From 5aae4a8185d1417028a4b22d43fbac851d51a843 Mon Sep 17 00:00:00 2001
From: Alex Auvolat <alex@adnab.me>
Date: Fri, 6 Oct 2023 15:32:59 +0200
Subject: add extra examples

---
 src/main.rs | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 67 insertions(+)

(limited to 'src')

diff --git a/src/main.rs b/src/main.rs
index ce352d4..8668a82 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -31,6 +31,7 @@ enum Cmd {
     Simplify,
     Cleanup,
     AddVocab,
+    AddExamples,
     Format,
 }
 
@@ -133,6 +134,34 @@ fn main() {
             )
             .expect("save");
         }
+        Cmd::AddExamples => {
+            let kanji_levels = read_kanji_levels().expect("read_kanji_levels");
+            let all_kanji = Charset::new(
+                kanji_levels
+                    .iter()
+                    .map(|(_, x)| x.to_string())
+                    .collect::<Vec<_>>()
+                    .join(""),
+            );
+
+            let mut ex = read_examples(&all_kanji).expect("read_examples");
+            ex.retain(|e| (5..=25).contains(&e.ja.chars().count()));
+
+            let mut batches: Vec<Batch> = fs::read("data/batches.json")
+                .map_err(anyhow::Error::from)
+                .and_then(|x| Ok(serde_json::from_slice(&x)?))
+                .unwrap_or_default();
+
+            add_examples(&mut batches, &ex);
+
+            fs::write(
+                "data/batches.json",
+                serde_json::to_string_pretty(&batches)
+                    .expect("serialize")
+                    .as_bytes(),
+            )
+            .expect("save");
+        }
         Cmd::Format => {
             let jmdict = fs::read_to_string("data/JMdict_e.xml").expect("read_jmdict");
             let jmdict = roxmltree::Document::parse_with_options(
@@ -495,6 +524,8 @@ struct Batch {
     examples: Vec<Example>,
     #[serde(default)]
     extra_vocab: Vec<JlptVocab>,
+    #[serde(default)]
+    extra_examples: Vec<Example>,
 }
 
 fn gen_batches(
@@ -953,6 +984,32 @@ fn add_vocab(all_batches: &mut [Batch], vocab: &[JlptVocab]) {
     }
 }
 
+fn add_examples(all_batches: &mut [Batch], examples: &[Example]) {
+    let mut chars = Charset::default();
+    for (i, batch) in all_batches.iter_mut().enumerate() {
+        chars = chars.union(&batch.chars);
+
+        let candidates = examples.iter()
+            .filter(|x| x.chars.inter_len(&batch.chars) > 0)
+            .filter(|x| x.chars.diff(&chars).len() == 0)
+            .filter(|x| batch.examples.iter().all(|y| y.ja != x.ja));
+        let mut cand_by_chars = HashMap::new();
+        for c in candidates {
+            cand_by_chars.insert(c.chars.to_string(), c.clone());
+        }
+        let mut candidates = cand_by_chars.into_iter().map(|(_, ex)| ex).collect::<Vec<_>>();
+        candidates.shuffle(&mut thread_rng());
+        candidates.truncate(20);
+        batch.extra_examples = candidates;
+
+
+        println!("---- BATCH #{:03} ----", i);
+        for ex in batch.extra_examples.iter() {
+            println!("{} - {}", ex.ja, ex.en);
+        }
+    }
+}
+
 // =====================================================================
 //                          FORMATTING TO HTML
 // =====================================================================
@@ -1075,6 +1132,16 @@ fn format_batch_aux<'a>(
         &batch.extra_vocab.iter().filter(|v| !batch.level.contains(&v.level)).collect::<Vec<_>>(),
         "Extra vocabulary (previous levels)")?;
 
+    writeln!(f, r#"<details><summary>Extra examples (reading practice)</summary><table class="extratable">"#)?;
+    for ex in batch.extra_examples.iter() {
+            writeln!(
+                f,
+                r#"<tr><td style="text-align: center"><span style="font-size: 1.2em">&nbsp;&nbsp;{}&nbsp;&nbsp;</span><br />{}</td></tr>"#,
+                ex.ja, ex.en
+            )?;
+    }
+    writeln!(f, r#"</table></details>"#)?;
+
     writeln!(f, "<hr />")?;
     writeln!(f, "<p>＼(≧▽≦)／</p>")?;
 
-- 
cgit v1.2.3