Learning with Texts

Learning with Texts (LWT) is a tool for language learning by reading. It is a self-hosted web application.

You feel that you won't learn much by translating dumb sentences or using grammar books? Learning With Texts offers you the possibility to learn by reading texts in your target language. Concept: when you don't know a word in a text, just click it. We show you the translation, and you will have regular tests to remember it. Ready to go?

THIS IS A THIRD PARTY VERSION. This version is not the official one, and brings many improvements and new features. It is quicker, has smaller database size, and is open for contributions. The official version is on source forge

Installation

As it is self-hosted, you will need a server, which can be your computer.

Get the latest GitHub release.
- You can also try to download the latest stable version if you want the cutting-edge updates (that may include some bugs).

Linux

Start a shell in the downloaded folder an run: ./INSTALL.sh. You may need to run chmod +x ./INSTALL.sh first.

Other Systems

Please see: docs/install.md for detailed instructions.
Create connect.inc.php with an existing database user. Everything is explained at docs/info.html.

And you are ready to go!

Please note that a composer repo is available, as well as an installation using Docker.

Description

LWT is a language learning web application. To learn a language, you need to practice, and we guide you in reading exercises.

First copy/paste any text you want to read. It can be raw text or an RSS feed.

Then, we parse the text. Unknown words will be displayed with different colors, just click them to see it in a dictionary.

Read as much as you want!

To make sure you memorize new words, you can take review exercises.

The difference with popular remembering software like Anki is that we keep track of the context to help you. By the way, we also ship an Anki exporter.

Features

Full list: docs/features.md

Features included from the official LWT software:

Support for almost 40 languages.
Text parsing for roman languages, right-to-left, and East-Asian ideogram systems
Translate words on-the-fly
Add an audio track and read it online
Practice words you don't remember
Statistics to record your progress

Features not in the official LWT

Full list: docs/newfeatures.md

Features that were added by the community:

Support for mobile
Automatically import texts from RSS feeds
Support for different themes
Display translations of terms with status in the reading frame
Multiwords selection (click and hold on a word → move to another word → release mouse button)
Bulk translate new words in the reading frame
Text to speech
Optional "ignore all" button in read texts
Key bindings in the reading frame
Selecting terms according to a text tag
Two database backup modes (new or old structure)

Improvements compared to the official LWT

Database improvements (db size is much smaller now)
Longer (>9) expressions can now be saved (up to 250 characters)
Save text/audio position in the reading frame
Google api (use 'ggl.php' instead of '*http://translate.google.com' for Google Translate)
Improved Search/Query for Words/Texts
Term import with more options (i.e.: combine translations, multiple tag import)
Support for MeCab for Japanese word-by-word automatic translation.
You can include video files from popular video platforms.
Code documentation.
Code is well organised, making debugging and contribution easier.

Contribute

Complete explanation: docs/contribute.md

To contribute, you need to clone or fork this repository, and Composer. The composer package is at hugofara/lwt.

Run git clone https://github.com/HugoFara/lwt

Next, got to the lwt folder and use composer install --dev.

In short:

git clone https://github.com/HugoFara/lwt
cd lwt
composer install --dev

Branches

The stable branch is master. Last commit on this branch is considered to be bug-free.
The dev branch is for unstable versions.
The official branch is for the official LWT Releases. Any other branch if considered under development.

Useful links

General documentation at docs/info.html.
Please find more help at docs/index.html.
You can also contact the community using GitHub or Discord.

Alternatives

See also: docs/links.md LWT is written for web, but you have many choices if you want a standalone installation:
LingL is a wonderful alternative written in Python, by gustavklopp.
FLTR ◆ Foreign Language Text Reader, a Java clone, by the same author as LWT (lang-learn-guy)

Unlicense

Under unlicense, view UNLICENSE.md, please look at [http://unlicense.org/].

Let's learn new languages!

gnebbia / lwt