Unable to retrieve values and instead getting reference pointer addresses on Oslc query csv export

[pauljosephwilson]

I am trying to pull comments and other fields that are showing up as rpXX:rtc_cm:column in my output data. I am wondering if there is a way to pull the string literals from these rather than the reference pointer addresses in the output csv. Any advice would be appreciated, thank you.

Command used oslcquery -J XXXXXXXXX -U XXXXXX -P XXXXXX -A ccm -p "XXXX" -V -s * -O output.csv --cacheable -q "dcterms:identifier= XXXX" where the identifier is a ticket that I know has comments. In the output however the internalComments field has values with reference pointers in them.

[barny]

Hi

I've never tried retrieving comments. When I do the same query as you, the CSV has an oslc:discussedBy column and it contains rp1:rtc_cm:comments - I guess that's what you're referring to. The full URL would be https://jazz.ibm.com:9443/ccm/oslc/workitems/_wTptg5d-Ee-nOc86QY6HcQ/rtc_cm:comments

The raw query result contains the URL of the comments feed for that artifact, so I guess the postprocessing which (attempts to) turn URLs into more meaningful names is somehow losing that.

You could try adding the -Q flag to disable the postprocessing - that results in the raw URL being in the oslc:discussedBy column. Then you'll have to do a separate step of retrieving the content from that URL.

If I get a chance I'll try see if I can figure out a general solution so that the URL isn't lost. What might work would be detecting if the prefix that gets generated (rp1 in my example) isn't one of the defaults then go back to the full URL - or something like that. That url->name conversion is done by resolve_uri_to_name in _ccm.py. Or maybe it's the conversion to a prefix that's problematic: to do this I have to split the url on the final / (which can't be in an xml tag) and the prefix/namespace is the first part, the tag is the second part.

But don't hold your breath waiting for me to look at this: the -Q option is probably the quickest way to get something working for you.

HTH

[barny]

Can't publish this for a while, but if you find _ccm.py at around line 280 (probably depending what version you're using) is this:
uri1 = rdfxml.uri_to_prefixed_tag(uri)
Change to this:
uri1 = rdfxml.uri_to_prefixed_tag(uri, oktocreate=False,noexception=True)
and now the comment URL won't be prefixed so should stay vanilla even without -Q