Handle special chars¶
Detect encoding¶
Text file encoding can be identified using, for example, file -b <file-path> or in vim
with :set fileencoding.
Here's an example of encoding that will fully work with dotdrop:
$ file -b <some-file>
UTF-8 Unicode text, with escape sequences
and another that will mislead the compare command and return false/inaccurate results:
$ file -b <some-file>
ISO-8859 text, with escape sequences
Special chars¶
CRLF¶
The use of dotfiles with DOS/Windows line endings (CRLF, \r\n) will result in
the comparison (compare) returning a difference where there is none.
This is due to Jinja2 stripping CRLF.
One solution is to use dos2unix to re-format the dotfiles before adding them
to dotdrop.
See https://github.com/deadc0de6/dotdrop/issues/42.
Non-Unicode chars¶
Jinja2 is not able to process non-Unicode chars (https://jinja.palletsprojects.com/en/2.11.x/api/). This means that dotfiles using non-Unicode chars can still be fully managed by dotdrop; however, when comparing the local file with the one stored in dotdrop, compare will return a difference even if there is none.
Either replace the non-Unicode chars (see below Re-encode) or accept the fact the comparison shows a difference while there's none.
See https://github.com/deadc0de6/dotdrop/issues/42.
Re-encode¶
To change an existing file's encoding, you can use recode UTF-8 <filename> (see recode) or in vim :set fileencoding=utf-8.