode().IsVertical()) {
// Below of the last text frame.
result.mRect.SetRect(0, kLastTextFrameRect.height,
kLastTextFrameRect.width, 0);
} else {
// Right of the last text frame (not bidi-aware).
result.mRect.SetRect(kLastTextFrameRect.width, 0,
0, kLastTextFrameRect.height);
}
result.mBaseFrame = lastTextFrame;
return result;
}
ContentEventHandler::FrameRelativeRect
ContentEventHandler::GuessFirstCaretRectIn(nsIFrame* aFrame)
{
const WritingMode kWritingMode = aFrame->GetWritingMode();
// Computes the font height, but if it's not available, we should use
// default font size of Firefox. The default font size in default settings
// is 16px.
RefPtr<nsFontMetrics> fontMetrics =
nsLayoutUtils::GetInflatedFontMetricsForFrame(aFrame);
const nscoord kMaxHeight =
fontMetrics ? fontMetrics->MaxHeight() :
16 * mPresContext->AppUnitsPerDevPixel();
nsRect caretRect;
const nsRect kContentRect = aFrame->GetContentRect() - aFrame->GetPosition();
caretRect.y = kContentRect.y;
if (!kWritingMode.IsVertical()) {
if (kWritingMode.IsBidiLTR()) {
caretRect.x = kContentRect.x;
} else {
// Move 1px left for the space of caret itself.
const nscoord kOnePixel = mPresContext->AppUnitsPerDevPixel();
caretRect.x = kContentRect.XMost() - kOnePixel;
}
caretRect.height = kMaxHeight;
// However, don't add kOnePixel here because it may cause 2px width at
// aligning the edge to device pixels.
caretRect.width = 1;
} else {
if (kWritingMode.IsVerticalLR()) {
caretRect.x = kContentRect.x;
} else {
caretRect.x = kContentRect.XMost() - kMaxHeight;
}
caretRect.width = kMaxHeight;
// Don't add app units for a device pixel because it may cause 2px height
// at aligning the edge to device pixels.
caretRect.height = 1;
}
return FrameRelativeRect(caretRect, aFrame);
}
nsresult
ContentEventHandler::OnQueryTextRectArray(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
LineBreakType lineBreakType = GetLineBreakType(aEvent);
const uint32_t kBRLength = GetBRLength(lineBreakType);
RefPtr<nsRange> range = new nsRange(mRootContent);
bool isVertical = false;
LayoutDeviceIntRect rect;
uint32_t offset = aEvent->mInput.mOffset;
const uint32_t kEndOffset = offset + aEvent->mInput.mLength;
bool wasLineBreaker = false;
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
// lastCharRect stores the last charRect value (see below for the detail of
// charRect).
nsRect lastCharRect;
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
// lastFrame is base frame of lastCharRect.
nsIFrame* lastFrame = nullptr;
while (offset < kEndOffset) {
nsCOMPtr<nsIContent> lastTextContent;
rv = SetRangeFromFlatTextOffset(range, offset, 1, lineBreakType, true,
nullptr, getter_AddRefs(lastTextContent));
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
// If the range is collapsed, offset has already reached the end of the
// contents.
if (range->Collapsed()) {
break;
}
// Get the first frame which causes some text after the offset.
FrameAndNodeOffset firstFrame = GetFirstFrameInRangeForTextRect(range);
// If GetFirstFrameInRangeForTextRect() does not return valid frame, that
// means that there are no visible frames having text or the offset reached
// the end of contents.
if (!firstFrame.IsValid()) {
nsAutoString allText;
rv = GenerateFlatTextContent(mRootContent, allText, lineBreakType);
// If the offset doesn't reach the end of contents yet but there is no
// frames for the node, that means that current offset's node is hidden
// by CSS or something. Ideally, we should handle it with the last
// visible text node's last character's rect, but it's not usual cases
// in actual web services. Therefore, currently, we should make this
// case fail.
if (NS_WARN_IF(NS_FAILED(rv)) || offset < allText.Length()) {
return NS_ERROR_FAILURE;
}
// Otherwise, we should append caret rect at the end of the contents
// later.
break;
}
nsIContent* firstContent = firstFrame.mFrame->GetContent();
if (NS_WARN_IF(!firstContent)) {
return NS_ERROR_FAILURE;
}
bool startsBetweenLineBreaker = false;
nsAutoString chars;
// XXX not bidi-aware this class...
isVertical = firstFrame->GetWritingMode().IsVertical();
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
nsIFrame* baseFrame = firstFrame;
// charRect should have each character rect or line breaker rect relative
// to the base frame.
AutoTArray<nsRect, 16> charRects;
// If the first frame is a text frame, the result should be computed with
// the frame's API.
if (firstFrame->GetType() == nsGkAtoms::textFrame) {
rv = firstFrame->GetCharacterRectsInRange(firstFrame.mOffsetInNode,
kEndOffset - offset, charRects);
if (NS_WARN_IF(NS_FAILED(rv)) || NS_WARN_IF(charRects.IsEmpty())) {
return rv;
}
// Assign the characters whose rects are computed by the call of
// nsTextFrame::GetCharacterRectsInRange().
AppendSubString(chars, firstContent, firstFrame.mOffsetInNode,
charRects.Length());
if (NS_WARN_IF(chars.Length() != charRects.Length())) {
return NS_ERROR_UNEXPECTED;
}
if (kBRLength > 1 && chars[0] == '\n' &&
offset == aEvent->mInput.mOffset && offset) {
// If start of range starting from previous offset of query range is
// same as the start of query range, the query range starts from
// between a line breaker (i.e., the range starts between "\r" and
// "\n").
RefPtr<nsRange> rangeToPrevOffset = new nsRange(mRootContent);
rv = SetRangeFromFlatTextOffset(rangeToPrevOffset,
aEvent->mInput.mOffset - 1, 1,
lineBreakType, true, nullptr);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
startsBetweenLineBreaker =
range->GetStartParent() == rangeToPrevOffset->GetStartParent() &&
range->StartOffset() == rangeToPrevOffset->StartOffset();
}
}
// Other contents should cause a line breaker rect before it.
// Note that moz-<br> element does not cause any text, however,
// it represents empty line at the last of current block. Therefore,
// we need to compute its rect too.
else if (ShouldBreakLineBefore(firstContent, mRootContent) ||
IsMozBR(firstContent)) {
nsRect brRect;
// If the frame is not a <br> frame, we need to compute the caret rect
// with last character's rect before firstContent if there is.
// For example, if caret is after "c" of |<p>abc</p><p>def</p>|, IME may
// query a line breaker's rect after "c". Then, if we compute it only
// with the 2nd <p>'s block frame, the result will be:
// +-<p>--------------------------------+
// |abc |
// +------------------------------------+
//
// I+-<p>--------------------------------+
// |def |
// +------------------------------------+
// However, users expect popup windows of IME should be positioned at
// right-bottom of "c" like this:
// +-<p>--------------------------------+
// |abcI |
// +------------------------------------+
//
// +-<p>--------------------------------+
// |def |
// +------------------------------------+
// Therefore, if the first frame isn't a <br> frame and there is a text
// node before the first node in the queried range, we should compute the
// first rect with the previous character's rect.
// If we already compute a character's rect in the queried range, we can
// compute it with the cached last character's rect. (However, don't
// use this path if it's a <br> frame because trusting <br> frame's rect
// is better than guessing the rect from the previous character.)
if (firstFrame->GetType() != nsGkAtoms::brFrame &&
aEvent->mInput.mOffset != offset) {
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
baseFrame = lastFrame;
brRect = lastCharRect;
if (!wasLineBreaker) {
if (isVertical) {
// Right of the last character.
brRect.y = brRect.YMost() + 1;
brRect.height = 1;
} else {
// Under the last character.
brRect.x = brRect.XMost() + 1;
brRect.width = 1;
}
}
}
// If it's not a <br> frame and it's the first character rect at the
// queried range, we need to the previous character of the start of
// the queried range if there is a text node.
else if (firstFrame->GetType() != nsGkAtoms::brFrame && lastTextContent) {
FrameRelativeRect brRectRelativeToLastTextFrame =
GuessLineBreakerRectAfter(lastTextContent);
if (NS_WARN_IF(!brRectRelativeToLastTextFrame.IsValid())) {
return NS_ERROR_FAILURE;
}
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
// Look for the last text frame for lastTextContent.
nsIFrame* primaryFrame = lastTextContent->GetPrimaryFrame();
if (NS_WARN_IF(!primaryFrame)) {
return NS_ERROR_FAILURE;
}
baseFrame = primaryFrame->LastContinuation();
if (NS_WARN_IF(!baseFrame)) {
return NS_ERROR_FAILURE;
}
brRect = brRectRelativeToLastTextFrame.RectRelativeTo(baseFrame);
}
// Otherwise, we need to compute the line breaker's rect only with the
// first frame's rect. But this may be unexpected. For example,
// |<div contenteditable>[<p>]abc</p></div>|. In this case, caret is
// before "a", therefore, users expect the rect left of "a". However,
// we don't have enough information about the next character here and
// this isn't usual case (e.g., IME typically tries to query the rect
// of "a" or caret rect for computing its popup position). Therefore,
// we shouldn't do more complicated hack here unless we'll get some bug
// reports actually.
else {
FrameRelativeRect relativeBRRect = GetLineBreakerRectBefore(firstFrame);
brRect = relativeBRRect.RectRelativeTo(firstFrame);
}
charRects.AppendElement(brRect);
chars.AssignLiteral("\n");
if (kBRLength > 1 && offset == aEvent->mInput.mOffset && offset) {
// If the first frame for the previous offset of the query range and
// the first frame for the start of query range are same, that means
// the start offset is between the first line breaker (i.e., the range
// starts between "\r" and "\n").
rv = SetRangeFromFlatTextOffset(range, aEvent->mInput.mOffset - 1, 1,
lineBreakType, true, nullptr);
if (NS_WARN_IF(NS_FAILED(rv))) {
return NS_ERROR_UNEXPECTED;
}
FrameAndNodeOffset frameForPrevious =
GetFirstFrameInRangeForTextRect(range);
startsBetweenLineBreaker = frameForPrevious.mFrame == firstFrame.mFrame;
}
} else {
NS_WARNING("The frame is neither a text frame nor a frame whose content "
"causes a line break");
return NS_ERROR_FAILURE;
}
for (size_t i = 0; i < charRects.Length() && offset < kEndOffset; i++) {
nsRect charRect = charRects[i];
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
// Store lastCharRect before applying CSS transform because it may be
// used for computing a line breaker rect. Then, the computed line
// breaker rect will be applied CSS transform again. Therefore,
// the value of lastCharRect should be raw rect value relative to the
// base frame.
lastCharRect = charRect;
Bug 1296217 ContentEventHandler::OnQueryTextRectArray() should apply CSS transform to each character or line breaker rect r=smaug Currently, ContentEventHandler::OnQueryTextRectArray() is used only in e10s mode (at caching necessary character rects in ContentCacheInChild). Therefore, this bug occurs only in e10s mode. ContentEventHandler::OnQueryTextRectArray() applies CSS transform only to each frame rect. Therefore, character rect's width and height are not applied. This patch makes the loop apply CSS transform to each character or line breaker rect (i.e., each item of charRects). Then, we need to rewrite the lastCharRect hack. It stores the last charRect value for computing next line breaker rect if next line breaker is caused by a block level element or something, i.e., not caused by a <br> frame. So, when brRect is computed with lastCharRect, the loop needs to apply CSS transform of the last text frame to the following brRect because it tries to compute a caret rect immediately after the last character. For doing this, this patch adds lastFrame which stores the last frame for lastCharRect and set it to baseFrame. Then, at applying CSS transform to each charRect, it can apply CSS transform of expected frame. Similarly, when brRect is computed with last text frame, this patch looks for the last text frame from lastTextContent and use it as base frame to apply to CSS transform. MozReview-Commit-ID: 5Yr2HMrooHd --HG-- extra : rebase_source : fa643f442036d9167a1df3a4383b0802a1729a3e
2016-09-06 05:46:31 +03:00
lastFrame = baseFrame;
rv = ConvertToRootRelativeOffset(baseFrame, charRect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
rect = LayoutDeviceIntRect::FromUnknownRect(
charRect.ToOutsidePixels(mPresContext->AppUnitsPerDevPixel()));
// Returning empty rect may cause native IME confused, let's make sure to
// return non-empty rect.
EnsureNonEmptyRect(rect);
aEvent->mReply.mRectArray.AppendElement(rect);
offset++;
// If it's not a line breaker or the line breaker length is same as
// XP line breaker's, we need to do nothing for current character.
wasLineBreaker = chars[i] == '\n';
if (!wasLineBreaker || kBRLength == 1) {
continue;
}
MOZ_ASSERT(kBRLength == 2);
// If it's already reached the end of query range, we don't need to do
// anymore.
if (offset == kEndOffset) {
break;
}
// If the query range starts from between a line breaker, i.e., it starts
// between "\r" and "\n", the appended rect was for the "\n". Therefore,
// we don't need to append same rect anymore for current "\r\n".
if (startsBetweenLineBreaker) {
continue;
}
// The appended rect was for "\r" of "\r\n". Therefore, we need to
// append same rect for "\n" too because querying rect of "\r" and "\n"
// should return same rect. E.g., IME may query previous character's
// rect of first character of a line.
aEvent->mReply.mRectArray.AppendElement(rect);
offset++;
}
}
// If the query range is longer than actual content length, we should append
// caret rect at the end of the content as the last character rect because
// native IME may want to query character rect at the end of contents for
// deciding the position of a popup window (e.g., suggest window for next
// word). Note that when this method hasn't appended character rects, it
// means that the offset is too large or the query range is collapsed.
if (offset < kEndOffset || aEvent->mReply.mRectArray.IsEmpty()) {
// If we've already retrieved some character rects before current offset,
// we can guess the last rect from the last character's rect unless it's a
// line breaker. (If it's a line breaker, the caret rect is in next line.)
if (!aEvent->mReply.mRectArray.IsEmpty() && !wasLineBreaker) {
rect = aEvent->mReply.mRectArray.LastElement();
if (isVertical) {
rect.y = rect.YMost() + 1;
rect.height = 1;
MOZ_ASSERT(rect.width);
} else {
rect.x = rect.XMost() + 1;
rect.width = 1;
MOZ_ASSERT(rect.height);
}
aEvent->mReply.mRectArray.AppendElement(rect);
} else {
// Note that don't use eQueryCaretRect here because if caret is at the
// end of the content, it returns actual caret rect instead of computing
// the rect itself. It means that the result depends on caret position.
// So, we shouldn't use it for consistency result in automated tests.
WidgetQueryContentEvent queryTextRect(eQueryTextRect, *aEvent);
WidgetQueryContentEvent::Options options(*aEvent);
queryTextRect.InitForQueryTextRect(offset, 1, options);
rv = OnQueryTextRect(&queryTextRect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
if (NS_WARN_IF(!queryTextRect.mSucceeded)) {
return NS_ERROR_FAILURE;
}
MOZ_ASSERT(!queryTextRect.mReply.mRect.IsEmpty());
if (queryTextRect.mReply.mWritingMode.IsVertical()) {
queryTextRect.mReply.mRect.height = 1;
} else {
queryTextRect.mReply.mRect.width = 1;
}
aEvent->mReply.mRectArray.AppendElement(queryTextRect.mReply.mRect);
}
}
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryTextRect(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
// If mLength is 0 (this may be caused by bug of native IME), we should
// redirect this event to OnQueryCaretRect().
if (!aEvent->mInput.mLength) {
return OnQueryCaretRect(aEvent);
}
LineBreakType lineBreakType = GetLineBreakType(aEvent);
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<nsRange> range = new nsRange(mRootContent);
nsCOMPtr<nsIContent> lastTextContent;
rv = SetRangeFromFlatTextOffset(range, aEvent->mInput.mOffset,
aEvent->mInput.mLength, lineBreakType, true,
&aEvent->mReply.mOffset,
getter_AddRefs(lastTextContent));
NS_ENSURE_SUCCESS(rv, rv);
rv = GenerateFlatTextContent(range, aEvent->mReply.mString, lineBreakType);
NS_ENSURE_SUCCESS(rv, rv);
// used to iterate over all contents and their frames
nsCOMPtr<nsIContentIterator> iter = NS_NewContentIterator();
iter->Init(range);
// Get the first frame which causes some text after the offset.
FrameAndNodeOffset firstFrame = GetFirstFrameInRangeForTextRect(range);
// If GetFirstFrameInRangeForTextRect() does not return valid frame, that
// means that there are no visible frames having text or the offset reached
// the end of contents.
if (!firstFrame.IsValid()) {
nsAutoString allText;
rv = GenerateFlatTextContent(mRootContent, allText, lineBreakType);
// If the offset doesn't reach the end of contents but there is no frames
// for the node, that means that current offset's node is hidden by CSS or
// something. Ideally, we should handle it with the last visible text
// node's last character's rect, but it's not usual cases in actual web
// services. Therefore, currently, we should make this case fail.
if (NS_WARN_IF(NS_FAILED(rv)) ||
static_cast<uint32_t>(aEvent->mInput.mOffset) < allText.Length()) {
return NS_ERROR_FAILURE;
}
// Look for the last frame which should be included text rects.
ErrorResult erv;
range->SelectNodeContents(*mRootContent, erv);
if (NS_WARN_IF(erv.Failed())) {
return NS_ERROR_UNEXPECTED;
}
nsRect rect;
FrameAndNodeOffset lastFrame = GetLastFrameInRangeForTextRect(range);
// If there is at least one frame which can be used for computing a rect
// for a character or a line breaker, we should use it for guessing the
// caret rect at the end of the contents.
if (lastFrame) {
if (NS_WARN_IF(!lastFrame->GetContent())) {
return NS_ERROR_FAILURE;
}
FrameRelativeRect relativeRect;
// If there is a <br> frame at the end, it represents an empty line at
// the end with moz-<br> or content <br> in a block level element.
if (lastFrame->GetType() == nsGkAtoms::brFrame) {
relativeRect = GetLineBreakerRectBefore(lastFrame);
}
// If there is a text frame at the end, use its information.
else if (lastFrame->GetType() == nsGkAtoms::textFrame) {
relativeRect = GuessLineBreakerRectAfter(lastFrame->GetContent());
}
// If there is an empty frame which is neither a text frame nor a <br>
// frame at the end, guess caret rect in it.
else {
relativeRect = GuessFirstCaretRectIn(lastFrame);
}
if (NS_WARN_IF(!relativeRect.IsValid())) {
return NS_ERROR_FAILURE;
}
rect = relativeRect.RectRelativeTo(lastFrame);
rv = ConvertToRootRelativeOffset(lastFrame, rect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
aEvent->mReply.mWritingMode = lastFrame->GetWritingMode();
}
// Otherwise, if there are no contents in mRootContent, guess caret rect in
// its frame (with its font height and content box).
else {
nsIFrame* rootContentFrame = mRootContent->GetPrimaryFrame();
if (NS_WARN_IF(!rootContentFrame)) {
return NS_ERROR_FAILURE;
}
FrameRelativeRect relativeRect = GuessFirstCaretRectIn(rootContentFrame);
if (NS_WARN_IF(!relativeRect.IsValid())) {
return NS_ERROR_FAILURE;
}
rect = relativeRect.RectRelativeTo(rootContentFrame);
rv = ConvertToRootRelativeOffset(rootContentFrame, rect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
aEvent->mReply.mWritingMode = rootContentFrame->GetWritingMode();
}
aEvent->mReply.mRect = LayoutDeviceIntRect::FromUnknownRect(
rect.ToOutsidePixels(mPresContext->AppUnitsPerDevPixel()));
EnsureNonEmptyRect(aEvent->mReply.mRect);
aEvent->mSucceeded = true;
return NS_OK;
}
nsRect rect, frameRect;
nsPoint ptOffset;
// If the first frame is a text frame, the result should be computed with
// the frame's rect but not including the rect before start point of the
// queried range.
if (firstFrame->GetType() == nsGkAtoms::textFrame) {
rect.SetRect(nsPoint(0, 0), firstFrame->GetRect().Size());
rv = ConvertToRootRelativeOffset(firstFrame, rect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
frameRect = rect;
// Exclude the rect before start point of the queried range.
firstFrame->GetPointFromOffset(firstFrame.mOffsetInNode, &ptOffset);
if (firstFrame->GetWritingMode().IsVertical()) {
rect.y += ptOffset.y;
rect.height -= ptOffset.y;
} else {
rect.x += ptOffset.x;
rect.width -= ptOffset.x;
}
}
// If first frame causes a line breaker but it's not a <br> frame, we cannot
// compute proper rect only with the frame because typically caret is at
// right of the last character of it. For example, if caret is after "c" of
// |<p>abc</p><p>def</p>|, IME may query a line breaker's rect after "c".
// Then, if we compute it only with the 2nd <p>'s block frame, the result
// will be:
// +-<p>--------------------------------+
// |abc |
// +------------------------------------+
//
// I+-<p>--------------------------------+
// |def |
// +------------------------------------+
// However, users expect popup windows of IME should be positioned at
// right-bottom of "c" like this:
// +-<p>--------------------------------+
// |abcI |
// +------------------------------------+
//
// +-<p>--------------------------------+
// |def |
// +------------------------------------+
// Therefore, if the first frame isn't a <br> frame and there is a text
// node before the first node in the queried range, we should compute the
// first rect with the previous character's rect.
else if (firstFrame->GetType() != nsGkAtoms::brFrame && lastTextContent) {
FrameRelativeRect brRectAfterLastChar =
GuessLineBreakerRectAfter(lastTextContent);
if (NS_WARN_IF(!brRectAfterLastChar.IsValid())) {
return NS_ERROR_FAILURE;
}
rect = brRectAfterLastChar.mRect;
rv = ConvertToRootRelativeOffset(brRectAfterLastChar.mBaseFrame, rect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
frameRect = rect;
}
// Otherwise, we need to compute the line breaker's rect only with the
// first frame's rect. But this may be unexpected. For example,
// |<div contenteditable>[<p>]abc</p></div>|. In this case, caret is before
// "a", therefore, users expect the rect left of "a". However, we don't
// have enough information about the next character here and this isn't
// usual case (e.g., IME typically tries to query the rect of "a" or caret
// rect for computing its popup position). Therefore, we shouldn't do
// more complicated hack here unless we'll get some bug reports actually.
else {
FrameRelativeRect relativeRect = GetLineBreakerRectBefore(firstFrame);
if (NS_WARN_IF(!relativeRect.IsValid())) {
return NS_ERROR_FAILURE;
}
rect = relativeRect.RectRelativeTo(firstFrame);
rv = ConvertToRootRelativeOffset(firstFrame, rect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
frameRect = rect;
}
// UnionRect() requires non-empty rect. So, let's make sure to get non-emtpy
// rect from the first frame.
EnsureNonEmptyRect(rect);
// Get the last frame which causes some text in the range.
FrameAndNodeOffset lastFrame = GetLastFrameInRangeForTextRect(range);
if (NS_WARN_IF(!lastFrame.IsValid())) {
return NS_ERROR_FAILURE;
}
// iterate over all covered frames
for (nsIFrame* frame = firstFrame; frame != lastFrame;) {
frame = frame->GetNextContinuation();
if (!frame) {
do {
iter->Next();
nsINode* node = iter->GetCurrentNode();
if (!node) {
break;
}
if (!node->IsNodeOfType(nsINode::eCONTENT)) {
continue;
}
nsIFrame* primaryFrame = node->AsContent()->GetPrimaryFrame();
// The node may be hidden by CSS.
if (!primaryFrame) {
continue;
}
// We should take only text frame's rect and br frame's rect. We can
// always use frame rect of text frame and GetLineBreakerRectBefore()
// can return exactly correct rect only for <br> frame for now. On the
// other hand, GetLineBreakRectBefore() returns guessed caret rect for
// the other frames. We shouldn't include such odd rect to the result.
if (primaryFrame->GetType() == nsGkAtoms::textFrame ||
primaryFrame->GetType() == nsGkAtoms::brFrame) {
frame = primaryFrame;
}
} while (!frame && !iter->IsDone());
if (!frame) {
break;
}
}
if (frame->GetType() == nsGkAtoms::textFrame) {
frameRect.SetRect(nsPoint(0, 0), frame->GetRect().Size());
} else {
MOZ_ASSERT(frame->GetType() == nsGkAtoms::brFrame);
FrameRelativeRect relativeRect = GetLineBreakerRectBefore(frame);
if (NS_WARN_IF(!relativeRect.IsValid())) {
return NS_ERROR_FAILURE;
}
frameRect = relativeRect.RectRelativeTo(frame);
}
rv = ConvertToRootRelativeOffset(frame, frameRect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
// UnionRect() requires non-empty rect. So, let's make sure to get
// non-emtpy rect from the frame.
EnsureNonEmptyRect(frameRect);
if (frame != lastFrame) {
// not last frame, so just add rect to previous result
rect.UnionRect(rect, frameRect);
}
}
// Get the ending frame rect.
// FYI: If first frame and last frame are same, frameRect is already set
// to the rect excluding the text before the query range.
if (firstFrame.mFrame != lastFrame.mFrame) {
frameRect.SetRect(nsPoint(0, 0), lastFrame->GetRect().Size());
rv = ConvertToRootRelativeOffset(lastFrame, frameRect);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
}
// Shrink the last frame for cutting off the text after the query range.
if (lastFrame->GetType() == nsGkAtoms::textFrame) {
lastFrame->GetPointFromOffset(lastFrame.mOffsetInNode, &ptOffset);
if (lastFrame->GetWritingMode().IsVertical()) {
frameRect.height -= lastFrame->GetRect().height - ptOffset.y;
} else {
frameRect.width -= lastFrame->GetRect().width - ptOffset.x;
}
// UnionRect() requires non-empty rect. So, let's make sure to get
// non-empty rect from the last frame.
EnsureNonEmptyRect(frameRect);
if (firstFrame.mFrame == lastFrame.mFrame) {
rect.IntersectRect(rect, frameRect);
} else {
rect.UnionRect(rect, frameRect);
}
}
aEvent->mReply.mRect = LayoutDeviceIntRect::FromUnknownRect(
rect.ToOutsidePixels(mPresContext->AppUnitsPerDevPixel()));
// Returning empty rect may cause native IME confused, let's make sure to
// return non-empty rect.
EnsureNonEmptyRect(aEvent->mReply.mRect);
aEvent->mReply.mWritingMode = lastFrame->GetWritingMode();
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryEditorRect(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
nsIContent* focusedContent = GetFocusedContent();
rv = QueryContentRect(IsPlugin(focusedContent) ?
focusedContent : mRootContent.get(), aEvent);
NS_ENSURE_SUCCESS(rv, rv);
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryCaretRect(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
// When the selection is collapsed and the queried offset is current caret
// position, we should return the "real" caret rect.
if (mSelection->IsCollapsed()) {
nsRect caretRect;
nsIFrame* caretFrame = nsCaret::GetGeometry(mSelection, &caretRect);
if (caretFrame) {
uint32_t offset;
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
rv = GetStartOffset(mFirstSelectedRange,
&offset, GetLineBreakType(aEvent));
NS_ENSURE_SUCCESS(rv, rv);
if (offset == aEvent->mInput.mOffset) {
rv = ConvertToRootRelativeOffset(caretFrame, caretRect);
NS_ENSURE_SUCCESS(rv, rv);
nscoord appUnitsPerDevPixel =
caretFrame->PresContext()->AppUnitsPerDevPixel();
aEvent->mReply.mRect = LayoutDeviceIntRect::FromUnknownRect(
caretRect.ToOutsidePixels(appUnitsPerDevPixel));
// Returning empty rect may cause native IME confused, let's make sure
// to return non-empty rect.
EnsureNonEmptyRect(aEvent->mReply.mRect);
aEvent->mReply.mWritingMode = caretFrame->GetWritingMode();
aEvent->mReply.mOffset = aEvent->mInput.mOffset;
aEvent->mSucceeded = true;
return NS_OK;
}
}
}
// Otherwise, we should guess the caret rect from the character's rect.
WidgetQueryContentEvent queryTextRectEvent(eQueryTextRect, *aEvent);
WidgetQueryContentEvent::Options options(*aEvent);
queryTextRectEvent.InitForQueryTextRect(aEvent->mInput.mOffset, 1, options);
rv = OnQueryTextRect(&queryTextRectEvent);
if (NS_WARN_IF(NS_FAILED(rv)) || NS_WARN_IF(!queryTextRectEvent.mSucceeded)) {
return NS_ERROR_FAILURE;
}
queryTextRectEvent.mReply.mString.Truncate();
aEvent->mReply = queryTextRectEvent.mReply;
if (aEvent->GetWritingMode().IsVertical()) {
aEvent->mReply.mRect.height = 1;
} else {
aEvent->mReply.mRect.width = 1;
}
// Returning empty rect may cause native IME confused, let's make sure to
// return non-empty rect.
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryContentState(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQuerySelectionAsTransferable(
WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
if (!aEvent->mReply.mHasSelection) {
aEvent->mSucceeded = true;
aEvent->mReply.mTransferable = nullptr;
return NS_OK;
}
nsCOMPtr<nsIDocument> doc = mPresShell->GetDocument();
NS_ENSURE_TRUE(doc, NS_ERROR_FAILURE);
rv = nsCopySupport::GetTransferableForSelection(
mSelection, doc, getter_AddRefs(aEvent->mReply.mTransferable));
NS_ENSURE_SUCCESS(rv, rv);
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryCharacterAtPoint(WidgetQueryContentEvent* aEvent)
{
nsresult rv = Init(aEvent);
if (NS_FAILED(rv)) {
return rv;
}
aEvent->mReply.mOffset = aEvent->mReply.mTentativeCaretOffset =
WidgetQueryContentEvent::NOT_FOUND;
nsIFrame* rootFrame = mPresShell->GetRootFrame();
NS_ENSURE_TRUE(rootFrame, NS_ERROR_FAILURE);
nsIWidget* rootWidget = rootFrame->GetNearestWidget();
NS_ENSURE_TRUE(rootWidget, NS_ERROR_FAILURE);
// The root frame's widget might be different, e.g., the event was fired on
// a popup but the rootFrame is the document root.
if (rootWidget != aEvent->mWidget) {
NS_PRECONDITION(aEvent->mWidget, "The event must have the widget");
nsView* view = nsView::GetViewFor(aEvent->mWidget);
NS_ENSURE_TRUE(view, NS_ERROR_FAILURE);
rootFrame = view->GetFrame();
NS_ENSURE_TRUE(rootFrame, NS_ERROR_FAILURE);
rootWidget = rootFrame->GetNearestWidget();
NS_ENSURE_TRUE(rootWidget, NS_ERROR_FAILURE);
}
WidgetQueryContentEvent eventOnRoot(true, eQueryCharacterAtPoint,
rootWidget);
eventOnRoot.mUseNativeLineBreak = aEvent->mUseNativeLineBreak;
eventOnRoot.mRefPoint = aEvent->mRefPoint;
if (rootWidget != aEvent->mWidget) {
eventOnRoot.mRefPoint += aEvent->mWidget->WidgetToScreenOffset() -
rootWidget->WidgetToScreenOffset();
}
nsPoint ptInRoot =
nsLayoutUtils::GetEventCoordinatesRelativeTo(&eventOnRoot, rootFrame);
nsIFrame* targetFrame = nsLayoutUtils::GetFrameForPoint(rootFrame, ptInRoot);
if (!targetFrame || !targetFrame->GetContent() ||
!nsContentUtils::ContentIsDescendantOf(targetFrame->GetContent(),
mRootContent)) {
// There is no character at the point.
aEvent->mSucceeded = true;
return NS_OK;
}
nsPoint ptInTarget = ptInRoot + rootFrame->GetOffsetToCrossDoc(targetFrame);
int32_t rootAPD = rootFrame->PresContext()->AppUnitsPerDevPixel();
int32_t targetAPD = targetFrame->PresContext()->AppUnitsPerDevPixel();
ptInTarget = ptInTarget.ScaleToOtherAppUnits(rootAPD, targetAPD);
nsIFrame::ContentOffsets tentativeCaretOffsets =
targetFrame->GetContentOffsetsFromPoint(ptInTarget);
if (!tentativeCaretOffsets.content ||
!nsContentUtils::ContentIsDescendantOf(tentativeCaretOffsets.content,
mRootContent)) {
// There is no character nor tentative caret point at the point.
aEvent->mSucceeded = true;
return NS_OK;
}
rv = GetFlatTextLengthInRange(NodePosition(mRootContent, 0),
NodePosition(tentativeCaretOffsets),
mRootContent,
&aEvent->mReply.mTentativeCaretOffset,
GetLineBreakType(aEvent));
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
if (targetFrame->GetType() != nsGkAtoms::textFrame) {
// There is no character at the point but there is tentative caret point.
aEvent->mSucceeded = true;
return NS_OK;
}
MOZ_ASSERT(
aEvent->mReply.mTentativeCaretOffset != WidgetQueryContentEvent::NOT_FOUND,
"The point is inside a character bounding box. Why tentative caret point "
"hasn't been found?");
nsTextFrame* textframe = static_cast<nsTextFrame*>(targetFrame);
nsIFrame::ContentOffsets contentOffsets =
textframe->GetCharacterOffsetAtFramePoint(ptInTarget);
NS_ENSURE_TRUE(contentOffsets.content, NS_ERROR_FAILURE);
uint32_t offset;
rv = GetFlatTextLengthInRange(NodePosition(mRootContent, 0),
NodePosition(contentOffsets),
mRootContent, &offset,
GetLineBreakType(aEvent));
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
WidgetQueryContentEvent textRect(true, eQueryTextRect, aEvent->mWidget);
WidgetQueryContentEvent::Options options(*aEvent);
textRect.InitForQueryTextRect(offset, 1, options);
rv = OnQueryTextRect(&textRect);
NS_ENSURE_SUCCESS(rv, rv);
NS_ENSURE_TRUE(textRect.mSucceeded, NS_ERROR_FAILURE);
// currently, we don't need to get the actual text.
aEvent->mReply.mOffset = offset;
aEvent->mReply.mRect = textRect.mReply.mRect;
aEvent->mSucceeded = true;
return NS_OK;
}
nsresult
ContentEventHandler::OnQueryDOMWidgetHittest(WidgetQueryContentEvent* aEvent)
{
NS_ASSERTION(aEvent, "aEvent must not be null");
nsresult rv = InitBasic();
if (NS_FAILED(rv)) {
return rv;
}
aEvent->mSucceeded = false;
aEvent->mReply.mWidgetIsHit = false;
NS_ENSURE_TRUE(aEvent->mWidget, NS_ERROR_FAILURE);
nsIDocument* doc = mPresShell->GetDocument();
NS_ENSURE_TRUE(doc, NS_ERROR_FAILURE);
nsIFrame* docFrame = mPresShell->GetRootFrame();
NS_ENSURE_TRUE(docFrame, NS_ERROR_FAILURE);
LayoutDeviceIntPoint eventLoc =
aEvent->mRefPoint + aEvent->mWidget->WidgetToScreenOffset();
nsIntRect docFrameRect = docFrame->GetScreenRect(); // Returns CSS pixels
CSSIntPoint eventLocCSS(
mPresContext->DevPixelsToIntCSSPixels(eventLoc.x) - docFrameRect.x,
mPresContext->DevPixelsToIntCSSPixels(eventLoc.y) - docFrameRect.y);
Element* contentUnderMouse =
doc->ElementFromPointHelper(eventLocCSS.x, eventLocCSS.y, false, false);
if (contentUnderMouse) {
nsIWidget* targetWidget = nullptr;
nsIFrame* targetFrame = contentUnderMouse->GetPrimaryFrame();
nsIObjectFrame* pluginFrame = do_QueryFrame(targetFrame);
if (pluginFrame) {
targetWidget = pluginFrame->GetWidget();
} else if (targetFrame) {
targetWidget = targetFrame->GetNearestWidget();
}
if (aEvent->mWidget == targetWidget) {
aEvent->mReply.mWidgetIsHit = true;
}
}
aEvent->mSucceeded = true;
return NS_OK;
}
/* static */ nsresult
ContentEventHandler::GetFlatTextLengthInRange(
const NodePosition& aStartPosition,
const NodePosition& aEndPosition,
nsIContent* aRootContent,
uint32_t* aLength,
LineBreakType aLineBreakType,
bool aIsRemovingNode /* = false */)
{
if (NS_WARN_IF(!aRootContent) || NS_WARN_IF(!aStartPosition.IsValid()) ||
NS_WARN_IF(!aEndPosition.IsValid()) || NS_WARN_IF(!aLength)) {
return NS_ERROR_INVALID_ARG;
}
if (aStartPosition == aEndPosition) {
*aLength = 0;
return NS_OK;
}
// Don't create nsContentIterator instance until it's really necessary since
// destroying without initializing causes unexpected NS_ASSERTION() call.
nsCOMPtr<nsIContentIterator> iter;
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
// Working with ContentIterator, we may need to adjust the end position for
// including it forcibly.
NodePosition endPosition(aEndPosition);
// This may be called for retrieving the text of removed nodes. Even in this
// case, the node thinks it's still in the tree because UnbindFromTree() will
// be called after here. However, the node was already removed from the
// array of children of its parent. So, be careful to handle this case.
if (aIsRemovingNode) {
DebugOnly<nsIContent*> parent = aStartPosition.mNode->GetParent();
MOZ_ASSERT(parent && parent->IndexOf(aStartPosition.mNode) == -1,
"At removing the node, the node shouldn't be in the array of children "
"of its parent");
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
MOZ_ASSERT(aStartPosition.mNode == endPosition.mNode,
"At removing the node, start and end node should be same");
MOZ_ASSERT(aStartPosition.mOffset == 0,
"When the node is being removed, the start offset should be 0");
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
MOZ_ASSERT(static_cast<uint32_t>(endPosition.mOffset) ==
endPosition.mNode->GetChildCount(),
"When the node is being removed, the end offset should be child count");
iter = NS_NewPreContentIterator();
nsresult rv = iter->Init(aStartPosition.mNode);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
} else {
RefPtr<nsRange> prev = new nsRange(aRootContent);
nsresult rv = aStartPosition.SetToRangeStart(prev);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
// When the end position is immediately after non-root element's open tag,
// we need to include a line break caused by the open tag.
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
if (endPosition.mNode != aRootContent &&
endPosition.IsImmediatelyAfterOpenTag()) {
if (endPosition.mNode->HasChildren()) {
// When the end node has some children, move the end position to before
// the open tag of its first child.
nsINode* firstChild = endPosition.mNode->GetFirstChild();
if (NS_WARN_IF(!firstChild)) {
return NS_ERROR_FAILURE;
}
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
endPosition = NodePositionBefore(firstChild, 0);
} else {
// When the end node is empty, move the end position after the node.
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
nsIContent* parentContent = endPosition.mNode->GetParent();
if (NS_WARN_IF(!parentContent)) {
return NS_ERROR_FAILURE;
}
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
int32_t indexInParent = parentContent->IndexOf(endPosition.mNode);
if (NS_WARN_IF(indexInParent < 0)) {
return NS_ERROR_FAILURE;
}
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
endPosition = NodePositionBefore(parentContent, indexInParent + 1);
}
}
if (endPosition.OffsetIsValid()) {
// Offset is within node's length; set end of range to that offset
rv = endPosition.SetToRangeEnd(prev);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
iter = NS_NewPreContentIterator();
rv = iter->Init(prev);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
} else if (endPosition.mNode != aRootContent) {
// Offset is past node's length; set end of range to end of node
rv = endPosition.SetToRangeEndAfter(prev);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
iter = NS_NewPreContentIterator();
rv = iter->Init(prev);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
} else {
// Offset is past the root node; set end of range to end of root node
iter = NS_NewPreContentIterator();
rv = iter->Init(aRootContent);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
}
}
*aLength = 0;
for (; !iter->IsDone(); iter->Next()) {
nsINode* node = iter->GetCurrentNode();
if (NS_WARN_IF(!node)) {
break;
}
if (!node->IsContent()) {
continue;
}
nsIContent* content = node->AsContent();
if (node->IsNodeOfType(nsINode::eTEXT)) {
// Note: our range always starts from offset 0
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
if (node == endPosition.mNode) {
*aLength += GetTextLength(content, aLineBreakType,
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
endPosition.mOffset);
} else {
*aLength += GetTextLength(content, aLineBreakType);
}
} else if (ShouldBreakLineBefore(content, aRootContent)) {
// If the start position is start of this node but doesn't include the
// open tag, don't append the line break length.
if (node == aStartPosition.mNode && !aStartPosition.IsBeforeOpenTag()) {
continue;
}
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
// If the end position is before the open tag, don't append the line
// break length.
if (node == endPosition.mNode && endPosition.IsBeforeOpenTag()) {
continue;
}
*aLength += GetBRLength(aLineBreakType);
}
}
return NS_OK;
}
nsresult
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
ContentEventHandler::GetStartOffset(nsRange* aRange,
uint32_t* aOffset,
LineBreakType aLineBreakType)
{
MOZ_ASSERT(aRange);
return GetFlatTextLengthInRange(
NodePosition(mRootContent, 0),
Bug 1304624 ContentEventHandler::GetFlatTextLengthInRange() shouldn't include a line break length caused by the end node's open tag when the given range ends before the open tag r=smaug This must be a regression of bug 1213589. When removing whole text from an editor, the editor removes moz-<br> after removing all nodes. Immediately before the remove, moz-<br> becomes <br>. So, IMEContentObserver::AttributeChanged() detects the native text length for the <br> element changing from 0 to 2 (on Windows) or 1 (on the other platforms). However, the call of ContentEventHandler::GetFlatTextLengthInRange() in IMEContentObserver::AttributeChanged() returns 2 (on Windows) or 1 (on the other platforms). Although, it tries to get the length from the start of the editor to before aElement, it always includes the length of the line breaker caused by the end node. The reason is, ContentEventHandler::GetFlatTextLengthInRange() always adds the line breaker's length for the end node without checking the range includes open tag of the end node. Therefore, this patch adds the check into ContentEventHandler::GetFlatTextLengthInRange(). So, ContentEventHandler::GetFlatTextLengthInRange() becomes to support NodePositionBefore() for the end of the range. Additionally, this patch fixes a bug at appending text node length. In the loop, |aEndPosition| shouldn't be referred because it may be hacked with |endPosition|. So, detecting the last text node should be compared with |endPosition|. For making the method code safer, this changes whole |aEndPosition| after defining |endPosition| is replaced with |endPosition|. MozReview-Commit-ID: FUp2nxuQHnO --HG-- extra : rebase_source : 611992faf95082ce8ae005b7bcce90a3bea4f99e
2016-09-29 08:04:15 +03:00
NodePosition(aRange->GetStartParent(), aRange->StartOffset()),
mRootContent, aOffset, aLineBreakType);
}
nsresult
ContentEventHandler::AdjustCollapsedRangeMaybeIntoTextNode(nsRange* aRange)
{
MOZ_ASSERT(aRange);
MOZ_ASSERT(aRange->Collapsed());
if (!aRange || !aRange->Collapsed()) {
return NS_ERROR_INVALID_ARG;
}
nsCOMPtr<nsINode> parentNode = aRange->GetStartParent();
int32_t offsetInParentNode = aRange->StartOffset();
if (NS_WARN_IF(!parentNode) || NS_WARN_IF(offsetInParentNode < 0)) {
return NS_ERROR_INVALID_ARG;
}
// If the node is text node, we don't need to modify aRange.
if (parentNode->IsNodeOfType(nsINode::eTEXT)) {
return NS_OK;
}
// If the parent is not a text node but it has a text node at the offset,
// we should adjust the range into the text node.
// NOTE: This is emulating similar situation of EditorBase.
nsINode* childNode = nullptr;
int32_t offsetInChildNode = -1;
if (!offsetInParentNode && parentNode->HasChildren()) {
// If the range is the start of the parent, adjusted the range to the
// start of the first child.
childNode = parentNode->GetFirstChild();
offsetInChildNode = 0;
} else if (static_cast<uint32_t>(offsetInParentNode) <
parentNode->GetChildCount()) {
// If the range is next to a child node, adjust the range to the end of
// the previous child.
childNode = parentNode->GetChildAt(offsetInParentNode - 1);
offsetInChildNode = childNode->Length();
}
// But if the found node isn't a text node, we cannot modify the range.
if (!childNode || !childNode->IsNodeOfType(nsINode::eTEXT) ||
NS_WARN_IF(offsetInChildNode < 0)) {
return NS_OK;
}
nsresult rv = aRange->Set(childNode, offsetInChildNode,
childNode, offsetInChildNode);
if (NS_WARN_IF(NS_FAILED(rv))) {
return rv;
}
return NS_OK;
}
nsresult
ContentEventHandler::GetStartFrameAndOffset(const nsRange* aRange,
nsIFrame*& aFrame,
int32_t& aOffsetInFrame)
{
MOZ_ASSERT(aRange);
aFrame = nullptr;
aOffsetInFrame = -1;
nsINode* node = aRange->GetStartParent();
if (NS_WARN_IF(!node) ||
NS_WARN_IF(!node->IsNodeOfType(nsINode::eCONTENT))) {
return NS_ERROR_FAILURE;
}
nsIContent* content = static_cast<nsIContent*>(node);
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<nsFrameSelection> fs = mPresShell->FrameSelection();
aFrame = fs->GetFrameForNodeOffset(content, aRange->StartOffset(),
fs->GetHint(), &aOffsetInFrame);
if (NS_WARN_IF(!aFrame)) {
return NS_ERROR_FAILURE;
}
return NS_OK;
}
nsresult
ContentEventHandler::ConvertToRootRelativeOffset(nsIFrame* aFrame,
nsRect& aRect)
{
NS_ASSERTION(aFrame, "aFrame must not be null");
nsPresContext* thisPC = aFrame->PresContext();
nsPresContext* rootPC = thisPC->GetRootPresContext();
if (NS_WARN_IF(!rootPC)) {
return NS_ERROR_FAILURE;
}
nsIFrame* rootFrame = rootPC->PresShell()->GetRootFrame();
if (NS_WARN_IF(!rootFrame)) {
return NS_ERROR_FAILURE;
}
aRect = nsLayoutUtils::TransformFrameRectToAncestor(aFrame, aRect, rootFrame);
// TransformFrameRectToAncestor returned the rect in the ancestor's appUnits,
// but we want it in aFrame's units (in case of different full-zoom factors),
// so convert back.
aRect = aRect.ScaleToOtherAppUnitsRoundOut(rootPC->AppUnitsPerDevPixel(),
thisPC->AppUnitsPerDevPixel());
return NS_OK;
}
static void AdjustRangeForSelection(nsIContent* aRoot,
nsINode** aNode,
int32_t* aNodeOffset)
{
nsINode* node = *aNode;
int32_t nodeOffset = *aNodeOffset;
if (aRoot == node || NS_WARN_IF(!node->GetParent()) ||
!node->IsNodeOfType(nsINode::eTEXT)) {
return;
}
// When the offset is at the end of the text node, set it to after the
// text node, to make sure the caret is drawn on a new line when the last
// character of the text node is '\n' in <textarea>.
int32_t textLength =
static_cast<int32_t>(static_cast<nsIContent*>(node)->TextLength());
MOZ_ASSERT(nodeOffset <= textLength, "Offset is past length of text node");
if (nodeOffset != textLength) {
return;
}
nsIContent* aRootParent = aRoot->GetParent();
if (NS_WARN_IF(!aRootParent)) {
return;
}
// If the root node is not an anonymous div of <textarea>, we don't need to
// do this hack. If you did this, ContentEventHandler couldn't distinguish
// if the range includes open tag of the next node in some cases, e.g.,
// textNode]<p></p> vs. textNode<p>]</p>
if (!aRootParent->IsHTMLElement(nsGkAtoms::textarea)) {
return;
}
*aNode = node->GetParent();
MOZ_ASSERT((*aNode)->IndexOf(node) != -1);
*aNodeOffset = (*aNode)->IndexOf(node) + 1;
}
nsresult
ContentEventHandler::OnSelectionEvent(WidgetSelectionEvent* aEvent)
{
aEvent->mSucceeded = false;
// Get selection to manipulate
// XXX why do we need to get them from ISM? This method should work fine
// without ISM.
nsCOMPtr<nsISelection> sel;
nsresult rv =
IMEStateManager::GetFocusSelectionAndRoot(getter_AddRefs(sel),
getter_AddRefs(mRootContent));
mSelection = sel ? sel->AsSelection() : nullptr;
if (rv != NS_ERROR_NOT_AVAILABLE) {
NS_ENSURE_SUCCESS(rv, rv);
} else {
rv = Init(aEvent);
NS_ENSURE_SUCCESS(rv, rv);
}
// Get range from offset and length
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<nsRange> range = new nsRange(mRootContent);
rv = SetRangeFromFlatTextOffset(range, aEvent->mOffset, aEvent->mLength,
GetLineBreakType(aEvent),
aEvent->mExpandToClusterBoundary);
NS_ENSURE_SUCCESS(rv, rv);
nsINode* startNode = range->GetStartParent();
nsINode* endNode = range->GetEndParent();
int32_t startNodeOffset = range->StartOffset();
int32_t endNodeOffset = range->EndOffset();
AdjustRangeForSelection(mRootContent, &startNode, &startNodeOffset);
AdjustRangeForSelection(mRootContent, &endNode, &endNodeOffset);
if (NS_WARN_IF(!startNode) || NS_WARN_IF(!endNode) ||
NS_WARN_IF(startNodeOffset < 0) || NS_WARN_IF(endNodeOffset < 0)) {
return NS_ERROR_UNEXPECTED;
}
mSelection->StartBatchChanges();
// Clear selection first before setting
rv = mSelection->RemoveAllRanges();
// Need to call EndBatchChanges at the end even if call failed
if (NS_SUCCEEDED(rv)) {
if (aEvent->mReversed) {
rv = mSelection->Collapse(endNode, endNodeOffset);
} else {
rv = mSelection->Collapse(startNode, startNodeOffset);
}
if (NS_SUCCEEDED(rv) &&
(startNode != endNode || startNodeOffset != endNodeOffset)) {
if (aEvent->mReversed) {
rv = mSelection->Extend(startNode, startNodeOffset);
} else {
rv = mSelection->Extend(endNode, endNodeOffset);
}
}
}
// Pass the eSetSelection events reason along with the BatchChange-end
// selection change notifications.
mSelection->EndBatchChangesInternal(aEvent->mReason);
NS_ENSURE_SUCCESS(rv, rv);
mSelection->ScrollIntoViewInternal(
nsISelectionController::SELECTION_FOCUS_REGION,
false, nsIPresShell::ScrollAxis(), nsIPresShell::ScrollAxis());
aEvent->mSucceeded = true;
return NS_OK;
}
nsRect
ContentEventHandler::FrameRelativeRect::RectRelativeTo(
nsIFrame* aDestFrame) const
{
if (!mBaseFrame || NS_WARN_IF(!aDestFrame)) {
return nsRect();
}
if (NS_WARN_IF(aDestFrame->PresContext() != mBaseFrame->PresContext())) {
return nsRect();
}
if (aDestFrame == mBaseFrame) {
return mRect;
}
nsIFrame* rootFrame = mBaseFrame->PresContext()->PresShell()->GetRootFrame();
nsRect baseFrameRectInRootFrame =
nsLayoutUtils::TransformFrameRectToAncestor(mBaseFrame, nsRect(),
rootFrame);
nsRect destFrameRectInRootFrame =
nsLayoutUtils::TransformFrameRectToAncestor(aDestFrame, nsRect(),
rootFrame);
nsPoint difference =
destFrameRectInRootFrame.TopLeft() - baseFrameRectInRootFrame.TopLeft();
return mRect - difference;
}
} // namespace mozilla