GUI -> Mess
===========
I just tried to use the vcf2csv [1] tool to convert my addressbook
to something simpler. The file was exported from the Ubuntu
Touch's default Contacts application and then imported into OwnCloud
Addressbook and edited (I just added some more adresses and deleted a
few older ones). Then a new VCF file was exported from the OwnCloud
thing.
When I converted it to a HTML then I first realised what mess was
inside: tons of useless fields, about 5 or 6 different files with phone
numbers and so. I was aware that the file structure will be far from
ideal but this is really terrible. Yes, it's user's fault because the
user (me) added data on several devices to different fieds (the first
contacts came from SIM card and the other were added later on the
Ubuntu phone and through the OwnCloud). But still, why these software
allow to use tons of strange fields?
When I converted such data for use on my Palm then I saw that some
things are strange. But it looks that it will be useful to clean the
data and export then once more...
At the moment I converted the VCF file to a SC format (for the sc(1)
spreadsheet program) and I'm going to study the data and make them more
consistent.
By the way, my workflow was:
-------------------------------------------
vcf2csv -i ~/default-2019-04-25.vcf -d "VERSION,X-REMOTE-ID,X-GOOGLE-ETAG,X-GROUP-ID,X-CREATED-AT,UID,N,X-EVOLUTION-FILE-AS,REV
,X-AVATAR-REV,X-URIS;TYPE=other,PHOTO;TYPE=jpeg;ENCODING=b,X-FOLKS-FAVOURITE,ROLE,TITLE" |iconv -f utf-8 -t latin2 | tee /tmp/
1.tsv
cd /tmp
psc -d " " <1.tsv |tee addresses.sc
-------------------------------------------
The vcf2csv makes Tab-separated file so the "Tab" delimiter must be
specified for the psc (the -d parameter: it's argument is not a comma
but a Tab character).
And I have learned that the character can be entered inside the
gVim by + sequence followed by the .
References:
[1] http://vcs2csv.sf.net