Currently, the search url parsing does not detect if the search is based on Content or Filename, and it just keeps the last selection which can be inconsistent with the actual search.
This patch add such detection, and since an advanced user can combine filename and content search (using the keyword filename:), now the parsing detects both items and handles the four possible cases:
Content | Filename | Search text | Search type |
---|---|---|---|
T | T | abc filename:"xyz" | Content |
T | F | abc | Content |
F | T | xyz | Filename |
F | F | do not set | |
Depends on: D25260