From 8140b51d3d75b2060dfdd0e88995ada8e07b84d3 Mon Sep 17 00:00:00 2001
From: tiennm99 <tiennm99@outlook.com>
Date: Thu, 23 Apr 2026 11:09:55 +0700
Subject: [PATCH] refactor: remove unused MODEL_VARIANT env var

Was set in Dockerfile/compose/.env/README but never read by the app.
Tokenization is inferred at lookup time by _variant_candidates trying
both spaced and underscore-joined forms.
---
 .env.example       | 3 +--
 Dockerfile         | 1 -
 README.md          | 2 --
 docker-compose.yml | 1 -
 4 files changed, 1 insertion(+), 6 deletions(-)

diff --git a/.env.example b/.env.example
index 7b31bdf..eca3534 100644
--- a/.env.example
+++ b/.env.example
@@ -10,6 +10,5 @@
 #   MODEL_URL=https://cloud.example.com/s/abc123XYZ/download
 MODEL_URL=https://cloud.example.com/s/REPLACE_ME/download
 
-# Variant — must match the filename inside the zip.
+# Must match the filename inside the zip.
 MODEL_PATH=/data/phow2v/word2vec_vi_words_300dims.txt
-MODEL_VARIANT=word
diff --git a/Dockerfile b/Dockerfile
index fffb065..3459e5f 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -16,7 +16,6 @@ COPY app ./app
 # PhoW2V's license forbids public redistribution, so every deployment
 # must point at its own private mirror (typically Nextcloud WebDAV).
 ENV MODEL_PATH=/data/phow2v/word2vec_vi_words_300dims.txt \
-    MODEL_VARIANT=word \
     PORT=8000
 
 EXPOSE 8000
diff --git a/README.md b/README.md
index 7050297..a389aea 100644
--- a/README.md
+++ b/README.md
@@ -110,7 +110,6 @@ Host the desired zip and update `.env`:
 ```bash
 MODEL_URL=https://cloud.example.com/s/<token-for-syllables-100>/download
 MODEL_PATH=/data/phow2v/word2vec_vi_syllables_100dims.txt
-MODEL_VARIANT=syllable
 ```
 
 Delete the `phow2v-cache` volume when switching, otherwise the stale
@@ -126,7 +125,6 @@ docker compose down -v && docker compose up --build
 |---|---|---|
 | `MODEL_URL` | *(required)* | Any URL that serves the zip via a plain GET. Bake any auth into the URL. |
 | `MODEL_PATH` | `/data/phow2v/word2vec_vi_words_300dims.txt` | Where the text-format vectors are persisted. |
-| `MODEL_VARIANT` | `word` | `word` or `syllable`. Declarative hint; must match the file you uploaded. |
 
 ## Using from doantu (miti99bot)
 
diff --git a/docker-compose.yml b/docker-compose.yml
index 8a1e8ad..784d3da 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -11,7 +11,6 @@ services:
       #   https://cloud.example.com/s/<shareToken>/download
       MODEL_URL: ${MODEL_URL:?set MODEL_URL in .env to your PhoW2V zip URL}
       MODEL_PATH: ${MODEL_PATH:-/data/phow2v/word2vec_vi_words_300dims.txt}
-      MODEL_VARIANT: ${MODEL_VARIANT:-word}
     volumes:
       - phow2v-cache:/data/phow2v
     restart: unless-stopped