Fix issue with forward slash in taxon name e.g. Fistulifera / Mayamaea

[fozy81]

No description provided.

[AugustT]

Hi Tim,

Good spot on this one. While your fix will work it seems to me that this is an issue on the NBN side. If there is an entry on the NBN for "Fistulifera / Mayamaea" (which there is) we should expect that a search for "Fistulifera / Mayamaea" should find it. It doesn't, either in rNBN or the search functionality on the gateway. I'm going to contact their support team to see if they can fix this at their end as that is preferable to your fix.

On another note, as you are the rNBN user I speak to the most it would be interesting to know what application you are using it for, just for my own interest.

Tom

[AugustT]

In fact it looks like a better fix would be:

q <- tolower(gsub('%20/%20','%20', q))

As this should result in only the genus you are interested being returned (rather than a list of 7)

[fozy81]

I've updated change to <- tolower(gsub('%20/%20','%20', q))

[fozy81]

Just on your first comment, I'm helping to update the taxa lists that SEPA hold internally (plants, inverts, algae etc)...which have incomplete NBN codes associated with them. This is working to a degree. But it also highlights some issues:

• Some of our taxa just aren't on the NBN - usually artificial groups or newly described species - so I guess we need to request these if possible from NBN - this process is a little opaque not that I have really looked into it. Some sort of Github for NBN taxon name requests...?
• Some taxon names have mulitple NBN codes - which is understandable but needs some thought how we handle that
• SEPA holds values against taxa names to indicate their tolerance to environmental impacts when the taxonomy changes, we sometimes need to think carefully about how these values need to be re-mapped.
• SEPA also manage other codes: maitland codes, whitton codes, AQEM etc - again taxonomic changes etc affect mapping of these too
• A side project looked at scraping Scottish Biodiversity Species records to look at distribution - I think SNH & SEweb were interested and thought I would try to demo. Not sure where this will go.
• Ultimately, we want NBN codes for all SEPA taxa and then we can share all our data quickly with NBN - the idea is to automate this - hoping to meet with NBN soon to discuss some of this (NBN have been in touch about this). Maybe POST api to NBN gateway account...I think iSpot can do that automatically to some extent. Or we publish linked data and NBN scrape that.

Anyway, rnbn very handy - thanks!

Tim

[AugustT]

Great stuff! Taxonomy can be a real headache.

I think POST functionality would be great, and we could integrate that into rNBN but there are issues over quality control. I know there is post functionality to iRecord, and data from there (once verified) goes to NBN so that is well worth a look into.

We (at the Biological Records Centre) do a lot of work with distribution maps (MCPs, Alpha Hulls, Frescalo etc) in R, so let me know if you run into any difficulties.

Tom

[sckott]

@fozy81 @AugustT if you need to do taxonomy things, our taxize package may be of interest. Also looks like you have unique identifiers for species? If so, we may want to add a method in taxize so users could get NBN identifiers if needed.

[fozy81]

Yes, had a look at taxize. But as I was looking for NBN identifiers so used rnbn. But be great if included in taxize. Adding obligatory xkcd cartoon http://xkcd.com/927/

[sckott]

hah! True, there are a gazillion different versions of IDs for the same taxon. I'll see if it could be put into taxize ... - issue opened, there's a link below to that issue

[AugustT]

I have had word from the team at the NBN that this issue has now been fixed, and indeed I just tried it out and it seems to work:

library(rnbn)

# Note that since this is a genera we need to add species_only = FALSE
FM <- getTVKQuery(query = 'Fistulifera / Mayamaea', species_only = FALSE)

FM$searchMatchTitle

[1] "Fistulifera / Mayamaea"       
[2] "Mayamaea agrestis"            
[3] "Mayamaea atomus"              
[4] "Fistulifera pelliculosa"      
[5] "Fistulifera saprophila"       
[6] "Mayamaea lacunolaciniata"     
[7] "Mayamaea atomus var. permitis"

@fozy81 If this addresses your issues I'll close this request

[fozy81]

Thanks, great.