Is there a way to disable keyboard shortcuts (hotkeys) in a firefox add-on?
If your addon has an overlay for browser.xul, an alternative to handling the "keypress" event is to use script in the "load" handler to remove the relevant elements from the browser XUL document like this:
var key = document.getElementById("key_openDownloads");
key.parentNode.removeChild(key);
In case it's not obvious, this removes the hotkey for opening the download manager. You can see all the keys here: http://mxr.mozilla.org/mozilla2.0/source/browser/base/content/browser-sets.inc. This is for Mozilla 2.0 (i.e. Firefox 4.0.x). You can change the tree using the dropdown at the top of the page, but I don't think these keys or their IDs change very often anyway.
Depending on your use case, this might be more convenient than catching "keypress".
Related
We have an application that is using Windows Ribbon Framework for an UI. The app itself is written in Delphi and uses Windows Ribbon Framework for Delphi to interface with the ribbon API.
Our ribbon XML places few commands into the Quick Access toolbar. Use can then remove/add commands either by using the built-in ribbon mechanism (selecting the drop/down button and clicking on a command name) or by selecting More commands command which opens the configuration dialog.
The problem I've encountered is that I cannot find a way to get the current state of commands in the QA collection (whether they are visible or not).
In the example above (picture) I would like to detect that first five commands are checked and that the last is not so I can prepare the configuration dialog accordingly.
I have no problems enumerating the IUICollection and accessing the items stored inside. I can also get the UI_PKEY_CommandId for each item. I cannot, however, find a way to read the checked/unchecked state. I tried reading UI_PKEY_BooleanValue and UI_PKEY_Enabled for all items in the collection, but they do not return that state.
I have also tried to monitor IUICommandHandler.UpdateProperty but it doesn't get called when such item is checked/unchecked (except that it is called with the UI_PKEY_Label key).
Does ribbon API even support this functionality?
I am writing code that capture hotkey after user press control-button in preferences of addon - I want to change the label when user push the button and change again when user push some keys (to display the new hotkey). Official documentation about simple-prefs not mention that I can change something visual in loaded preferences view and in debug I looked into object prefs and saw that my preferences is just a strings - only the values and types.
I do not think you can unless you hack the preferences page. I saw your other question and the example add-on from the answer. It's not using simple-prefs and I don't think it's a great user experience either. I have the same problem and was thinking to use a control button to bring up a panel on which you can listen to the keystrokes and print them there. A panel is all html so you can do whatever you like.
I have an xul overlay based Firefox Add-on that works. It has inline options defined in content/options.xul and they display/function correctly when the user goes to Firefox->Add-ons->Extensions->{The Name of My Extension}->Options
The question, can I a create a button on my Add-on that will launch my options either by taking the user to Firefox->Add-ons->Extensions->{The Name of My Extension}->Options with one click or by launching a dialog based on options.xul?
Right now I am maintaining a separate options.html and options.js that gets and sets the same settings that options.xul handles when the user navigates there via the Firefox button but I would much rather dump options.html and options.js and only maintain my main.js and options.xul.
Any comments or code examples will be greatly appreciated.
This is easiest to do via a global function BrowserOpenAddonsMgr() defined in the browser window, like this:
BrowserOpenAddonsMgr("addons://detail/" + encodeURIComponent(addonID));
This function takes care of focusing the existing add-ons tab if there is one or opening a new tab. It will not scroll down to the options however which is an issue if the add-on has a lengthy description. Starting with Firefox 12 this can be solved by adding "/preferences" to the view identifier:
BrowserOpenAddonsMgr("addons://detail/" + encodeURIComponent(addonID) + "/preferences");
With older Firefox versions you are out of luck (and you shouldn't use this suffix there, it won't work). The other issue is that there can be a lengthy "loading" phase where the add-ons manager fetches metadata for all extensions.
I'm developing an internal Google Chrome Extension that needs a way to initiate print the current page to the printer. I do not want the default Print dialog to come up (so, javascript:window.print() is out of question).
As far as I understand, this is not possible just with the JS + HTML plug-in, so I'm also open to using the NPAPI plugin also (with a dummy mime-type). And I'm concerned for Windows platform only.
I'm also open for various hacks / workarounds if possible, though a standard solution would be nice.
If you think this is not possible, let me know if you know any feature request logged for it?
Any suggestions/clarifications are welcome..
In chrome (v18+) we have the --kiosk --kiosk-printing switches. One can print automatically to default printer without print confirmation.
You can see it from this video http://www.youtube.com/watch?v=D6UHjuvI7IE
Since NPAPI allows you to create native C++ plugins that you can interact with through an object tag (which you can use from an extension), that would probably be the way to do it.
The tricky bit is that I don't know of a good way to get the bits for printing the page. The only person I know of who has done something similar to this actually got the window handle for the browser (available through NPAPI) and scraped the bits off of it to print that way, but that won't take into account print stylesheets or anything. You could also try using automation events to try to control the print dialog, but I have no idea if that would work or not.
By design, the browsers try not to let you do something like this, as it could open some serious vulnerabilities if any website could just start printing things to your printer without confirmation...
Anyway, if you find a way to do it with C++ you can use FireBreath to ease the creation of the NPAPI plugin.
There are various extensions that take snapshots of the current web page (for example, this one); you could adapt one to send the image to a printer via an NPAPI plugin.
I've recently been looking for a similar ability, and it seems like it would be quite possible using Chrome's new native messaging api.
https://developer.chrome.com/extensions/nativeMessaging
There are plenty of examples of this with C#, but here is one quick example of troubleshooting Chrome native messaging with a basic C# application
Native messaging from chrome extension to native host written in C#
I realize this may be a day late and a dollar short, but in case anyone else comes across this question, this is the solution that worked for me. From inside a C# app, you can directly print to installed printers using the PrintDocument class. If you figured out a way to get the page image, this would be far easier than using firebreath or NPAPI.
Disable print preview in Google Chrome on Mac
Quit Google Chrome
Launch Terminal on your Mac. (Search “Terminal” using the Search box)
Type defaults write com.google.Chrome DisablePrintPreview -bool true
Close Terminal and open Google Chrome
Disable print preview in Google Chrome on Windows
Close Google Chrome
From your desktop, right click Google Chrome
Click Properties
In the dialog box, add ‘ –disable-print-preview‘ at the end of the Target field sans the apostrophe (make sure to include the space before –)
Click Apply
Disable print preview in Firefox on Mac
In the address bar type “about:config” and press Enter.
Right click on the page, hover over ‘New’ and click on ‘Boolean’
Type ‘print.always_print_silent’ as the preference name and click ‘OK’
Click on ‘true’ and click ‘OK’.
Close the about:config window.
Disable print preview in Firefox on Windows
In the address bar type “about:config” and press Enter.
Right click on the page, hover over ‘New’ and click on ‘Boolean’ Type
‘print.always_print_silent‘ as the preference name and click ‘OK’
Click on ‘true’ and click ‘OK’.
Close the about:config window.
https://support.dryfta.com/how-to-disable-print-preview-in-chrome-firefox-on-windows-mac/
I put a TLinkLabel on my form, filled it in with a caption including a valid HTML link, and got some nice blue underlined text. When I ran the program, I expected it to invoke Firefox (my default browser) and open the link automatically. Apparently that's not the case.
The helpfile says I have to code this in an OnLinkClick event handler. It doesn't say anything about how to do that, though. It'll pass in a string value called "Link". How do I say "invoke the default browser and have it open Link"?
You can call ShellExecute.
I wrote this method for generic calls, and should works in your case.
procedure ShellOpen(const Url: string; const Params: string = '');
begin
ShellAPI.ShellExecute(0, 'Open', PChar(Url), PChar(Params), nil, SW_SHOWNORMAL);
end;
In your code you should call this
procedure TForm1.LinkLabelClick(Sender: TObject);
begin
ShellOpen(LinkLabel.Caption);
end;
I have all sorts of problems with TLinkLabel that ships with delphi 2010.
a) The control does not render as a hyperlink but as a simple label text on the form. b) the cursor does not change to point out this is a link even though I set the Cursor property. c) the OnLinkClick event does not fire at all.
I am working on windows 7.
So, as far as I am concerned, TLinkLabel does nothing as it should and is useless. ShellExecute is the only solution and must be placed in the OnClick event.
TLinkLabel provides a label that looks like a link. It's your job as the programmer to make it act like a link because only you can know what links are supposed to act like in your program. You wanted the label to automatically open the user's default Web browser using the URL in the label, but that's not the only thing links do. For example:
Internet Explorer is not my default browser, but when I click a link in Internet Explorer, I do not expect the linked page to open in Firefox.
When I click a link in the help program, I expect the linked topic to appear in the help program, not in any Web browser at all.
The preference pages in Eclipse are very complicated. Settings on one page are sometimes related to settings on another page. There are links on those pages that take the user directly to the related page. There is no URL and no HTML involved in this case, and yet they're still labels with underlined text.
Some programs try to offer a choice between opening links in new windows versus re-using old windows. You can't implement that feature without knowing which browser is in use. Your program might offer the user a choice to ignore the default browser setting and always use a specific one. To do that, your UI control can't make too many assumptions about what the program is supposed to do.
I'm guessing you're referring to a TLinkLabel control that comes with Delphi. (My versions don't have such a component.) I imagine that the Delphi control is meant to mimic the one in the .Net class library. It can hold multiple links, and each link can do something different.
If you want a control that always does the shell's default action for URLs, then consider using a different TLinkLabel; the one by Alexander Bach does exactly what you expected. It's from Delphi 3, but it should work unmodified in all later versions as well, including Delphi 2009. If you look at the code, you'll see how it works. It simply calls ShellExecute, as Cesar's answer demonstrates.
LOL, it's funny. So instead of setting crHandPoint as cursor, colored and underlined font and filling the OnClick event to standard TLabel we have component that knows link tag and which at all I need to supply with same On(Link)Click event :))
Only thing it is good for is that it makes easier to embed link into some text and that it is using system style of link...
p.s.: really you have to put Some text with link into the Caption and setup OnLinkClick to that ShellExecute...
I use a control called TInternetLabel instead. It does exactly what you want: on click it opens the browser so you don't have to put code in the OnClick event.
I tried this solution but it still gave problems in Delphi XE4, probably becasue ShellOpen does not understand the HTML-code in the Caption.
What worked for me was a combination of Cesar Romero (the basic code), Adam Feistner (The HTML-code in the Caption) and an older solution:
Put the URL in the HINT field.
Change the line: ShellOpen(LinkLabel.Caption);
to
ShellOpen(LinkLabel.Hint);
This worked for me.