Skip to content

Conversation

@eoudejans
Copy link
Collaborator

Tif files openened/created for writing were opened using append mode "a", which caused consequtive writes to grow the tiffile usually by the given domain of the dataitem. This was never a problem as geodms handled temporary (.tmp) files itself and deleted these accordingly. Changing "a" to "w" makes sure that a new tif file is always created on write, which is in line with the new state: new writes to storage targets are always done to fresh files.

…ing append mode "a", which caused consequtive writes to grow the tiffile by the given domain of the dataitem. This was never a problem as geodms handled temporary (.tmp) files itself and deleted these accordingly. Changing "a" to "w" makes sure that a new tif file is always created on write, which is in line with the new state: new writes always to fresh files.
@eoudejans eoudejans added the bug label Apr 29, 2025
@eoudejans eoudejans self-assigned this Apr 29, 2025
@MaartenHilferink
Copy link
Collaborator

MaartenHilferink commented May 5, 2025

Wat als in dezelfde dan wel opvolgende sessies GridData en PaletteData worden berekend en gecommit, of eerst PaletteData en dan GridData?
Was het niet zo dat het Committen van GridData ook het committen van PaletteData impliceert?
Is dit onderdeel van unit-test-case(s)?

@eoudejans
Copy link
Collaborator Author

eoudejans commented May 6, 2025

Release unit tests all passed. So the question becomes: is palette data properly tested in the unit tests?

The best way would be to collect palette data if necessary and do the write once.

@jipclaassens jipclaassens removed the bug label Oct 20, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants