Synch API and CSV format

[Xavier_G_Sabate]

Hi all, 

I have been playing with the SYNCH API for a while with excellent results. I found this capability very useful for creating and updating records. So far so good.

My next challenge is to fill a table which one of its fields can hold a text coming from a legacy system. The problem I'm facing is that those texts include "," (commas) since they are descriptions and side information related to every record.

I would like to keep the original format, otherwise it can be difficult to understand and read, however, when I try to use the synch API to move the data, I just crash with the CSV definition. I have not found a way to scape the commas in the string that is passed to the synch API, so it cannot process the request for those records.

Has anybody faced the same situation? any hint? workarounds?... any help is welcome 🙂

[ScottWorld]

It should work if you put quotation marks around the strings that contains commas.

Otherwise, you can switch to a different field delimiter (such as a semicolon), and use Make’s CSV tools to sync your data with Airtable.

i show how to do this on this episode of the BuiltOnAir podcast.

[Xavier_G_Sabate]

Hi Scott,

Thank you for your quick help. I would say that I have already tried this workaround (is the first solution that came to my mind) but didn't work.
However I will give it a second try again, maybe it was my fault when coding... 

[ScottWorld]

If you’ve already tried quotation marks and it fails, it might be a bug in Airtable’s CSV sync API, so you may need to report this to them as a bug.

[Xavier_G_Sabate]

Definitely there is something wrong with the API. I tried again both using quotation marks and double quotation marks to wrap the original text. The API call returns a success when sending the request, however the table is not synchronized.
In the tab name of the target table I can see the "danger" icon and hovering the mouse on top,  this message is displayed: "Column names must be non empty strings" which makes no sense unless the process is messing with the commas.

As a workaround I will substitute the commas in the source by pipes "|" and then revert the change using a formula field that I will use to display the original data. Not ideal but will do the job until this pipeline works better...

[ScottWorld]

Ugh, that’s very sad to hear. I would definitely alert Airtable to this major bug.

Also, did you consider going the Make CSV route that I mentioned above? You have tons more power & control & customizability & flexibility by building your own custom CSV sync.