2014-01-20 17:36 UTC+0100 Przemyslaw Czerpak (druzus/at/poczta.onet.pl)

[Przemyslaw Czerpak]

2014-01-20 17:36 UTC+0100 Przemyslaw Czerpak (druzus/at/poczta.onet.pl)
* src/rtl/hbsocket.c
* removed not longer valid TODO note

* contrib/hbmisc/bbabble.c
! fixed buffer size calculation (it fixes buffer overflow for small
strings, i.e. 2 bytes and does not allocate more memory then necessary
for long ones)

* contrib/hbct/dattime2.c
! fixed CToMonth() and CToDoW() to respect case intensive search for
strings with national characters
* variable name synced with Viktor branch

* contrib/hbct/strswap.c
! fixed NULL reference when neither 1-st nor 2-nd parameter is passed
by reference

* contrib/hbct/fcopy.prg
! fixed source file positioning in case of partial write
! use hb_FGetDateTime() / hb_FSetDateTime() instead of
FileDate()+FileTime() / SetFDaTi()
It's faster and unlike FileDate() and FileTime() it does
not leave directory open.
! fixed typo (RTE)
% minor simplification

* src/rtl/disksphb.c
! use dynamic binding for GetDiskFreeSpaceEx() function

; modifications below comes from Viktor fork:

* contrib/hbct/ctwin.c
% removed unnecessary declaration

* contrib/hbct/envparam.c
! do not call FreeEnvironmentStrings() is it's not necessary

* contrib/hbodbc/odbc.c
! fixed potential buffer overflow in SQLERROR()/SQLGETDIAGREC()

* contrib/hbnetio/netiosrv.c
! fixed possible NULL reference

* contrib/hbct/ctnet.c
! added missing check for result of WinAPI function call

* contrib/hbct/cttime.prg
! SecToTime(): fixed regression from 2007-11-18 13:46 UTC+0100

* contrib/hbwin/hbolesrv.c
! fixed char -> TCHAR after recent unicodification


best regards
Przemek

[Qatan]

Hello Przemek,

Thanks for the fixes.

BTW would you mind to explain a bit how it is working right now?
I mean, seems that Viktor updates a different fork and he gets fixes from
the mainstream Harbour but you also get fixes from his fork... why it's like
that?
Is it just an option from Viktor's part? I am not sure but seems that he
wanted to do more "radical" changes and was not feeling free to do it on the
mainstream? Is that? Please correct me if I am wrong.
I am just curious.
Thanks for any info. If you are willing to say anything, of course.
Regards,

Qatan

[Przemyslaw Czerpak]

On Mon, 20 Jan 2014, Qatan wrote:

Hi,

You should ask Viktor about it.
I haven't talked to him about his fork.

best regards,
Przemek

[ToninhoFWi]

Hi Przemek.

First, thanks for all and second, I have this error with bcc:

..\contrib\hbct\ctnet.c:
Error E2451 ..\contrib\hbct\ctnet.c 198: Undefined symbol 'NO_ER' in
function HB_FUN_NETRMTNAME
Error E2377 ..\contrib\hbct\ctnet.c 197: If statement missing ) in function
HB_FUN_NETRMTNAME
Error E2054 ..\contrib\hbct\ctnet.c 199: Misplaced else in function
HB_FUN_NETRMTNAME
Warning W8004 ..\contrib\hbct\ctnet.c 209: 'dwLen' is assigned a value that
is never used in function HB_FUN_NETRMTNAME
*** 3 errors in Compile ***


Best Regards,

Toninho.

[Qatan]

Hello Przemek,

>You should ask Viktor about it.
>I haven't talked to him about his fork.

OK, Thanks. I will do it.

Viktor, please? Do you want to say something?
I am not questioning what you do. I am just curious... if you want to say
anything about it of course.

Regards,

Qatan

[Heinz V Bergen]

I get same errors under MSVC and MINGW

I think ctnet.c line 197 got truncated somehow and should be:

      if( dwSize > 0 && dwSize <= dwLen && WNetGetConnection( lpLocalName, lpRemoteDevice, &dwSize ) == NO_ERROR )

Regards,
Heinz V Bergen

[vszakats]

Hi Qatan,

On Monday, January 20, 2014 7:54:27 PM UTC+1, Qatan wrote:

> Viktor, please? Do you want to say something?

> I am not questioning what you do. I am just curious... if you want to say

> anything about it of course.

Elch summed it up best in a funny way (but he deleted it).

Internet communities are certainly useful, but there came

a point (few years ago) where this one became too much of

a burden to me. Working hard with diminishing benefits

makes crowds angry and suspicious while wanting more of it.

Both of above makes me feel bad and angry too. And the

cycle just becomes worse, till it feels a total failure.

I've found the resolution was to pull away from it.

I still enjoy fixing (or reporting) bugs or making the

code better (or seldom fixing stuff that I need for

actual production work), so I keep doing it in my own

sandbox, without the burden of politics, management,

implied responsibility or the various other forms of

external pressure.

Since the patches and source code is fully public, still

everybody may benefit from it, by its own merits. If that's

the case, attribution and donations are welcome (and

issues got opened).

[ BTW contrary to guesses I don't plan to change to any

more "agressive" development style. Unless you count here

some marginal incompatible changes, plus turning off

HB_LEGACY_LEVEL4 after many years. ]

[ Oh and GitHub now recognizes .prg/.ch and .hb files

as xBase and Harbour respectively. Or any projects that

uses Linguist, where these patches went and were all

accepted just recently. ]

[ The .pot handling bug is also fixed in the __i18n_*()

API. I hope it will be pulled by Przemek. ]

[Qatan]

Hello Viktor,

 

>> Viktor, please? Do you want to say something?

>> I am not questioning what you do. I am just curious... if you want to say

>> anything about it of course.

>

>Elch summed it up best in a funny way (but he deleted it).

>

>Internet communities are certainly useful, but there came

>a point (few years ago) where this one became too much of

>a burden to me. Working hard with diminishing benefits

>makes crowds angry and suspicious while wanting more of it.

>Both of above makes me feel bad and angry too. And the

>cycle just becomes worse, till it feels a total failure.

>I've found the resolution was to pull away from it.

 

Yes, I understand you.

I think I am starting to learn how to deal with bad feelings.

Sometimes we just have to ignore it and think about the good.

But I am not the best example... as I said I am not good doing it.

After all we all have more important things besides Harbour, don’t we?

Harbour is fun!

 

 

>I still enjoy fixing (or reporting) bugs or making the

>code better (or seldom fixing stuff that I need for

>actual production work), so I keep doing it in my own

>sandbox, without the burden of politics, management,

>implied responsibility or the various other forms of

>external pressure.

 

If it helps you to enjoy it than I can only say that you are doing good.

I really hope you can enjoy this nice project.

Right now Harbour is very powerful and it’s becoming more and more stable.

 

 

>Since the patches and source code is fully public, still

>everybody may benefit from it, by its own merits. If that's

>the case, attribution and donations are welcome (and

>issues got opened).

>

>[ BTW contrary to guesses I don't plan to change to any

>more "agressive" development style. Unless you count here

>some marginal incompatible changes, plus turning off

>HB_LEGACY_LEVEL4 after many years. ]

 

That’s what I meant based on a message you wrote lately.

 

 

>[ Oh and GitHub now recognizes .prg/.ch and .hb files

>as xBase and Harbour respectively. Or any projects that

>uses Linguist, where these patches went and were all

>accepted just recently. ]

>

>[ The .pot handling bug is also fixed in the __i18n_*()

>API. I hope it will be pulled by Przemek. ]

 

Nice! Thanks for fixing it.

I hope Przemek pulls it to mainstream Harbour.

I will test it on Qalculator. 

 

Well, I really desire you may enjoy all your work and time dedicated to this project.

It saved many programmers with their old programs that would not work anymore as the computers and O.Ses. are upgraded.

Not to say the countless extensions that gave to Harbour so many nice possibilities (many that I don’t even know yet).

And... Thank you for taking time to answer.

 

Best Regards,

 

Qatan

[elch]

Hi Qatan,

>Elch summed it up best in a funny way (but he deleted it).

>

>Internet communities are certainly useful, but there came

>a point (few years ago) where this one became too much of

>a burden to me. Working hard with diminishing benefits

>makes crowds angry and suspicious while wanting more of it.

>Both of above makes me feel bad and angry too. And the

>cycle just becomes worse, till it feels a total failure.

>I've found the resolution was to pull away from it.

 

Yes, I understand you.

i fear the opposite is the case.

 

After all we all have more important things besides Harbour, don’t we?

wait, whut ? - what ?  - fishing ??

Maybe you, but at very least not me - as getting daily food and clothes yearly seem a 'somehow' important thing, isn't it ?

Robbing retired maybe a funny alternative to reach same aim ... 8-)

 

Harbour is fun!

Harbour MAKES fun -- aside the naked work.

I strong believe the intention of an illuminated expert like Viktor was not to sit down days for days to work on Harbour,

so that some other have a toy for wasting some of their useless sparetime.

 

>I still enjoy fixing (or reporting) bugs or making the

>code better (or seldom fixing stuff that I need for

>actual production work), so I keep doing it in my own

>sandbox, without the burden of politics, management,

>implied responsibility or the various other forms of

>external pressure.

 

If it helps you to enjoy it than I can only say that you are doing good.

:-) !

hehe, british alike understatement, yes ? -- even with more decades training i won't reach his experience horizon.

 

I really hope you can enjoy this nice project.

'Enjoy' seem wrong words in your sentences - replace them with 'benefit'.

But what can be expected from poor professionals like me and a bunch hobbyists.

Wish me moooneey and i will donate a bit ;-)

 

Right now Harbour is very powerful and it’s becoming more and more stable.

:-) !

maybe the 'cleanest' piece of software i experienced: not even a single memory leak in ready app.

It's hard to find bugs, or better spoken 'misbehavior', which can't be corrected on user level.

Up now i have not even a single time saw this software breaking down !

Okay, there are about 2K ToDos left in source for further ideas, but that is no topic of 'stable' ..

 

.. 

It saved many programmers with their old programs that would not work anymore as the computers and O.Ses. are upgraded.

Wrong view of point: it opens the possibility to continue by enhancing your formerly time-proven work, instead of starting from scratch.

No topic for a single calculator app, but very sure a k.o. criteria for a piece in which you invested years !

See no real problem to let old software run. E.g. for Linux exists a now nearly forgotten famous dosemu (i mean not this poor 'dosbox' named one).

Clipper programs run well in that 'virtual machine', and there are similar ways to get your stone age software to run ...

THANK you Viktor !

wish as many 'coconuts' you can carry :-)

elch

[Qatan]

Hello Elch,

 

    Surely your point of view is valid and very important.

    It’s obviously different than mine but I really appreciated it.

    You have to keep in mind that I don’t use Harbour professionally (meaning to make a living) besides helping me to pay some expensive dentist bills (I participated on a tiny portion on a small ERP project). BTW I never got a bug from the program I made after it got out of “Beta” what shows that Harbour is like a “tank”. Very stable and reliable programming language.

    I can assure you that I am very glad with Harbour and with Viktor dedication. I am not mentioning many others that are also very dedicated but I want to express my gratitude for all of them. Harbour became an unique tool.

    Best wishes to you all!

 

Qatan