AH!! Thanks for reminding me, I did not add that fix from T4 into T3. I how have and re-uploaded v3.47. NOTE - Download v3.47 again if you want this fix.T3's been that way since I started using it. It wanted semicolons when switching to a new chapter. So now we don't need a semicolon?
OK - there have been 2 fixes in T3 and T4. The one I think you are speaking about is the "AND" issue. "AND", "and", and "And" were interpreted as "semi-colons". This is the only thing that has been changed with semi-colons. "AND" is now interpreted as a "comma", and only with "Chapter Tooltips" turned on.
John 3:16 and 17 is now interpreted as John 3:16,17. The old tooltip formatter produced the correct result, the new formatter with Chapter Tooltips was producing the wrong result. John 3:16; 17. is not the same as John 3:16, 17. The former implies chapter 3 verse 16, and chapter 17, while the latter implies chapter 3, verses 16 and 17 to the Chapter formatter. Of course, you can still use John 3:16 and 17:3, and this will give the expected results of Joh_3:16; Joh_17:3. So that John 3:16, 17:3 is the same as John 3:16; 17:3. but John 3:16; 17, the semi-colon is implying the whole of chapter 17. This behavior with respect the "AND" has changed with repect to Chapter tooltips.
The other issue was a bit deeper. I give the example of tooltipping this reference: "Deut 7:2-5, 25, 12:2-3, 30, 18:9-13, 30:17-18". This reference would work fine with the old tooltip formatter. With Chapter tooltips turned one, in T4 it would would be marked as invalid, and in T3, it would get underlined, but not colored, and also marked as invalid. But is it invalid? YES! With Chapter tooltips turned off, it was tooltipped just fine. So tracing out the parsing of this reference while tooltipping in the new algorithm, I found I had left out the final step of one line which was taking the reference string apart. That is, one typo error in the source code. The problem was, that the segment, "12:2-3, 30" was interpreting the "30" as being in Chapter "21" and not chapter "12". There is no Deuteronomy 21:30, so it was flagged as invalid. HOWEVER, lets say the 30 was instead 5. The tooltip results would have been all flaged as valid, yet you would have had a reference to Deuteronomy "21:5" instead of the expected result of "Deu_12:5". Nasty, eh? This has been fixed now in both T3 and T4. Note, this problem only manifested itself if there were several compound references combined. If the reference was Deut 12:2-3, 5, this would have tooltipped correctly. Once the second block was added, did the problem surface. However, it would have appied to both string blocks. If the chapter was a single character, then one would have never noticed an issue. Only when there were 2 or more chapter characters. Now you know... :-)
Edited by BH., 29 January 2012 - 02:21 PM.
