Is there a way to use fill.reads to color alignments by read strand?  I would like to plot these on a single track

library(Gviz)
alTrack <- AlignmentsTrack(system.file(package="Gviz", "extdata", 
  "gapped.bam"), isPaired=FALSE)
plotTracks(alTrack, from=2997000, to=2998200, chromosome="chr12", 
  fill.reads=c("red", "blue"))

I tried using the strandedBamImport function from a previous question (Gviz - show plus and minus strand from a bam) but get an Error in .Call2("cigar_ranges" ...  

The way how Gviz handles colouring for AlignmentsTrack is a bit special. It recycles colours specified in the fill.reads along the length of unique read ids (uid column). On top of it, if you plot reads as arrows, it also sorts the reads before plotting according to the strand.

Maybe this piece of code does what you want. It is similar to what you suggested but still retain all plotting functionalities for the AlignmentsTrack class. I tested it on few regions (even with gapped reads) and it seems to work.

library(Gviz)

bf <- system.file(package="Gviz", "extdata", "gapped.bam")

chrom <- "chr12"
from <- 2997500
to <- 2998200

alTrack <- Gviz:::.import.bam.alignments(bf, GRanges(chrom, IRanges(from, to)))
alTrack <- AlignmentsTrack(sort(alTrack))

displayPars(alTrack)$fill.reads <- c(plus="#ECAFAE", minus="#AFAEED")[as.numeric(sort(ranges(alTrack)$readStrand))]

plotTracks(alTrack, from=from, to=to, chromosome=chrom)

Let me know if that works for you.

I can color by read strand if I subset a data.frame before creating the alignments track.  

bam1 <- readGAlignments(BamFile(system.file(package="Gviz", "extdata", "gapped.bam")))
bamdf <- data.frame(bam1)
names(bamdf)[1] <- "chromosome"
bamdf <- subset(bamdf, start > 2997000 & end < 2998200)
clrs <- ifelse(bamdf$strand == "+", "#ECAFAE", "#AFAEED")
alTrack <- AlignmentsTrack(bamdf, name="Alignments", fill = clrs)
plotTracks(alTrack)