ESICUP nesting test: swim

Source: ESICUP data sets, github.com/ESICUP/datasets (CC0). Real data set from the textile industry, coordinates (rounded to the nearest integer) stated in Oliveira J. F., Gomes A. M. and Ferreira S., 2000, "A new constructive algorithm for nesting problems", OR Spectrum 22 (2), 263-284.
The original file is in esicup/.

The test
  Strip width:      5752
  Sheet length:     50000 (ESICUP's board, long enough for every part)
  Parts:            48, from 10 different shapes
  Total part area:  25441305
  Rotations:        0, 180 degrees only
  Mirroring:        none
  Spacing and kerf: 0
  Units:            ESICUP's numbers as they are; the DXF files say millimetres, so a program reads them 1:1

What is in this archive
  parts/            one DXF per shape; parts.csv says how many of each
  parts.csv         file, quantity, width, height, corners and area of each shape
  setup.json        the same facts for a program to read
  swim.velmarjob  the test as a Velmar Cut job, with the sheet and rules already set
  swim-shapes.svg a picture of the shapes

How to run it
  1. Import the parts at 1:1 with the quantities in parts.csv. In Velmar Cut, open swim.velmarjob instead.
     The job opens at the Fast level, and the results on our benchmarks page were not made at that level's settings,
     so a figure from the job as it opens will be different from them.
  2. Set a sheet or strip of the width above, long enough for every part.
  3. Allow only the rotations above. No mirroring, no spacing, no kerf.
  4. Run for the time you choose, and note the length of strip the layout uses.
  5. Density = total part area / (strip width x used length).
