Importing .vcf file with long UTF-8 name causes messy characters
Categories
(Thunderbird :: Untriaged, enhancement)
Tracking
(Not tracked)
People
(Reporter: ku, Unassigned)
References
Details
Attachments
(2 files)
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.97 Safari/537.36
Steps to reproduce:
Import .vcf file export by Windows Live Mail.
In my case, it has Japanese characters.
For reproduction purpose, I have replaced the actual name with dummy Zenkaku alphabet characters.
Actual results:
The imported person name made messy:
ＴｈｉｓＩｓＡＬBDｎｇＮａｍｅＰ85ｒｓｏｎ
Ｅｘｐ8FｒｔｅｄＢｙＬｉEFｅＭａｉｌ
Expected results:
In Windows Live Mail:
ＴｈｉｓＩｓＡＬｏｎｇＮａｍｅＰｅｒｓｏｎ
ＥｘｐｏｒｔｅｄＢｙＬｉｖｅＭａｉｌ
|
||
Comment 2•4 years ago
|
||
This relates to #1330189.
Any entry with name or address that contains non-ASCII characters is, when exported to vcf file, encoded in UTF-8, but importing it back to address book inteprets them as ISO8859-1 and become messy characters. This is quite problematic for non English users.
The bug persists in version 78.5.0 (32-bit). Moreover, I could not solve the problem, because when translating the cyrillic UTF8 into the 8859-1 encoding, the Cyrillic letters are replaced with question marks, which, by the way, is logical.
Rolling back to Thunderbird 68?
Moreover, I could not solve the problem
Hi I have attached another vcf file
ThisIsALongNamePerson ExportedByLiveMailWithoutLF.vcf.
How about this file?
This another file doesn't contain = mark at end of line which means:
A soft line break consists of an = at the end of an encoded line, and does not appear as a line break in the decoded text.
Good afternoon! In the case of your file, everything is perfectly imported: due to the Latin encoding, try adding a contact in Japanese or Russian.
Perhaps this solution, but in the form of a rake - a support. This bug must be resolved at the level of the import module code.
Контакт
Отображаемое имя: ＴｈｉｓＩｓＡＬｏｎｇＮａｍｅＰｅｒｓｏｎ ＥｘｐｏｒｔｅｄＢｙＬｉｖｅＭａｉｌ
Hi
ＴｈｉｓＩｓＡＬｏｎｇＮａｍｅＰｅｒｓｏｎ ＥｘｐｏｒｔｅｄＢｙＬｉｖｅＭａｉｌ
This is a UTF-8 encoded text:
BEGIN:VCARD
VERSION:2.1
N;CHARSET=utf-8;ENCODING=QUOTED-PRINTABLE:=EF=BC=B4=...
FN;CHARSET=utf-8;ENCODING=QUOTED-PRINTABLE:=EF=BC=B4=...
REV:20200609T113612Z
END:VCARD
=EF=BC=B4 Ｔ
=EF=BD=88 ｈ
=EF=BD=89 ｉ
=EF=BD=93 ｓ
=EF=BC=A9 Ｉ
=EF=BD=93 ｓ
=EF=BC=A1 Ａ
=EF=BC=AC Ｌ
=EF=BD=8F ｏ
=EF=BD=8E ｎ
=EF=BD=87 ｇ
=EF=BC=AE Ｎ
=EF=BD=81 ａ
=EF=BD=8D ｍ
=EF=BD=85 ｅ
=EF=BC=B0 Ｐ
=EF=BD=85 ｅ
=EF=BD=92 ｒ
=EF=BD=93 ｓ
=EF=BD=8F ｏ
=EF=BD=8E ｎ
;
=EF=BC=A5 Ｅ
=EF=BD=98 ｘ
=EF=BD=90 ｐ
=EF=BD=8F ｏ
=EF=BD=92 ｒ
=EF=BD=94 ｔ
=EF=BD=85 ｅ
=EF=BD=84 ｄ
=EF=BC=A2 Ｂ
=EF=BD=99 ｙ
=EF=BC=AC Ｌ
=EF=BD=89 ｉ
=EF=BD=96 ｖ
=EF=BD=85 ｅ
=EF=BC=AD Ｍ
=EF=BD=81 ａ
=EF=BD=89 ｉ
=EF=BD=8C ｌ
It may be problem of QuotedPrinterable decoder that reading ThisIsALongNamePerson ExportedByLiveMail.vcf contains invalid characters.
BEGIN:VCARD
VERSION:2.1
N;CHARSET=utf-8;ENCODING=QUOTED-PRINTABLE:=EF=BC=B4=EF=BD=88=EF=BD=89=EF=BD=93=EF=BC=A9=EF=BD=93=EF=BC=A1=EF=BC=AC=EF=
=BD=8F=EF=BD=8E=EF=BD=87=EF=BC=AE=EF=BD=81=EF=BD=8D=EF=BD=85=EF=BC=B0=EF=BD=
=85=EF=BD=92=EF=BD=93=EF=BD=8F=EF=BD=8E;=EF=BC=A5=EF=BD=98=EF=BD=90=EF=BD=
=8F=EF=BD=92=EF=BD=94=EF=BD=85=EF=BD=84=EF=BC=A2=EF=BD=99=EF=BC=AC=EF=BD=89=
=EF=BD=96=EF=BD=85=EF=BC=AD=EF=BD=81=EF=BD=89=EF=BD=8C
FN;CHARSET=utf-8;ENCODING=QUOTED-PRINTABLE:=EF=BC=B4=EF=BD=88=EF=BD=89=EF=BD=93=EF=BC=A9=EF=BD=93=EF=BC=A1=EF=BC=AC=EF=
=BD=8F=EF=BD=8E=EF=BD=87=EF=BC=AE=EF=BD=81=EF=BD=8D=EF=BD=85=EF=BC=B0=EF=BD=
=85=EF=BD=92=EF=BD=93=EF=BD=8F=EF=BD=8E =EF=BC=A5=EF=BD=98=EF=BD=90=EF=BD=
=8F=EF=BD=92=EF=BD=94=EF=BD=85=EF=BD=84=EF=BC=A2=EF=BD=99=EF=BC=AC=EF=BD=89=
=EF=BD=96=EF=BD=85=EF=BC=AD=EF=BD=81=EF=BD=89=EF=BD=8C
REV:20200609T113612Z
END:VCARD
As Wikipedia says, The tail = mark means that it should erase tail CR+LF or LF codes.
https://en.wikipedia.org/wiki/Quoted-printable
It is like multiline define in C language
#define DEBUG_BREAK(a)\
if ((a)) \
__asm int 3;
Description
•