@PERL and ^C

[Jim Cook]

TCC 9.02.152 Windows XP [Version 5.1.2600]
TCC Build 152 Windows XP Build 2600 Service Pack 3

When I do this and type "hello<enter>" it does what I'd expect.
However, if I press ^C when it is waiting, the whole window vanishes.
It happens whether a stand-alone TCC or within a TCMD tab.

Can others reproduce this? I fear maybe not, because I have a
suspicion that the original fault comes from within my machine's
perl.dll

echo "%@perl[$_=<>]"

--
2008 Fridays: 4/4, 6/6, 8/8, 10/10, 12/12 and 5/9, 9/5, 7/11, 11/7.
Next year they're Saturday.
Measure wealth by the things you have for which you would not take money.

 

[rconn]

Jim Cook wrote:

> TCC 9.02.152 Windows XP [Version 5.1.2600]
> TCC Build 152 Windows XP Build 2600 Service Pack 3
>
> When I do this and type "hello" it does what I'd expect.
> However, if I press ^C when it is waiting, the whole window vanishes.
> It happens whether a stand-alone TCC or within a TCMD tab.
>
> Can others reproduce this? I fear maybe not, because I have a
> suspicion that the original fault comes from within my machine's
> perl.dll

That's a long-standing bug in the Perl DLL. There is no workaround at
this time; it will require fixing the Perl code and releasing our own
version, something I'm highly reluctant to do.

Rex Conn
JP Software

 

[p.f.moore]

2008/9/26 JP Software Forums <[email protected]>:

> That's a long-standing bug in the Perl DLL. There is no workaround at
> this time; it will require fixing the Perl code and releasing our own
> version, something I'm highly reluctant to do.

Rex,
Can you explain the exact nature of the bug in Perl, and in broad
terms what would have to be done to the Perl code to fix it?

The reason I ask is that I reported a very similar problem in the
"Plugins" forum some time ago (the only reference I can give is the
tag attached to the email header: Plugins-t-338) where my own Lua
plugin was crashing the TCC instance when Ctrl-C was pressed while the
plugin was in a GetLine call waiting for input. I reproduced the
problem in a small test case and determined that it was related to
(the lack of) break handling in my plugin - it appears that TCC
doesn't provide a default break handler to catch issues (which I'd
consider a bug, but I suspect would be deemed as WAD). Nevertheless,
if you can explain how to write "correct" break handlers for a plugin
(I'm guessing from the symptoms that the Perl issue is similar) that
would be immensely useful. (And yes, I know that reading the MS
documentation should give me the details I need, but I have done so,
and experimented endlessly, but I cannot work out what is needed to
write a clean break handler that regains control from within GetLine
in a way that allows me to resume processing).

Paul.

 

[vefatica]

On Fri, 26 Sep 2008 03:42:48 -0500, "JP Software Forums"
<[email protected]>,"p.f.moore" <> wrote:

>Can you explain the exact nature of the bug in Perl, and in broad
>terms what would have to be done to the Perl code to fix it?

On 10/6/2006, Rex Conn wrote:

"That's a bug in the Active State Perl -- they're intercepting the ^C and
unconditionally shutting down the parent app (4NT in this case)."

At the time, I investigated the Perl source code. It's true.

 

[Jim Cook]

>> However, if I press ^C when it is waiting, the whole window vanishes.

> That's a long-standing bug in the Perl DLL. There is no workaround at
> this time; it will require fixing the Perl code and releasing our own

I was pretty sure that was the case, and you may even be able to find
historical messages where I've asked the same question and been
answered the same way. I hope not, but ....

Thanks again for the information.

--
2008 Fridays: 4/4, 6/6, 8/8, 10/10, 12/12 and 5/9, 9/5, 7/11, 11/7.
Next year they're Saturday.
Measure wealth by the things you have for which you would not take money.