- Work in progress; any comment/question/etc. is welcome
- yaml or xml will probably be more than an option
- Easy to read/write by human/machine plain text
- Parsimonious specifications
- If any, covariates included
- If required, easy to extend
A line can consist of:
- space-separated numbers: coordinates in each dimension;
- a word and a number/word: covariate name and value;
- a single word: new partition of coordinates;
- nothing or something else: ignored.
Also:
- Only coordinates are mandatory;
- Covariates can be grouped on top of files;
- A single partition does not need to be introduced, yet
ldk
,cur
andout
can help.
An outline:
name brahma
type beer
out
37 561
40 540
40 529
...
33 624
34 603
35 593
An open outline with more covariates/cofactors:
var Aglan
domes cult
view VD
ind O10
mass 1.123
cur
-0.5000 0.00000
-0.4967 0.01035
-0.4935 0.02414
...
0.4928 0.02396
0.4964 0.01026
0.5000 0.00000
Landmarks and 2 partitions of semi-landmarks:
id 571
taxa T. mono
ldk
697 977
766 991
704 1046
629 1008
...
731 926
541 962
835 911
cur
541 962
542 965
543 967
...
591 1070
592 1073
593 1075
cur
541 949
542 952
542 954
...
591 1070
592 1073
593 1075
A shape with nothing else but coordinates:
200 91
187 95
173 105
...
242 105
228 95
214 91
Examples adapted from Momocs: bot[1]
, olea[1]
, charring[1]
, shapes[1]
.
.coo
is intended to be a exchange file format.- R functions for conversions to/from/via
.coo
are being developped as a standalone package - Momocs will use it.