Search found 34 matches

by Solbianca
Feb 10, 2017 13:38
Forum: General
Topic: Any reason for not allowing dim byref in UDT?
Replies: 20
Views: 3660

Re: Any reason for not allowing dim byref in UDT?

Is this principle could be also allowed outside a Type block, in order to be able to separate the reference declaration and the reference assignment? I think that the principle should be allowed inside type block only. Because, when use outside type block, it don't meet the requirement of reference...
by Solbianca
Feb 10, 2017 8:31
Forum: General
Topic: Any reason for not allowing dim byref in UDT?
Replies: 20
Views: 3660

Re: Any reason for not allowing dim byref in UDT?

Yea, I suppose a rule like "all references must be assigned in the constructor" could work. It's probably better than adding a new syntax, and the compiler can do some basic checking to ensure all references are assigned at least once in the constructor. You'd just have to be careful not ...
by Solbianca
Feb 08, 2017 11:14
Forum: General
Topic: Any reason for not allowing dim byref in UDT?
Replies: 20
Views: 3660

Re: Any reason for not allowing dim byref in UDT?

If FB initializes normal fields too without calling of the copy/assign operator, we need the member initializer like C++. Do you have any good ideas about that? I thought about that in my own way. type Foo public: constructor(byref n as const Bar) '' init constructor. ... private: as Bar value = any...
by Solbianca
Feb 07, 2017 20:50
Forum: General
Topic: Any reason for not allowing dim byref in UDT?
Replies: 20
Views: 3660

Re: Any reason for not allowing dim byref in UDT?

I think that FB does't need a new syntax for it. For example, use ANY keyword. type Foo public: declare constructor(byref n as const integer, byref m as integer) private: as const integer constValue = any '' no initialize yet. byref as integer byrefValue = any '' no initialize yet. end type construc...
by Solbianca
Oct 05, 2016 21:07
Forum: Community Discussion
Topic: 1.05.0
Replies: 43
Views: 10065

Re: 1.05.0

@dkl
Thank you for fixing it :3
by Solbianca
Oct 05, 2016 10:07
Forum: Community Discussion
Topic: 1.05.0
Replies: 43
Views: 10065

Re: 1.05.0

Hello there, I think that wstring type has a bug of the surrogate pair conversion. When it convert code point into a surrogate pair, the high surrogate has a wrong value. Probably the operation does not subtract &h10000ul from code point. Test Code: type SurrogatePair as ushort high as ushort lo...
by Solbianca
Mar 02, 2016 22:23
Forum: Community Discussion
Topic: 1.05.0
Replies: 43
Views: 10065

Re: 1.05.0

@dkl @fxm The reason for this is that if the base class has a Let overload, then the derived class must have one too, otherwise assignment would be broken: the base's Let operator wouldn't be called when assigning objects of the derived class. Now I get it! The compiler generated the error because o...
by Solbianca
Mar 02, 2016 16:04
Forum: Community Discussion
Topic: 1.05.0
Replies: 43
Views: 10065

Re: 1.05.0

Hi, This class can not inherit. type Foo public: declare constructor() declare destructor() ... private: declare operator let(byref as const Foo) ... end type Result: error 201: Illegal member access, Foo.operator.let The derived class is copying the object of the base class internally. Why? Thanks,
by Solbianca
Sep 18, 2015 11:03
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

Hi,
Yes, I confirmed that the problem was solved.
Thank you again :D
by Solbianca
Sep 17, 2015 16:38
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

As you're japanese (it seems) the string $$__compilinfo__$$.... is stored in the exe with an Unicode format. But fbdebugger is searching the string in Ascii format (in the memory area where the constants are stored). For information, when loading the variable named compil_info a function is called ...
by Solbianca
Sep 17, 2015 11:17
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

Hi,
Oh, I'm sorry to have caused you trouble.

Try this:
http://fileup.jp/up/9590.zip.html
It is no password.
by Solbianca
Sep 17, 2015 8:35
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

Umm... It is surely strange.

File: http://fileup.jp/up/9589.zip.html
Download pass: abc
(I'm sorry with a Japanese server)

I try it a little more in various ways.
Thanks
by Solbianca
Sep 16, 2015 14:46
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

I'm using v1.03.0, 32bit and GAS backend. The variable 'comil_info' in the proc/var tab is getting the information, but the compile info dialog does not display it. The test code: dim as string compil_info="$$__COMPILINFO__$$"+__FB_VERSION__+"/"+__FB_BUILD_DATE__+"/"+__...
by Solbianca
Sep 16, 2015 11:43
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

Not so hard ;-) Now position/size of the main window are saved when exiting and restored when restarting. Good job! the debugger found the own place to stay. I added the following code or your specified code into my source code, but I wasn't able to get the information. Is a way wrong? if so, It is...
by Solbianca
Sep 15, 2015 11:44
Forum: Projects
Topic: FB debugger : 3.02 32/64 BIT WDS/LNX..... (2023/07/05)
Replies: 762
Views: 362771

Re: FB debugger : 2.90 B2 32/64 BIT ..... (sept 14th, 2015)

Hi SARG,
It works without problems on my PC too, but I found a indication bug in `compile info`.
I wish that the debugger memorizes the position of the own window, if possible.

Thanks