[sword-devel] Odd behavior in ListKey

[Troy A. Griffitts]

DM,

Sorry for the late reply and thank you for tracking this down. Yes!  
This looks like a reasonable solution.  Please feel free to commit if 
all the sword/tests/testsuite/run-all.sh returns no problems with your 
change.

It's been a bit crazy these days and I haven't been able to keep up with 
all the great stuff going on here.

Troy


On 6/16/25 9:27 PM, DM Smith wrote:
> Troy,
> In VerseKey::checkBounds()
> void VerseKey::checkBounds() {
>
>         long i = getIndex();
>
>         initBounds();
>         if (i > upperBound) {
>                 setIndex(upperBound);
>                 i = getIndex();
>                 error = KEYERR_OUTOFBOUNDS;
>         }
>         if (i < lowerBound) {
>                 setIndex(lowerBound);
>                 error = KEYERR_OUTOFBOUNDS;
>         }
> }
>
> i, upperBound and lowerBound are all -1, when working with a chapter 
> that is not in the book. This is good. The proc getIndex() can’t 
> compute the index since the chapter is not known so it rightfully 
> returns -1.
>
> I think there should be a check in here like there is in 
> VerseKey.setIndex(int iindex).
>
> So
> void VerseKey::checkBounds() {
>
>         long i = getIndex();
>
> *        // assert we're sane*
> *        if (i < 0) {*
> *                error = KEYERR_OUTOFBOUNDS;*
> *                return;*
> *        }*
>
>         initBounds();
>         if (i > upperBound) {
>                 setIndex(upperBound);
>                 i = getIndex();
>                 error = KEYERR_OUTOFBOUNDS;
>         }
>         if (i < lowerBound) {
>                 setIndex(lowerBound);
>                 error = KEYERR_OUTOFBOUNDS;
>         }
> }
>
> What do you think?
>
> DM
>
>> On Jun 15, 2025, at 8:11 AM, DM Smith <dmsmith at crosswire.org> wrote:
>>
>> Troy,
>>
>> I’ve narrowed it down to turning Auto Normalize off.
>>
>> In examples/cmdline/parseverselist.cpp, add 
>> parser.setAutoNormalize(false) to see the error with Gen.51.1. Note 
>> Gen.50.99 (last chapter with bad verse number) works.
>>
>> DM
>>
>>> On Jun 12, 2025, at 7:03 PM, DM Smith <dmsmith at crosswire.org> wrote:
>>>
>>> Troy,
>>>
>>> I’m working on an infinite loop bug in osis2mod.
>>>
>>> I’ve narrowed it down to ListKey containing a verse with a chapter 
>>> which is beyond the end of a book.
>>>
>>> When this happens list.increment(1) never sets an error.
>>>
>>> Simplest test case (bit incomplete):
>>>
>>> int i = 0;
>>> ListKey list = new ListKey();
>>> list.add(“Gen.51.1”);
>>> for (list = TOP; !list.popError(); list.increment(1)) {
>>> if (i++ > 5) break;
>>> cout << i << list << endl;
>>> }
>>>
>>> If I change the verse reference to Gen.1.99 (valid chapter, invalid 
>>> verse), it works as expected.
>>>
>>> Can you figure out the problem?
>>>
>>> I’ve got a work around but I’d rather not do that.
>>>
>>> Thanks,
>>> DM
>>>
>>> _______________________________________________
>>> sword-devel mailing list: sword-devel at crosswire.org
>>> http://crosswire.org/mailman/listinfo/sword-devel
>>> Instructions to unsubscribe/change your settings at above page
>>
>> _______________________________________________
>> sword-devel mailing list: sword-devel at crosswire.org
>> http://crosswire.org/mailman/listinfo/sword-devel
>> Instructions to unsubscribe/change your settings at above page
>
>
> _______________________________________________
> sword-devel mailing list:sword-devel at crosswire.org
> http://crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://crosswire.org/pipermail/sword-devel/attachments/20250619/4aede030/attachment.htm>