|
One solution is to design a custom-draw.
http://www.codeproject.com/listctrl/xlistctrl.asp
Kuphryn
|
|
|
|
|
Hi,
I am sorry about this, since I am working on a big project that involves ListBoxes & ListCtrl's ... I did mixed-up my message ListBox (as ListCtrl) ... I appologize for this!
Actually, I could not set BackGround color & Text color for my ListBox (not a ListCtrl)
You still can help?
Sorry & regards
|
|
|
|
|
The only way I believe to change the color of a listbox is via owner-draw. You have to derived a class from CListBox or whatever and draw your own boxes. There are classes available that allows you to change its color.
Kuphryn
|
|
|
|
|
Hello,
I'm trying to write a small program that connects to an Oracle database.
Oracle delivers an oci.lib file when installing the Oracle client.
I'm also using a stream template to facilitate the working with Oracle API.
I include the oci.lib in my project's link tab, specify the location where the lib file can be found and everything works as expected. So far so good.
When I transfer my small program to a computer which doesn't have an Oracle client installed, the program complains about missing dll's. This, I can understand.
My question is: is it possible to statically link all the libraries/dll's in the lib file to my executable?
If I look at the dependecies, my program requires a lot of (Oracle) dll's. Will all of these dll's be statically linked to my executable?
And what exactly is a lib file. Just someting Visual C++ exports when you create a dll so other people can easily make use of the dll?
Thanks in advance
Wim
|
|
|
|
|
is it possible to statically link all the libraries/dll's in the lib file to my executable?
No, there's no way to have DLLs bundled into your .EXE. However, you can install them along with your executable (possibily in the same directory) if you use some sort of installation program to ship your app.
And what exactly is a lib file
There are two types of static libraries:- Regular libs are just compiled pieces of code you can link and use without further dependencies from DLLs.
- Import libs are static libraries you can use to dynamically link functions from a DLL. These import libs search the DLL at tun-time, locate the functions and load them for the app to use, roughly speaking. DLLs can be used without this aid, though usually is best to rely on import libs to do the job.
Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo
|
|
|
|
|
As pointed out... a .lib file could be a static library, or simply an import library to a DLL.
Import libraries have the advantage that you can call DLL functions as if they were regular functions. You don't need to call LoadLibrary to get the DLL, and GetProcAddress to get functions. And you get type checking. The disadvantage is that if the DLL doesn't exist, your code won't run at all. And your code will show an ugly, and not very user-friendly error message complaining that it can't find the DLL.
A good rule I follow: use import libaries for DLLs when you are certain the DLL will exist (e.g., you install it.) Use the regular DLL functions (like LoadLibrary, GetProcAddress) if there is a chance the DLL might not exist, and you want to handle that situation gracefully.
It is technically possible to include DLL's in an EXE, but it is highly ugly, and the only time I've ever done it was to create a self-extracting installer. (Basically, you include the DLL as a resource in your EXE. Yuck!)
Just make sure your installer puts these DLLs in the proper location when installing your product, and you should be good to go.
"Time spent with cats is never wasted." - Colette
|
|
|
|
|
Navin wrote:
The disadvantage is that if the DLL doesn't exist, your code won't run at all. And your code will show an ugly, and not very user-friendly error message complaining that it can't find the DLL.
one trick is to use delayed loading. when your app starts, do a LoadLibrary on the DLL. if it isn't found, complain nicely but bail. if it is present, do nothing, and the import lib will do its thing normally.
-c
No matter how fast light travels it finds the darkness has always got there first, and is waiting for it.
-- Terry Pratchett,
|
|
|
|
|
Hi,
Thanks for your answer. Everything is very clear to me now, but I still have one question.
As Oracle moves along to new versions, they tend to rename the dll's too.
So I used a Oracle 8i lib file, but I tried the program on a computer which had a Oracle 7 client installed and the program failed, because some dll's had the version number in their name.
So, the best way to go is to use LoadLibrary and GetProcAddress as you pointed out.
Now, my question is: is it possible to generate some sort of a header file out of a lib file. Although the dll names changed over time, the actual function names stayed the same.
So if I should be able to generate a header file, which maps the dll functions to a GetProcAddress(..., ...), the only thing I have to write then is a LoadLibrary part.
Wim
|
|
|
|
|
Hello,
the subject says everything...
does anybody know how to do it? Do this is possible?
If not how should I do it?
Moreover I need them in the first subitem...
Thank you in advance...
|
|
|
|
|
the only two ways of doing this, (that I know of)
are , creating the check box at runtime as child of the listctrl and show/hide them on top of the item when nedded, or use a CImageList with icons to fake the checkboxes. 
Dont know any outher way  but the topic is interesting (I will do some testing  )
Casa.Sapo.pt
|
|
|
|
|
Hi,
I have a design/OOP question. I am trying to find the most elegant solution or design of the object. I want to create several instances of an object, and then have each of those objects perform the same function at the same time when called upon to do that task. An example of what my question is this: for instance you wanted to create a bunch of door objects and you want them to open simultaneously at the appropriate time.
How would one design something like that? A real life example is in a horse race and the gates the horses are kept in. You want all the gates to open at once at the start of the race instead of having one person walk to each of them and open them manually. I’ve tried several ideas with the static keyword, but unfortunately, my design skills are not up to par.
If there is anyone who could show me a c++ example of how to design such an object, you would have helped me a lot.
|
|
|
|
|
Use a static member variable or (more generally) forward to a static member function, just like this:
class gate
{
public:
void open(){static_open();}
bool is_open(){return static_is_open();}
private:
static bool m_open;
static void static_open();
static bool static_is_open();
};
bool gate::m_open=false;
void gate::static_open()
{
m_open=true;
}
bool gate::static_is_open()
{
return m_open;
}
Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo
|
|
|
|
|
So how would I use this?
int main(){<br />
<br />
gate x;<br />
gate y;<br />
gate z;<br />
<br />
bool gate::m_open = true;<br />
<br />
return 0;<br />
}
??? Thanks in advance for your help!
|
|
|
|
|
In Joaquín's example you would do this:
<br />
int main(){<br />
<br />
gate x;<br />
gate y;<br />
gate z;<br />
<br />
x.open();<br />
<br />
return 0;<br />
}<br />
I personally would remove the member functions and the gate class would look like this:
<br />
class gate{<br />
public: <br />
static void static_open(); <br />
static bool static_is_open();<br />
private:<br />
static bool m_open; <br />
};<br />
bool gate::m_open=false;<br />
void gate::static_open()<br />
{<br />
m_open=true;<br />
}<br />
<br />
bool gate::static_is_open()<br />
{<br />
return m_open;<br />
}<br />
Then your program could use it like this:
<br />
int main(){<br />
<br />
gate x;<br />
gate y;<br />
gate z;<br />
<br />
gate::open();<br />
<br />
return 0;<br />
}<br />
This is closer to what is really happening design-wise.
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life!
|
|
|
|
|
Is this weird?
I have an array with some entries I want to suppress from showing in the CListCtrl. WIthout thinking about index mismatch and conflict(and all I reallly need here is to populate the ctrl), I did:
for (int i3 = 0; i3 < pDoc->m_nNumItems; i3++)
{
fClearAllForList1 = TRUE;
if(pDoc->m_docFldHolder[i3] == "CDE")continue;
if(pDoc->m_docFldHolder[i3] == "XYZ")continue;
int nItem = m_list1.InsertItem(i3,pDoc->m_docFldHolder[i3]);
}
The entries all show up with no blank rows where the item was suppressed. SO if I didnt put in element # 3 of the arrray, the fourth element shows up where the third one shhould have been though if you look at the indices above, technically it should have skipped a row...
Is this safe or should I put use two indices, one for the rows and one for the array? AgainI really need to populate it only. I'm not doing any processing with the item number in the ctrl.
Appreciate your help,
ns
|
|
|
|
|
Why not insert void rows for the "hidden" items?
for (int i3 = 0; i3 < pDoc->m_nNumItems; i3++)
{
fClearAllForList1 = TRUE;
if(pDoc->m_docFldHolder[i3] == "CDE"||
pDoc->m_docFldHolder[i3] == "XYZ"||
pDoc->m_docFldHolder[i3] == "_"){
m_list1.InsertItem(i3,"");
}
int nItem = m_list1.InsertItem(i3,pDoc->m_docFldHolder[i3]);
}
Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo
|
|
|
|
|
And these void rows wont show ?
Appreciate your help,
ns
|
|
|
|
|
I have a SDI type app with 2 rich edit controls in the view (One for receiving text, and one for sending text).. How can I tell which edit control has the focus so when I do a "ctrl + c" to copy data to the clipboard MainFrame knows which edit control to copy the data from? GetFocus doesn't seem to work.
Any ideas?
|
|
|
|
|
Doesn't the richedit handle the clipboard stuff automatically??
--
Please state the nature of your medical emergency.
|
|
|
|
|
Doesn't seem to.. Unless when I create it I need to add something.. Here is how I create the controls..
if (CView::OnCreate(lpCreateStruct) == -1)
return -1;
if( !m_wndMonitor.Create(WS_CHILD|WS_VISIBLE|WS_VSCROLL|ES_MULTILINE|ES_READONLY, CRect(), this, 100) )
return -1;
if(!m_wndSendText.Create(WS_CHILD|WS_VISIBLE|WS_VSCROLL|ES_MULTILINE, CRect(), this, 101))
return -1;
m_wndMonitor.ModifyStyleEx(WS_EX_CLIENTEDGE, 0);
m_wndMonitor.SendMessage(WM_SETFONT, (WPARAM)GetStockObject(SYSTEM_FIXED_FONT));
unsigned mask = m_wndMonitor.SendMessage(EM_GETEVENTMASK,0,0);
m_wndMonitor.SendMessage(EM_SETEVENTMASK,0,mask | ENM_LINK);
m_wndMonitor.SendMessage(m_wndMonitor.SetEventMask(ENM_LINK)) ;
m_wndMonitor.SendMessage(EM_AUTOURLDETECT,TRUE,0);
m_wndSendText.ModifyStyleEx(0, WS_EX_STATICEDGE);
m_wndSendText.SendMessage(WM_SETFONT, (WPARAM)GetStockObject(SYSTEM_FIXED_FONT));
return 0;
m_wndMonitor is one rich edit control
m_wndSendText is the other
|
|
|
|
|
Hi,
I am trying to convert a small piece of c++ code to c# however I am not familiar with c++.
I have a BSTR called sToChange, and an unsigned int called i
the string sToChange is "ZYX123".
in c++ when i is set to sToEncode[0] (which is 90, the char code for Z) it equals 12:
i = sToEncode[0]
//i now equals 12
in c# the same code i equals 90.
What is happening in the c++ code?
Thanks,
John
|
|
|
|
|
90 is ASCII for 'Z'.
IIRC, BSTR's are unicode, right? And they don't necessarily map one to one on ASCII. Have you tried converting the BSTR to a char*? It may also be the case that your BSTR string does not contain what you think it contains.
Do a CString str; str.Format("%S", bstrString); and check the contents of str. I'm assuming you are using MFC or ATL7. If you're using ATL7, try ATLTRACE("%S\n", bstrString); to see what the string contains. And if you're not using MFC, ATL nor ATL7, then use WideCharToMultiByte to convert the BSTR to char*.
--
Please state the nature of your medical emergency.
|
|
|
|
|
I am sending a command out my serial port and waiting for a reply. All replies end with '\r'. I am using WaitCommEvent() with EV_RXFLAG as the event, to wait for a '\r' received.
But with the serial port receive queue, if a '\r' has already been received and is in the queue before I call WaitCommEvent(), will it still return that the event has happened? Or will it timeout? Anyone know?
Thanks,
dan.
|
|
|
|
|
I'd say the event will be triggered for buffered characters --otherwise the function is nearly unusable in non-overlapped mode. To make sure, try the following:- Issue the command.
- Wait for a sufficiently long time (say 10 sec.) with
Sleep to make sure the '\r' has been received by the time Sleep returns.
- Call
WaitCommEvent and see what happens.
If WaitCommEvent works the bad way (it doesn't set the event), come back here and we'll try to search for a solution.
Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo
|
|
|
|
|
Hi,
I'm trying to plot a non linear graph using MSChart. I set up the array and val stores the result of the calculation. However, the graph I plot appears linear and it's plotted val vs val whereas I want index[0] vs val?
double val=0;
double step=0.5;
DWORD numElements[] = {10, 2}
saRet.Create(VT_VARIANT, 2, numElements);
long index[2];
for(index[0]=0; index[0]<10; index[0]++) {
for(index[1]=0; index[1]<2; index[1]++) {
val = index[0]*index[0];
saRet.PutElement(index, COleVariant( val ));
val += step;
}
}
m_chart.SetChartData(saRet.Detach());
m_chart.Refresh();
}
Any ideas?
Cheers
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
