[Yum-devel] [PATCH] when writing out the config data in add_on - do the to_unicode() bit so we don't trace out rh bug https://bugzilla.redhat.com/show_bug.cgi?id=622301

[James Antill]

On Mon, 2010-08-09 at 10:12 -0400, Seth Vidal wrote:
> ---
>  yum/history.py |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/yum/history.py b/yum/history.py
> index 502b908..66861c8 100644
> --- a/yum/history.py
> +++ b/yum/history.py
> @@ -610,7 +610,7 @@ class YumHistory:
>              # open file in append
>              fo = open(data_fn, 'w+')
>              # write data
> -            fo.write(data)
> +            fo.write(to_unicode(data))
>              # flush data
>              fo.flush()
>              fo.close()

 Have you tested that with random utf-8 crack in the config? ACK if you
have.

 I'd have thought you wanted "to_utf8()" ... or to change the open()
call like:

fo = codes.open(data_fn, "w+", "utf-8")

...or even utf-8-sig! Personally I think the to_utf8() is probably more
obvious for us mortals, but I guess the later is the best practice, or
something.

http://stackoverflow.com/questions/934160/write-to-utf-8-file-in-python

etc.