Skip to content

Metadata parsing improvement #61

Description

@priit111

Development asar-xarray for Envisat file format support began with using GDAL. Initially it was not obvious that GDAL does not read all of the metadata required for proper processing with sarsen. However as the project continued it became necessary to parse the input dataset with self written parser. Unfortunately this means the input dataset is read twice, once by gdal and once by our own parser with negative performance implications.

I will list here discussion points:

  • Completely remove gdal, however much of the metadata handling code was done with gdal in mind.
  • Improve the current dual reading approach by reading only parts of the input file. However it must be clarified that the file I/O is not the bottleneck in processing chain.
  • WSS mode not supported by gdal.
  • Potentially try to make changes in GDAL itself, however it is completely unknown how fast these could be done and how open to contributions is the GDAL project.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions