Open m-hope opened 5 years ago
temi
commented
5 years ago @m-hope This appears to be cause by this issue - https://github.com/AtlasOfLivingAustralia/bie-index/issues/223. BIE test seems to be working as expected. I am hoping next time @charvolant deploys to prod, this issue will be fixed. Robina encountered the same issue when testing https://github.com/AtlasOfLivingAustralia/profile-hub/issues/661.
m-hope
commented
5 years ago Thanks @temi
m-hope
commented
5 years ago Thanks @temi ... There is also and issue of searching on synonyms within the FOA and not getting profiles that have been manually matched to the currently accepted name for that synonym, but we should probably investigate this further once the BIE has been corrected.
In the meantime, I'm happy for this issue to be put on the Backlog awaiting future testing once BIE is fixed.
m-hope
commented
5 years ago I have noticed that the matching of a name for a new profile (ie: when a match is provided to a name entered in the top field of the 'matching' dialog) can sometimes give a 'different' result to when the lower fields (ie: using the "Match to a different name") even though the same name is used.
Eg: In a collection, create a new profile and enter "Poaceae"... it should match correctly to Poaceae Barnhart. If you click on "Create Profile", the new profile correctly displays the breadcrumb, map and images. However, if instead you click on "Match to different name:" and then enter "Poaceae" (and select "Poaceae" in the autocomplete list, the new profile, the breadcrumb is incorrect (places it in Plantae), there are dots on the map and no images.
In looking at the logs, it appears that the search for the name in biocache is returning the NZOR version of the name, which is a synonym of the APC version of the name and will have a different taxonomic identifier... This seems to indicate that Temi is correct that the old ALA name-index is faulty.
Note: I have tried applying the latest name index to aws-profiles, but it was incompatible.
If a profile is created that is not a known name (eg: Acacia notknown), there are a number of names that don't seem to bring across the correct information (eg: the taxonomic hierarchy visible in the breadcrumb). For these species names, the profile is placed in Plantae. However if a profile is created specifically for this species name (ie: the user puts the name in the name of the profile), the hierarchy is imported (and the breadcrumb is displayed) correctly.
This seems to occur to 'random' species names, however it most often (but not always) occurs when the authorship is complex (ie: more than one name), but not all species names with 'complex' names are affected.
A list of affected names (ie: not rematched properly) include: Acacia anceps DC. Acacia aneura F.Muell. ex Benth. Acacia leptophleba F.Muell. ex Benth. Acacia prominens A.Cunn. ex G.Don Acacia saligna (Labill.) H.L.Wendl. Callistemon citrinus (Curtis) Skeels Callistemon pityoides F.Muell. Eucalyptus dumosa A.Cunn. ex J.Oxley Eucalyptus flocktoniae (Maiden) Maiden
A list of species names that rematch correctly (ie: the breadcrumb displays okay) include: Acacia amanda G.J.Leach Acacia amblygona A.Cunn. ex Benth. Acacia ommatosperma (Pedley) Pedley Acacia spania Pedley Acacia stenophylla A.Cunn. ex Benth. Acacia verticillata (L'Hér.) Willd. Acacia victoriae Benth. Callistemon flavovirens (Cheel) Cheel Callistemon hemistictus (S.T.Blake ex Craven) Udovicic & R.D.Spencer Callistemon montanus C.T.White ex S.T.Blake Eucalyptus albopurpurea (Boomsma) D.Nicolle Eucalyptus deglupta Blume Eucalyptus alba Reinw. ex Blume Eucalyptus chloroclada (Blakely) L.A.S.Johnson & K.D.Hill Eucalyptus ecostata (Maiden) D.Nicolle & M.E.French Eucalyptus falciformis (null) Rule Eucalyptus globulus Labill.