Close file handle before passing edit's tempfile to editor.

[minchinweb]

Fix for #415 on Windows.

[xemoka]

Will this ever get merged or is there some conflict for doing it this way?

[mrlo]

Also looking to see if this is going to get merged...

[arisAlexis]

bump

[wren]

This looks like a great fix, and we'd love to merge it. Would you mind resolving the merge conflict?

[minchinweb]

@wren Will do!

[minchinweb]

@wren: the pull request has been recreated and passes the test suite. I'm don't think there was a test in place for this particular issue.

[wren]

@minchinweb It looks like this was maybe already fixed by another pr? Sorry for the back and forth, but I hadn't seen the other one when I commented here.

Is there some reason the os.close call needs to be moved up a few lines like this PR does?

[minchinweb]

@wren: yes, this is almost identical to #564, except that that one got applied to the 1.9.x branch, but has never been merged into the 2.0 branch; this one applies directly to the 2.0 branch. So this should be applied to the 2.0 branch, or #564 should be cherry-picked to the 2.0 branch.

As for the location of closing the filehandle, first here is the code:

    filehandle, tmpfile = tempfile.mkstemp(prefix="jrnl", text=True, suffix=".txt")
    os.close(filehandle)  # in PR 564
    with codecs.open(tmpfile, 'w', "utf-8") as f:
        if template:
            f.write(template)
    os.close(filehandle)  # in PR 416 (this one)

Personally, I like it lower, as it maintains the file lock if we decide to write the template file to our temp file.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[sergiorgiraldo]

Hi, it seems this continues in version 1.9.8. Can this be merged?

[wren]

@sergiorgiraldo v2.0 is out now, and I believe has a fix for this already. Can you please upgrade from v1.9.8 and confirm the fix?

[wren]

Already fixed.

[wren]

@sergiorgiraldo Sorry, this wasn't actually fixed, apparently! It looks like @micahellison has submitted the actual fix here.

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.