# nopytest_test_coadds.py throws warnings, and should be fixed

XMLWordPrintable

## Details

• Type: Story
• Status: Done
• Resolution: Done
• Fix Version/s: None
• Component/s:
• Labels:
None
• Templates:
• Story Points:
1
• Team:
Data Release Production

## Description

nopytest_test_coadds.py throws the following warnings in many places:

 lsst.afw.image WARN: Could not parse options; writing with defaults:  File "src/PropertySet.cc", line 189, in T lsst::daf::base::PropertySet::get(const string&) const [with T = std::shared_ptr; std::string = std::basic_string]  image not found {0} lsst::pex::exceptions::NotFoundError: 'image not found' 

This seems to be coming from:
https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/posixStorage.py#L642
which calls and executes to:
https://github.com/lsst/afw/blob/a4679bc1e1d40d112e1484359a353fee477c1331/python/lsst/afw/image/image/fitsIoWithOptions.py#L97

Where it expects a property of the property set to be called image. However, the options that are passed through the call chain are:

['visit', 'ccd']

This optional prameter is generated:
https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/posixStorage.py#L637
and that variable comes from:
https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/posixStorage.py#L249
which in turn comes from the butler here:
https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/butler.py#L1421
the location variable is created here:
https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/butler.py#L1297
That optional data ultimately is initialized here:
https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/butlerLocation.py#L207
So whatever is creating butler locations (Some repository or mapper) is not adding the correct fields, or afw should be looking for different things to write.

These are just notes for what I have found so far, to help anyone who gets to this ticket. I am tagging Paul Price because I think he was the last one to work with the additionalData code, and Jim Bosch because he may be familiar with this work, or how it impacts gen3 middleware and will know if this is even worth pursuing. Also tagging Yusra AlSayyad, as I am not sure if I put this in the right epic.

## People

• Assignee:
Jim Bosch
Reporter:
Nate Lust
Reviewers:
John Parejko
Watchers:
Jim Bosch, John Parejko, Nate Lust, Yusra AlSayyad