readland.tps only converts one set of negative landmark coords to NA

[bkaraceylan]

Hello!

I have this weird issue with geomorph 3.2.0. When using readland.tps with negNA = True, if the specimen has more than one missing landmark, the function only converts the first set of missing coords the NA.

Example tpsdata:

LM=6
2854.00000 2968.00000
-1.00000 -1.00000
-1.00000 -1.00000
2400.00000 2854.00000
2267.00000 1937.00000
2968.00000 1091.00000
IMAGE=ANT_30.jpg
ID=ANT_30
SCALE=0.000935

After importing with readland.tps:

         [,1]      [,2]
[1,]  2.668490  2.775080
[2,]        NA        NA
[3,] -0.000935 -0.000935
[4,]  2.244000  2.668490
[5,]  2.119645  1.811095
[6,]  2.775080  1.020085

[antkal]

This was a small bug in readland.tps when a single specimen with NAs was present. I´ve fixed this in the develop branch of github, could you please reinstall from there via devtools::install_github("geomorphR/geomorph", ref = "Develop", build_vignettes = TRUE) and confirm it does the trick for your dataset, please? Thanks for catching this! Antigoni

…

On 14/01/2020 20:40, bkaraceylan wrote: Hello! I have this weird issue with geomorph 3.2.0. When using readland.tps with negNA = True, if the specimen has more than one missing landmark, the function only converts the first set of missing coords the NA. Example tpsdata: LM=6 2854.00000 2968.00000 -1.00000 -1.00000 -1.00000 -1.00000 2400.00000 2854.00000 2267.00000 1937.00000 2968.00000 1091.00000 IMAGE=ANT_30.jpg ID=ANT_30 SCALE=0.000935 After importing with readland.tps: |[,1] [,2] [1,] 2.668490 2.775080 [2,] NA NA [3,] -0.000935 -0.000935 [4,] 2.244000 2.668490 [5,] 2.119645 1.811095 [6,] 2.775080 1.020085 | — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#26?email_source=notifications&email_token=AEVLB7MKSQPZ2D3N52IKCGLQ5YPKPA5CNFSM4KGZHC72YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IGFLZGQ>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEVLB7JRYIV5EAOJQVFUJQDQ5YPKPANCNFSM4KGZHC7Q>.Web Bug from https://github.com/notifications/beacon/AEVLB7JLCASOFJOXUWOZ7O3Q5YPKPA5CNFSM4KGZHC72YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IGFLZGQ.gif [ { ***@***.***": "http://schema.org", ***@***.***": "EmailMessage", "potentialAction": { ***@***.***": "ViewAction", "target": "#26?email_source=notifications\u0026email_token=AEVLB7MKSQPZ2D3N52IKCGLQ5YPKPA5CNFSM4KGZHC72YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IGFLZGQ","url": "#26?email_source=notifications\u0026email_token=AEVLB7MKSQPZ2D3N52IKCGLQ5YPKPA5CNFSM4KGZHC72YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IGFLZGQ", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { ***@***.***": "Organization", "name": "GitHub", "url": "https://github.com" } } ]

-- Antigoni Kaliontzopoulou Auxiliary Researcher CIBIO/InBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos University of Porto Campus Agrário de Vairão, 4485-661 Vairão PORTUGAL https://sites.google.com/view/akaliontzopoulou/home https://cibio.up.pt/research-groups-1/details/phenevol http://cibio.up.pt/people/details/akaliont http://www.seh-herpetology.org/

[bkaraceylan]

In fact my tps file contains 30 specimens, not a single one. I've only posted one of the specimens with missing landmarks to here. 3 out of 30 specimens has 2 missing landmarks and only the first missing landmark in these specimens are converted to NA's.

Development branch returns an error for the same tps file:

test <- readland.tps(file = "ANT.TPS", specID = "ID", negNA = TRUE)
Error in lm.nas[[which(ind.nas == i)]] : subscript out of bounds