snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2025-10-31 16:06:31 +00:00
Code Issues Packages Projects Releases Wiki Activity

Introduce a parameter to skip the first documents

Browse Source
This commit is contained in:
Kerollmops
2025-02-06 10:20:39 +01:00
parent 2ea5c57871
commit 86fcad788e
1 changed files with 12 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
crates/meilitool/src/main.rs
View File

@@ -83,6 +83,10 @@ enum Command {
/// Do not export vectors with the documents. /// Do not export vectors with the documents.
#[arg(long)] #[arg(long)]
ignore_vectors: bool, ignore_vectors: bool,
/// The number of documents to skip.
#[arg(long)]
offset: Option<usize>,
}, },
/// Attempts to upgrade from one major version to the next without a dump. /// Attempts to upgrade from one major version to the next without a dump.
@@ -131,8 +135,8 @@ fn main() -> anyhow::Result<()> {
Command::ExportADump { dump_dir, skip_enqueued_tasks } => { Command::ExportADump { dump_dir, skip_enqueued_tasks } => {
export_a_dump(db_path, dump_dir, skip_enqueued_tasks, detected_version) export_a_dump(db_path, dump_dir, skip_enqueued_tasks, detected_version)
} }
Command::ExportDocuments { index_name, ignore_vectors } => { Command::ExportDocuments { index_name, ignore_vectors, offset } => {
export_documents(db_path, index_name, ignore_vectors) export_documents(db_path, index_name, ignore_vectors, offset)
} }
Command::OfflineUpgrade { target_version } => { Command::OfflineUpgrade { target_version } => {
let target_version = parse_version(&target_version).context("While parsing `--target-version`. Make sure `--target-version` is in the format MAJOR.MINOR.PATCH")?; let target_version = parse_version(&target_version).context("While parsing `--target-version`. Make sure `--target-version` is in the format MAJOR.MINOR.PATCH")?;
@@ -468,6 +472,7 @@ fn export_documents(
db_path: PathBuf, db_path: PathBuf,
index_name: String, index_name: String,
ignore_vectors: bool, ignore_vectors: bool,
offset: Option<usize>,
) -> anyhow::Result<()> { ) -> anyhow::Result<()> {
let index_scheduler_path = db_path.join("tasks"); let index_scheduler_path = db_path.join("tasks");
let env = unsafe { EnvOpenOptions::new().max_dbs(100).open(&index_scheduler_path) } let env = unsafe { EnvOpenOptions::new().max_dbs(100).open(&index_scheduler_path) }
@@ -491,8 +496,12 @@ fn export_documents(
let all_fields: Vec<_> = fields_ids_map.iter().map(|(id, _)| id).collect(); let all_fields: Vec<_> = fields_ids_map.iter().map(|(id, _)| id).collect();
let embedding_configs = index.embedding_configs(&rtxn)?; let embedding_configs = index.embedding_configs(&rtxn)?;
if let Some(offset) = offset {
eprintln!("Skipping {offset} documents");
}
let mut stdout = BufWriter::new(std::io::stdout()); let mut stdout = BufWriter::new(std::io::stdout());
for ret in index.all_documents(&rtxn)? { for ret in index.all_documents(&rtxn)?.skip(offset.unwrap_or(0)) {
let (id, doc) = ret?; let (id, doc) = ret?;
let mut document = obkv_to_json(&all_fields, &fields_ids_map, doc)?; let mut document = obkv_to_json(&all_fields, &fields_ids_map, doc)?;