v0.2.0

utu.rotation

Adds utu.rotation.rotate, which carries helioprojective coordinates along
the solar surface to a new time. It is
sunpy.coordinates.propagate_with_solar_surface expressed in named arrays,
and takes a time per point, so a raster may give one per step.

• Correct for array times. A multidimensional obstime silently returns
  NaN from sunpy, and get_earth raises on one, so everything is flattened
  on the way in and put back into shape afterwards.
• Fast. sunpy builds one rotation matrix per point in a Python loop
  whenever it is given an array obstime, which is hundreds of times slower
  for no gain in accuracy. This only ever hands it a scalar time, and
  interpolates between times an hour apart. A 400-step raster takes half a
  second rather than three minutes.
• Accurate over any span. How far apart those times are is what costs
  accuracy, and it costs the square of it, so the count is chosen from the
  span rather than fixed. The error stays between 0.001 and 0.010 arcsec
  from one hour to a whole disk transit.
• Off-limb points. off_disk decides whether material above the limb,
  which has no surface to be carried along, comes back as NaN or is left
  where it was seen.

v0.1.1

Fixes to the placement of the labels drawn by utu.spectrum.stem, all found by looking at the figure of the passband in the ESIS instrument article.

• A label of a faint line lay along the bottom of the plot, across the axis. The lines are given to the solver as points along their length, but they stand on the baseline without covering it, so the strip of axis between two of them held no points at all — and that is where the label of a short line starts. A row of points along the baseline says otherwise.
• A label covered the leader of its neighbour, and another overhung a line. A label here is an ion and a wavelength, several times wider than it is tall, so two of them need far more room beside one another than above; and the solver measures a label by its glyphs while what is drawn is the box around them, larger by its padding.

Both are now measured, against the boxes which are drawn rather than the glyphs inside them, and the measurement is itself checked against a case which should fail.

v0.1.0

What's Changed

• Build the documentation by @roytsmart in #2
• Upload coverage with the version of the action the other packages use by @roytsmart in #3
• Give the matrix one check to report through by @roytsmart in #4
• Add utu.spectrum, the emission lines of an optically thin plasma by @roytsmart in #1

New Contributors

• @roytsmart made their first contribution in #2

Full Changelog: https://github.com/sun-data/utu/commits/v0.1.0