search

nickmckay / LiPDverse

Creating interactive dashboards for LiPD files, datasets and libraries
GNU General Public License v3.0
3 stars 1 forks source link

MS09DTNA does not display in/on LiPDverse #35

Open lukasjonkers opened 5 years ago

lukasjonkers commented 5 years ago

but LiPD file has the data... I also updated the LiPD file to correct the measurement material. All paleomeasurement data (two tabs in google sheet) have been merged into a single paleomeasurement in LiPD. Not clear why, but this means that there is more than a single age column in the LiPD file, which I guess is the reason why it doesn't display. Columns 10-13 are the old ocean2k data. I don't know how to separate them from the new iso2k data. Can you do this please? (or if not, delete those data, as they are already in the o2k database)

nickmckay commented 5 years ago

Wow. This dataset got really screwed up. I think I fixed it, but we'll need to double check this.

lukasjonkers commented 5 years ago

OK, paleo 0 measurement appears to have almost the correct data: column 3 (inferred d18Osw from inflata) is the same as column 0 and can be removed column 5 is the temperature from MgCa from G. inflata, can you add this? column 9 is the temperature from MgCa from G. bulloides, can you add this?

the file appears to have many redundant publications

nickmckay commented 5 years ago

I can't tell if this has been fixed. Can you fix in R or lipdverse and send to me?

lukasjonkers commented 5 years ago

MS09DTNA.RData.zip

OK, here's a list of all the TS from that site. I removed the redundant column and sorted the material, primary TS etc. Let me know if this works. I couldn't do it in playground since the remove column still only removes the last (rightmost) column.

nickmckay commented 5 years ago

Does this look correct now?

lukasjonkers commented 5 years ago

it's almost there. All the data is there, no duplications. However it seems you have only updated the metadata for the d18O time series. Some of the QC comments have also ended up with the wrong time series. Would it help to have an updated LiPD file, or can you work with the R data I sent earlier?

Also is TSid in the serialisation?

nickmckay commented 5 years ago

I tried to use the Rdata - but due to a quirk of the R lipd utilities, it's a little difficult to do it this way, which is likely where the mixup came in. TSid should be in the serialization ("paleoData_TSid" in TS). Can you just copy/paste metadata for the columns that need it? Sorry.

Also, for the futture, I think that deleting columns in lipd.net does work, just not the way you were trying. If you pick the column in the metadata (not the table where you see the column). See screenshot below. If this doesn't work, we should figure out why.

image

lukasjonkers commented 5 years ago

Alright. I'll send you an updated LiPD file soon. I see where it went wrong with the delete/remove column. The 'delete' as you showed works fine indeed. The 'remove' (in the spreadsheet, see below) only removes the last. Seems I've been using the wrong button. Could have saved you so much trouble... screen shot 2019-02-04 at 18 40 41

lukasjonkers commented 5 years ago

almost there. Upon validation I get the following error: Invalid data type: pub1.DOI Expected: array, Given: object. What does that mean, how do I correct?

lukasjonkers commented 5 years ago

MS09DTNA.lpd.zip still contains the error above, but I want to go home... sorry

nickmckay commented 5 years ago

Interesting. Probably not a big deal though. Thanks!

On Mon, Feb 4, 2019 at 11:00 AM Lukas notifications@github.com wrote:

MS09DTNA.lpd.zip https://github.com/nickmckay/LiPDverse/files/2829231/MS09DTNA.lpd.zip still contains the error above, but I want to go home... sorry

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/nickmckay/LiPDverse/issues/35#issuecomment-460348302, or mute the thread https://github.com/notifications/unsubscribe-auth/AK73Z_jTKAuCT9hey-Rejb98FwCeXV7rks5vKHVNgaJpZM4ZPgoJ .