ERROR
The request could not be satisfied.
The Amazon CloudFront distribution is configured to block access from your country.
Generated by cloudfront (CloudFront)
Request ID: 1kgdo0O7td23ZodieyG8Cobs0nCNH4v0Mj8rPQ4mRnCq4gENCNRhVg==
[snipped...]
Please test:
- If you can edit forms in the designer
- rename components / change properties in Object inspector / Add new events
[snipped...]
Use your keyboard to move to Default property (or any Boolean property)
It seems that the focus in Object Inspector moves past Boolean properties on [arrow] key up event. Using the mouse, I can select any property with no problem.
Am I the only person having this issue?
Use your keyboard to move to Default property (or any Boolean property)
What do you mean by Default property?
QuoteIt seems that the focus in Object Inspector moves past Boolean properties on [arrow] key up event. Using the mouse, I can select any property with no problem.
Am I the only person having this issue?
No you are not!
Use your keyboard to move to Default property (or any Boolean property)So, this phrase should look like:
Using MS Windows:
Drop a button on a form
Hit F11 to move the focus to the Object Inspector window
Use your keyboard to move to Default property (or any Boolean property)
QuoteUse your keyboard to move to Default property (or any Boolean property)So, this phrase should look like:
"Use your keyboard to move to any Boolean property"
Using MS Windows:
Drop a button on a form
Hit F11 to move the focus to the Object Inspector window
Use your keyboard to move to Default property (or any Boolean property)
Yep, same in trunk (only tested Windows).
Reported as issue #29406 (http://bugs.freepascal.org/view.php?id=29406).
Unlike many other controls checkboxes handle arrows own for select next/previous control. The problem somewhere here.
Why did I not notice it earlier? ...
"Object Inspector: The Boolean property editor now uses a CheckBox by default. A ComboBox with "False", "True" can still be selected in options. The control is now a custom drawn TCheckBoxThemed which is used for both selected and non-selected properties, thus making them visually identical. "
It is much better to leave these kind of overengineered features off by default.
No need to shove them down our throats.
Hi
Just done a fresh install of 1.6rc2 on osx 10.6.8.
At first it seems ok, so I tried to just rebuild lazarus
and get the following error.
LAZUTILs error code 256
and cant rebuild lazarus.
Obviously I have not tried to install any packages as they will fail..
Any Ideas
Josh
Blaazen, are you reading this? Any idea why TCheckBoxThemed affects the key presses?
Anyway I will look at this during the weekend.
Just wondering... Why did all my previous installed component-palletes dissapeare?
Just done a fresh install of 1.6rc2 on osx 10.6.8.
At first it seems ok, so I tried to just rebuild lazarus
and get the following error.
LAZUTILs error code 256
and cant rebuild lazarus.
Any steps/tests you want me to do let me know
Why do you feel that a CheckBox for a boolean value is overengineering?I don't feel that way.
CheckBox is the most natural control for booleans. That is what CheckBox was created for in the first place.I agree to some extent.
That's why I changed it to a more natural and logical control.No problem with the change by itself, as long as the options to fall back to the original, more efficient behavior, are enabled by default (indeed a checkbox in the right location).
Anyway, I did not quite understand your reasoning but fortunately you can configure the boolean property editor to your liking.See above and yes, that is fortunate indeed.
@joshProbably the symlink inside the bin folder points to the wrong version of fpc. You should remove fpc2.6.4/Lazarus 1.4.4 completely:
I could create a simple GUI app with a tbutton, and this would compile and run, but I cannot re-build lazarus, or add any components that require a rebuild of lazarus to function.
I have tried re-freshing lazarus files from menu, and removing all pp files as a last ditch attempt, but this did not help.
It maybe something in my config, but odd as to 1.4.4 running.
the main difference is fpc 3.0.0 over 2.6.4..
Just wondering... Why did all my previous installed component-palletes dissapeare?
You must rebuild Lazarus, just like when installing new component packages.
Tools -> Configure "Build Lazarus" ...
The OI is now broken with an inconsistent look and feel:
the checkboxes create a very 'busy', irregular OI window; that by itself is enough reason to not activate such a feature by default;
the checkboxes are no longer consistent with the way the other options work (black text, white background and blue highlight color when selected/clicked);
also the color to indicate a non-default value is inconsistent;
when going through the options with the cursor keys the checkboxes are skipped; in case of 2 adjacent true/false options the cursor stops at the 2nd one :(
and the piece de resistance: for people that do not know the meaning of a v-shaped symbol in a small white box, the text '(true)' is placed to the right of it %) (if a text would have been required for whatever reason it should have been yes/no or on/off and no parentheses).
No problem with the change by itself, as long as the options to fall back to the original, more efficient behavior, are enabled by default (indeed a checkbox in the right location).
'more natural and logical control' does not apply to his particular situation.
Well IMHO both concepts (checkbox and dropdown) are valid (regardless of implementation details)The OI is now broken with an inconsistent look and feel:
the checkboxes create a very 'busy', irregular OI window; that by itself is enough reason to not activate such a feature by default;
Our minds work in different ways. For me it appears now as a more functional OI window.
Hint: (11030) Start of reading config file /etc/fpc.cfg
Hint: (11031) End of reading config file /etc/fpc.cfg
Free Pascal Compiler version 3.0.0 [2015/11/14] for i386
Copyright (c) 1993-2015 by Florian Klaempfl and others
(1002) Target OS: Darwin for i386
Free Pascal Compiler version 3.0.0 [2015/11/14] for i386
Copyright (c) 1993-2015 by Florian Klaempfl and others
(1002) Target OS: Darwin for i386
(3104) Compiling alllclunits.pp
(3104) Compiling lclproc.pas
<stdin>:4284:2: error: invalid instruction mnemonic 'fistpq'
fistpq -12(%ebp)
^~~~~~
<stdin>:4303:2: error: invalid instruction mnemonic 'fistpq'
fistpq -12(%ebp)
^~~~~~
<stdin>:4326:2: error: invalid instruction mnemonic 'fistpq'
fistpq -16(%ebp)
^~~~~~
<stdin>:4350:2: error: invalid instruction mnemonic 'fistpq'
fistpq -16(%ebp)
^~~~~~
(9001) Assembling (pipe) /Developer/lazarus/lcl/units/i386-darwin/lclproc.s
lclproc.pas(3221) Fatal: (10026) There were 1 errors compiling module, stopping
Fatal: (1018) Compilation aborted
Our minds work in different ways. For me it appears now as a more functional OI window.That's why programmers normally are not involved with designing an application (and I'll skip all other remarks that are programmer driven.)
What drugs are you on?Quotealso the color to indicate a non-default value is inconsistent;That is false information. The color is the same brown as can be seen easily.
That looks like it is compiling for 32 bit.
Target OS: Darwin for i386
if double click - not available Copy and Cut menu items in the editor. :-\
But if select word then items is available.
QuoteWhat drugs are you on? Look at the attachment.Quotealso the color to indicate a non-default value is inconsistent;That is false information. The color is the same brown as can be seen easily.
Look at my number of messages. It matches the current year (until I write another message). Heh!
if double click - not available Copy and Cut menu items in the editor. :-\
But if select word then items is available.
Works here. Please explain how to reproduce.
if double click - not available Copy and Cut menu items in the editor. :-\
But if select word then items is available.
Works here. Please explain how to reproduce.
see attachment. 2 screens.
if double click - not available Copy and Cut menu items in the editor. :-\I can confirm it. Also if doubleclick or simple right mouse click on some identifier then does not work "Find declaration" and "Refactoring"->"Rename identifier...". Looks like Lazarus can't determine that keyword is active but hotkeys works as expected so it is only context menu problem. I am using this function very often so it is very annoying for me.
if double click - not available Copy and Cut menu items in the editor. :-\I can confirm it.
Anyway, fixed in r51318. Have you reported in mantis?No, only here. Thanks, I will check it. I hope it will be merged to fixes_1_6 branch.
I hope it will be merged to fixes_1_6 branch.
Anyway, fixed in r51318. Have you reported in mantis?No, only here.
Anyway, fixed in r51318.I checked it, works as expected now.
It's on the merge listOK, thanks for info.
Please report in mantis the next time. Issues reported on forum and mailing list get easily (and often) forgotten.OK
When the DockMaster layout is restored the height of docked windows increases. This is very unpleasant, particularly if the application is opened and closes several times per day...
What's the status of this bug ?
(I remember having briefly read something about this last week in the forum, but no luck with the search feature tonight)
Will it be fixed in 1.6 ?
When the DockMaster layout is restored the height of docked windows increases. This is very unpleasant, particularly if the application is opened and closes several times per day...
What's the status of this bug ?
This?
http://bugs.freepascal.org/view.php?id=18538
Is the "Choose parent > GroupBox" new child object jumping away (at times out of the forms current area) phenomenom fixed in new release. Noticed it today in 1.2.6 w7 i386. I'm a noob and noticed it today, may be already ironed out in 1.4.x
Please, look at this bug and insert bugfix to RC3:
http://bugs.freepascal.org/view.php?id=29141
Hi
Just done a fresh install of 1.6rc2 on osx 10.6.8.
At first it seems ok, so I tried to just rebuild lazarus
and get the following error.
LAZUTILs error code 256
and cant rebuild lazarus.
Obviously I have not tried to install any packages as they will fail..
Any Ideas
Josh
When the DockMaster layout is restored the height of docked windows increases. This is very unpleasant, particularly if the application is opened and closes several times per day...
What's the status of this bug ?
This?
http://bugs.freepascal.org/view.php?id=18538
QuoteThis?Actually it's very close:
http://bugs.freepascal.org/view.php?id=18538
https://vimeo.com/152515956
Can someone please get this fix into 1.6. It's fixed in trunk.
I have been waiting 9 months for this one to get into a release:
http://bugs.freepascal.org/view.php?id=27959
Can someone please get this fix into 1.6. It's fixed in trunk.
I have been waiting 9 months for this one to get into a release:
http://bugs.freepascal.org/view.php?id=27959
Added r50763 to the list of revisions to be merged:
http://wiki.freepascal.org/Lazarus_1.6_fixes_branch
The commit message talks about FPC 2.6.4 but I guess it fixes things also for FPC 3.0.
Thank you so much...I will be sure to test it when RC3 get posted.
QuoteThis?Actually it's very close:
http://bugs.freepascal.org/view.php?id=18538
https://vimeo.com/152515956
Looks very similar to what Lazarus had but was fixed in r51097, r51098 and r51099.
Can you check what revision caused your regression?
Why is it not possible to reproduce in a simple example project? Did you make workarounds in your code?
There's no real workarounds
I've noticed that "Source->Refactor->Unused units..." is not working properly. It opens up the unused units dialog but unused units do not appear.
Also, something I've noticed recently is that JCF (pressing CTRL+D to automatically indent code) does not work correctly when a generic class is present in the code.
Inserting something as the following (even a complete class definition) causes JCF to fail in both 1.4.4 and 1.6 versions of Lazarus.
TThing = class(specialize TFPGList<integer>);
I've noticed that "Source->Refactor->Unused units..." is not working properly. It opens up the unused units dialog but unused units do not appear.
Damn, you are right! When did that break? I will bisect for the guilty revision tomorrow.
[Edit] Reported:
http://bugs.freepascal.org/view.php?id=29486
It works with basic generic syntax. See:
http://bugs.freepascal.org/view.php?id=16128
Your usage is weird. Remove the "class" in front of specialize and it might work.
More patches for JCF's generics support are welcome. It is not actively maintained.
I just noticed that in 1.4.4 that when selecting a font in the Object Inspector it does not offer 'default' in the either the list or the dialog. It is very minor, because 'default' can be typed into the Font|Name editbox, but might confuse beginners who can't find a way back.
I do not have 1.6 RC2, so could someone check this.
I am using Lazarus 1.6 RC2 64 bits on Linux Mint. I found this issue: in Lazarus 1.4.4 this code worked fine:
procedure TForm1.DateFieldKeyPress(Sender:TObject; var Key:char);
begin
if (key >=#32) and not (key in ['0'..'9']) then key:=#47; // turn key into a '/'
end;
But in Lazarus 1.6 RC2, that code has no effect...
I am using Lazarus 1.6 RC2 64 bits on Linux Mint. I found this issue: in Lazarus 1.4.4 this code worked fine:
procedure TForm1.DateFieldKeyPress(Sender:TObject; var Key:char);
begin
if (key >=#32) and not (key in ['0'..'9']) then key:=#47; // turn key into a '/'
end;
But in Lazarus 1.6 RC2, that code has no effect...
It may be related to some valid change DateFields. If you have time you could bisect with trunk to find the revision that caused it.
In any case you should write also to Lazarus mailing list. The people may read it more than this forum.
A bug report with an example program is OK, too.
Reported issue #0029536 "Changes made in OnKeyPress to a Key value take no effect, in a TEdit component".
I tested it on Windows 7 32 bits, Lazarus 1.6 RC2, and this issue does not occur.
@Josh
I just installed Lazarus 1.6RC, FPC 3.0 on Mac OS X 10.8.5 and successfully compiled and ran a large (10,000s line) program with no problems. I did not notice IDE issues, but have done limited testing yet.
I then tried rebuilding Lazarus by attempting to install a package, and also using Tools > Configure "Bulid Lazarus", but could not succeed.
Build IDE: Exit code 512, Errors: 1, Warnings 1
VTwin
tried to install the 1.6RC2 on Scientific LInux 6.6 and the RPM seems to be corrupted:
Running Transaction
Installing : lazarus-1.6RC2-0.i686 1/1
Error unpacking rpm package lazarus-1.6RC2-0.i686
error: unpacking of archive failed on file /usr/lib/lazarus/components/turbopower_ipro/units/i386-linux/gtk2/iphtml.o;56ab2822: cpio: read
Probably my fault.
The file on sourceforge was only the first 46MB of the file.
I uploaded the file again.
I have noticed over the last few releases of Lazarus that after install the rebuild of Lazarus with added packages (typically Anchordocking etc.) fails on the first try, but will succeed on 2nd try. This is on Win7 pro 64 bit, 32-bit version of Lazarus.
1.6 RC2 on OSX. Project>Project Options>Paths>Other Files/Units
....
Press the speed button again and another browse window that you can't interact with comes up.
Is it a known problem?
Hi,
It look like there is a severe problem when using local functions in class methods:
Once I create a local function in a class method strange things happen. It seems like the pointer to a methods stack frame is lost/currupted or points to some improper place. Local variables of the method shows nonsense values during debugging, it is not possible to see some properties of the class (No symbol in current context reported) and the code is using clearly not proper values of the properties.
I am starting to be a little confused about the problem. It is probably a debugger related one (GDB showing improper values of local variables, classRead my update on the wiki.
When the array was declared as having 2^24 rows debuggerGDB when told to inspect static arrays attempts to allocate the full memory.
*Now in designer select the form
And hit Ctrl+Z or select Undo from Edit in main menu.
An (Unhandled?) exception occurs with message: "(false)" is not a valid boolean
[Edit] Please try with r51634 (trunk).Tried and it works, thank you!