Angus Leeming <[EMAIL PROTECTED]> writes:
| On Thursday 02 May 2002 6:03 pm, Herbert Voss wrote:
>> Angus Leeming wrote:
>> > On Thursday 02 May 2002 5:17 pm, Herbert Voss wrote:
>> >>I don't who committed this:
>> >>
>> >>2002-05-02 Herbert Voss <[EMAIL PROTECTED]>
>> >>
>> >> * filetools.C: getline -> std::getline.
>> >>
>> >>
>> >>maybe that I'm getting old, but this is in my ChangeLog, but
>> >>not changed in my filetools.C and I never send such patch.
>> >
>> > I too am getting old. I've applied so many of your patches recently that
>> > my subconscious must have thought I'd changed my name.
>> >
>> > I'll change the ChangeLog entry to blame myself, but you should find a
>> > using::std::getline directive on line 70 of filetools.C.
>>
>> ok, I misunderstand the changelog; looked for the one in the function.
>> Could you please apply the patch with the changed lyxerr line,
>> which nobody hurts. it's annoying when I'm fixing bugs and get
>> that output.
>>
>> Herbert
>
| I don't think it's safe. What if str has zero length? You need
>
| string const print_str = (str.size() > 60) ? str.substr(0,60) : str;
| lyxerr[Debug::GRAPHICS] << "Scanstring: " << print_str << endl;
>
| To the C++-experts out there:
>
| Am I right? Is str.substr(0,60) unsafe if str has zero size?
I think you are wrong. But easy to check... just try it.
I think that
str.substr(0) == str
str.substr(0, <number larger than leght of str>) == str
easy to check though.
--
Lgb
