I would like to program a WYSIWYG editor for HTML. I'm looking for a high level approach, which I will eventually be implementing in C++.
My initial approach is to create a hierarchy of classes which extend a common base class (node). So object "body" would contain object "p" which would contain object "b" which would contain some text.
class node {
node *parent;
vector<node> children;
string name;
map<string,string> attributes;
string text;
virtual void render(const rect &rect, const point &offset) = 0;
virtual void onEvent(const event &e);
}
The main engine would call something like body.render(screen, point(0, 0)), which would recursively render its children.
The cursor would be represented by a pointer into the object hierarchy, and each node would have its own internal cursor state, and would respond to keyboard events when it is the selected node.
For example, if the user hits the left-arrow, and the "p" node is selected, the "p" node's reaction to the keypress might be to change the current node to the parent of "p".
Abstractly, it seems like this could work, the closest thing I can find to what I'm looking for is Sigil, which at first glance seems pretty intimidating to study (main.cpp is 70k).
Before I go down this road, I was wondering if anyone had a simpler approach, or can see any pitfalls with this method.
A simpler approach, if you can do this, is to embed a web browser in your application and set contentEditable="true" on <body>. If you're on Windows, you can use the built-in one, or you can embed an engine like Gecko or WebKit.
You can look at doing color picking.
Basically you render everything twice. Once is what is displayed to the user.
The other is a bunch of bounding boxes done in unique random colours that you can turn back into pointers via a hashtable lookup or whatever. This allows you to quickly locate what is under a mouse cursor without having to recursively look up every object. You might want to miss the currently selecvted object from the color picking that way you can click twice to get the object below you current one (if you want to keep going for lower you will need to mess around with zorder depth somehow. Maybe don't render anything in the selected elements bounding box region that is over the zorder of your currently selected object).
Finally once you have selected the specific element you can enter a sub editing mode. For example turning a block of text into an editable text box with a blinking cursor, drawing borders/handles around your element to allow you to resize/manipulate it and showing a list of properties.
Realistically if your looking at making a HTML WYSIWYG editor, that will also require a HTML rendering engine. Unless you plan to make your own from scratch (which is a massive undertaking by itself), or only plan on implementing a very limited subset of HTML (which would still be fairly time-consuming). Most HTML renderer engines will include some kind of support for tapping into things like object picking. It might be worth grabbing webkit and seeing what it supports. Maybe look at the Chromium source code (Chrome includes an inbuilt webpage inspector that isn't too far off being an editor (it allows object picking and runtime editing of properties but the editing isn't totally WYSIWYG since its done in an external window, but that might be enough for what you want.
You might even do better to look at writing you editor (or parts of it) as some kind of JavaScript/Grease Monkey extension that you can load over the page being edited. You could write your whole editor in HTML or just add some support handle wrapper script that communicates to your native program.
Related
I have a CEdit-Box containing some text, which is being refreshed by a thread every ~0.25 seconds. The problem is that everytime the text is being refreshed, a possible selection of text is being erased.
I found 2 ways to avoid this so far:
My implementation right now (1) :
Use quite a passage of logic in order to assure that the text is actually really changing, and not just refreshing itself with the exact same string. This is kind of avoiding the problem, but it feels very clunky to be honest.
Another idea (2) :
Every cycle, before the text is being refreshed, we need to grab the selection we currently have, store it, and try to reconstruct it after our text has been refreshed. However, I don't know how this would turn out if the new text doesn't contain our old string at all. I guess the functionality is implemented in WTL, but I don't think this is a very good approach.
Is there any other way? Something like a control setting which would do something like this?
I may be bordering on a discussion type question here so I apologise if the question is not specific enough.
I would like to know if I my current application design is inherently weak/flawed. My background is C so I am not using clever c++ patterns to their fullest extent, of this I am sure.
My application is similar to a 3D modelling package, without geometry creation (e.g setting up animations using existing models). The user imports a geometry and can set various parameters on pre existing geometry, and there are time dependent values that relate to the whole system. The output is a text file to be processed by another application.
I am using a QTreeview to render a QStandardItemModel. I simply store pointers to my core classes in the model's items. These have specific UI for each class type, and are all descended from a common base class. They all have a QWidget which is their "mainwidget"
When the user clicks on part of the treeview, the stored class is retrieved and its mainwidget is displayed on a pane on the UI. So - treeview on the left, pane to the right with the current item's contents, and a 3D view showing the geometry.
Most of my data is stored in the classes UI elements themselves; I don't have a central database which stores anything, and when it's time to save the project, I traverse the tree and let every item write itself to a QSettings file. This feels quite naive but it does work, and the reverse happens on project load. The project class generates new classes based on the type information in the settings file and they then read the contents out of the file themselves.
Similarly when writing the output file, each item knows how to write itself and does so. Where other classes can influence the output of others (for example, start and end times), higher level classes process the children and will set start and end times based on the order and duration of each child.
Should I be storing more data in the QStandardItemModel itself, or defining my own model perhaps? Does it sound like I have set myself up for future problems?
At the moment I have modified this system a couple of times to provide customised applications, but I am about to try and make it more generic. I welcome suggestions for improving my design. Go easy, please!
You should try to avoid creating god objects. Split your tasks and duties into smaller chunks. It makes it much easier to maintain and also much easier to extend if you need to.
Your specific use-case would benefit a lot from a more complete use of the Model-View-Controller pattern.
What doesn't make sense in your design is that your data objects hold a UI element. Since only one item can be shown in the right pane, this seems like a waste of resources. It makes more sense to have a single object that then gets passed a data object to display.
What I suggest for your program is the following:
Splitting your data into classes that only have functions for reporting and modifying values. There should be no knowledge of how to display data or store to a file.
Create separate class that handles the reading and writing from a file. If your model is very simple, you could just use single functions to do this using the method shown in the documentation for QDataStream or QTextStream.
Use your QTreeView and QStandardItemModel as Adaptor class between your data objects and the left pannel.
Create a controller class that gets informed by the QTreeView if data needs to be displayed in the right panel. The controller will then retrieve the data item and pass it to the right panel in order to be displayed.
The right panel should act like another View class with the sole responsibility of graphically displaying the data passed in from the controller class.
One advantage of doing it this way is that, if there are different categories of data that get displayed differently in the right panel, the controller could examine the selected data item, determine what the category is, create a view widget to put in the right panel, and then pass it the data to display.
This pattern is open-ended as far as extendability is concerned because you do not need to change your data classes if you need a new display, you merely need to create a new right-panel widget, and then teach your controller class how to determine when the new view should be used.
Another advantage of this pattern is that you only ever need to have a single widget created to display data in your right panel. If you change your selected item, you can just pass it to the view class that already exists and get it to refresh its display with the newly selected data. You only need to distroy the right-panel view widget if a different category of data object is selected and its data needs to be drawn in a different way. The controller class can determine whether a right-panel view widget can be re-used or whether it needs to be swapped out for a different widget.
We have a list that uses a custom renderer containing a label, a checkbox and two icons (which have click events). This list needs to be made WCAG 2.0 compliant and in order to do that we need the list to be keyboard navigable.
The problem is with being able to move from one list item to the next and have the focus move to the label for the next/previous list item. Specifically, when the user enters the list using TAB button, the label for the first list item receives focus (highlighted box around text) and the entire row in the list is highlighted as the selected item.
However, when the user then presses the down arrow key to move to the next list item, the next row becomes highlighted (is now the selected item) but the focus remains on the label of the previous row (highlight still shown around label for row 1). The only way to get the focus to move to the newly selected row is to tab through the checkbox and two icons. This isn't a big deal if there are only a couple list items but would be a pain if there are 20+ rows in the list.
Is there a way to get the focus to move to the label of the newly selected row as soon as the user moves (using up/down cursor keys) to the new list item? I know a picture would help but I don't have anyway of posting a screenshot online. Any help would be greatly appreciated.
You're going to have to dig into how focus works in Flex. This is not a complete answer, but hopefully you can put together a solution that works for you. I did this about 4-5 years ago in Flex 3, but it should be similar in Flex 4.
How Focus Works in Flex
The main things to know are the FocusManager singleton class and the IFocusManagerComponent interface.
The FocusManager moves the focus around the UI based on user interactions (mouse clicks, keyboard navigation, etc.).
If a component implements the IFocusManagerComponent interface, then the FocusManager will include it in the "tab" loop and allow the component to be focused via keyboard navigation.
How Focus Works With Flex List Components
You've already stumbled onto the peculiarities of how focus works with the List component and item renderers. The Flex List components implement IFocusMangerComponent and so when you tab through the UI the FocusManager sends the focus to the list.
The List may or may not focus the item renderers. In Flex 3 you had to be using editable item renderers for this to happen, it may or may not be the same in Flex 4.
Some Ideas for Solutions to Your Problem
I think there are numerous ways to solve this. Use some combination of these techniques:
override the protected keyDownHandler() method of the List component. I don't have the code handy, but if you look at it's implementation in the List class you should be able to make your overridden version set the focus on the next renderer.
use methods of the FocusManager to find components in the tab loop: getNextFocusManagerComponent(), findFocusManagerComponent(). Check the docs there are others that will be useful. For example, when the user presses the down arrow, you can let the next item renderer get selected, then use findFocusManagerComponent() (passing in the newly selected renderer) and then tell the FocusManager to focus it with the setFocus() method. This is probably not exactly the right approach ;)
By the way, the FocusManger is a Flex singleton object, every UIComponent in Flex has a focusManager property you can use to get a reference to it.
consider disabling focus on objects that don't need to receive focus (like the Label in your item renderer). There are numerous properties to do this: focusEnabled, hasFocusableChildren, mouseFocusEnabled, tabEnabled, tabChildren etc.
consider disabling focus on the List component, but then making your item renderers implement the IFocusManagerComponent interface. Implementing the interface is simple, you just declare it in your class (there's no actual methods to implement). The tricky part will be now your item renderers need to have key down handlers (just override the protected keyDownHandler() method that all UIComponent objects have).
I think there are other techniques you can use, it's just been too long since I did this. I'd be happy to provide more help if you get stuck somehwere...
I am making a new graphical menu interface for a project I am making. I don't want to use the menu system provided by windows APIs and want to make one from scratch.
My question is, what is the best method for setting up the structure?
I'm thinking I will need a menu item object, each of which will have to have their own item array list, etc...
Is it considered sloppy to have recursive coding like that? (Ie an object which contains objects of itself, which contains objects of itself, etc...)
I'm thinking I can give the item object a draw interface which checks itself to see if it has an item array that is not null. If it does, it executes the draw command all the way down, thereby giving me a menu with (for my purposes) unlimited submenu level
In my opinion your approach is fine. In nearly all UI frameworks, views contain views as subviews after all.
But the thing is that writing drawing code is too much work for small projects I think. I would consider using a UI framework such as QT and use its view mechanism as a starting point. You can write your own Menu class which will be a subclass of generic View class in the framework.
I have been away from C++ for a couple of years now doing AS3/Flex work. I have gotten used to being able to create a component and place it in design mode with very little fuss and I am struggling to get my head around the C++ Builder way of doing the same thing.
I have written many components for C++ Builder in the past, but none of them have been visual. What I would like to do now is create a component for customer search and another for order processing because I want to be able to create a new instance of these on the fly. What I don't want to do is have to place each of the components like the dbgrid and search fields manually in code. I would like to do this (as well as set their properties) in design mode.
How do I go about this? I have browsed the source for other Delphi components and I notice they have dfm files which seems to be what I need. How do I do this in C++ Builder? The only option I see is to add a new form if I want a dfm, but this isn't what I want as my components will be based on TPanel.
Is there a way to do this or do I have to resort to doing it all in code with no visual reference?
Pursuing the DFM idea I did a test this morning where I created a component based on TPanel and added a new form to it which I create and parent in the constructor of the component. In design mode I set the form border to none and placed a grid on it. This all looks OK until I place the component in my application, at that point it looks like a panel with a standard looking form in it and the grid is missing. If I run the app the panel shows as expected, borderless and with a grid. The DFM appears to be ignored in design mode for some reason.
If you know a better way to do this than using components then please give me some pointers.
Any help and advice will be appreciated beyond words
You might want to have a look at frames (look for "Frame objects"). They are "subforms" you can design visually and then place on forms.
Just as an FYI item, you can also drop the panel on a form, put any other controls on it, position them, set captions, etc..
Now, click the panel to select it, and use Component->Create Component Template from the IDE's main menu. This creates a unit you can install as a component which will add the panel and it's sub-controls (as a single component) to the IDE's component palette.
Of course, you can then modify the source for that new component like any other component source.