Repository navigation
Fix issue #11486 - #11487
Fix issue #11486#11487
Conversation
It is possible in certain cases for a commentary card (HISTORY or COMMENT) to be stored in a single Card object with a value too long to fit in a single physical FITS card. In these cases the logical card's "image" (its physical representation) is multiple cards concentated together, and neither the _itersubcards() nor _split() methods handled this case properly. For some reason these code paths were previously avoided for these long commentary cards, but after astropy#11108 tried to reuse _split() in more places it did not handle this case properly.
|
Now to figure out how to generate a changelog entry... |
|
Will there be a v4.2.2 with this fix? |
|
Well... We just got done with 4.2.1... I am not sure, maybe @eteq have an idea when the next one will get out. In the meantime, are you able to do a workaround, like breaking up your very long HISTORY entry into multiple lines? I had done that myself in my work in the past, where I break it up but indent the subsequent lines, so I know they belong to the same entry. For future reference, would be nice if you test against |
|
Was it completely fixed? Use the 6.0.1 version of astropy and have the same error |
|
@sbelkin01 , please open new issue with minimum reproducible example and the affected header info. Thanks! |
5 participants
Fixes #11486.
As usual, FITS card parsing/verification is a mess that should be cleaned up thoroughly, but this provides a quick fix to the regression.