View Single Post
Old 12-30-2014, 05:42 AM   #820
davidfor
Grand Sorcerer
davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.
 
Posts: 24,905
Karma: 47303824
Join Date: Jul 2011
Location: Sydney, Australia
Device: Kobo:Touch,Glo, AuraH2O, GloHD,AuraONE, ClaraHD, Libra H2O; tolinoepos
Quote:
Originally Posted by chaley View Post
You don't need to run CC to have the error.

The wireless device driver is tripping over finding a value in the field_metadata structure, something that should not be there. The problem is to find what put the value there. You can see the value that is inserted by looking at the file metadata_db_prefs_backup.json. My suspicion is that if you start calibre, do nothing, and then quit then metadata_db_prefs_backup.json will not contain any '#value#' attributes. If you start calibre, do whatever it is that PeterT is doing then quit, metadata_db_prefs_backup.json will contain something like
Code:
      "#value#": {
        "__class__": "datetime.datetime", 
        "__value__": "2014-12-22T23:47:57+00:00"
      },
Interesting: I just noticed that the metadata_db_prefs_backup.json file contains #value# attributes in three metadata entries: #kobolastread (the value above), #koboreadpct ("#value#": 100), and #koboreadloc ("#value#": "(0)OEBPS/xhtml/cover.xhtml"). Don't know if that helps.
Yes, that is what is happening. And the problem is my changes when the new DB API was implemented. The handling for the old API was in the wrong place and still executed the update. I suppose that I should remove that completely now.

My changes seem to be working, I'll post an update as soon as I look at the other reported problems.
davidfor is offline