[sword-devel] Bug: Inverted selection of pri/sec (Re: Primary/secondary reading...)

Karl Kleinpaste karl at kleinpaste.org
Mon Jan 21 08:47:12 MST 2008


I figured out my glitch in use of "Textual Variants" this morning.

What I've learned in the process of fixing the GnomeSword code to select
variants properly is that the Sword lib functions for OSISVariants and
ThMLVariants get the result exactly backwards, seemingly on the basis of
the local variable "hide", though I don't claim to have analyzed it fully.

For an example failure, start up the Windows UI and navigate to WHNU
Matt 1:18, where the 3rd word contains this markup (line breaks for
readability):

<div type="variant" class="1">
[ιησου]
<sync type="Strongs" value="G2424" />
<sync type="morph" class="Robinson" value="N-GSM" />
</div>
<div type="variant" class="2">
ιησου
<sync type="Strongs" value="G2424" />
<sync type="morph" class="Robinson" value="N-GSM" />
</div>

Note the intended primary -vs- secondary choices above, and then observe
that the actual display is inverted from that.



More information about the sword-devel mailing list