how to save qundocommand to file and reload it? - qt

I am using qt's undo framework , which use qundocommand to do some application support undo.
Is there an easy way I can use to save those qundocommand to a file and reload it?

There's no built-in way. I don't think it's very common to save the undo stack between sessions. You'll have to serialize the commands yourself by iterating through the commands on the stack, and saving each one's unique data using QDataStream. It might look something like this:
...
dataStream << undoStack->count(); // store number of commands
for (int i = 0; i < undoStack->count(); i++)
{
// store each command's unique information
dataStream << undoStack->command(i)->someMemberVariable;
}
...
Then you would use QDataStream again to deserialize the data back into QUndoCommands.
You can use QFile to handle the file management.

Use Qt's serialization as described here:
Serialization with Qt
Then within your QUndoCommands you can use a temp file to write the data to it:
http://qt-project.org/doc/qt-4.8/qtemporaryfile.html
However this might cause you an issue since each file is kept open and so on some platforms (Linux) you may run out of open file handles.
To combat this you'd have to create some other factory type object which handles your commands - then this could pass in a reference to a QTemporaryFile automatically. This factory/QUndoCommand care taker object must have the same life time as the QUndoCommands. If not then the temp file will be removed from disk and your QUndoCommands will break.
The other thing you can do is use QUndoCommand as a proxy to your real undo command - this means you can save quite a bit of memory since when your undo command is saved to file you can delete the internal pointer/set it to null. Then restore it later.

Here's a PyQt solution for serializing/pickling QUndoCommands. The tricky part was getting the parent to call __init__ first, then the children. This method relies on all the children's __setstate__ to be called before the parent's, which happens upon pickling as children are returned in the parent's __getstate__.
class UndoCommand(QUndoCommand):
"""
For pickling
"""
def __init__(self, text, parent=None):
QUndoCommand.__init__(self, text, parent)
self.__parent = parent
self.__initialized = True
# defined and initialized in __setstate__
# self.__child_states = {}
def __getstate__(self):
return {
**{k: v for k, v in self.__dict__.items()},
'_UndoCommand__initialized': False,
'_UndoCommand__text': self.text(),
'_UndoCommand__children':
[self.child(i) for i in range(self.childCount())]
}
def __setstate__(self, state):
if hasattr(self, '_UndoCommand__initialized') and \
self.__initialized:
return
text = state['_UndoCommand__text']
parent = state['_UndoCommand__parent'] # type: UndoCommand
if parent is not None and \
(not hasattr(parent, '_UndoCommand__initialized') or
not parent.__initialized):
# will be initialized in parent's __setstate__
if not hasattr(parent, '_UndoCommand__child_states'):
setattr(parent, '_UndoCommand__child_states', {})
parent.__child_states[self] = state
return
# init must be called on unpickle-time to recreate Qt object
UndoCommand.__init__(self, text, parent)
for child in state['_UndoCommand__children']:
child.__setstate__(self.__child_states[child])
self.__dict__ = {k: v for k, v in state.items()}
#staticmethod
def from_QUndoCommand(qc: QUndoCommand, parent=None):
if type(qc) == QUndoCommand:
qc.__class__ = UndoCommand
qc.__initialized = True
qc.__parent = parent
children = [qc.child(i) for i in range(qc.childCount())]
for child in children:
UndoCommand.from_QUndoCommand(child, parent=qc)
return qc

Related

How to solve a tkinter memory leak?

I have a dynamic table with a fixed row number (like a FIFO Queue), which updates continuously through tkinter's after() function. Inside the table is a Button, which text is editable.
To make the Button's text editable I used the solution of BrenBarn and reference a loop variable into a function call at the command-attribute.
When the function update_content_items() is cycled, I found, that the memory usage is increasing MB by MB per second. I can confirm that after commenting out the lambda expression, the memory leak was gone. (as seen live running 'top' in the terminal)
It seems I have to use the lambda, otherwise the Button will have a wrong index and the user edits the wrong row, when I just used self.list_items[i], though the user clicked the right one.
Is there a way to solve the problem? How can the user click the right button and edit it while having the right index and getting rid of the leak?
The corresponding code:
def update_content_items(self):
"""
Continuously fills and updates the Table with rows and content.
The size of the table rows is initially fixed by an external value at config.ini
:return: nothing
"""
if len(self.list_items) > self.queueMaxlen:
self.queueMaxlen = len(self.list_items)
self.build_table()
try:
for i in range(len(self.list_items)):
item = self.list_items[i]
self.barcodeImgList[i].image = item.plateimage
orig_image = Image.open(io.BytesIO(item.plateimage))
ein_image = ImageTk.PhotoImage(orig_image)
self.barcodeImgList[i].configure(image=ein_image)
# keeps a reference, because somehow tkinter forgets it...??? Bug of my implementation???
self.barcodeImgList[i].image = ein_image
orig_image = None
ein_image = None
#FIXME Memory LEAK?
self.numberList[i].configure(text=item.number,
command=lambda K=i: self.edit_barcode(self.list_items[K]))
self.timestampList[i].configure(text=item.timestamp)
self.search_hitlist[i].config(bg='white', cursor="xterm")
self.search_hitlist[i].unbind("<Button-1>")
if item.queryresult is not None:
if item.queryresult.gesamtstatus != 'Gruen':
self.search_hitlist[i].insert(tk.END, item.queryresult.barcode +
'\n' + item.queryresult.permitlevel)
self.search_hitlist[i].configure(bg='red', cursor="hand2")
self.search_hitlist[i].bind("<Button-1>", item.url_callback)
else:
self.search_hitlist[i].configure(bg='green', cursor="xterm")
self.search_hitlist[i].configure(state=tk.DISABLED)
self.on_frame_configure(None)
self.canvas.after(10, self.update_content_items)
except IndexError as ie:
for number, thing in enumerate(self.list_items):
print(number, thing)
raise ie
def edit_barcode(self, item=None):
"""
Opens the number plate edit dialogue and updates the corresponding list item.
:param item: as Hit DAO
:return: nothing
"""
if item is not None:
new_item_number = EditBarcodeEntry(self.master.master, item)
if new_item_number.mynumber != 0:
item.number = new_item_number.mynumber
self.list_items.request_work(item, 'update')
self.list_items.edit_hititem_by_id(item)
self.parent.master.queryQueue.put(item)
else:
print("You shouldn't get here at all. Please see edit_barcode function.")
EDIT: It seems there is indeed a deeper memory leak (python itself). The images won't get garbage collected. Memory is slowly leaking in Python 3.x and I do use PIL. Also here: Image loading by file name memory leak is not properly fixed
What can I do, because I have to cycle through a list with records and update Labels with images? Is there a workaround? PhotoImage has no explicit close() function, and if I call del, the reference is gc'ed and no configuring of the Label possible.
an example of my proposed changes, with indentation fixed:
def update_content_items(self):
"""
Continuously fills and updates the Table with rows and content.
The size of the table rows is initially fixed by an external value at config.ini
:return: nothing
"""
if len(self.list_items) > self.queueMaxlen:
self.queueMaxlen = len(self.list_items)
self.build_table()
try:
for i in range(len(self.list_items)):
item = self.list_items[i]
self.barcodeImgList[i].image = item.plateimage
orig_image = Image.open(io.BytesIO(item.plateimage))
ein_image = ImageTk.PhotoImage(orig_image)
self.barcodeImgList[i].configure(image=ein_image)
# keeps a reference, because somehow tkinter forgets it...??? Bug of my implementation???
self.barcodeImgList[i].image = ein_image
orig_image = None
ein_image = None
self.numberList[i].configure(text=item.number) # removed lambda
self.numberList[i].bind("<Button-1>", self.edit_barcode_binding) # added binding
self.timestampList[i].configure(text=item.timestamp)
self.search_hitlist[i].config(bg='white', cursor="xterm")
self.search_hitlist[i].unbind("<Button-1>")
if item.queryresult is not None:
if item.queryresult.gesamtstatus != 'Gruen':
self.search_hitlist[i].insert(tk.END, item.queryresult.barcode +
'\n' + item.queryresult.permitlevel)
self.search_hitlist[i].configure(bg='red', cursor="hand2")
self.search_hitlist[i].bind("<Button-1>", item.url_callback)
else:
self.search_hitlist[i].configure(bg='green', cursor="xterm")
self.search_hitlist[i].configure(state=tk.DISABLED)
self.on_frame_configure(None)
self.canvas.after(10, self.update_content_items)
except IndexError as ie:
for number, thing in enumerate(self.list_items):
print(number, thing)
raise ie
def edit_barcode_binding(self, event): # new wrapper for binding
K = self.numberList.index(event.widget) # get index from list
self.edit_barcode(self.list_items[K]) # call the original function
def edit_barcode(self, item=None):
"""
Opens the number plate edit dialogue and updates the corresponding list item.
:param item: as Hit DAO
:return: nothing
"""
if item is not None:
new_item_number = EditBarcodeEntry(self.master.master, item)
if new_item_number.mynumber != 0:
item.number = new_item_number.mynumber
self.list_items.request_work(item, 'update')
self.list_items.edit_hititem_by_id(item)
self.parent.master.queryQueue.put(item)
else:
print("You shouldn't get here at all. Please see edit_barcode function.")

Difference between two files view in HTML using Java or any jar

I want to write a script which compare two files in java and see there difference in html page ( side by side ), can someone help me out how to write ( where to start). I am pulling my hair out for this....
I want to use this script in beanshell postprocessor so that I can compare the standard output files with result files easily
I don't think you should be asking people for writing code for you here, consider hiring a freelancer instead.
Alternatively you can use the following approach:
Add JSR223 Assertion as a child of the request which you would like to fail if files won't be equal
Put the following code into "Script" area:
def file1 = new File('/path/to/file1')
def file2 = new File('/path/to/file2')
def file1Lines = file1.readLines('UTF-8')
def file2Lines = file2.readLines('UTF-8')
if (file1Lines.size() != file2Lines.size()) {
AssertionResult.setFailure(true)
AssertionResult.setFailureMessage('Files size is different, omitting line-by-line compare')
} else {
def differences = new StringBuilder()
file1Lines.eachWithIndex {
String file1Line, int number ->
String file2Line = file2Lines.get(number)
if (!file1Line.equals(file2Line)) {
differences.append('Difference # ').append(number).append('. Expected: ')
.append(file1Line).append('. Actual: ' + file2Line)
differences.append(System.getProperty('line.separator'))
}
}
if (differences.toString().length() > 0) {
AssertionResult.setFailure(true)
AssertionResult.setFailureMessage(differences.toString())
}
}
If there will be differences in files content you will see them listed one by one in the JSR223 Assertion
See Scripting JMeter Assertions in Groovy - A Tutorial for more details.

When dragging multiple items from QListWidget, non-draggable items get removed

I have two QListWidgets. The user can select multiple items from one list and drag them to the other list. But within each list, some items are draggable and some are not. If the selection contains both draggable and non-draggable items, a problem happens. Only the draggable items appear in the second list, which is correct. But all the items disappear from the first list.
In the animated image above, items 00, 01, and 02 are selected. Only items 00 and 02 are drag enabled. After the drag-and-drop, all three items are gone from the first list. How can I fix this?
Here is some code to reproduce the problem:
import random
import sys
from PySide import QtCore, QtGui
class TestMultiDragDrop(QtGui.QMainWindow):
def __init__(self, parent=None):
super(TestMultiDragDrop, self).__init__(parent)
centralWidget = QtGui.QWidget()
self.setCentralWidget(centralWidget)
layout = QtGui.QHBoxLayout(centralWidget)
self.list1 = QtGui.QListWidget()
self.list1.setDragDropMode(QtGui.QAbstractItemView.DragDrop)
self.list1.setDefaultDropAction(QtCore.Qt.MoveAction)
self.list1.setSelectionMode(QtGui.QListWidget.ExtendedSelection)
self.list2 = QtGui.QListWidget()
self.list2.setDragDropMode(QtGui.QAbstractItemView.DragDrop)
self.list2.setDefaultDropAction(QtCore.Qt.MoveAction)
self.list2.setSelectionMode(QtGui.QListWidget.ExtendedSelection)
layout.addWidget(self.list1)
layout.addWidget(self.list2)
self.fillListWidget(self.list1, 8, 'someItem')
self.fillListWidget(self.list2, 4, 'anotherItem')
def fillListWidget(self, listWidget, numItems, txt):
for i in range(numItems):
item = QtGui.QListWidgetItem()
newTxt = '{0}{1:02d}'.format(txt, i)
if random.randint(0, 1):
item.setFlags(QtCore.Qt.ItemIsSelectable | QtCore.Qt.ItemIsEnabled)
else:
# If the item is draggable, indicate it with a *
newTxt += ' *'
item.setFlags(QtCore.Qt.ItemIsSelectable | QtCore.Qt.ItemIsEnabled | QtCore.Qt.ItemIsDragEnabled)
item.setText(newTxt)
listWidget.addItem(item)
def openMultiDragDrop():
global multiDragDropUI
try:
multiDragDropUI.close()
except:
pass
multiDragDropUI = TestMultiDragDrop()
multiDragDropUI.setAttribute(QtCore.Qt.WA_DeleteOnClose)
multiDragDropUI.show()
return multiDragDropUI
if __name__ == '__main__':
app = QtGui.QApplication([])
openMultiDragDrop()
sys.exit(app.exec_())
Here I have some suspicion on setDefaultDropAction(QtCore.Qt.MoveAction)
Read below para from documentation: Specially the bold line
In the simplest case, the target of a drag and drop action receives a copy of the data being dragged, and the source decides whether to delete the original. This is described by the CopyAction action. The target may also choose to handle other actions, specifically the MoveAction and LinkAction actions. If the source calls QDrag::exec(), and it returns MoveAction, the source is responsible for deleting any original data if it chooses to do so. The QMimeData and QDrag objects created by the source widget should not be deleted - they will be destroyed by Qt.
(http://doc.qt.io/qt-4.8/dnd.html#overriding-proposed-actions)
First give a try with QtCore.Qt.CopyAction
Second, if MoveAction is mandatory, try creating QMimeData and QDrag objects in your source list widget's mouseMoveEvent.
Here in below link, you can find some help for creating QMimeData and QDrag objects in your source list widget's mouseMoveEvent. (code is in C++, My intention is to get conceptual idea).
http://doc.qt.io/qt-4.8/dnd.html#overriding-proposed-actions
I think Kuba Ober is right that this is a Qt bug. In the C++ source code, there is a function void QAbstractItemViewPrivate::clearOrRemove(). It deletes all selected rows, but it does not look at whether each item is drag-enabled or not.
That being the case, I came up with a few workarounds:
Method 1: Make all non-draggable items non-selectable as well
This is the easiest method. Just remove the QtCore.Qt.ItemIsEnabled flag from all non-draggable items. Of course if you want all of your items to be selectable, this won't work.
Method 2: Recreate the "startDrag" function
Since the clearOrRemove function belongs to a private class, I cannot override it. But that function is called by the startDrag function, which can be overridden. So I essentially duplicated the function in Python and replaced the call to clearOrRemove with my own function removeSelectedDraggableItems.
The problem with this method is that startDrag contains calls to a few other functions belonging to a private class. And those functions call other private class functions. Specifically, these functions are responsible for controlling how the items are drawn during the drag event. Since I didn't want to recreate all the functions, I just ignored those. The result is that this method results in the correct functionality, but it loses the graphical indication of which items are being dragged.
class DragListWidget(QtGui.QListWidget):
def __init__(self):
super(DragListWidget, self).__init__()
def startDrag(self, supportedDragActions):
indexes = self.getSelectedDraggableIndexes()
if not indexes:
return
mimeData = self.model().mimeData(indexes)
if not mimeData:
return
drag = QtGui.QDrag(self)
rect = QtCore.QRect()
# "renderToPixmap" is from a private class in the C++ code, so I can't use it.
#pixmap = renderToPixmap(indexes, rect)
#drag.setPixmap(pixmap)
drag.setMimeData(mimeData)
# "pressedPosition" is from a private class in the C++ code, so I can't use it.
#drag.setHotSpot(pressedPostion() - rect.topLeft())
defaultDropAction = self.defaultDropAction()
dropAction = QtCore.Qt.IgnoreAction
if ((defaultDropAction != QtCore.Qt.IgnoreAction) and
(supportedDragActions & defaultDropAction)):
dropAction = defaultDropAction
elif ((supportedDragActions & QtCore.Qt.CopyAction) and
(self.dragDropMode() != self.InternalMove)):
dropAction = QtCore.Qt.CopyAction
dragResult = drag.exec_(supportedDragActions, dropAction)
if dragResult == QtCore.Qt.MoveAction:
self.removeSelectedDraggableItems()
def getSelectedDraggableIndexes(self):
""" Get a list of indexes for selected items that are drag-enabled. """
indexes = []
for index in self.selectedIndexes():
item = self.itemFromIndex(index)
if item.flags() & QtCore.Qt.ItemIsDragEnabled:
indexes.append(index)
return indexes
def removeSelectedDraggableItems(self):
selectedDraggableIndexes = self.getSelectedDraggableIndexes()
# Use persistent indices so we don't lose track of the correct rows as
# we are deleting things.
root = self.rootIndex()
model = self.model()
persistentIndices = [QtCore.QPersistentModelIndex(i) for i in selectedDraggableIndexes]
for pIndex in persistentIndices:
model.removeRows(pIndex.row(), 1, root)
Method 3: Hack "startDrag"
This method changes the drop action from "MoveAction" to "CopyAction" before calling the built-in "startDrag" method. Then it calls a custom function for deleting the selected drag-enabled items. This solves the problem of losing the graphical dragging animation.
This is a pretty easy hack, but it comes with its own problem. Say the user installs an event filter that changes the drop action from "MoveAction" to "IgnoreAction" in certain cases. This hack code doesn't get the updated value. It will still delete the items as though the action is "MoveAction". (Method 2 does not have this problem.) There are workarounds for this problem, but I won't go into them here.
class DragListWidget2(QtGui.QListWidget):
def startDrag(self, supportedDragActions):
dropAction = self.defaultDropAction()
if dropAction == QtCore.Qt.MoveAction:
self.setDefaultDropAction(QtCore.Qt.CopyAction)
super(DragListWidget2, self).startDrag(supportedDragActions)
if dropAction == QtCore.Qt.MoveAction:
self.setDefaultDropAction(dropAction)
self.removeSelectedDraggableItems()
def removeSelectedDraggableItems(self):
# Same code from Method 2. Removed here for brevity.
pass

boost::multi_index_container in QMap

Is it possible to use constructions like QMap< QString, boost::multi_index_container<...> >?
On the one hand, we have private:
BOOST_COPYABLE_AND_MOVABLE(multi_index_container)
in the containers declaration. It should tell us not to put multi_index_container<> into other stl-like container.
On the other hand, such construction
QMap< QString, boost::multi_index_container<...> > _map;
map.insert("bla-bla", container1);
...
auto tmp = _map.value(QString("bla-bla")).get<keyVal>();
//keyVal corresponds to one for the multi_index_container instance
is compiled well with Visual Studio 2012 (+update4 + boost v1.55 + qt v4.8.5).
Experimentally it was discovered that:
auto tmp = _map.value(QString("bla-bla")).get<keyVal>();
makes tmp to handle deleted data.
while
auto tmp = _map.value(QString("bla-bla"));
auto tmp_1 = tmp.get<keyVal>();
leaves tmp_1 with valid data.
Does anybody know what is the proper treatment with boost::multi_index_container<> if we need to put it in another container?
Are there any differences for Qt5.3?
As you can see in the docs for QMap:
const T QMap::value(const Key & key, const T & defaultValue = T()) const
The value is returned by value, not by reference. Besides being very inefficient for large values (such as multi index containers, maybe?) it also returns a temporary.
Now, get<keyVal>(); does return a reference to the first index, but it's the one that belongs to the temporary, which dies at the end of the containg full-expression.
To get a modifiable reference, use
T & QMap::operator[](const Key & key)
Now you can:
auto& tmp = _map[QString("bla-bla")].get<keyVal>();
Note the &.
Adding to #sehe's answer on how to retrieve references rather than temporary values, boost::multi_index_container is copyable and movable. This BOOST_COPYABLE_AND_MOVABLE macro you refer to has to be placed on the class private section (as specified here), but that does not affect visibility of copy/movement ctors. Moral of the story: consult docs before diving into the code.
In my particular case, the problem was that compiler used const T operator[](const Key & key) const instead of T & operator[](const Key & key) because the method of the class was a const one and the map was a field of the class.

Filtering tab completion in input task implementation

I'm currently implementing a SBT plugin for Gatling.
One of its features will be to open the last generated report in a new browser tab from SBT.
As each run can have a different "simulation ID" (basically a simple string), I'd like to offer tab completion on simulation ids.
An example :
Running the Gatling SBT plugin will produce several folders (named from simulationId + date of report generaation) in target/gatling, for example mysim-20140204234534, myothersim-20140203124534 and yetanothersim-20140204234534.
Let's call the task lastReport.
If someone start typing lastReport my, I'd like to filter out tab-completion to only suggest mysim and myothersim.
Getting the simulation ID is a breeze, but how can help the parser and filter out suggestions so that it only suggest an existing simulation ID ?
To sum up, I'd like to do what testOnly do, in a way : I only want to suggest things that make sense in my context.
Thanks in advance for your answers,
Pierre
Edit : As I got a bit stuck after my latest tries, here is the code of my inputTask, in it's current state :
package io.gatling.sbt
import sbt._
import sbt.complete.{ DefaultParsers, Parser }
import io.gatling.sbt.Utils._
object GatlingTasks {
val lastReport = inputKey[Unit]("Open last report in browser")
val allSimulationIds = taskKey[Set[String]]("List of simulation ids found in reports folder")
val allReports = taskKey[List[Report]]("List of all reports by simulation id and timestamp")
def findAllReports(reportsFolder: File): List[Report] = {
val allDirectories = (reportsFolder ** DirectoryFilter.&&(new PatternFilter(reportFolderRegex.pattern))).get
allDirectories.map(file => (file, reportFolderRegex.findFirstMatchIn(file.getPath).get)).map {
case (file, regexMatch) => Report(file, regexMatch.group(1), regexMatch.group(2))
}.toList
}
def findAllSimulationIds(allReports: Seq[Report]): Set[String] = allReports.map(_.simulationId).distinct.toSet
def openLastReport(allReports: List[Report], allSimulationIds: Set[String]): Unit = {
def simulationIdParser(allSimulationIds: Set[String]): Parser[Option[String]] =
DefaultParsers.ID.examples(allSimulationIds, check = true).?
def filterReportsIfSimulationIdSelected(allReports: List[Report], simulationId: Option[String]): List[Report] =
simulationId match {
case Some(id) => allReports.filter(_.simulationId == id)
case None => allReports
}
Def.inputTaskDyn {
val selectedSimulationId = simulationIdParser(allSimulationIds).parsed
val filteredReports = filterReportsIfSimulationIdSelected(allReports, selectedSimulationId)
val reportsSortedByDate = filteredReports.sorted.map(_.path)
Def.task(reportsSortedByDate.headOption.foreach(file => openInBrowser((file / "index.html").toURI)))
}
}
}
Of course, openReport is called using the results of allReports and allSimulationIds tasks.
I think I'm close to a functioning input task but I'm still missing something...
Def.inputTaskDyn returns a value of type InputTask[T] and doesn't perform any side effects. The result needs to be bound to an InputKey, like lastReport. The return type of openLastReport is Unit, which means that openLastReport will construct a value that will be discarded, effectively doing nothing useful. Instead, have:
def openLastReport(...): InputTask[...] = ...
lastReport := openLastReport(...).evaluated
(Or, the implementation of openLastReport can be inlined into the right hand side of :=)
You probably don't need inputTaskDyn, but just inputTask. You only need inputTaskDyn if you need to return a task. Otherwise, use inputTask and drop the Def.task.

Resources