linuxsda/lexicon_project

Lexicon: An Instant dictionary

Lexicon_Project

Lexicon: Dictionary, anywhere

Lexicon is a native linux dictionary made for providing meaning of a word anywhere, instantly. It fetches the meaning from database and pushes it through notification. No need to Google the word or open a dictionary.

• Highlight the word.

• Press ALT + CTRL + Z.

---

---

Features:

• Non-intrusive interface
• Synonym support
• Works offline
• Online support
• Translation support for major languages.
• Auto-correct (beta)
• Works everywhere on the desktop.

---

How to install?:

• Download zip and place scripts in relevent folders.
• Install dependencies: yad, xsel, xbindkeys, sqlite3, html2text, notify-osd, agrep and Leolik notify-osd patch
• Execute lexicon and click launch

/usr/bin - has main lexicon script. The Launcher.

/usr/lib/lexicon - has all the source code. AutoCorrect and Lexicon named scripts contain all the functioning.

/usr/share/lexicon - has all the database.

---

For more help run:

man lexicon    

Originally posted: here and here

---

For further information and installation instructions, follow here or here.

---

Project has been shutdown due to lack of time and interest. Thank you for your support.

---

You can contact me at [email protected]

lexicon_project's Issues

Offline dictionary from wiktionary

Issue for tracking offline support in lexicon_project. It might be more efficient to use sql over xml export since we could import directly into sqlite3, however, wikimedia uses 'wikitext' for the article content so not sure how that factors in.

Wiktionary's DB can be exported as SQL and XML:
https://dumps.wikimedia.org/enwiktionary/latest/enwiktionary-latest-page.sql.gz
https://dumps.wikimedia.org/enwiktionary/latest/enwiktionary-latest-pages-articles.xml.bz2

DB layout & parsing:
https://www.mediawiki.org/wiki/Manual:Database_layout
https://phabricator.wikimedia.org/diffusion/MW/browse/master/maintenance/tables.sql
https://www.mediawiki.org/wiki/Help:Export#Example
https://www.mediawiki.org/wiki/Help:Export#Processing_XML_export

Consideration (at least for XML):

• Each page is separated with <page>
• When parsing we should ignore titles with namespaces (ie. Talk:) (XML, we want: <ns>0</ns> when parsing <page>)
• Select last revision from db (first <revision> tag in list)
  • Select only <text>
• Do we index the sqlite3 db or let wikitionary do this? (XML <id>int value for sql</id>)
• Do we parse wikimedia tags when we download & index the db or at runetime, like using an offline wiki parser?
  • We would need to index templates then (Template:)