Felipe Monteiro de Carvalho schrieb:
It the case of TTabControl one idea is custom drawing the entire
control, with themes it looks native, and that's how it already works
now. Finishing the current TTabControl would probably be easier to do
then finishing your new implementation.
I couldn't
Mark Morgan Lloyd schrieb:
The thing that seems really painful to me is that a program which (say)
uses database facilities would have to have a way of asking the target
OS's package system for the name of the package that contained the
appropriate libraries. This probably implies talking to
The implementation of the TTabControl seems to cause headaches, and
reveals wrong assumptions of the devlopers about the behaviour of that
control :-(
When I recently submitted an patch, with a new implementation of the
TTabControl (named TTabs in the patch), this implementation was tested
Leonardo M. Ramé schrieb:
On 2011-07-31 15:58:26 +0200, Bart wrote:
As an end-user I would get rather confused if my mouse cursor didn't
move anymore when I was moving the mouse.
Bart
I agree, but some times is better to fix the cursor to a specific
position. In my case, I have a N x M grid
Felipe Monteiro de Carvalho schrieb:
IMO we should accept that a Delphi compatible TTabControl doesn't exist yet,
for widgetsets other than Win32. What can/should we do, when it turns out
that such a control can not be implemented in all widgetsets?
I bet that it can be implemented in all
Graeme Geldenhuys schrieb:
On 31 July 2011 22:39, Henry Vermaak wrote:
It's an interface to the Windows Installer where you define all the
specifics of your project in an XML file. It the compiles and links
into an msi file.
I can't comment on WiX (never knew about it), but I can say our
Joshua Lim schrieb:
set OLDDIR=%CD%
set PATHDEST=C:\Develop
set PATH=%PATHDEST%\fpc\binutils\;%PATH%
I have only the really required folders in the path, when building FPC
or Lazarus.
REM svn co http://svn.freepascal.org/svn/fpc/tags/release_2_4_2 fpc
svn co
Felipe Monteiro de Carvalho schrieb:
There is something, however, that I would like to clarify.
What exactly is wrong with the current TTabControl that you want to
fix? What exactly does it do that you expect it to do differently?
I started with a test program in Delphi, and used it to test
Felipe Monteiro de Carvalho schrieb:
Concrete questions:
- What's the purpose of the page widget, required in many gtk_notebook
procedures?
Could you show the concrete place in our code base which you are talking about?
By gtk_notebook* do you mean Gtk rountines?
Yes. When it's required to
For my attempt to fix the problems in the gtk2 TabControl implementation
I need some background information. Can anybody help me, or continue my
started update?
In the first step I try to keep the dummy page, that is created together
with the notebook widget, and use it for all tabs. I also
Hans-Peter Diettrich schrieb:
For my attempt to fix the problems in the gtk2 TabControl implementation
I need some background information. Can anybody help me, or continue my
started update?
In the first step I try to keep the dummy page, that is created together
with the notebook widget
Graeme Geldenhuys schrieb:
On 07/27/2011 09:03 AM, Felipe Monteiro de Carvalho wrote:
That's what I though initially, but it is wrong. In Delphi this
control allows other controls to be placed on it, so it effectively
has 1 sheet.
And I believe that to be a bug in Delphi. See attachment:
Graeme Geldenhuys schrieb:
As for the TabControl vs PageControl vs NoteBook vs whatever Delphi
has... I understand [which might be wrong] is that the controls are
described as follows:
TabControl - this just paints the tabs part, and presents the
developer with a OnTabChange
Martin schrieb:
In Delphi Child.Top = zero means overlap the tabs (as does it for
TGroupBox).
That's the behaviour of the underlying COM control, and since Delphi
(VCL) only supports Win32 widgets, this has been okay in the past.
IMHO I see no unsolvable issue, if that was different in
In the last days I tried to install Lazarus on various (virtual)
machines, on Ubuntu 10.04 (LTS), with different problems preventing the
use of the SVN trunk. Finally I was successful only in an older
installation.
I'm not sure what's going on, but I suspect that some Ubuntu security
update
silvioprog schrieb:
Several peoples use a script to a automated installation with sucess, see:
A script doesn't help when dependencies cannot be resolved :-(
Actually SVN cannot be installed, due to missing files.
DoDi
--
___
Lazarus mailing
Graeme Geldenhuys schrieb:
On 27 July 2011 17:00, Hans-Peter Diettrich wrote:
Yet when you set the Memo1.Align := alClient, then in
maximizes the memo to fit underneath the tabs. This is very conflicting
behaviour -
IMO this is just one solution for above problems, where do you see any
Mattias Gaertner schrieb:
Several peoples use a script to a automated installation with sucess, see:
A script doesn't help when dependencies cannot be resolved :-(
Actually SVN cannot be installed, due to missing files.
Have you install libgtk2.0-dev ?
Please try and see yourself.
Artur Stuczynski schrieb:
If you cannot install SVN I suspect your OS installation went titsup.
Right, the current Ubuntu distribution is quite broken. Older
installations continue to work, but newer ones have problems.
DoDi
--
___
Lazarus
Hans-Peter Diettrich schrieb:
I just tried to install Lazarus (trunk) on Ubuntu 10.4, and came across
errors with fpcres.
More errors:
A missing output directory ~/.lazarus/units/... made fail both make
bigide and rebuilding the IDE from the Tools menu.
Is this an FPC or Lazarus (script
Marco van de Voort schrieb:
On Tue, Jul 26, 2011 at 12:42:28AM +0100, Hans-Peter Diettrich wrote:
the bashrc entry should read
export PATH=~/lib/fpc/2.5.1:~/bin:$PATH
It should only contain ~/bin and in ~/bin there should be symlinks to
various ppc* binaries in ~/lib/fpc/version. Cross
Felipe Monteiro de Carvalho schrieb:
There is also a new class: TNewTabControl, which is not registered and
is ment to receive patches to implement the new tab control. This new
tab control might then, when finished, be renamed to TTabControl and
the old control might be renamed to
Andreas Schneider schrieb:
At Tuesday, 26.07.2011 on 17:04 Hans-Peter Diettrich wrote:
In the meantime I added AutoSize to the T[New]TabControl, and when
my Linux version becomes usable, I'll try to fix the gtk2 etc.
errors.
Why not simply work in an SVN Branch (or if you use Git simply
Felipe Monteiro de Carvalho schrieb:
On Tue, Jul 26, 2011 at 5:04 PM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
Why not name these controls TOldTabControl and TNewTabControl, and set
TTabControl to one of both? This will allow to switch easily between both
implementations, without
Graeme Geldenhuys schrieb:
On 07/26/2011 04:10 PM, Felipe Monteiro de Carvalho wrote:
TTabControl for the more
Delphi-compatible version, TNewTabControl for the one with the correct
class hierarchy.
My I suggest
TTabControl- The Delphi-compatible one
Which one should that be?
Felipe Monteiro de Carvalho schrieb:
On Mon, Jul 25, 2011 at 12:35 AM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
Your implementation did not work,
Which widgetset?
I think that all of them, or at least gtk2 and carbon. These steps
failed for me:
1 Drop a TTabControl in the form
2
I just tried to install Lazarus (trunk) on Ubuntu 10.4, and came across
errors with fpcres.
First the path was missing the ~/bin directory, which should be
mentioned properly in the wiki. Instead of
export PATH=~/lib/fpc/2.5.1:$PATH
the bashrc entry should read
export
Thanks for the attempt to replace the old TTabControl by the new
implementation :-)
There seem to exist some issues with the new implementation, so that
I'll give some background information, about the new TCustomTabControl
and descendants.
The basic TCustomTabControl does not have pages
Marc Santhoff schrieb:
Am Donnerstag, den 21.07.2011, 02:09 +0100 schrieb Hans-Peter Diettrich:
Marc Santhoff schrieb:
Nice idea, but before starting work remember that some things are
already invented.
All that does not help much in managing and editing the FPCDocs.
I think I'll have
Felipe Monteiro de Carvalho schrieb:
drdiettri...@aol.com wrote:
There seem to exist some issues with the new implementation, so that I'll
give some background information, about the new TCustomTabControl and
descendants.
Your implementation did not work,
Which widgetset?
so I deleted
Mattias Gaertner schrieb:
The base units are independent of the widgetset units and the widgetset
units use the base units. So the package of widgetset units must use
the package of the base units.
Theoretically: If the base package would be named 'LCL', then
all existing projects would no
The recent change in the package names has invalidated all #LCL links in
the documentation :-(
IMO the package names should be exchanged, so that LCL contains the
high-level components (equivalent to the Delphi VCL). The other package
can be renamed into e.g. Widgetsets, to make clear what it
Mattias Gaertner schrieb:
Hans-Peter Diettrich drdiettri...@aol.com hat am 18. Juli 2011 um
15:20 geschrieben:
Is it normal that the codetools try to insert e.g. end; into comments,
while editing the comment?
No. Please create a bug report with an example.
I couldn't reproduce this one
During the development of the doc tracker I stumbled over several issues:
The English wording often violates my feeling for the language. Can some
native English speakers proofread the documentation, and correct
stylistic flaws?
I don't want to discourage non-native speakers to contribute to
Mattias Gaertner schrieb:
Often I come across circles, referring to circular unit references.
IMO the correct term instead would be loops, in e.g. avoid loops.
AFAIK the correct term in graph theory is cycle.
Sounds good.
Many descriptions only describe the obvious, like method names
Mattias Gaertner schrieb:
What if the method just does what the name says?
Then no extra documentation is needed.[...]
How to distinguish an item that needs documentation and an
item that does not need documentation?
Give an example?
DoDi
--
Howard Page-Clark schrieb:
I also can assign doc files to those people, which are willing to
proofread the documentation, so that we can avoid duplicate work.
I'm a native English speaker, and I'm willing to proofread
documentation, and make suggestions for improvements. However, I suspect
Mattias Gaertner schrieb:
On Sat, 23 Jul 2011 16:59:14 +0100
Hans-Peter Diettrich drdiettri...@aol.com wrote:
[...]
As already mentioned in another mail, all links to #LCL are broken now.
I changed the #LCL to #LCLBase in the lcl xml files.
Argh :-(
?
Why do you break existing
Currently the project title is shown everywhere, instead of the project
file name. IMO at least Project Options should show the project (file)
name in its Caption, since the title is shown and editable in this
dialog. Also the Project Manager should show the project name in its
Caption,
The code tools seem to have problems in parsing case statements, and
with debugging while statements. See this code snippet:
//determine XML lines to copy
while PFirst XLast do begin
//copy until PFirst
while XFirst PFirst do begin
dst.Add(Strings[XFirst]); inc(XFirst);
end;
Mattias Gaertner schrieb:
Can you reproduce it? For example edit something in front, so that the
parser reparses the code in question.
Bingo, a change to the file solved the problem :-)
I think I saw something fishy with undo that sometimes the source
editor does not commit the changes to
Marc Santhoff schrieb:
Nice idea, but before starting work remember that some things are
already invented.
All that does not help much in managing and editing the FPCDocs.
Roadmap:
I'm just about to show the help entries element .../ in the XML docs,
that have been changed since the last
Is it normal that the codetools try to insert e.g. end; into comments,
while editing the comment?
DoDi
--
___
Lazarus mailing list
Lazarus@lists.lazarus.freepascal.org
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
Felipe Monteiro de Carvalho schrieb:
On Sun, Jul 17, 2011 at 3:33 AM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
Ok, so I issued the command and it seams to have worked! But where are
the html files?
Have a look into the docs/html directory?
user-188-33-177-8:fpcdocs felipe$ cd docs
Martin schrieb:
One solution were another directory, holding all old versions of the
primary files, so that these files can be compared to the current
versions. The comparison can be improved by checksums or similar
means, so that a direct file-by-file comparison can be avoided.
Perhaps also
Michael Van Canneyt schrieb:
A raw text output (without attributes) would be helpful, in addition
to HTML etc. formats.
A text output format exists.
Ah, I didn't notice that.
I think you should use attributes instead of tags. fpdoc is not very
forgiving of 'unknown' tags. It does not
Graeme Geldenhuys schrieb:
Also the indentation of the nodes in the XML files
sometimes affect the documentation output as well. Yes Latex and HTML
output writers ignore whitespace, but not all other output formats do.
Right, indentation and unlogical blank lines around var are an
issue.
Michael Van Canneyt schrieb:
Can namespaces serve the same purpose?
I'm not sure I understand what you mean ?
If you're talking about XML namespaces, forget it. Way too complicated
and bloated.
But maybe you meant something else?
No, I meant exactly the XML namespaces, for which I found
Graeme Geldenhuys schrieb:
Experience with the FPC website shows that chances of having multi-language
documentation are virtually 0. Basically, I think the FPC/Lazarus community
is too small for the effort it takes.
+1
-1
Most people seriously underestimate the huge amount of effort it
Felipe Monteiro de Carvalho schrieb:
Ok, so I issued the command and it seams to have worked! But where are
the html files?
Have a look into the docs/html directory?
DoDi
--
___
Lazarus mailing list
Lazarus@lists.lazarus.freepascal.org
Graeme Geldenhuys schrieb:
On 16 July 2011 21:52, Hans-Peter Diettrich drdiettri...@aol.com wrote:
*We* do, but nowadays nobody else knows Pascal.
That shouldn't really be an excuse. Pascal is so easy to read and
understand. Also if you can program in any other language, picking up
a new
I'm about to write kind of a content management system for the Lazarus
and FPC docs (FPDoc xml format). This system shall allow to track
changes to the source documentation, so that updates can be applied e.g.
to the FPDocs in other languages. I need such a CMS for updating my
personal
Martin schrieb:
The 2ndary file could have a checksum of the node n the primary file.
then it can check, if the primary file was modified. The checksum must
be in a way, that ignores formatting, etc. (maybe even minor spelling
fixes)
A raw text output (without attributes) would be helpful,
Tommi Prami schrieb:
I just updated the latest SVN truck version (Revision : 31692)
Compiling of the our winCE project will fail with message :
C:\lazarus\lcl\interfaces\wince\wincewscomctrls.pp(69,20) Error: There
is no method in an ancestor class to be overridden: class
How can events be added to frames?
I want to add an OnFileClick event, that reacts on an click into a
listbox in the frame. The new event is typed as
TOnFileClick = procedure(Sender: TMyFrame; p,d: string) of object;
and the handler in the frame provides the strings and calls the assigned
Bart schrieb:
It is Felipe's work in progress I think.
Yes, based on my supplied patch, that's why I can explain what remains
to do.
The mapping between logical Pages and visible Tabs in the
TCustomNotebook was scattered into multiple unrelated places, e.g.
TCustomPage and the
Fabio Luis Girardi schrieb:
I searched on web and lazarus examples, but I don't found any example
of how use TListView.OnCustomDrawItem to color each row using latest
lazarus version.
It should not be different from any other CustomDraw procedures. Do a
FillRect with the desired color, before
Felipe Monteiro de Carvalho schrieb:
On Thu, Jul 14, 2011 at 4:19 PM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
Eventually another Delphi compatible rename should be made. As I just found
out, Delphi defines PageIndexFromTabIndex for my TabToPageIndex. But since
this is a private method
Mattias Gaertner schrieb:
3) Sometimes it helps to remove the frames, and to add them again, but
not always.
Helps for what?
A general question, Before I start submitting bug reports:
Does there exist documentation about frame *development*?
When I have a form with embedded frames,
Mattias Gaertner schrieb:
That means: before you change a property of a frame, close all open
forms using that frame.
Very good idea :-)
Here is a start:
http://wiki.lazarus.freepascal.org/IDE_Development#Frames
Fine :-)
Now I'll resume my current work and keep all suggestions in mind.
Sven Barth schrieb:
On 12.07.2011 18:38, leledumbo wrote:
Seems like codetools parser bug. Try writing it type by type:
type
Tp2pMessageClass = class of Tp2pMessage;
Tp2pMessageClassArray = array [0 .. 1] of Tp2pMessageClass;
const
MsgTypes: Tp2pMessageClassArray = (
...
);
I came across some weird behaviour of TDirectoryEdit events, and traced
these down into the behaviour of the more basic components. IMO the
events, related to Edit fields, deserve reconsideration.
From the user VP, an Edit field should *not* trigger events while or
immediately after it has
In working with an frame I get any number of errors :-(
[Win32]
It looks to me as if the (main) form resource is not updated properly,
at the right time:
1) Frame.alClient (alLeft...) is not really honored. The frame is marked
as occupying the according space in the container form, but its
Mattias Gaertner schrieb:
1) Frame.alClient (alLeft...) is not really honored. The frame is marked
as occupying the according space in the container form, but its
components are not (always) resized.
Can you provide an example?
Screenshot or project?
2) An event handler,
Graeme Geldenhuys schrieb:
On Sunday, 10 July 2011, Mattias Gaertner nc-gaert...@netcologne.de wrote:
But I think DoDi was referring to floating windows, that means top level
windows, not child windows.
I'll read the thread again to try and better understand the problem.
Background
zeljko schrieb:
Qt have docking windows , and we have QDockWidget in our bindings (I've
played with it few months ago and it works .. .but of course there's no
support in LCL for such ...). Also QDockWidget works fine on Linux, Mac
and Win (and others where qt supports it).
I wonder how
Mattias Gaertner schrieb:
The listbox contents change while typing, no units changed or anything else.
Well, the current unit changes.
This was not an issue with the old (correct) behaviour.
I found a unneeded clean up.
Please test.
Now it seems to work again :-)
DoDi
--
Florian Klämpfl schrieb:
Am 08.07.2011 23:42, schrieb Hans-Peter Diettrich:
How can I create and apply patches to Lazarus?
First problem is SVN, which does not allow for incremental patches,
i.e. a big patch cannot be split into smaller steps, as frequently
requested by the developers
Martin Schreiber schrieb:
On Friday 08 July 2011 19:46:41 Hans-Peter Diettrich wrote:
Right. On all other platforms the widgetsets sit on top of an window
manager, which influences dragging and docking entire windows (forms).
The current (platform independent) workaround is a dock header
Florian Klämpfl schrieb:
svn-apply can be used to apply patches to svn created by git
format-patch:
http://svn.freepascal.org/svn/fpcbuild/scripts/florian/unix/svn-apply
Tested on unix only.
How do I use it at all? On Windows?
No idea, I use it only on linux.
I just wrote an converter for
Martin schrieb:
I don't know if Microsoft does this, but in visual studio floating
windows have a different border color, different looking icons, and no
shadow
Windows allows for all such stuff - problem is only all the other window
managers out there.
DoDi
--
Martin Schreiber schrieb:
On Saturday 09 July 2011 10:08:11 Hans-Peter Diettrich wrote:
I cannot see how MSEgui handles dockable (floating) forms.
In order to dock a floating window LeftClick into the drag grip, move the
mouse into the destination window, release the mouse button. The drag
Martin Schreiber schrieb:
A window with border, with title bar and an additional drag handle in client
area, see dock2.png in my previous mail.
Such a feature would have to be built into the LCL, it's not feasable in
user code.
DoDi
--
___
Mattias Gaertner schrieb:
Sometimes after having typed a few characters if I press backspace the
list is suddenly empty and won't be populated anymore. Also sometimes
it won't complete when I press enter on the selected word, it will
just close the list and nothing else happens.
Sometimes
Graeme Geldenhuys schrieb:
On 07/07/2011 08:21 PM, Hans-Peter Diettrich wrote:
I can provide you an patch, that makes docking work again.
So why not apply it to Trunk?
Don't ask /me ;-)
BTW: what platform?
Isn't that irrelevant? LCL is supposed to be cross-platform.
Delphi dragdock
Sven Barth schrieb:
Am 08.07.2011 08:34, schrieb Graeme Geldenhuys:
On 07/07/2011 08:21 PM, Hans-Peter Diettrich wrote:
I can provide you an patch, that makes docking work again.
So why not apply it to Trunk?
Hans-Peter already reported those patches (or are those new ones
Tommi Prami schrieb:
Now it works perfectly only I can't get unfocused App to come to the
Screen to the top most app. ..
If there is some force way to do it. Any will do, would be good...
Perhaps SetFocus could help? (untested)
DoDi
--
___
How can I create and apply patches to Lazarus?
First problem is SVN, which does not allow for incremental patches,
i.e. a big patch cannot be split into smaller steps, as frequently
requested by the developers. Solutions?
Next problem are patches created by Git, which seem to be
Lee Jenkins schrieb:
Is docking support production ready in Lazarus at this time?
Recent changes and outstanding bugs prevent the use of dragdock.
I can provide you an patch, that makes docking work again.
BTW: what platform?
DoDi
--
___
Felipe Monteiro de Carvalho schrieb:
General concepts:
1 Avoid changing the current TPageControl as much as possible,
because it is extensively used and changes here might cause a pletora
of problems
2 I think it is no problem to introduce heavy changes to TTabControl
because it is rarely used
Felipe Monteiro de Carvalho schrieb:
And I would make smaller patches implementing the following actions:
1 Make sure TCustomNotebook can implement TTabControl, add an option
that it can work without tabs and test in all widgetsets (maybe it can
already do that, you would need to check this)
2
Zaher Dirkey schrieb:
Hi,
I like to build my packages and add it to Lazarus outside of IDE,
Do you mean a registry for *available* runtime packages?
In another way, i like to send my packages sources to my
friends/coworker and run that batch to auto add this packages to Lazarus
without need
Max Vlasov schrieb:
On Mon, Apr 18, 2011 at 10:57 AM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
ATN Vincenct Snijders, Dmitry Boyarintsev, chronos...
Mantis #18820, #18898, #18961...
The ListView component is in a bad state, and a review of the current item
handling is required, before
Felipe Monteiro de Carvalho schrieb:
I'd second that idea, currently I'm working on such an improvement for
tabbed controls. The TCustomNotebook suffers from the added pages, which
should be added only in a derived control, as special kind of owner data.
Ummm, not really. Please note that
Felipe Monteiro de Carvalho schrieb:
Yes, I realized that misinterpretation already. If somebody is willing to
update the widgetsets accordingly, I can do all the required LCL and IDE
updates right now. (see below)
I already started that a long time ago, I already did the following steps:
1
Felipe Monteiro de Carvalho schrieb:
On Mon, Jul 4, 2011 at 4:53 PM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
I just finished the revised implementation of TCustomNotebook, so that it
can work with and without pages. In that approach I didn't rename
TCustomNotebook, because
Martin schrieb:
On 01/07/2011 03:49, Hans-Peter Diettrich wrote:
Currently I observe the following flaws (Win7):
1) An editor page in a secondary editor window *only* can be cloned
into a new window. Why does the handling of secondary editor windows
differ from the handling of the primary
Martin schrieb:
Actually, I just checked on windows vista, and it works (2nd window
allows cloning back to 1st window). I tested before, but tricked myself.
Oops, *this* seems to be the reason!
Please not when you test, that there is another restriction. Each unique
file can only have one
Marcelo Silva (IG) schrieb:
I do not know the name of that little window that opens when you press
Ctrl + Space lazarus.
But I know I know what I mean.
In Delphi Right clicking it gives you the option to sort by name or scope,
this way is easier to find a particular property or function.
Martin schrieb:
BTW, CTRL is the wrong modifier, it's intended to *ignore* all
possible targets.
Is there an official definition for that somewhere?
Is Delphi official enough? ;-)
But you may be right, above applies to drag-dock, not to drag-drop. In
notebook-docking the notebook tabs are
Martin schrieb:
BTW, you are away of the page locking feature ?
http://lazarus-dev.blogspot.com/2010/04/page-locking-or-multiply-editor-windows.html
Wow, really great. It will become one of my favorite Lazarus features,
now that I know about it :-)
DoDi
--
Martin schrieb:
On 01/07/2011 17:58, Hans-Peter Diettrich wrote:
Martin schrieb:
The display disabled menu entry, instead of display none makes sense.
Hm, on further thought. displaying source-editors to which you can not
move/clone as disabled menu entry, brings up a few questions
Lukasz Sokol schrieb:
On 29/06/2011 16:11, Michael Schnell wrote:
On 06/28/2011 08:14 PM, Hans-Peter Diettrich wrote:
I just found a very nice description of some aspects of thread safety in
http://pubs.opengroup.org/onlinepubs/9699919799/xrat/V4_xsh_chap02.html#tag_22_02_09_08
titled Thread
Lukasz Sokol schrieb:
Objects (classes) add further topics to the list of items, that make
up thread-safety in general.
After compilation, it's just code, isn't it?
Classes may have the 'hidden' variables but they're not 'hidden' to the
OS/CPU ?
Objects include information, that may be
Currently I observe the following flaws (Win7):
1) An editor page in a secondary editor window *only* can be cloned into
a new window. Why does the handling of secondary editor windows differ
from the handling of the primary window???
2) Code completion still gets stuck all the time. The
Marco van de Voort schrieb:
On Mon, Jun 27, 2011 at 03:05:36PM +0100, Hans-Peter Diettrich wrote:
OOP is just a convenient abstraction, it's for people needed only :)
OOP is for designers, not for hackers ;-)
When some thread-safe code must never bring a program (automaton,
system
Andrew Brunner schrieb:
On Mon, Jun 27, 2011 at 2:51 PM, Hans-Peter Diettrich
drdiettri...@aol.com wrote:
Absolutely incorrect.
Why?
Two reasons.
1.) Visibility. A critical section does not particularly protect
visibility as with 2 threads accessing a piece of memory.
This is exactly
Andrew Brunner schrieb:
This is relatively new theory which required low-level CPU code to
perform such locks.
This is not new. String operations (CMPS...) can be interrupted since a
long time, and it took some more time until even the single-core CPUs
could resume such an interrupted
Henry Vermaak schrieb:
You'll have to read the generated assembler for Windows critical
sections. My hunch would be that they use memory barriers, too.
When a critical section object is owned, the only other threads affected
are the threads that are waiting for ownership in a call to
Michael Schnell schrieb:
On 06/27/2011 07:52 PM, Hans-Peter Diettrich wrote:
You forget the strength of protection.
Maybe I seem to see what you mean.
You think, a critical section will not prevent that the data that is
accessed by both threads is duplicated in a register or in the Data
901 - 1000 of 1724 matches
Mail list logo
