Question: 503 Backend fetch failed

[Naomivanes]

Hi!

I am trying the following command:

curl -s -H "Accept: text/csv" -H "Content-Type: text/tsv" --data-binary @boldnames.csv "https://api.checklistbank.org/dataset/201891/nameusage/match"

This gives me a 503 Backend fetch failed (Guru meditation: XID: 761826565). A few weeks ago this was working and as far as I know I didn't change the layout of the input file. This file ccontains some testdata. Here are the first few rows of boldnames.csv:

scientificName Lutzomyia cruciata Lutzomyia longipalpis Dampfomyia deleoni Brumptomyia mesai Dampfomyia beltrani Psathyromyia shannoni Bichromomyia olmeca Therevidae Sarcophaga australis

Is there someyhing that stands out that I am doing wrong?

[mdoering]

Hi @Naomivanes. The API method was just very recently created and we redesigned it to be safer to match large number of names. It is now placed somewhere else and you can only do asynchroneous match jobs for bulk matching.

Single matches can still be done with:

GET /dataset/{key}/match/nameusage?q=Lutzomyia

Bulk matches now work similar to downloads. You will get an email when they are completed. Requesting a new match job can be done in 2 ways:

a) upload a list of names like you used to do with an authenticated POST and the requested format in a query param:

curl -s --user USERNAME:PASSWORD -H "Content-Type: text/tsv" --data-binary @boldnames.csv -X POST "https://api.checklistbank.org/dataset/201891/match/nameusage/job?format=CSV"

b) match an existing list of names against another existing dataset in ChecklistBank:

curl -s --user USERNAME:PASSWORD -X POST "https://api.checklistbank.org/dataset/201891/match/nameusage/job?format=CSV&sourceDatasetKey=3351"

I guess the last one is for you if you want to match all of BOLD against OTOL or the other way around?

I am just about to deploy some of these changes, so please give me a few hours today before trying that out.

[mdoering]

This repo is for the old, GBIF implementation. For checklistbank.org please use https://github.com/CatalogueOfLife/backend

[Naomivanes]

Hi @mdoering ,

Thanks for your response. In the future I will use the correct repository. For now I made a script to map BOLD taxons to Open Tree of Life taxons and their respective ids through the Open Tree of Life API. I will look at these next week, thank you!

[jar398]

@mdoering Maybe put a link to CatalogueOfLife/backend from the checklistbank README? I would not have known about the change.

[mdoering]

Thanks, good idea!