Step #1: Like previous test campaign, I first started with a "dirty" DB, no luck. I am pretty confident that the "dirty_flag" will never be triggered since with existing broken databases, since what was broken were ids in columns "content_id" and "collection_id" in "collections" table, not book ids.
Step #2: I then deleted all collections to see if it would be better (from T1), plugged the T1 : I ended with empty collections.
Step #3: Finally, I formatted the SD card, plugged the T1 and started adding books from calibre.
Conclusion : Like previous time, repair procedure does not work. Starting with empty database seemed to work fine, but ids are low ones.
I hope I do not raise a false alarm and did perform the tests properly. I added a few print statements to see what was going on, you will find attached the traces of step #3 along with the db.
Hello,
I am glad to see there is more progress on it, but for me it does not seem to work any better than previous version.
I am still using 0.8.53, with patched driver http:// bazaar. launchpad. net/~kovid/ calibre/ trunk/view/ head:/src/ calibre/ devices/ prst1/driver. py (dating from 2012-06-04 04:25:47 UTC).
Step #1: Like previous test campaign, I first started with a "dirty" DB, no luck. I am pretty confident that the "dirty_flag" will never be triggered since with existing broken databases, since what was broken were ids in columns "content_id" and "collection_id" in "collections" table, not book ids.
Step #2: I then deleted all collections to see if it would be better (from T1), plugged the T1 : I ended with empty collections.
Step #3: Finally, I formatted the SD card, plugged the T1 and started adding books from calibre.
Conclusion : Like previous time, repair procedure does not work. Starting with empty database seemed to work fine, but ids are low ones.
I hope I do not raise a false alarm and did perform the tests properly. I added a few print statements to see what was going on, you will find attached the traces of step #3 along with the db.