<html> <head> <base href="https://bugs.documentfoundation.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_UNCONFIRMED " title="UNCONFIRMED - ISLOGICAL function depends on column formatting EDITING" href="https://bugs.documentfoundation.org/show_bug.cgi?id=115351#c6">Comment # 6</a> on <a class="bz_bug_link bz_status_UNCONFIRMED " title="UNCONFIRMED - ISLOGICAL function depends on column formatting EDITING" href="https://bugs.documentfoundation.org/show_bug.cgi?id=115351">bug 115351</a> from <a class="email" href="mailto:erack@redhat.com" title="Eike Rathke <erack@redhat.com>"> Eike Rathke</a> <pre>(In reply to C Hemingway from <a href="show_bug.cgi?id=115351#c5">comment #5</a>) > When I save an XLS or XLSX spreadsheet with a Logical field, when I later > reload it, it loses the logical designation and becomes that > [>0]"T";[<0]"T";"F" format. There is no boolean logical type for D2:D3, even in the file format the cell type is office:value-type="float" calcext:value-type="float". That is because the [>0]"T";[<0]"T";"F" format is of type number and not boolean. > Clarification - on that 4th column, I changed the display format of the > Boolean field to T;;F. What do you expect? T;;F works only by accident (if T and F don't have some other meaning as format keywords) and then is the same as "T";;"F" Which when re-loaded becomes [>0]"T";[<0]"";"F" that is identical in meaning. Note the second subformat has an empty string, not "T" as your other format. I don't know how you got to the [>0]"T";[<0]"T";"F" > When LibreOffice opens the file later, the boolean > field has become numeric There never was a boolean "field" if you applied the T;;F format. > and the column becomes [>0]"T";[<0]"T";"F". > Should assigning a custom display change the column type on reload? There is no column type, there are cell formats which lead to numeric/date/boolean display cell types. > For the > ISLOGICAL test, is LibreOffice remembering (caching?) that the column is > Boolean until the spreadsheet is reloaded? No, as explained previously already, there is no distinct boolean type, it is all about number display formats. Btw, there's an odd user-defined number format in that document which loads as [>0]"T"[~jewish]YYYY"U"YY;[<0]"T"[~jewish]YYYY"U"YY;"FAL"S[~jewish]YY listed as T5663U663 and currently displays a FAL0660 preview on an empty cell or 0. Seems to be some result of your previous experiments, but it illustrates how almost any combination of valid keywords yields some result ;-)</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> </ul> </body> </html>