[O] [bug] Separate clocksum format for durations >= 1 day

[Sebastien Vauban]

Nicolas, Toby,

Nicolas Goaziou wrote:
> I've applied your patch (with some small changes in a docstring). Thank
> you again for all that work.

Here an ECM which gave correct results before commit a00a7b2, and which does
not anymore:

--8<---------------cut here---------------start------------->8---
#+TITLE:     ECM Prestas
#+Time-stamp: <2012-11-30 Fri 12:12>
#+LANGUAGE:  en
#+OPTIONS:   H:4 num:t toc:t

* Tasks

Some tasks with CLOCK lines.

** Design

*** TODO Do this
    :LOGBOOK:
    CLOCK: [2012-11-19 Mon 09:00]--[2012-11-19 Mon 10:11] =>  1:11
    :END:
    :PROPERTIES:
    :Effort:   2:00
    :END:

** Develop

*** TODO Do that
    :LOGBOOK:
    CLOCK: [2012-11-19 Mon 10:00]--[2012-11-19 Mon 12:22] =>  2:22
    CLOCK: [2012-11-20 Tue 10:01]--[2012-11-20 Tue 12:35] =>  2:34
    :END:

* Reporting

** Summary

#+TBLNAME: prestasTotal
#+BEGIN: clocktable :maxlevel 4 :scope file :block 2012-11 :lang "en" :narrow 
80! :indent t
#+CAPTION: Clock summary at [2012-11-30 Fri 12:12], for November 2012.
| Headline            | Time |      |      |
|---------------------+------+------+------|
| Total time          | 6:07 |      |      |
|---------------------+------+------+------|
| Tasks               | 6:07 |      |      |
| \__ Design          |      | 1:11 |      |
| \_____ TODO Do this |      |      | 1:11 |
| \__ Develop         |      | 4:56 |      |
| \_____ TODO Do that |      |      | 4:56 |
#+END:

** Bill

| Total of worked hours                  | 6.12 | hours |
| Total of worked days (in days of 7:36) | 0.81 | days  |
#+TBLFM: @1$2=remote(prestasTotal,@2$2);t::@address@hidden/7.6;%.2f
--8<---------------cut here---------------end--------------->8---

Now, with current master, the "bill" table returns:

--8<---------------cut here---------------start------------->8---
| Total of worked hours                  | 0.00 | hours |
| Total of worked days (in days of 7:36) | 0.00 | days  |
#+TBLFM: @1$2=remote(prestasTotal,@2$2);t::@address@hidden/7.6;%.2f
--8<---------------cut here---------------end--------------->8---

I don't know (yet -- lack of time) if it's possible that, by setting some vars
to non-default values, the above problem would disappear; but I would expect
the defaults to conserve the old behavior anyway, wouldn't I?

Best regards,
  Seb

--
Sebastien Vauban