Merge lp:~mhr3/gwibber/proper-lens-search into lp:gwibber
Status: | Merged |
---|---|
Merged at revision: | 1281 |
Proposed branch: | lp:~mhr3/gwibber/proper-lens-search |
Merge into: | lp:gwibber |
Diff against target: |
363 lines (+175/-114) 1 file modified
lens/src/daemon.vala (+175/-114) |
To merge this branch: | bzr merge lp:~mhr3/gwibber/proper-lens-search |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Ken VanDine | Pending | ||
Review via email: mp+93315@code.launchpad.net |
Description of the change
This branch creates a TreeIndex on top of the model, which is used for searching. By using Dee's TextAnalyzer it properly tokenizes all the data and even performs folding of utf8 characters into ascii (which means you can search for accented characters using their non-accented counterparts).
Another benefit is a proper multiword search, so searching for "one two three" will find items containing for example "oh dear the timer shows three.. two.. one..".
One drawback is that single letter searches might be slower, but basically anything else should be faster.
Please note that this code revealed a few issues in DeeTreeIndex implementation (namely #933111 and #933131) and fixes for these aren't yet in upstream dee.
Looks great, thanks!