On 04/12/06, Troy A. Griffitts <[EMAIL PROTECTED]> wrote:
> Hi Martin,
> You cannot take out the line you have taken out without changing the
> logic of the code. Not having written the code, I cannot confirm that
> you have not broken anything. Please note that SWBuf::setSize will be
> sure
Hi Martin,
You cannot take out the line you have taken out without changing the
logic of the code. Not having written the code, I cannot confirm that
you have not broken anything. Please note that SWBuf::setSize will be
sure the data buffer can handle at least 'size' number of byte. Since
Hi Troy.
Thanks for taking notice of my patch.
The strlen should not pose a problem, because I removed one that was there but
seemed superfluous:
-inBuf.setSize(strlen(inBuf.c_str()));
So we still have only one strlen. Nothing should be slower.
God bless.
Martin
Am Montag, 4. Dezember 2006
Thanks Daniel,
If you take this on, please note:
Do not include this:
+ cacheBuf[0]='\0'; //just in case len==0
Please see if the added strlen is much of a speed hit. This code should
probably be changed to use SWBuf, which keeps track of len and doesn't
need to do strl
Thanks Daniel.
> Assuming (ok, big assumption ;) that there will be an imminent release
> can the frontends please try out with current sword svn (preferably +
> Martins patch). I can't see how anything would have been broken since
> 1.5.9 but better safe than sorry.
BibleTime is ready.
mg
On 01/12/06, Martin Gruner <[EMAIL PROTECTED]> wrote:
> The attached patch fixes the stability problems experienced with Sword and
> ciphered and locked modules.
>
> To see unpatched Sword crash, just run
>
> mod2imp GerHfa2002 and
> mod2imp GerHfaLex2002
>
> without the correct unlock
