searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-12-26 06:30:05 +01:00
Code Issues Releases Wiki Activity

Add the words grouped by four positions in the infos binary

Browse Source
This commit is contained in:
Clément Renault 2020-08-29 18:12:31 +02:00
parent ad5cafbfed
commit 605f75b56f
No known key found for this signature in database
GPG Key ID: 92ADA4E935E71FA4
1 changed files with 16 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
src/bin/infos.rs
View File

@ -78,7 +78,7 @@ enum Command {
full_display: bool, full_display: bool,
/// The words you want to display the values of. /// The words you want to display the values of.
words: Vec<String>, words: Vec<String>,
} },
} }
fn main() -> anyhow::Result<()> { fn main() -> anyhow::Result<()> {
@ -190,6 +190,7 @@ fn biggest_value_sizes(index: &Index, rtxn: &heed::RoTxn, limit: usize) -> anyho
let main_name = "main"; let main_name = "main";
let word_positions_name = "word_positions"; let word_positions_name = "word_positions";
let word_position_docids_name = "word_position_docids"; let word_position_docids_name = "word_position_docids";
let word_four_positions_docids_name = "word_four_positions_docids";
let word_attribute_docids_name = "word_attribute_docids"; let word_attribute_docids_name = "word_attribute_docids";
let mut heap = BinaryHeap::with_capacity(limit + 1); let mut heap = BinaryHeap::with_capacity(limit + 1);
@ -200,11 +201,16 @@ fn biggest_value_sizes(index: &Index, rtxn: &heed::RoTxn, limit: usize) -> anyho
if heap.len() > limit { heap.pop(); } if heap.len() > limit { heap.pop(); }
} }
if let Some(documents) = index.main.get::<_, ByteSlice, ByteSlice>(rtxn, b"documents")? { if let Some(documents) = index.main.get::<_, Str, ByteSlice>(rtxn, "documents")? {
heap.push(Reverse((documents.len(), format!("documents"), main_name))); heap.push(Reverse((documents.len(), format!("documents"), main_name)));
if heap.len() > limit { heap.pop(); } if heap.len() > limit { heap.pop(); }
} }
if let Some(documents_ids) = index.main.get::<_, Str, ByteSlice>(rtxn, "documents-ids")? {
heap.push(Reverse((documents_ids.len(), format!("documents-ids"), main_name)));
if heap.len() > limit { heap.pop(); }
}
for result in index.word_positions.as_polymorph().iter::<_, Str, ByteSlice>(rtxn)? { for result in index.word_positions.as_polymorph().iter::<_, Str, ByteSlice>(rtxn)? {
let (word, value) = result?; let (word, value) = result?;
heap.push(Reverse((value.len(), word.to_string(), word_positions_name))); heap.push(Reverse((value.len(), word.to_string(), word_positions_name)));
@ -219,6 +225,14 @@ fn biggest_value_sizes(index: &Index, rtxn: &heed::RoTxn, limit: usize) -> anyho
if heap.len() > limit { heap.pop(); } if heap.len() > limit { heap.pop(); }
} }
for result in index.word_four_positions_docids.as_polymorph().iter::<_, ByteSlice, ByteSlice>(rtxn)? {
let (key_bytes, value) = result?;
let (word, lower_position) = StrBEU32Codec::bytes_decode(key_bytes).unwrap();
let key = format!("{} {}..{}", word, lower_position, lower_position + 4);
heap.push(Reverse((value.len(), key, word_four_positions_docids_name)));
if heap.len() > limit { heap.pop(); }
}
for result in index.word_attribute_docids.as_polymorph().iter::<_, ByteSlice, ByteSlice>(rtxn)? { for result in index.word_attribute_docids.as_polymorph().iter::<_, ByteSlice, ByteSlice>(rtxn)? {
let (key_bytes, value) = result?; let (key_bytes, value) = result?;
let (word, attribute) = StrBEU32Codec::bytes_decode(key_bytes).unwrap(); let (word, attribute) = StrBEU32Codec::bytes_decode(key_bytes).unwrap();