uneven number of tokens

[SamGG]

Hi,

As I am parsing a lot of FCS files from Cytosplore, the "uneven number of tokens:" appears many times. I dived in the question and noticed that in fact the end of the TEXT is erroneously defined at the beginning of DATA. This leads in appending one extra byte to the TEXT, which in turns leads to misleading the count of keywords. What is interesting is that if the extra byte is 0x00, there is no error as it is considered as the end of a C string.
So I added this chunk of code between those two lines

flowCore/R/IO.R

Lines 659 to 660 in 14781e7

	}
	}

	# update textend according to datastart
	if (offsets["textend"] == offsets["datastart"]) {
	  offsets["textend"] <- offsets["datastart"]-1
	}

In fact, this removes half of the warnings as there are two warnings for each "erroneous" FCS file. The remaining warning arises from the first call of readFCStext with uncorrected offsets. This could be also solved but I did find a nice way and short way to do it.

Finally, I tried to compile flowCore with RStudio and its tools under Windows OS. I lost hours at it. I finally ended with a shell command, which does not please me in fact. I faced errors as the compiler included RccpArmadillo .h files, which seems to have no sense. If you have a nice way to compile flowCore under Windows, let me know please.

Best,
Samuel