Skip to content

feat(sessions): BM25-ranked session search with search-index feature Refs #758 #831

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
AlexMikhalev merged 8 commits into from
Apr 22, 2026
Merged

feat(sessions): BM25-ranked session search with search-index feature Refs #758 #831

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
aba067d
feat(types): extract BM25 score module to terraphim_types Refs #758
AlexMikhalev Apr 22, 2026
419276b
refactor(service): replace score module with re-export from terraphim…
AlexMikhalev Apr 22, 2026
7fe4540
feat(sessions): add search-index feature with BM25 session search ada…
AlexMikhalev Apr 22, 2026
0ea97e9
feat(sessions): replace brute-force search with BM25-ranked results R…
AlexMikhalev Apr 22, 2026
3a22077
feat(agent): add /sessions index CLI command for BM25 search index st…
AlexMikhalev Apr 22, 2026
a923103
fix(types): add Default derives for all scorer structs (clippy) Refs …
AlexMikhalev Apr 22, 2026
d362f36
feat(sessions): hybrid search with KG concepts first, BM25 fallback R…
AlexMikhalev Apr 22, 2026
2fd0c2b
fix(sessions): UTF-8 boundary panic in build_body truncation, log BM2…
AlexMikhalev Apr 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion crates/terraphim_agent/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -79,7 +79,7 @@ terraphim_hooks = { path = "../terraphim_hooks", version = "1.0.0" }
terraphim_tracker = { path = "../terraphim_tracker", version = "1.0.0" }
terraphim_orchestrator = { path = "../terraphim_orchestrator", version = "1.0.0" }
# Session search - uses workspace version (path for dev, version for crates.io)
terraphim_sessions = { path = "../terraphim_sessions", version = "1.6.0", optional = true, features = ["tsa-full", "aider-connector"] }
terraphim_sessions = { path = "../terraphim_sessions", version = "1.6.0", optional = true, features = ["tsa-full", "aider-connector", "search-index"] }

[dev-dependencies]
serial_test = "3.3"
Expand Down
10 changes: 9 additions & 1 deletion crates/terraphim_agent/src/repl/commands.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,6 +191,8 @@ pub enum SessionsSubcommand {
Files { session_id: String, json: bool },
/// Find sessions by file path
ByFile { file_path: String, json: bool },
/// Build search index and show index statistics
Index { verbose: bool },
}

#[cfg(feature = "firecracker")]
Expand Down Expand Up @@ -1309,8 +1311,14 @@ impl FromStr for ReplCommand {
subcommand: SessionsSubcommand::ByFile { file_path, json },
})
}
"index" => {
let verbose = parts.contains(&"--verbose") || parts.contains(&"-v");
Ok(ReplCommand::Sessions {
subcommand: SessionsSubcommand::Index { verbose },
})
}
_ => Err(anyhow!(
"Unknown sessions subcommand: {}. Use: sources, list, search, stats, show, concepts, related, timeline, export, enrich, files, by-file",
"Unknown sessions subcommand: {}. Use: sources, list, search, stats, show, concepts, related, timeline, export, enrich, files, by-file, index",
parts[1]
)),
}
Expand Down
168 changes: 131 additions & 37 deletions crates/terraphim_agent/src/repl/handler.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1904,49 +1904,105 @@ impl ReplHandler {
}

SessionsSubcommand::Search { query } => {
let sessions = svc.search(&query).await;
#[cfg(feature = "enrichment")]
{
let thesaurus = if let Some(ref tui_service) = self.service {
let role_name: terraphim_types::RoleName = self.current_role.clone().into();
tui_service.get_thesaurus(&role_name).await.ok()
} else {
None
};
let sessions = svc.search_with_thesaurus(&query, thesaurus).await;

if sessions.is_empty() {
println!("{} No sessions match '{}'", "ℹ".blue().bold(), query.cyan());
return Ok(());
}
if sessions.is_empty() {
println!("{} No sessions match '{}'", "ℹ".blue().bold(), query.cyan());
return Ok(());
}

println!(
"\n{} sessions match '{}':",
sessions.len().to_string().green(),
query.cyan()
);
let mut table = Table::new();
table
.load_preset(UTF8_FULL)
.apply_modifier(UTF8_ROUND_CORNERS)
.set_header(vec![
Cell::new("ID").add_attribute(comfy_table::Attribute::Bold),
Cell::new("Source").add_attribute(comfy_table::Attribute::Bold),
Cell::new("Title").add_attribute(comfy_table::Attribute::Bold),
]);
println!(
"\n{} sessions match '{}':",
sessions.len().to_string().green(),
query.cyan()
);
let mut table = Table::new();
table
.load_preset(UTF8_FULL)
.apply_modifier(UTF8_ROUND_CORNERS)
.set_header(vec![
Cell::new("ID").add_attribute(comfy_table::Attribute::Bold),
Cell::new("Source").add_attribute(comfy_table::Attribute::Bold),
Cell::new("Title").add_attribute(comfy_table::Attribute::Bold),
]);

for session in sessions.iter().take(10) {
let title = session
.title
.as_ref()
.map(|t| {
if t.len() > 50 {
format!("{}...", &t[..50])
} else {
t.clone()
}
})
.unwrap_or_else(|| "-".to_string());
for session in sessions.iter().take(10) {
let title = session
.title
.as_ref()
.map(|t| {
if t.len() > 50 {
format!("{}...", &t[..50])
} else {
t.clone()
}
})
.unwrap_or_else(|| "-".to_string());

table.add_row(vec![
Cell::new(&session.external_id[..8.min(session.external_id.len())]),
Cell::new(&session.source),
Cell::new(title),
]);
table.add_row(vec![
Cell::new(&session.external_id[..8.min(session.external_id.len())]),
Cell::new(&session.source),
Cell::new(title),
]);
}

println!("{}", table);
}

println!("{}", table);
#[cfg(not(feature = "enrichment"))]
{
let sessions = svc.search(&query).await;

if sessions.is_empty() {
println!("{} No sessions match '{}'", "ℹ".blue().bold(), query.cyan());
return Ok(());
}

println!(
"\n{} sessions match '{}':",
sessions.len().to_string().green(),
query.cyan()
);
let mut table = Table::new();
table
.load_preset(UTF8_FULL)
.apply_modifier(UTF8_ROUND_CORNERS)
.set_header(vec![
Cell::new("ID").add_attribute(comfy_table::Attribute::Bold),
Cell::new("Source").add_attribute(comfy_table::Attribute::Bold),
Cell::new("Title").add_attribute(comfy_table::Attribute::Bold),
]);

for session in sessions.iter().take(10) {
let title = session
.title
.as_ref()
.map(|t| {
if t.len() > 50 {
format!("{}...", &t[..50])
} else {
t.clone()
}
})
.unwrap_or_else(|| "-".to_string());

table.add_row(vec![
Cell::new(&session.external_id[..8.min(session.external_id.len())]),
Cell::new(&session.source),
Cell::new(title),
]);
}

println!("{}", table);
}
}

SessionsSubcommand::Stats => {
Expand Down Expand Up @@ -2546,6 +2602,44 @@ impl ReplHandler {
println!("{}", table);
}
}

SessionsSubcommand::Index { verbose } => {
let sessions = svc.list_sessions().await;
let count = sessions.len();
let total_messages: usize = sessions.iter().map(|s| s.message_count()).sum();

println!("\n{} Search Index Status", "🔍".bold());
println!("{}", "─".repeat(40));
println!(" Sessions indexed: {}", count.to_string().green());
println!(" Total messages: {}", total_messages.to_string().green());
println!(" Scorer: {}", "BM25 (Okapi)".cyan());

if verbose {
let sources: std::collections::HashMap<&str, usize> = {
let mut map = std::collections::HashMap::new();
for s in &sessions {
*map.entry(s.source.as_str()).or_default() += 1;
}
map
};
println!("\n {} By source:", "▸".blue());
for (source, cnt) in sources {
println!(" {}: {}", source.magenta(), cnt);
}

let total_chars: usize = sessions
.iter()
.map(|s| {
s.messages.iter().map(|m| m.content.len()).sum::<usize>()
+ s.title.as_ref().map(|t| t.len()).unwrap_or(0)
})
.sum();
println!(
"\n Total text size: {} chars",
total_chars.to_string().yellow()
);
}
}
}

Ok(())
Expand Down
Loading
Loading