Created attachment 66827 [details] OS X crash log +++ This bug was initially created as a clone of Bug #45596 +++ Problem description: While cutting and pasting from one cell to other, spreadsheet became unresponsive and crashed when doing paste. The spreadsheet was relatively new but some (not much) copy-pasting had been done earlier. OS X Mountain Lion 10.8.1. Crash log attached.
I think this bug may be more widespread than first indicated. I've been trying to attach a simple LO calc document to illustrate this, but every time I hit the submit button, I get a server error. However, this bug is easy to reproduce (at least for me). Create a new Calc (spreadsheet) document. Enter the number 1 in a cell. Select the cell below that first cell and create a formula to add 1 to the first cell. You should have now have 1 followed by 2 in the cell below. Finally, select that second cell and copy it down so that you have a list of 10 more numbers, in order. If I do that in LO 3.6.1.2 and then select that group of cells created above, LO will immediately crash if I either: 1) Do Edit -> Cut; hit the Mac delete key; or 3) Attempt to drag and drop that group of cells to any place else in that document. I can reproduce this in dozens of existing LO calc documents I have. Unfortunately, I haven't done any serious editing of my calc documents in the past week. During that time I updated Java from 2012-004 to 2012-005 using the Apple supplied (from Oracle) Java versions. Keep in mind that for Mac users, the version of Java reported in the LO preferences and used by LO will be the full Java version supplied by Apple. If a user later updates just the JRE version using an Oracle-supplied update, only the new JRE used by Web browsers will be updated. LO will still use the full SE version that a user probably got from Apple. If a user installs a full Oracle Java SE update, then that update WILL be used by LO. My gut tells me this is a Java issue (don't ask me why). In the past I've tried to manually find and delete Java files on a Mac system in an effort to downgrade Java versions, but that has often caused problems because there are so many Java files installed in a Mac OS X install. Thus, later today, I'll reinstall my Mac OS 10.8.1 and then reload an older Java version to see if that's the source of the problem. In the meantime, can any Mac users reading this attempt to recreate this bug using the instructions above? Report if you can or if you can't and also report your Mac OS version as well as the Java version reported in the LibreOffice Java preference. Thanks. This is killing my productivity this morning. I have a library of hundreds of LO Calc docs that I normally use when creating new documents. I do a lot of dragging and dropping.
Thank you very much for your bug report! @ juha.siukonen@iki.fi : If the problem described by noibsguy in comment #1 is the same as your original issue (and that seems rather probable), this bug would be the same as bug 54674, and therefore in turn the same as bug 53364. What does this mean? Good news: this bug is already fixed, and the fix will appear in LibreOffice 3.6.2, which will be released soon*). Until then, we are sorry for all inconveniences this bug may cause for you, and thank you for your patience! *) At least a first pre-release should be available end of next week; you can download it then from http://www.libreoffice.org/download/pre-releases/ and use it to confirm if this issue is really fixed. Maybe you can even use it for your real work -- this is not recommended, of course, but at least I myself use pre-releases most times for my work, and in 90% of the time they work well, so you could give it a try... Just my 2 cents ;-) *** This bug has been marked as a duplicate of bug 53364 ***
Sorry, I forgot to mention: Feel free to reopen this bug report, if you can still reproduce this problem with LibreOffice 3.6.2.1 or a current Master build. Thank you! For the record: Reset version number to original value ("3.6.0.4"); please note that the Version field should always contain the FIRST version in which a bug is known to exists, NOT the last one ...