Initial SFERA platform baseline

rust/crates/query-parser/Cargo.toml

@@ -0,0 +1,4 @@
+[package]
+name = "query-parser"
+version = "0.1.0"
+edition = "2021"

rust/crates/query-parser/src/lib.rs

@@ -0,0 +1,143 @@
+#[derive(Debug, Clone, PartialEq, Eq)]
+pub struct ParsedQueryText {
+    pub tables: Vec<String>,
+}
+
+pub fn parse_query_text(query_text: &str) -> ParsedQueryText {
+    ParsedQueryText {
+        tables: extract_tables(query_text),
+    }
+}
+
+pub fn extract_tables(query_text: &str) -> Vec<String> {
+    let mut tables = Vec::new();
+    let lines: Vec<String> = query_text.lines().map(clean_query_line).collect();
+
+    for (index, line) in lines.iter().enumerate() {
+        if let Some(table) = table_after_from(line) {
+            push_unique(&mut tables, table);
+        } else if is_standalone_from(line) {
+            if let Some(next_line) = lines.get(index + 1) {
+                if let Some(table) = first_table_token(next_line) {
+                    push_unique(&mut tables, table);
+                }
+            }
+        }
+
+        if let Some(table) = table_after_join(line) {
+            push_unique(&mut tables, table);
+        }
+    }
+
+    tables
+}
+
+fn clean_query_line(line: &str) -> String {
+    let mut value = line
+        .trim()
+        .trim_end_matches(';')
+        .trim()
+        .trim_matches('"')
+        .trim();
+    if let Some(stripped) = value.strip_prefix('|') {
+        value = stripped.trim();
+    }
+    value.trim_matches('"').trim().to_string()
+}
+
+fn is_standalone_from(line: &str) -> bool {
+    line.eq_ignore_ascii_case("ИЗ") || line.eq_ignore_ascii_case("FROM")
+}
+
+fn table_after_from(line: &str) -> Option<String> {
+    let trimmed = line.trim();
+    let upper = trimmed.to_uppercase();
+    let rest = if upper.starts_with("ИЗ ") {
+        &trimmed["ИЗ".len()..]
+    } else if upper.starts_with("FROM ") {
+        &trimmed["FROM".len()..]
+    } else {
+        return None;
+    };
+    first_table_token(rest)
+}
+
+fn table_after_join(line: &str) -> Option<String> {
+    let normalized = line.replace('\t', " ");
+    let parts: Vec<&str> = normalized.split_whitespace().collect();
+    let index = parts.iter().position(|part| {
+        part.eq_ignore_ascii_case("JOIN") || part.eq_ignore_ascii_case("СОЕДИНЕНИЕ")
+    })?;
+    let table = parts.get(index + 1)?.trim_matches(',');
+    if table.is_empty() {
+        None
+    } else {
+        Some(table.to_string())
+    }
+}
+
+fn first_table_token(line: &str) -> Option<String> {
+    let table = line.trim().split_whitespace().next()?.trim_matches(',');
+    if table.is_empty() {
+        None
+    } else {
+        Some(table.to_string())
+    }
+}
+
+fn push_unique(values: &mut Vec<String>, value: String) {
+    if !values.iter().any(|existing| existing == &value) {
+        values.push(value);
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::{extract_tables, parse_query_text};
+
+    #[test]
+    fn extracts_table_after_standalone_from() {
+        let tables = extract_tables(
+            r#"
+ВЫБРАТЬ
+    Остатки.Номенклатура
+ИЗ
+    РегистрНакопления.ОстаткиТоваров КАК Остатки
+"#,
+        );
+
+        assert_eq!(tables, vec!["РегистрНакопления.ОстаткиТоваров"]);
+    }
+
+    #[test]
+    fn extracts_inline_from_and_join_tables() {
+        let query = parse_query_text(
+            r#"
+SELECT
+    Orders.Ref,
+    Customers.Description
+FROM Document.CustomerOrder AS Orders
+LEFT JOIN Catalog.Customers AS Customers
+    ON Orders.Customer = Customers.Ref
+"#,
+        );
+
+        assert_eq!(
+            query.tables,
+            vec!["Document.CustomerOrder", "Catalog.Customers"]
+        );
+    }
+
+    #[test]
+    fn cleans_pipe_prefixed_1c_query_lines() {
+        let tables = extract_tables(
+            r#"
+|ВЫБРАТЬ
+|Контрагенты.Ссылка
+|ИЗ Справочник.Контрагенты КАК Контрагенты
+"#,
+        );
+
+        assert_eq!(tables, vec!["Справочник.Контрагенты"]);
+    }
+}