Hi Simon, Yes, it seems so. Because column width is based on bytes, it may happen that e.g. a 2-byte character is split in half which produces a non-recognizable or control character (represented in the terminal as ???).
The solution would be to check the column width based on text length but respecting file encoding. This may not be trivial, since (AFAIK) sqlite3 is not informed of the file encoding at import time. Thanks, -- adam > On 26 Feb 2015, at 12:24, Simon Slavin <slavins at bigfraud.org> wrote: > > > On 26 Feb 2015, at 10:58am, Adam Podstawczy?ski <adam at podstawczynski.com> > wrote: > >> Thank you Ryan. I will provide the files if still necessary, but I have just >> discovered it has to do with column width. > > This suggests that the column width setting is being checked against the > number of bytes in the text value rather than the width of the text value > when being printed out. Thanks for figuring it out. > > Simon. > _______________________________________________ > sqlite-users mailing list > sqlite-users at mailinglists.sqlite.org > http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
