https://bugs.freedesktop.org/show_bug.cgi?id=50575

--- Comment #4 from Terrence Enger <[email protected]> 2012-06-07 
11:54:13 PDT ---
But sqlite 3 *does* support fractional seconds in at least some
contexts.  From a terminal session, first the command interpreter that
comes with sqlite3 ...

    $ sqlite3 ../bug_047520/bug_047520.db 
    SQLite version 3.7.4
    Enter ".help" for instructions
    Enter SQL statements terminated with a ";"
    sqlite> .mode columns
    sqlite> select * from byTs ;
    1           2012-04-06 12:34:56.654321  Friday    
    2           2012-04-05 13:45:57.123456  Thursday  
    sqlite> 

and then the command interpreter from unixodbc ...

    $ isql bug_047520
    +---------------------------------------+
    | Connected!                            |
    |                                       |
    | sql-statement                         |
    | help [tablename]                      |
    | quit                                  |
    |                                       |
    +---------------------------------------+
    SQL> select * from byTs
    +-----------+---------------------------------+-----------+
    | nr        | ts                              | word      |
    +-----------+---------------------------------+-----------+
    | 1         | 2012-04-06 12:34:56.654321      | Friday    |
    | 2         | 2012-04-05 13:45:57.123456      | Thursday  |
    +-----------+---------------------------------+-----------+
    SQLRowCount returns 2
    2 rows fetched
    SQL> 


"The timestamps in my tables are have no fractions of seconds" is
Heinz Repp talking about *his* data, not the data that I offered in
<https://bugs.freedesktop.org/attachment.cgi?id=59688>.


<asides>

  *Hoping* to be more helpful than a nuisance, ...

  (*) It is possible to retrieve the data successfully by disguising the
      key column:

          select nr, cast( "ts" as char(26) ), word from byTs

      Still, I suppose there must be *some* reason for the existence
      of class OKeySet.

  (*) One entry on my list of things to try (back before I got stuck in
      build difficulties) was a table keyed by a DECIMAL column with
      non-zero precision.  I may get back to that someday.

  (*) Bug 50849 "SIGSEGV: ODBC to PostgreSQL, renaming column in SELECT
      list" shows another case of LO plus ODBC giving results different
      from isql.

</asides>

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
_______________________________________________
Libreoffice-bugs mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

Reply via email to