[PATCH] milestone groups do not adhere to 'overall_completion = False'

[Andrew C Martin <andrew.c.martin@…>]

As I hinted at in #9718, milestone groups defined in TracIni do not adhere to 'overall_completion = False'. roadmap.py wraps the value of this field in a bool(), which always returns True for a non-empty string.

I have written attachment:overall_completion.patch​ which I believe corrects this issue.

Reproducible by adding these lines to your trac.ini:

[milestone-groups]
closed = closed
closed.order = 0
closed.overall_completion = true
new = new
new.order = 1
new.css_class = new
new.label = new
new.overall_completion = false
active = *
active.order = 2
active.css_class = open
active.label = in progress

Then observe that 'new' tickets do count towards your completion total.

This is an issue in the current trunk, and I can confirm it existed as of the 0.11.6 branch.

[Andrew C Martin <andrew.c.martin@…>]

Replying to Andrew C Martin <andrew.c.martin@…>:

Then observe that 'new' tickets don't count towards your completion total.

That should read "Then observe that 'new' tickets do count towards your completion total."

[rblank]

Thanks for the patch.

[Stefan]

can confirm it's a 0.11.6 issue as well and hence updated the version to reflect that being a 0.11.6 bug.

[rblank]

The fix is actually quite a bit simpler if you accept to import the "private" variable _TRUE_VALUES from trac.config. I have applied this simple fix in [10307].

But _TRUE_VALUES is used too much (and often badly, as case is often neglected). We already have trac.util.as_bool(), so I'll adapt it to be "compatible" and just use it everywhere.

[rblank]

The uses of _TRUE_VALUES have been refactored in [10308].

[Christopher A. Stelma <chris@…>]

fix for two inverted booleans from r10308

[Christopher A. Stelma <chris@…>]

Replying to rblank:

The uses of _TRUE_VALUES have been refactored in [10308].

this inverted two of the booleans, I noticed this because now only hidden repositories show in the timeline filter list (the opposite of expected behavior).
fix in trac-0.12-stable-r10436-inverted-booleans.diff​

[rblank]

Replying to Christopher A. Stelma <chris@…>:

this inverted two of the booleans,

Oops, good catch! Patch applied in [10437]. Thanks!