Problems with certain linear barcodes

[thokall]

Generating linear barcodes seem to have issues with certain strings, whereas most strings are working as expected.

This code block works

library(baRcodeR)
custom_create_PDF(Labels = "SEP0164581", type = "linear")

but this does not generate a valid linear barcode

library(baRcodeR)
custom_create_PDF(Labels = "SEP0164580", type = "linear")

Both examples will generate valid qr-codes if changing the type to matrix, but since I am using external tools that can only read linear barcodes I need to find a tool for generating these.

I have generated around 10000 codes of the same type and by testing a subset there seem to be between 5-10% of the generated barcodes that are not valid.

I have tested on both a linux and a mac os x machine with the same results. Below is my R session information for the linux machine.

sessionInfo()
R version 3.5.1 (2018-07-02)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.1 LTS

Matrix products: default
BLAS: /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.7.1
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.7.1

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=sv_SE.UTF-8       
 [4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=sv_SE.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=sv_SE.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=sv_SE.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] baRcodeR_0.1.2 qrcode_0.1.1  

loaded via a namespace (and not attached):
 [1] compiler_3.5.1    magrittr_1.5      tools_3.5.1       yaml_2.2.0        stringi_1.2.4    
 [6] R.methodsS3_1.7.1 grid_3.5.1        stringr_1.3.1     R.utils_2.8.0     R.oo_1.22.0

[thokall]

Looking at one example the checksum part of the barcode looks to be the problem, but it is beyond me to why some work whereas other do not.

[thokall]

Not sure anyone is reading this, but I think the problems is related to the checksum calculation and/or translation from these to the barcode.

For the string "SEP0164580" the checksum in the below function is 1338 and the check_character hence 102. The binary_label will then try to extract value the barcode corresponding to the 102 + 32 from the barcodes128 object supplied with the package. Since there is no entry in the ASCII column with value 134 this will not add a valild checksum to the barcode

code_128_make <- function(Labels){
  ## labels is a character string
  ## read in dict 
  Barcodes <- barcodes128
  ## double check Labels
  Labels <- as.character(Labels)
  Labels <- iconv(Labels, from = "utf-8", to = "ascii", sub = "-")
  start_code <- 209
  lab_chars <- unlist(strsplit(Labels, split = ""))
  lab_values <- sapply(lab_chars, function(x) utf8ToInt(x))
  # ascii to code 128 is just a difference of 32, this line keeps clarity
  code_values <- lab_values - 32
  # 104 is the start value for start code b, hardcoded right now
  check_sum <- 104 + sum(code_values * 1:length(code_values))
  check_character <- check_sum %% 103
  Binary_code <- sapply(lab_values, function(x, Barcodes) Barcodes$Barcode[x == Barcodes$ASCII], Barcodes = Barcodes)
  ## create quiet zone
  quiet_zone <- paste(c(1:(10)*0),collapse="")
  ## paste together in order: quiet zone, start code binary, binary label, checksum character
  ## stop code, and quiet zone
  binary_label <- paste(quiet_zone, 
                        Barcodes$Barcode[Barcodes$ASCII == start_code],
                        paste(Binary_code, collapse=""),
                        Barcodes$Barcode[Barcodes$ASCII == check_character + 32],
                        "1100011101011",
                        quiet_zone,
                        collapse = "", sep ="")
  ## split binary apart for 
  bar_values <- as.numeric(unlist(strsplit(binary_label, split = "")))
  barcode_bars <- grid::rasterGrob(t(!as.matrix(bar_values)), width = 1, height = 1, interpolate = F)
  return(barcode_bars)
}

[ColauttiLab]

Thanks for the detective work! I can confirm that I was able to re-create the problem using your codes -- a barcode is generated but won't scan for certain codes. I agree it seems to be the checksum character.

I guess one workaround might be to use (e.g. 98 = shift A) if Checksum > 107? I think this would prevent any checksum ASCII characters that are not in the binary table.

[thokall]

Hi I rewrote and use the row number instead of matching in the ASCII column. A limited test suggest that it works, but I will test more extensively and do a pull request if you find it useful. The only change is in the code_make_128 function.

binary_label <- paste(quiet_zone, 
                           Barcodes$Barcode[Barcodes$ASCII == start_code], 
                            paste(Binary_code, collapse = ""), 
                            Barcodes$Barcode[check_character +1], 
                            "1100011101011", 
                            quiet_zone, 
                            collapse = "", 
                            sep = "")

I am not sure I have the necessary understanding to make full use of all three 128 versions, even though this should be the correct way implementing a solution

[thokall]

Since the modulo 103 is never larger than 102 it should be enough to test up to checksum 102 (eg row 1-103)

[yihanwu]

Hi @thokall Would you mind making this a pull request? It's great you were able to troubleshoot this!

@ColauttiLab Let's avoid switching 128 versions. I think code A cannot encode lower case letters.