http://www.purebasic.com
https://www.purebasic.fr/english/
That would be different already at the first byte written to it. It would not tell whether or not the creation has finished. There should really be some kind of decent notification.coder14 wrote:You could check the modified date of the exe file properties.
Then it should be "Executable created successfully!"DarkDragon wrote:That would be different already at the first byte written to it. It would not tell whether or not the creation has finished. There should really be some kind of decent notification.coder14 wrote:You could check the modified date of the exe file properties.
(That mean new EXE is created at 08:25:06)[08:25:06] Waiting for executable to start...
[08:25:06] Executable type: Windows - x86 (32bit, Unicode)
[08:25:06] Executable started.
Code: Select all
EnableExplicit
Define EXE_Name$ = ProgramParameter(0)
Define EXE_Timestamp$ = FormatDate("%dd/%mm/%yyyy - %hh:%ii:%ss",
GetFileDate(EXE_Name$, #PB_Date_Modified))
MessageRequester("Info", "EXE: " + EXE_Name$ + #CRLF$ + EXE_Timestamp$)Depends on your output device. On SD Cards or USB Sticks the output might be a bit slow. Also for big projects it must be slower than for small projects.Mistrel wrote:Compilation is so fast though that compilation will be done by the time he brings his file explorer into the foreground.
These are workarounds and no solutions. I would move this topic to bug reports in all OS.Marc56us wrote:2 Solutions:
marcoagpinto wrote:There is a need to show in the status part when one creates an executable:
"05:28 Executable created"
because sometimes I can't tell if I saved a newer executable over the old one.
This is not a bug, this is a feature request, so still good here in "Feature Requests and Wishlists" (IMHO)DarkDragon wrote:I would move this topic to bug reports in all OS.
Do you know the halting problem? I see a similarity between this and the humans sitting in front of the executable.Marc56us wrote:This is not a bug, this is a feature request, so still good here in "Feature Requests and Wishlists" (IMHO)DarkDragon wrote:I would move this topic to bug reports in all OS.
That doesn't do anything when you're creating a final executable, which was the request.Marc56us wrote:File > Preferences > Debugger > [X] Display Timestamp in Error log
You're right,Dude wrote:That doesn't do anything when you're creating a final executable, which was the request.Marc56us wrote:File > Preferences > Debugger > [X] Display Timestamp in Error log
Apart from small programs, always work with the project manager (even with one single source file)This behaviour exists on OSX.Marc56us wrote:What's missing is a confirmation message when the EXE already exists and will be overwrited...
Code: Select all
CompilerSelect #PB_Compiler_Processor
CompilerCase #PB_Processor_x86
Structure MY_NOTIFYICONDATA
cbSize.l
hwnd.i
uId.l
uFlags.l
uCallbackMessage.l
hIcon.i
szTip.s{128}
dwState.l
dwStateMask.l
szInfo.s{256}
StructureUnion
uTimeout.l
uVersion.l
EndStructureUnion
szInfoTitle.s{64}
dwInfoFlags.i
guidItem.GUID
hBalloonIcon.i
EndStructure
CompilerCase #PB_Processor_x64
Structure MY_NOTIFYICONDATA
cbSize.l
PB_Alignment1.b[4]
hWnd.i
uID.l
uFlags.l
uCallbackMessage.l
PB_Alignment2.b[4]
hIcon.i
szTip.s{128}
dwState.l
dwStateMark.l
szInfo.s{256}
StructureUnion
uTimeout.l
uVersion.l
EndStructureUnion
szInfoTitle.s{64}
dwInfoFlags.l
guidItem.GUID
hBalloonIcon.i
EndStructure
CompilerEndSelect
ExeFile.s = ProgramParameter(0)
Procedure.s GetMachineArchHack(file.s)
Protected dos_stub_end
If ReadFile(0,file)
FileSeek(0,$84)
pe_mach_arch.s = ReadString(0, #PB_Ascii)
CloseFile(0)
If Len(pe_mach_arch) > 0
If pe_mach_arch = Chr(76) + Chr(1) + Chr(5)
ProcedureReturn "x86"
ElseIf pe_mach_arch = Chr(100) + Chr(8224) + Chr(6)
ProcedureReturn "x64"
Else
ProcedureReturn "Unknown machine type."
EndIf
Else
ProcedureReturn "Failed to read Machine Architecture. "
EndIf
EndIf
EndProcedure
If Len(ExeFile) > 0
If Bool(FindString(ExeFile, "PureBasic_Compilation")) = #False
LoadImage(0, #PB_Compiler_Home + "\Examples\Sources - Advanced\Waponez II\Waponez.ico")
ni.MY_NOTIFYICONDATA
ni\cbSize = SizeOf(MY_NOTIFYICONDATA)
ni\uID = 10
ni\uFlags = #NIF_ICON| #NIF_INFO | #NIF_MESSAGE |#NIF_TIP
ni\hIcon = ImageID(0)
ni\uCallbackMessage = #WM_NOTIFYICON
ni\uTimeout = 10
ni\uVersion = #NOTIFYICON_VERSION
ni\szInfoTitle = GetMachineArchHack(ExeFile) + " Executable Created"
ni\dwInfoFlags = #NIIF_INFO
Shell_NotifyIcon_(#NIM_ADD, ni)
ni\szInfo = "Compilation of " + GetFilePart(ExeFile) + " Complete " + #CRLF$ + FormatDate("%dd/%mm/%yyyy - %hh:%ii:%ss", GetFileDate(ExeFile, #PB_Date_Modified)) + #CRLF$ +"OK."
Shell_NotifyIcon_(#NIM_MODIFY, ni)
Shell_NotifyIcon_(#NIM_DELETE, ni)
EndIf
EndIf
EndAnd blocking the user is not a bug? I mean the user won't know when it is finished. He can only guess it. There is no deterministic way for a usual user to tell whether or not the creation has finished. In theory he could wait an infinite amount of time and never get a hint that it is finished. At least not without external tools.Marc56us wrote:You're right,Dude wrote:That doesn't do anything when you're creating a final executable, which was the request.Marc56us wrote:File > Preferences > Debugger > [X] Display Timestamp in Error log
but a bug is something that blocks a program or do error(s)
When I compile I get that popup with the progress bar. Does this not include the time it takes to write the executable to disk?DarkDragon wrote:There is no deterministic way for a usual user to tell whether or not the creation has finished.