Import contacts from CSV into addrbook fails: Column mismat

[pstover]

I tested the import of contacts from a CSV file into TBs address book.

Therefore I created a super simple CSV test file with the following content:

Last Name,Display Name,Primary Email,Custom 1,Notes
mylastname,mydisplayname,myemail@foobar.com,mycustom,mynotes

Now I started the import process into TB.

When I came to the step after csv file selection the import Address Book dialog pops up (see the following snapshot:

http://www.picfront.org/d/9AUV?language=en

Much to my surprise Thunderbird does NOT recognize the first line as a column description although I ticked the checkbox "First record contains field names".

Why?

I tried the CSV file in UTF-8 and in ANSI encoding (does this matter?). But the non-detection of field names persist.

What else can I do to let TB recognize successfully the field names?

Is this a bug?

Peter

[mgagnonlv]

There is indeed a column mismatch, so you need to customize the field list. Right now, only some fields are matched and Thunderbird complains when it tries to import a display name that contains an "@" (it is matched with Primary email) and a primary email that probably contains strange characters (notes contains dots, semicolons, accents, carriage returns, etc.)

Assuming the file you describe and the image attached, I say you should do the following :

1. In the leftmost column of the Import window, uncheck everything except the following (i.e. uncheck all fields you don't have. Keep only the following checked :
   – last name
   – display name
   – primary email
   – custom 1
   – notes
2. Click on the checked "Address book fields" and move them up so they align to the "Record data to import" in the right column. Your screen should now look like (the "x" corresponds to a checked column) :

   Code: Select all

   x  Last Name      Last Name
   x  Display Name   Display Name
   x  Primary Email  Primary Email
   x  Custom 1       Custom 1
   x  Notes          Notes
      First Name
      Nickname
      Secondary email
      Screen Name
   etc.

3. As far as encoding is concerned, you'll see a difference if you have names in languages other than English (i.e. with accents) and notes with accents or other strange characters. One easy way to check is to click on next a few times to see records 1, 2, 3, 4, etc. So if you see something strange, try to import with "system encoding" instead of UTF-8 encoding (or vice versa).
4. When you have checked everything, click on OK. The import will proceed.

[pstover]

Hello Michel,

thank you for the reply.
However it is disappointing that this tedious workaround is necessary.
Even worse: One could say that the column-match feature is a bug.
I cannot imgine how this mess could pass a verification test by the Thunderbird team.

How can TB claim that it works if even the most simple constellation does not work?

[DanRaisch]

A) Because that (importing new address data from outside of Thunderbird) is probably not a very high frequency activity after initial conversion from a previous client.
B) Because the simple remedy to the column mismatch is to start by exporting data from Thunderbird's address book so one can see and duplicate the column layout before creating or using a CSV file for import.