Missing functionality: plist read/write

initWithContentsOfURL implemented for CPDictionary in the attached patch. I suppose there’s no use for initWithContentsOfFile in cappuccino, tell me if I’m wrong.

by Farcaller

Missing functionality: plist read/write

We could just alias initWithContentsOfFile to initWithContentsOfURL. Is there a reason for puting the logic in the class method rather than the init?

It would be great to see this finished up for CPArray, and with the writeTo methods (which should just send a PUT request to the URL with the contents of the plist as the body.

It would also be interesting to see asynchronous versions of these, since synchronous requests are, generally, a terrible idea in the browser.

by Ross Boucher

NSDictionary commited at farcaller/cappuccino@8ea10ab

Any ideas how to make async version and keep it cocoa way?

TabView add item

Could you show me the code snippet?

by Francisco Tolmasky

TabView add item

var tabset=[[CPTabView alloc] initWithFrame:CPRectMake(10,10,300,300)];

var item1=[[CPTabViewItem alloc] initWithIdentifier:@"0"];
[item1 setLabel:@"test"];
[tabset addTabViewItem:item1];
var item2=[[CPTabViewItem alloc] initWithIdentifier:@"1"];
[item2 setLabel:@"test2"];
[tabset addTabViewItem:item2];

[label setStringValue:[[tabset numberOfTabViewItems] stringValue]];

[contentView addSubview:tabset];
[contentView addSubview:label];

by TG

TabView add item

var tabset=[[CPTabView alloc]
initWithFrame:CPRectMake(10,10,300,300)];

var item1=[[CPTabViewItem alloc] initWithIdentifier:@"0"];
[item1 setLabel:@"test"];
[tabset addTabViewItem:item1];
var item2=[[CPTabViewItem alloc] initWithIdentifier:@"1"];
[item2 setLabel:@"test2"];
[tabset addTabViewItem:item2];

[label setStringValue:[[tabset numberOfTabViewItems] stringValue]];

[contentView addSubview:tabset];
[contentView addSubview:label];

-----Original Message----- From: Lighthouse [mailto:support@lighthouseapp.com]
Sent: Thursday, September 04, 2008 11:41 PM
To: Tom Giannulli
Subject: [Cappuccino #10] TabView add item

by TG

TabView add item

Ah yes it’s because the tab view item doesn’t have an associated view ([item setView:myView]). I will check whether the expected behavior should be to raise an internsalinconsistencyexception or to just show nothing in the tabview.

For right now though, if you just do [item setView:[[CPView alloc] initWithFrame:CGRectMakeZero()]]] it should work though.

by Francisco Tolmasky

TabView add item

The error is solved but the tabs/labels are not displayed

by TG

TabView add item - labels/tabs not displayed

var tabset=[[CPTabView alloc] initWithFrame:CPRectMake(10,10,300,300)];

var x=@"tab1";
var y=@"tab2";
var item1=[[CPTabViewItem alloc] initWithIdentifier:x];
[item1 setLabel:@"test1"];
[item1 setView:[[CPView alloc] initWithFrame:CPRectMake(0,0,0,0)]];
[tabset addTabViewItem:item1];
var item2=[[CPTabViewItem alloc] initWithIdentifier:y];
[item2 setLabel:@"test2"];
[item2 setView:[[CPView alloc] initWithFrame:CPRectMake(0,0,200,200)]];
[tabset addTabViewItem:item2];

[tabset setTabViewType:CPTopTabsBezelBorder];
[label setStringValue:[[tabset numberOfTabViewItems] stringValue]];
[contentView addSubview:tabset];

by TG

TabView add item - labels/tabs not displayed

I confirm this as well. To be on the safe side I checked on Safari and FireFox, and neither display anything. I looked through the source a bit, and couldn’t find any trickery to force it to display the labels.

by Will Larson

TabView add item - labels/tabs not displayed

Could you try setting the tabViewType earlier, like right after you create the tab view itself. Not that this wouldn’t be a bug, just trying to identify where the problem is.

by Francisco Tolmasky

TabView add item - labels/tabs not displayed

Tried your suggestion, but isn’t working for me. Here is my source:

@@@ javascript

-(void)setupTabView: (CPView)contentView { var bounds = [contentView bounds]; var frame = CGRectMake(CGRectGetMinX(bounds)+75,

         CGRectGetMinY(bounds)+100,
         CGRectGetWidth(bounds)-150,
         CGRectGetHeight(bounds)-200);

tabView = [[CPTabView alloc] initWithFrame:frame]; [tabView setTabViewType:CPTopTabsBezelBorder]; [tabView layoutSubviews]; [tabView setAutoresizingMask: CPViewHeightSizable | CPViewWidthSizable];

webSearchTabItem = [[CPTabViewItem alloc] initWithIdentifier:@"web"]; imageSearchTabItem = [[CPTabViewItem alloc] initWithIdentifier:@"image"]; newsSearchTabItem = [[CPTabViewItem alloc] initWithIdentifier:@"news"];

webView = [[CPView alloc] initWithFrame:CPRectMakeZero()]; imageView = [[CPView alloc] initWithFrame:CPRectMakeZero()]; newsView = [[CPView alloc] initWithFrame:CPRectMakeZero()];

[webSearchTabItem setLabel:@"Web"]; [imageSearchTabItem setLabel:@"Image"]; [newsSearchTabItem setLabel:@"News"];

[webSearchTabItem setView:webView]; [imageSearchTabItem setView:imageView]; [newsSearchTabItem setView:newsView];

[tabView addTabViewItem:webSearchTabItem]; [tabView addTabViewItem:imageSearchTabItem]; [tabView addTabViewItem:newsSearchTabItem];

//[tabView _createBezelBorder]; [contentView addSubview:tabView]; }

@@@

by Will Larson

TabView add item - labels/tabs not displayed

Sorry for formatting: http://dpaste.com/77487/ is formatted paste.

by Will Larson

TabView add item - labels/tabs not displayed

I’ve been able to narrow down the issue (and find a work around): if you add the tabView to its parent view before adding the tab items, then it works correctly (creates labels and displays the tabViewItem’s view when the tab is selected).

However, if you add the tabViewItems before adding the tabView to the parentView, then it neither creates the labels nor displays the tabViewItem’s view as appropriate.

by Will Larson

TabView add item - labels/tabs not displayed

Great reduction WIll, I think I should have a fix for this soon.

by Francisco Tolmasky

TabView add item - labels/tabs not displayed

by Francisco Tolmasky

TabView add item - labels/tabs not displayed

The issue here is that we delay creating the tab view borders and items as long as possible, so that we prevent unnecessarily loading images that we may not need.

As a side effect, there’s a view we haven’t created yet, and we’re just sending messages to nil. This is why it’s not working. There’s also a related bug with trying to change a view out from under it later.

Essentially, the way this all works needs to be redesigned to work more like the menu system. Since it isn’t a trivial change, I’m moving it off 0.5.2. I’m also not planning on committing a stop gap measure, since the workaround is noted above. Just add items after adding the tab view to your content view.

by Ross Boucher

TabView add item - labels/tabs not displayed

I’ve been able to reproduce this in 0.7b.
Here is the sample code :
http://gist.github.com/83994

by Thomas Balthazar

Didn't I already fix this?

http://github.com/nciagra/cappuccino/commit/f4aa7032f10107858a75971b6a83560bb1d09cd9

-hash is broken on CPArray

Why do both array’s need to return the same hash value? That doesn’t seem to be documented anywhere in the Cocoa docs.

This code seems incredibly dangerous to me. You’re essentially saying that all kinds of arrays that have nothing to do with each other could be considered equal. This will break anything that uses hash as a unique identifier (for example, KVO/Bindings, and I’m sure other code in Cappuccino).

by Ross Boucher

-hash is broken on CPArray

I spoke before I understood what I was talking about.

From the documentation for hash:

@"If two objects are equal (as determined by the isEqual: method), they must have the same hash value. This last point is particularly important if you define hash in a subclass and intend to put instances of that subclass into a collection."

Our implementation of hash does not behave this way. Perhaps we should consider changing our hash to become address, and re-implementing hash the Cocoa way?

by Ross Boucher

-hash is broken on CPArray

I’m not sure I understand you correctly, -[CPObject hash] does return __adress, I haven’t checked all the other objects, but all objects who redefine -isEqual: need their own implementation that matches the contract.

The one which I provided for CPArray - and I checked it against the two other Cocoa implementations I know, which do it the same way. (Also it makes sense: The Implementation needs to be FAST, and it needs to be the same when two objects are -isEqual:

I also checked against some examples on Cocoa and it behaves the same way.

by Martin Häcker

-hash is broken on CPArray

My point was that -hash works differently in Cappuccino than in Cocoa (something I previously had not realized).

We need access to __address as a method, which is what hash has been doing for us. To make -hash actually correct, we’ll need to add -address and update all of the cappuccino code that relies on hashes being globally unique.

by Ross Boucher

-hash is broken on CPArray

Too late to finish sentences...

The one which I provided does fit the contract and also behaves the same way that the others do.

by Martin Häcker

-hash is broken on CPArray

Yes, but it breaks parts of Cappuccino that expect the current behavior. For example, KeyValueObserving and CPWindow both rely on the value of -hash as a unique index into various data structures.

by Ross Boucher

-hash is broken on CPArray

Oh damn.

I just had a look around in the source and lots of objects like CPDictionary, CPData, CPDate and CPNull don’t overide -hash but should. (And most should probably also overide isEqual:)

Uh oh....

Yes, this is going to take some work.

by Martin Häcker

-hash is broken on CPArray

Can’t resist...

As a good person once said: In God we trust - for everything else we have UnitTests.

:)

by Martin Häcker

See also issue #66 which is related to this.

I've taken the first steps to remedy this by replacing all current instances of -hash with the new -UID method, so that when we change hash in the future, it won't break anything.

@import "json2.js" breaks

I’d like to add that just copying the source in a obj-j file also breaks the interpreter - it seems the parser somehow gets confused on the correct javascript code in that file.

by Martin Häcker

@import "json2.js" breaks

by Thomas Balthazar

There are known problems with certain regexes confusing the Objective-J parser.

A workaround is to use the "new RegExp()" form of creating regular expressions (properly escaping \ " ' etc). Alternatively you could include json2.js using a

We now include json2.js in Objective-J so I'm changing this title to the reflect the more general problem with regexes breaking the Objective-J parser.

Partial, working CPScanner and CPCharacterSet.

Sorry I haven’t gotten around to reviewing/checking this in yet, I’ll try to do so tomorrow (today).

by Francisco Tolmasky

Partial, working CPScanner and CPCharacterSet.

by Francisco Tolmasky

Partial, working CPScanner and CPCharacterSet.

Hey millenomi, could you do a pass over this and make it so it conforms to our new style guidelines ( http://cappuccino.org/contribute/coding-style.php ). If not I could do so, but I figure you’d prefer to do it yourself. Also, before we get this in we’ll need you to sign a contributor agreement, which we should hopefully have by the end of today or tomorrow.

by Francisco Tolmasky

Partial, working CPScanner and CPCharacterSet.

Any update on this?

by Ross Boucher

Partial, working CPScanner and CPCharacterSet.

This patch would be very useful with some trivial fixes in the following methods of CPScanner.

1/
-(BOOL)_performScanWithSelector:withObject:into: (see comments in the code about how to handle references)

2/
- (CPString) scanUpToString:(CPString)s should be rather - (BOOL) scanUpToString:(CPString)s intoString:(CPString)value and set the captured string into value (value is a reference).

millenomi, we need you, please come back to finish this important contribution.

by cacaodev

CPCharacterSet and CPScanner are in this repository : http://github.com/cacaodev/cappuccino/tree/master

It's mainly working. Missing methods are included in the file in case you want to finish the job.

I know this sounds obnoxious, but one of the things needed here is some style improvements. In particular, spacing is pretty inconsistent, as are brackets, same line returns with single if statements, some more minor stuff. Accessing window explicitly is generally not a great idea (window._CPCharacterSetTrimAtBeginning = ...).

Francisco can probably comment better on more technical concerns.

I fixed the same line return (or anything) with single line statement. Can you explain to me what is the problem with spacing and brackets because i don't see. I'm following the rule "one bracket = one line" except maybe for one while loop. Also accessing global from window seems to be a workaround done by the original author (millenomi). He says that globals are not exposed to other files. Maybe because globals have to be at the beginning of the file ?

Well, for example, many lines use 2 spaces for a tab, while others use 3, and still others use 4. Some lines are not properly indented at all. Cappuccino prefers that all code be properly intended, using 4 spaces in place of a tab.

Global variables are accessible outside of the scope of the file, var'd variables are not. Putting a var indicates file level scope, while leaving the var out will make it global.

I'm also not sure what those two variables are for. They have an underscore, indicating that they aren't meant to be "public", and yet they also need to be global? I don't see a corresponding NS* in the Cocoa docs, but maybe I just missed it.

setAllowsMultipleSelection: and setSelectable: do not allow multiple selections in CPCollectionView

toggleIndexInSelection:

http://github.com/camwest/cappuccino/commit/04abd8bad1f3ab870e7d01287de04c7434aebb6e

selectRangeFromIndex:toIndex:

http://github.com/camwest/cappuccino/commit/8a471cc838c029fd0c90e1b13012214a720930e9

Minor Code Tweaks:
http://github.com/camwest/cappuccino/commit/e674dbd99629504502f7c3ddf974ce7a944cdb5e
http://github.com/camwest/cappuccino/commit/7bc879b20d1e4627a1cc178cc4c221c94377bf7b

by Cameron Westland

CPTextField focus doesn’t work through different windows

Attached test case.

by nciagra

CPTextField focus doesn’t work through different windows

Temporary fix committed.

by Ross Boucher

CPTextField focus doesn’t work through different windows

by Ross Boucher

Commited to http://github.com/stevegeek/CappuccinoBugs , however the behaviour seems to now have been corrected. Unless this is only the temporary fix mentioned above.

This is the temporary fix.

I am working on this. Using a div and javascript to look exactly the same as the normal tooltip.

CPPredicate suite

Very basic Test case for CPPredicate.

Creates various CPExpression (constant, keypath, function, aggregate)
Creates a CPCompoundPredicate from 2 CPComparisonPredicate
Filters an array with the predicate.

by cacaodev

CPPredicate suite

Improved version.
- fixed bugs in CPComparisonPredicate evaluation for strings - Support for CPMacthPredicateOperatorType (javascript regex) - Partial support for CPDiacriticInsensitive option. Not covering all diacritics. - cleaned code in evaluation of compound predicates. - Added aggregate expression

Coming next: unionset, minusset, subquery expressions. (10.5 new methods in NSPredicate).

Feel free to implement predicateWithFormat: . We need a parser for that so it could be CPScanner or another way to parse (lex ?).

I will post a ojunit test later if needed.

by cacaodev

CPPredicate suite

This is the last version with a lot of code style cleaning.
Also supports for unionset, intersectset, minusset.
predicateWithFormat: still missing.

These are raw files are in a zip. I can also give you a patch against Foundation but anyway you will have to add import directives in Foundation.j yourself because I’m not sure what’s the right way.

Included: CPPredicateTest.j importing CPPredicate from Foundation.

If you have any comments, even code style, I prefer that you post it bellow.

by cacaodev

CPPredicate suite

An updated version of CPPredicate & al with a test case have been pushed into my fork.

http://github.com/cacaodev/cappuccino/tree/master

by cacaodev

CPPredicate suite

Please forget the previous comment.

Here is the commit url in my fork:
http://github.com/cacaodev/cappuccino/commit/6ecc557abb06a21f044778bc3a114d1f2ad8c1c4

And attached is a new patch matching this commit:

by cacaodev

Added documentation for CPPredicate, CPComparisonPredicate, CPCompoundPredicate and CPExpression.

http://github.com/cacaodev/cappuccino/commit/95550024e9a592feba92099d861187bc53313cd9

Need for a CPOpenPanel

This is a challenging task. The first goal is essentially impossible, at least, sort of.

Web apps can’t talk to the native file system, and that isn’t likely to change in the near future. The only exception is using the "File Upload" mechanism (input type="file").

We do need to build support for this into Cappuccino (we have it in 280 Slides, and have the sample code in our gist pastebin). However, beyond that, there’s essentially nothing we can do with respect to local file system files.

As for a generic open panel for remote server files, this is a worthwhile goal, but it isn’t obvious how it would work. The interaction is very server component specific. How do I get a list of available files? Do I support folders? Are there previews for these files available? Metadata?

It may be that we’ll make a set of default assumptions, built a somewhat generic looking open panel, and then see where it takes us, but right now I think the project is better served by focusing on slightly more generic tasks (specifically ones that don’t require us to worry about server interaction).

All that being said, if someone wants to volunteer to work on this, then that’s great. I’m happy to think things through with someone.

We can also consider releasing the 280 Slides open panel as sample code. It would not be directly reusable (its fairly specifically oriented towards slides), but it would be a reasonable starting point for anyone.

by Ross Boucher

Need for a CPOpenPanel

The Import Button on the 280slides.com does not seem to work for Opera.

by Thrinz

Need for a CPOpenPanel

It also doesn’t work in some versions of IE, but I haven’t been able to determine which versions....

It’s not part of Cappuccino though, and for several reasons.

by Ross Boucher

CPSplitView doesn’t honor splitView:constrainMinCoordinate:ofSubviewAt: when resizing window

The docs on Apple’s page say the following:

This method is invoked before the NSSplitView begins tracking the cursor to position a divider. You may further constrain the limits that have been already set, but you cannot extend the divider limits. proposedMin is specified in the NSSplitView’s flipped coordinate system. If the split bars are horizontal (views are one on top of the other), proposedMin is the top limit. If the split bars are vertical (views are side by side), proposedMin is the left limit. The initial value of proposedMin is the top (or left side) of the subview before the divider. offset specifies the divider the user is moving, with the first divider being 0 and going up from top to bottom (or left to right).

So it sounds like it should only be called during repositioning of the divider. Could you possibly do a test against Cocoa to see whether this is actually the case?

by Francisco Tolmasky

CPSplitView doesn’t honor splitView:constrainMinCoordinate:ofSubviewAt: when resizing window

Ok, I wrote a little cocoa app and this method does get called when the window is resized as well as when the divider is repositioned.

I attached the project if you care to see for yourself...

by Patrick Crosby

CPSplitView doesn’t honor splitView:constrainMinCoordinate:ofSubviewAt: when resizing window

Cool, I’ll try to fix this tomorrow then

by Francisco Tolmasky

CPSplitView doesn’t honor splitView:constrainMinCoordinate:ofSubviewAt: when resizing window

This hasn’t been fixed in 0.7b yet.
Here is a test case :
http://gist.github.com/83999

by Thomas Balthazar

http://github.com/stevegeek/CappuccinoBugs/tree/master/i0041/ I have commited the test here, and I do not see the problem anymore

in which class?

Add references to Objective-J @reference?

by Francisco Tolmasky

Improve the parser to detect syntax errors

By all means take a stab at it.

by Francisco Tolmasky

Improve the parser to detect syntax errors

Ok, no promises but I’ll give it a shot and see if I can provide a patch.

by Chris Corbyn

Objective-J preprocessor chokes on certain RegExp literals

(from [abbbc6e]) Replaced regex literals with RegExp objects, due to bug [#117] http://github.com/280north/cappuccino/commit/abbbc6e8f09b9bee14f7b794c83712492ee1cffd

by admin (at 280north)

Objective-J preprocessor chokes on certain RegExp literals

Yes it does

by Francisco Tolmasky

Is this still valid? which expressions cause it to choke?

CPWindow setMinSize does not set overflow properly

CPWindow can not (or at least should not be able to) be resized smaller than its minimum size, or larger than its maximum size. I’m not sure what you’re proposing.

by Ross Boucher

CPWindow setMinSize does not set overflow properly

Sorry, I meant to specify this is relating only to the bridge window. By default, you remove the browsers scroll bars. However, I feel that the bridge window’s setMinSize should implement special behavior to add the browser’s scroll bars back if the user resizes their browser to less than the minimum size (which is not prevented by setMinSize).

by nciagra

CPWindow setMinSize does not set overflow properly

We should just prevent this from happening, that can be done right?

by Francisco Tolmasky

Multiple non Javascript-native instance variable declarations per-line don’t work

This is actually due to us not supporting the comma notation for multiple declarations of the same time. It’s surprising that any of them worked.

by Francisco Tolmasky

I'm guessing the int one didn't actually work, and the failure case just wasn't detected.

I played around with this a bit today (not explicitly WebViews and iFrames, though). It appears that an overflow property being set to anything other than hidden within the iFrame causes this problem.

Also, the issue is still relevant in 0.7.

I feel like this is a bug in Firefox that we won't be able to work around, or were you able to in your experiments find some way to actually solve the problem that keeps webviews working?

The problem is that press is not aware of the classes used in cibs, so they may be stripped.

Workarounds:

• use the "--nostrip" option if you only care about flattening, not reducing file size

• explicitly import all classes used in the cibs, possibly at the top of AppController.j (don't just import "AppKit/AppKit.j" since it is ignored)

I got things working with the --nostrip option, which is great. I'll also try the the explicit import to see if I can get that working. Many thanks.

I've commited a patch for this at: http://github.com/maport/cappuccino/commit/93cc5700c8e39e84fded1a4b4ba3d69f03a12dd8

This adds a new orientation property which can be set to CPCollectionViewHorizontalOrientation to "orient" the view horizontally rather than the standard vertical arrangement:

- (void)setOrientation:(CPCollectionViewOrientation)anOrientation

In this orientation the view has fixed height, items are arranged in columns starting on the left and the width is varied.

It is particularly useful if you want to render a one row list that scrolls horizontally.

The patched CPCollectionView should be completely backward compatible and users who don't use setOrientation shouldn't encounter any differences. However I have deprecated setVerticalMargin in favour of a new setDivisionMargin method since the "vertical margin" is the margin between rows in the standard vertical orientation but becomes the margin between columns in the the horizontal orientation.

Any better suggestions than setDivisionMargin welcome; a "division" is the general term I am using to represent a row or column.

This isn't a split view bug, its a bug with CPWebView, which doesn't correctly pass events in Firefox it seems.

What is the status of CPSound? I would say this is a crucial part of some application. I took a look at your branch and looked through CPSound.j but it's hard to tell what's missing in it (if anything).

I just uploaded the latest changes to my own branch. I have implemented most of the useful stuff. I currently use quicktime, and the audio tag if QT is not supported. However, more addons should be added. Also the pause/stop behaviour seems to be different depending on what tag is used. Also I do not know what browsers this works in, would help if somebody checked.

I can see the desired behavior. I'm not sure what the best approach to implement it would be though. If you have a more concrete proposal, or better yet code, I'm happy to try it out and provide some feedback.

Objective-J runtime Method js object holds the selector (name), IMP (js function) and types of the parameters (not managed for now).
So the runtime does the job : when sending a message on a receiver object, it searchs in its methods for a method with same selector/name, and types (not done for now) and returns the function. It is the same in ObjC.

In Java the programmer has to do that, either by static/compiled call to the function (passing parameters objects values), or by using reflection APIs (pass the parameters types in a Class array).

I am working on this - proposed fix at http://github.com/saikat/cappuccino/commit/8a2788f7ad4d9a2423a004f8b991bf42e3588efe

Some additional discussion: http://github.com/280north/cappuccino/issues/#issue/447

My proposed fix link is broken, but the proposed fix was to just get rid of the nextResponder propagation in CPResponder for mouseEntered and mouseExited. Discussed here - http://groups.google.com/group/objectivej-dev/browse_thread/thread/b37e35412c4104/b09f7db0347698cf?lnk=gst&q=mouseEntered#b09f7db0347698cf . This would follow the Cocoa implementation but not docs (which I think is more intuitive).