Bug in org-time-stamp?

Bug in org-time-stamp?

[bostjanv]

[-- Attachment #1: Type: text/plain, Size: 2067 bytes --]

Section 8.2 (node: Creating timestamps) of the info manual contains the
following description:

`C-c .'
     Prompt for a date and insert a corresponding time stamp.  When the
     cursor is at a previously used time stamp, it is updated to NOW.
     When this command is used twice in succession, a time range is
     inserted.  

In my opinion, the second sentence does not correspond to the actual
operation. To see this, one can perform a test on a single-line org file,
for example:

* TODO <2008-08-28 Thu> test todo item

We encounter (at least) the following types of behavior:

(1) point is within the timestamp, mouse cursor is either inside or outside
    (if inside, do not click). In that case entering `C-c .' will result in
    a timestamp update query, and hitting RETURN will produce no change in
    the timestamp.
(2) point is outside the timestamp, mouse cursor inside. In that case the
    `C-c .' command and RETURN will result in an updated timestamp at the
    point position while the original timestamp remains unchanged.
(3) point is within the timestamp. If after `C-c .' and the timestamp query
    one clicks on a date in the calendar, then the original timestamp will
    be changed to the selected date.

On examining the first two cases one concludes that the origin of the problem
is quite simple: In Case (1) the default answer to the update query is the
ORIGINAL VALUE OF TIMESTAMP while in Case (2) it is NOW. In my opinion it
should be NOW in both cases. Case (3) does not require comment as the
corresponding behavior is expected.

Additionally, if the previously suggested change be accepted, I propose that
the above description in the info manual be changed to the following:
    
`C-c .'
     Prompt for a date and insert a corresponding time stamp.  When the
     point or mouse cursor is at a previously used time stamp, the prompt
     requests an updated value, and the latter is inserted at point
     (default is NOW). When this command is used twice in succession, a
     time range is inserted.  

Regards,
bostjanv

[-- Attachment #2: Type: text/plain, Size: 204 bytes --]

_______________________________________________
Emacs-orgmode mailing list
Remember: use `Reply All' to send replies to the list.
Emacs-orgmode@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-orgmode

Re: Bug in org-time-stamp?

[Carsten Dominik]

Hi Bostjanv,


first of all, I do not understand why you seem to think it matters if  
the mouse is inside or outside the time stamp when `C-c .' is called.   
I believe it does not matter at all.

On Aug 31, 2008, at 3:56 AM, bostjanv@alum.mit.edu wrote:

> Section 8.2 (node: Creating timestamps) of the info manual contains  
> the
> following description:
>
> `C-c .'
>     Prompt for a date and insert a corresponding time stamp.  When the
>     cursor is at a previously used time stamp, it is updated to NOW.
>     When this command is used twice in succession, a time range is
>     inserted.
>
> In my opinion, the second sentence does not correspond to the actual
> operation. To see this, one can perform a test on a single-line org  
> file,
> for example:
>
> * TODO <2008-08-28 Thu> test todo item
>
> We encounter (at least) the following types of behavior:
>
> (1) point is within the timestamp, mouse cursor is either inside or  
> outside
>    (if inside, do not click). In that case entering `C-c .' will  
> result in
>    a timestamp update query, and hitting RETURN will produce no  
> change in
>    the timestamp.

This is correct, and it is a bug in the documentation.  The  
documentation shows how this function behaved a long time ago,  but  
since then we decided that the current value of the timestamp should  
be the default instead as this application seems to be more common.   
This allows you, for example, to use this command to quickly change or  
add the time component of a stamp by typing 15:33 or so.

If you want a shortcut to shift the time stamp to today, use "+0" at  
the prompt.

Thank you for the report, I have updated the documentation to reflect  
the real behavior.

- Carsten

>
> (2) point is outside the timestamp, mouse cursor inside. In that  
> case the
>    `C-c .' command and RETURN will result in an updated timestamp at  
> the
>    point position while the original timestamp remains unchanged.
> (3) point is within the timestamp. If after `C-c .' and the  
> timestamp query
>    one clicks on a date in the calendar, then the original timestamp  
> will
>    be changed to the selected date.
>
> On examining the first two cases one concludes that the origin of  
> the problem
> is quite simple: In Case (1) the default answer to the update query  
> is the
> ORIGINAL VALUE OF TIMESTAMP while in Case (2) it is NOW. In my  
> opinion it
> should be NOW in both cases. Case (3) does not require comment as the
> corresponding behavior is expected.
>
> Additionally, if the previously suggested change be accepted, I  
> propose that
> the above description in the info manual be changed to the following:
>
> `C-c .'
>     Prompt for a date and insert a corresponding time stamp.  When the
>     point or mouse cursor is at a previously used time stamp, the  
> prompt
>     requests an updated value, and the latter is inserted at point
>     (default is NOW). When this command is used twice in succession, a
>     time range is inserted.
>
> Regards,
> bostjanv
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode