I Cannot tweak author_sort_copy_method anymore

[lrpirlet]

I am running 5.10.0 and I suspect a problem with tweaks author_sort_copy_method...

A (not so) long time ago, under version 4, I did tweak the author_sort_copy_method by changing from comma to

author_sort_copy_method = 'nocomma'

.. . All that worked perfectly and still work that way today...

Well, today I observed a few bad author sorts so I changed from false

author_use_surname_prefixes = True

and I also added 'ANTHOLOGIE' in the author_name_copywords.

The author value "ANTHOLOGIE par Asimov" does copy as such in the author_sort field, BUT

The author "Alfred E Van Vogt" still sort automatically to "Vogt Alfred E Van"... I suspected a problem with capital letter so I added "Van" to

author_surname_prefixes = ['da', 'de', 'di', 'la', 'le', 'van', 'von', 'Van']

. No success.

So I removed ANTHOLOGIE from the author_name_copywords and I got a different author from author_sort value... So it's working?!?

So I changed 'nocomma' to 'comma' (in fact I did a reset) and the author sort did NOT change (I still have "fn ln" -> "ln fn" without the comma)

I have been looking around but have not been able to find anything similar... I have the calibre source from git and follow the evolution, I have seen a recent commit named 'Add a new tweak under "Author sort name algorithm" to optionally recognize common surname prefixes such as von, van, de etc.)' But I cannot (yet) understand python class and/or object programming...

Oh yes, I am running calibre in french if that matters and I have a few extentions... Tell me if you need more informations from me or if there is a better place to report this.

[kovidgoyal]

It works fine for me, are you sure you are running calibre 5.10.1??

[lrpirlet]

Thanks for your interest...
I was at 5.10.0... So I went to update my version of calibre. I had to follow the old version link to find the 5.10.1 64 bits version that I downloaded and installed...
Well same problem...

So, from under the preference tab, I activated the restart in debug mode.

The result is as follow:

calibre Journal de débogage
calibre 5.10.1 [64bit]  embedded-python: True is64bit: True
Windows-10-10.0.19041 Windows ('64bit', 'WindowsPE')
('Windows', '10', '10.0.19041')
Python 3.8.5
Windows: ('10', '10.0.19041', '', 'Multiprocessor Free')
Interface language: fr
Successfully initialized third party plugins: Gather KFX-ZIP (from KFX Input) (1, 36, 0) && Package KFX (from KFX Input) (1, 36, 0) && Find Duplicates (1, 8, 4) && KFX metadata reader (from KFX Input) (1, 36, 0) && KFX Input (1, 36, 0) && Set KFX metadata (from KFX Output) (1, 48, 0) && KFX Output (1, 48, 0) && KePub Input (3, 4, 3) && KePub Metadata Reader (3, 4, 3) && KePub Metadata Writer (3, 4, 3) && KePub Output (3, 4, 4) && Kindle Collections (1, 7, 29) && Kobo Books (1, 8, 0) && Kobo Utilities (2, 12, 3) && KoboTouchExtended (3, 4, 4) && Manage Series (1, 2, 11) && View Manager (1, 6, 0) && Wiki Reader (2, 2, 1)
calibre 5.10.1 [64bit]  embedded-python: True is64bit: True
Windows-10-10.0.19041 Windows ('64bit', 'WindowsPE')
('Windows', '10', '10.0.19041')
Python 3.8.5
Windows: ('10', '10.0.19041', '', 'Multiprocessor Free')
Interface language: fr
Successfully initialized third party plugins: Gather KFX-ZIP (from KFX Input) (1, 36, 0) && Package KFX (from KFX Input) (1, 36, 0) && Find Duplicates (1, 8, 4) && KFX metadata reader (from KFX Input) (1, 36, 0) && KFX Input (1, 36, 0) && Set KFX metadata (from KFX Output) (1, 48, 0) && KFX Output (1, 48, 0) && KePub Input (3, 4, 3) && KePub Metadata Reader (3, 4, 3) && KePub Metadata Writer (3, 4, 3) && KePub Output (3, 4, 4) && Kindle Collections (1, 7, 29) && Kobo Books (1, 8, 0) && Kobo Utilities (2, 12, 3) && KoboTouchExtended (3, 4, 4) && Manage Series (1, 2, 11) && View Manager (1, 6, 0) && Wiki Reader (2, 2, 1)
Turning on automatic hidpi scaling
devicePixelRatio: 1.0
logicalDpi: 96.0 x 96.0
physicalDpi: 117.58408679927668 x 117.60771704180064
Using calibre Qt style: True
[0.00] Starting up...
[0.01] Showing splash screen...
[0.38] splash screen shown
[0.38] Initializing db...
[0.42] db initialized
[0.42] Constructing main UI...
DEBUG:    0.0 KoboUtilites::action.py - loading translations
DEBUG:    0.0 KoboUtilites::dialogs.py - loading translations
DEBUG:    0.0 KoboUtilites::action.py - loading translations
DEBUG:    0.6 No Kobo Touch, Glo or Mini appears to be connected
DEBUG:    0.6 rebuild_menus - self.supports_ratings=None, self.supports_tiles=None
DEBUG:    0.6 KoboUtilities:set_toolbar_button_tooltip - start: text='None'
DEBUG:    0.6 KoboUtilities:set_toolbar_button_tooltip - setting to text='Utilities to use with Kobo ereaders

Pilote : KoboTouchExtended'
[1.17] main UI initialized...
[1.17] Hiding splash screen
Starting QuickView
[5.08] splash screen hidden
[5.08] Started up in 5.08 seconds with 1455 books

So I went to my other computer (an old slow one) and installed calibre 5.10.1 fresh... I left it in english... created a new library.. imported 2 books one from A.E. Van Vogt, the other from Adélaïde de Clermont-Tonnerre... Well it seems to work... sort of... It work when I import the book, it work when I wipe clear the field author sort in the edit metadata individualy... .

Ok I deleted the book completely and reset "Author sort name algorithm" before re-importing the books...

I set the name to be Adélaïde de Clermont-Tonnerre and A.E. Van Vogt on both books... author sort became "Adélaïde de Clermont-Tonnerre, Adélaïde de and Vogt, A.E. Van ...

Now I changed the "Author sort name algorithm" to "nocomma" and the "author_surname_prefixes" to "True"... I had to set BOTH "author sort" field to blank before to be able to get rid of the comma and/or to correctly order the prefixe correctly...

I am under the impression that all and every reference to an author name must be deleted in the WHOLE db before it takes the new name... That would be fine if the autor_sort field in the bulk metadata could be wiped before to be calculated again...

(note that A.E Van Vogt is NOT the same as A.E. van Vogt, same for Adélaïde de Clermont-Tonnerre is different from Adélaïde De Clermont-Tonnerre... so changing from upper to lower case does make the author_sort field goes red untill known at some other place...)

Well I hope to be somewhat clear... I still doubt.. I am sure that changing the "author sort algorithm" is a sort of adventure on a big library (the one above only contains part of my books as I use it for development)

Anyway thanks to take care (BTW, Noosfere.org, a french db of books, author, critics, etc.... does make the distinction between the name and the first name using uppercase character for the name. So when referring to the author A;E; Van Vogt the name becomes: VAN VOGT Alfred Elton)

[lrpirlet]

Ooops.

Sorry Kovidgoyal, it is documented and I did read it, but really could not understand the comment... I thought I had to edit via the metadata editor...That metadata bulk editor also has a recalculate button. I guess, now, that this button does NOT cover the whole db but only the current book if not "protected".

So the tag browser is where one can select book by author, series,etc. at one clik of the mouse. (this for me so I won't forget).

Go to the tag browser, right clik author, activate "manage authors" in the menu... Inside this box, in the bottom left corner clik "Recalculate all author sort values".. It takes a few seconds to flag all updated author sort... Yes author Alfred E Van Vogt become Van Vogt Alfred E... just like I expected.

Sorry to be a pain and thanks for this very (too) rich software .