The library¶

The library is where wuDict keeps its own indexes. Its default folder is ~/.wudict/db.

The three stages¶

Preview starts the moment wuDict finds the file. Preparation starts on the first search and runs in the background. Full-text and contains search indexes can be enabled on-demand only for the dictionaries where you need them, via the ☰ dictionary panel.

Headword indexing is automatic by default. Set AUTO_INDEX to off if you want to disable auto-indexing (not recommenced — while it might save a few megabytes of disk space this will cause higher CPU and RAM usage).

By default only one dictionary is indexed at a time to preserve RAM and CPU. You can increase the number of parallel indexing threads via the INDEX_WORKERS parameter.

What a library folder holds¶

~/.wudict/db/
  Webster/
    text.db     articles and the search indexes
    media.db    audio and images - only if you clicked "pack media"
    info.txt    what this is and where it came from
    res/        optional: your replacement files

text.db alone is a complete dictionary. Without media.db, audio and images are read from the original files if they are present.

Article text is compressed, so a prepared folder is usually smaller than the file it came from. Set NO_COMPRESS to disable compression for the definitions stored in the SQLite3 database. That makes the database roughly three times larger.

Move a dictionary to another machine¶

1. Copy the whole folder, for example ~/.wudict/db/Oxford.
2. Put it in the other machine's library folder, or in any folder that machine scans.
3. Search. The original dictionary file is not needed.

Prepare from the terminal¶

wudict ingest ~/Dicts

ingest skips what is already prepared. Add -headwords for the small index only, -contains for the substring index, and -full to pack media as well.

Removing a dictionary¶

Open the ☰ panel and find the dictionary. Click the file row (e.g. oald10.mdx), just above About this dictionary — to expand it. It lists every path that belongs to this dictionary, and the 🗑 Remove… button.

Clicking it turns the row into a confirmation that names what each choice takes:

Deleting everything is the default on purpose. With AUTO_INDEX on, deleting only the index of a dictionary whose file is still in a scanned folder frees the space until the next search rebuilds it.

↩ cancel backs out and puts the file list back as it was.

This is available on every platform. Deleting from another machine over the network is a separate switch — ALLOW_REMOTE_DELETE, OFF by default.

Removing FTS and contains indexes¶

Turn an index off with the same ☰ panel switch that turned it on.

wudict clean            # list broken or leftover library items
wudict clean -f         # delete them
wudict rm Webster        # list what removing "Webster" would delete
wudict rm -f Webster     # delete the library folder AND the original files

Both commands list what they would do and delete nothing until you add -f.

rm takes -keep-source to delete only the prepared folder, or -keep-index to delete only the originals.

Full CLI reference

Dictionaries whose originals are gone¶

wuDict lists a prepared dictionary only when its original file is still in a scanned folder. To list every prepared dictionary regardless, switch USE_CACHED on, or click Use these dictionaries on the setup page.