Fix parsing #15

[dedan]

this should fix the problem described in #15.

I did not change much but it should already do the trick.

You think this is enough? please comment..

[maebert]

Looks good. Not sure what the best practise regarding line separators is, though - should we use the system's default, or check what newline sequence the journal file uses and use that one for writing as well (ie. if the journal was created on Windows, use CR+LF even os.linesep is LF on *nix?

[dedan]

first I thought system separator should be fine. But you are right,
then we would run into problems when
someone uses the same file via dropbox on a win and a *nix machine. But who uses
commandline applications on on win anyway?

I just saw that you already merged in my latest changes, I think this
was a bit premature
because I had some strange behavior yesterday. Will look into it as
soon as I'll find some time.
Maybe you want to revert the merge until then?

And the last thing: should we maybe start adding tests?

On Tue, May 15, 2012 at 11:57 AM, Manuel Ebert
reply@reply.github.com
wrote:

Looks good. Not sure what the best practise regarding line separators is, though - should we use the system's default, or check what newline sequence the journal file uses and use that one for writing as well (ie. if the journal was created on Windows, use CR+LF even os.linesep is LF on *nix?

---

Reply to this email directly or view it on GitHub:
https://github.com/maebert/jrnl/pull/26#issuecomment-5712345

[dedan]

false alarm, strange behavior is not reproducable