Skip to content

Conversation

@x4base
Copy link

@x4base x4base commented Oct 14, 2014

The design of database has been changed in commit 4fa966b.
This method helps update the database by altering the db and generating
the tabkeys from m0,m1,m2,m3,m4 in the old schema.

I don't know whether there is any existing function here. I just can't find one. I am using the Boshiamy Chinese input method. After I upgraded from Fedora 18 to Fedora 20, ibus-table keeps crashing because of this exception: 'tabsqlitedb.py:781:select_words:sqlite3.OperationalError: no such column: tabkeys'. Therefore I wrote this function.

But I don't really know in which function should this method be called. I tried to call it in tabsqlitedb.init(), but you need the root permission to modify the sysdb file. Another way is to wrap it into a command line tool to allow the user upgrade the db by himself/herself, but I don't know where to put it either. So maybe please accept this pull request and help me finish it, or tell me what can I do. Thanks!

The design of database has been changed in commit 4fa966b.
This method helps update the database by altering the db and generating
the tabkeys from m0,m1,m2,m3,m4 in the old schema.
@mike-fabian mike-fabian self-assigned this Oct 16, 2014
@x4base
Copy link
Author

x4base commented Oct 17, 2014

It can be downloaded from http://boshiamy.com/member_download.php#tab4 . I don't know if you can read Chinese. It's in the 4th tab (with 'IBus' in its name). However, it's for paid member only.
It is in the proprietary license. It mentioned something like 'Copyright by Liu company', 'You cannot convert and re-distribute it'. I think they might release the new version only when they are asked to. Maybe they just haven't known there is a new db schema yet.

I tried to implement the basic part of ibus-table-updatedb. Please review it. Thanks!

@x4base
Copy link
Author

x4base commented Oct 19, 2014

mike-fabian > Or maybe an option to /usr/bin/ibus-table-createdb to reconstruct the source from the .db file again.

Is it really better? I'm not sure in what situation it would be useful. After you reconstructed the source, you still have to generate the db, right? Isn't just updating more straightforward? Or do you mean you think it is useful to have both features? Maybe having both src-to-db and db-to-src conversions would make the code harder to maintain?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants