Skip to content

Latest commit

 

History

History
358 lines (240 loc) · 11.5 KB

version_5.7_updates.rst

File metadata and controls

358 lines (240 loc) · 11.5 KB

Version 5.7 Updates

Version 5.7.5

  • Externally released on 2020-01-14
  • Became metview/new at ECMWF on 2020-01-14 (Linux desktops, ecgate, lxc, lxop)
  • At ECMWF:
    • Installed 2020-01-14
    • Built with Magics 4.2.4
    • Built with ecCodes 2.16.0
    • Built with ODB_API version 0.19.2
    • Built with emoslib 000462
    • Includes version 1.3.3 of the Python interface

Fixes:

  • fixed issue where regridding of previously retrieved spectral U/V fields was not taking into account the latitude weighting
  • fixed issue where the BUFR filtering in the :ref:`Bufr Picker <bufr_picker_icon>` icon and the Bufr Examiner did not correctly handle the case where the coordinate value was missing
  • fixed issue where :ref:`FLEXPART Prepare <flexpart_prepare_icon>` could fail with a "missing level" error
  • fixed issue where the :ref:`ml_to_hl <macro_fieldset_fn>` function failed when run from Python
  • fixed issue where netCDF data with values encoded as unsigned short (ushort) could not be read
  • fixed issue where the :ref:`Contouring <mcont_icon>` icon did not allow the parameter Contour Reference Level to be set if parameters Contour or Contour Highlight were "Off"
  • fixed issue in the :ref:`Table Reader <read_table_icon>` where Table Column Types only worked with uppercase "STRING"; now it works with any case
  • improved the inline editor tips for the :ref:`Table Reader <read_table_icon>`
  • fixed issue running the "metview -p /path/to/file" command to plot a file directly from the command line
  • fixed issue where a first-time startup of Metview on MacOS 10.15 would fail to create the default user directory

Version 5.7.4

  • Externally released on 2019-12-06
  • Became metview/new at ECMWF on 2019-12-09   (Linux desktops, ecgate, lxc, lxop)
  • At ECMWF:
    • Installed 2019-12-06
    • Built with Magics 4.2.3
    • Built with ecCodes 2.15.0
    • Built with ODB_API version 0.19.2
    • Built with emoslib 000462
    • Includes version 1.3.2 of the Python interface

Fixes:

  • fixed issue where the :ref:`Percentile <percentile_icon>` module did not correctly handle missing values in the input data. Added a new parameter:
    • compute_if_missing: 'on'  (this is the default)
      • for each grid point, the percentiles are computed using only the non-missing values
    • compute_if_missing: 'off'
      • for each grid point, if there are any missing values, the result will be a missing value
  • fixed issue where the main user interface crashed if a Macro Parameters icon is run more than once
  • fixed issue in :ref:`FLEXPART Run <flexpart_run_icon>` module when dealing with high resolution grids
  • fixed issue where the FLEXPART trajectory output file is not correctly parsed
  • added macro function to convert raw trajectory output to Metview CSV format. The function is as follows:
    • flexpart_convert_trajectory(raw_tr_file, outprefix)
    • The ASCII file generated by FLEXPART (raw_tr_file) is split according to releases and converted into a CSV format that is better suited to Metview.
    • The output filenames are as follows: outprefix_rRRR.csv where RRR is the release number (starts from one) with leading zeros.

Version 5.7.3

  • Externally released on 2019-11-28
  • Became metview/new at ECMWF on 2019-11-28   (Linux desktops, ecgate, lxc, lxop)
  • At ECMWF:
    • Installed 2019-11-28
    • Built with Magics 4.2.3
    • Built with ecCodes 2.15.0
    • Built with ODB_API version 0.19.2
    • Built with emoslib 000462
    • Includes version 1.3.0 of the Python interface

Fixes:

  • fixed issue where NetCDF variables encoded with signed byte arrays were not properly handled
  • fixed issue with the title generated from plots of FLEXPART data when the time steps are smaller than one hour
  • fixed issue with the RPMs not finding the correct libraries
  • implemented small change in Magics 4.2.3 that could help overcome a rare problem where text appeared upside down in the interactive plot window
    • if you experience this problem, please ensure that you use Magics 4.2.3, and set the environment variable MV_FORCE_TEXT_PEN=1 before starting Metview, and please give us your feedback!

Version 5.7.2

  • Externally released on 2019-11-06 (get the 5.7.2.1 tarball to avoid an unnecessary message being printed when importing Metview's Python module)
  • Became metview/new at ECMWF on 2019-11-06 (Linux desktops, ecgate, lxc, lxop)
  • At ECMWF:
    • Installed 2019-11-06
    • Built with Magics 4.2.0
    • Built with ecCodes 2.14.0
    • Built with ODB_API version 0.19.1
    • Built with emoslib 000462
    • Includes version 1.3.0 of the Python interface

Fixes:

  • fixed issue where the title of a :ref:`cross section <mcross_sect_icon>` could show the wrong parameter
  • fixed issue where the function surrounding_points_indexes() gave a wrong result when the longitude of the target point was 360 and the grid was reduced Gaussian
  • fixed issue where the Thermo Data module could write null time values into its result if the input was BUFR
  • fixed issue where the :ref:`Observation Filter <obsfilter_icon>` did not correctly handle BUFR data with empty station IDs
  • fixed issue where Metview's MARS client module generated suplerfluous messages when run

Version 5.7.1

This was just a rebuild of Metview on ECMWF's HPC with plotting enabled.

Version 5.7.0

  • Externally released on 2019-10-04
  • Became metview/new at ECMWF on 2019-10-04 (Linux desktops, ecgate, lxc, lxop)
  • At ECMWF:
    • Installed 2019-10-04
    • Built with Magics 4.2.0
    • Built with ecCodes 2.14.0
    • Built with ODB_API version 0.19.1
    • Built with emoslib 000462
    • Includes version 1.3.0 of the Python interface

Desktop:

Macro/Python:

  • it is now possible to view the online documentation in a web browser for all the functions/icons listed in the Code Editor sidebar:

    /_static/release/version_5.7_updates/image1.png

    A detailed description about the usage of the the Code Editor sidebar was made vailable at: Integrated help in the Code Editor

  • :ref:`geopoints functions <macro_geopoints_fn>` create_geo() and set_dates() now accept date variables in Macro or datetime variables in Python

  • geopoints can now have missing values for latitudes and longitudes, and they will be ignored by most operations

  • new :ref:`geopoints function <macro_geopoints_fn>`: mask()

  • new fieldsets function: ml_to_hl(...) - interpolates fieldsets from model levels to height levels above ground or sea level

  • new function: percentile(vector)

    • p = percentile(vdata, |2, 99, 60|) # compute 2nd, 99th and 60th percentiles of vdat
  • the grib_get_xxx() functions no longer print warning messages if the requested key does not exist (they return nil/None, which can be checked)

  • fixed issue where the nearest_gridpoint() function did not produce correct results with certain reduced Gaussian grids on sub-areas

Plotting:

  • added new option called **Map Layer Mode **to the :ref:`Coastlines <mcoast_icon>` icon to control how the map is rendered into the foreground or/and background of the plot. The possible values are as follows: split, foreground, background.
  • added new context menu option to the icons in the Layer sidebar, "Save as" to save a copy of visdef icons modified directly in the plot window
  • fixed issue when the polar_north and polar_south projections resulted in empty plots when specified via certain corner coordinates.

MARS:

  • set the environment variable METVIEW_MARS_LOG=1 before starting Metview in order to obtain useful logging information from the MARS client without setting -slog

BUFR:

  • the BUFR Examiner contains a new tab called "Descriptors" showing all the expanded descriptors (only metadata) of the given message.

Spectra:

Percentile:

  • the :ref:`Percentile <percentile_icon>` module (that acts on GRIB data) previously had a restriction that the number of percentiles computed could not exceed the number of fields in the GRIB data. This restriction has now been lifted.

Potential Temperature: