We are planning the next release: Lazarus 2.0.6

[jamie]

I looked at the current list of fixes, I don't see my changes I requested in there unless it was a midnight silent run?

 1. On two of my systems the Html Help example can't locate the default browser, I showed the
defective code and what to do to fix it.
 
 2. The ShellTreeView is basically useless by itself unless it has a OnEntryEvent to select weather we can reject the entry or even alter its name or place a Filter property in it that can do the same, but the Event would be better, more flexible.
 
  With the ShellTreeView its only logical use is with the ShellListView, the two should have been married together as one component. But currently they are not and it just sits there on the component bard wasting space by them selves in my view. The two of them should be flexible to be used stand alone in any manner as we please.

 Thanks for listening to my whining.

[Martin_fr]

Do you have the link to the bug in mantis?

Or -assuming it is fixed in trunk - do you know who fixed it when (revision number)?

[Bart]

2. The ShellTreeView is basically useless by itself unless it has a OnEntryEvent to select weather we can reject the entry or even alter its name or place a Filter property in it that can do the same, but the Event would be better, more flexible.

It has OnAddItem event in trunk.
I implemented that, based on your complaints in another forum thread.
See Lazarus 2.2.0 release notes.
Mutatis mutandis for TShellTreeView, which in trunk now also supports case-insensitive masks on *nix.

It's not a bugfix but an added functionality, it's not Delhi compatible.
Therefore it was not put on the merge list for the 2.0 fixes branch.

Bart

[jamie]

Thank you very much, you are a gentleman and a scholar.

 Now we need to address the issue of the HTMLHELP example failing with those that have Microsoft Edge in their PC's

 I have two PC's with Win10 that has Edge in it, the code to fine the default browser fails.

 If I disable this code frag that looks for Edge specifically then the system follows through and loads Edge anyways if that Is the default browser.

 If this hasn't been fixed I can point you to the source of the hardship.

[Thaddy]

Now we need to address the issue of the HTMLHELP example failing with those that have Microsoft Edge in their PC's

AFAIK OpenUrl works? Can you show the issue?

[ptvs]

We would invite everyone to provide their feedback to help us improve this upcoming release. Please let as know in particular:
- Any bug-fixes already made to trunk, that you believe should still be merged to the fixes branch

0035043: Add checkbox to show/hide property filter in object inspector
0035044: Hide toolbar buttons's captions in Call Stack window
0035270: Handling mouse events through shaped TWinControl

[Thaddy]

None of those are showstoppers. Bugs are, feature requests not. Focus plz.

For noobs and ill-informed:
A release candidate  is feature complete, new features will NOT be added.
You should focus on bugs. Nothing more. Maybe Martin should add a sticky?

As you can see and read for yourself, some of your issues are solved in higher - not release - versions.

[ptvs]

A release candidate  is feature complete, new features will NOT be added.
You should focus on bugs. Nothing more. Maybe Martin should add a sticky?

Ok, I can agree that 35043 and 35044 are features, though they only fix some minor UI inconsistency.
But 35270 is clearly a bug fix.

As you can see and read for yourself, some of your issues are solved in higher - not release - versions.

That's only one issue, others have empty TargetVersion.

[Thaddy]

So in that case focus on the bug...? That's all, and good testing. Devels want to know about real bugs.
Since it is already fixed there is a slight chance of making it into the release. (border case in my view).

OTOH if it is not a clear feature and the feature set was already frozen, you will have to wait for the release after the current proposed one.

[JuhaManninen]

But 35270 is clearly a bug fix.

Yes. I now added it to the list of revisions to be merged.
I anticipated some regressions. The original code had comments and all to describe its behavior. It was there for some reason at least in the past.
Now half a year has passed without complaints, I guess it is safe to merge it.

[Thaddy]

Thanks for explaining and we hope for a nice release. (Which it usually is)

I am sure ptvs is now satisfied too?

[Akira1364]

The presence of an official, normally-downloadable 64-bit Cocoa installer is by far the most urgent thing for this release, I'd say.

AFAIK the Carbon-based IDE simply doesn't work on the latest OSX release, and cannot be made to work.

Otherwise, for example, when someone inevitably posts the announcement on Reddit, the thread will just devolve into a bunch of people commenting about how Lazarus "isn't really as cross-platform as it claims" or similar sentiments, which is hardly good PR.

[Thaddy]

The presence of an official, normally-downloadable 64-bit Cocoa installer is by far the most urgent thing for this release, I'd say.

And the last one to be finished... (by logic)

Otherwise, for example, when someone inevitably posts the announcement on Reddit, the thread will just devolve into a bunch of people commenting about how Lazarus "isn't really as cross-platform as it claims" or similar sentiments, which is hardly good PR.

You are perfectly capable to compile from source. Not everyone is.
This is something that hunts all open source projects: expectations because in the future there are more options.
Better to educate our users in what to expect. ( and I am just a simple user).

[Akira1364]

You are perfectly capable to compile from source. Not everyone is.

Right, that's the point. Many people won't bother with Lazarus at all if they can't just download a working version directly, regardless of the various ways it's technically capable of being compiled.

At the very least, if it's really the case that a normal installer can't be put together for OSX in time, it would likely be a good idea to include a mention of FPCUpDeluxe in the announcement so that people are aware they have other easy options.

Moreover, FPCUpDeluxe is useful enough and widely used enough that permanently mentioning it on the actual websites for FPC and Lazarus would probably be a good idea too...

[dbannon]

Now, please forgive my ignorance but isn't the cocoa issue 'just' one of packaging ?

If the install chooses to put fpc and its libraries in user space, $HOME/somwhere/, in the same way many of us (in particular) Linux user put our svn built Lazarus, that would no longer offend Apple and with appropriate PATH and LDPATH all would work ?

Maybe we could argue that the installer is not actually part of 2.0.6 or any other Lazarus version ?

I agree its important, any claim to being cross platform depends on Mac.  Carbon no longer convinces anyone. But the installer could happen quite independently of 2.0.6

Davo