DM-18203: Set encoding for reading text catalogs. #158
Conversation
tests/test_readTextCatalog.py
Outdated
| config = ReadTextCatalogTask.ConfigClass() | ||
| config.encoding = "ascii" | ||
| task = ReadTextCatalogTask(config=config) | ||
| # This will generate a ResourceWarning due to a NumPy bug. |
There was a problem hiding this comment.
If you want to make the test output clean consider using a warnings.catch_warnings with a simplefilter to disable warning output for that command.
There was a problem hiding this comment.
I don't think you can, at least not trivially — I tried this, but the warning isn't generated in this code, but deep in the guts of Python: it comes from /Users/jds/Projects/LSST/stack/python/miniconda3-4.5.4/envs/lsst-scipipe-fcd27eb/lib/python3.6/traceback.py:216 sometime after this test case has finished, so a context in the test can't catch it.
I considered putting in a filter to drop all ResourceWarnings from traceback.py:216, but that seems less than optimal. Other suggestions welcome!
There was a problem hiding this comment.
Maybe it's fixed in the new numpy that we will be using from tomorrow...
There was a problem hiding this comment.
Pretty sure it's still broken in the latest version. :(
There was a problem hiding this comment.
Thanks for filing a numpy ticket on this and linking it here. However, should we have a jira ticket to remove this comment once numpy is fixed? It might be confusing in "future numpy is fixed" land...
jdswinbank
force-pushed
the
tickets/DM-18203
branch
2 times, most recently
from
df2264f
to
2c19f89
Compare
| encoding = pexConfig.Field( | ||
| dtype=str, | ||
| default="utf-8", | ||
| doc="Encoding of text reference file." |
There was a problem hiding this comment.
Do we really need to make this configurable? Can't we just make it utf-8 forever and forget about it?
tests/test_readTextCatalog.py
Outdated
| config = ReadTextCatalogTask.ConfigClass() | ||
| config.encoding = "ascii" | ||
| task = ReadTextCatalogTask(config=config) | ||
| # This will generate a ResourceWarning due to a NumPy bug. |
There was a problem hiding this comment.
Thanks for filing a numpy ticket on this and linking it here. However, should we have a jira ticket to remove this comment once numpy is fixed? It might be confusing in "future numpy is fixed" land...
|
Your new commit is fine. Squash it. |
Commas are needed to separate fields in the header.
jdswinbank
force-pushed
the
tickets/DM-18203
branch
from
0f72a82
to
ea4250c
Compare
No description provided.