zoekt/testdata/shards/ at aae71e599361f1d6ec09f68f2b299f51dd5e1dd6 · boltless.me/zoekt · Tangled

boltless.me / zoekt

0

fork of https://github.com/sourcegraph/zoekt

0

zoekt / testdata / shards /

at aae71e599361f1d6ec09f68f2b299f51dd5e1dd6 5 files

Stefan Hengl ranking: add document ranks to shards (#449) 3y ago

ctagsrepo_v16.00000.zoekt

ranking: add document ranks to shards (#449) We persist document ranks in the shards and sort file matches based on the rankings determined by the document ranks and match scores. Co-authored-by: Keegan Carruthers-Smith <keegan.csmith@gmail.com>

3 years ago

ctagsrepo_v17.00000.zoekt

all: compound shard support (#95) This commit adds support for compound shards. A shard now has multiple repositories associated with it, rather than always one. Most of zoekt is document based, so minimal changes are required in the core search evaluation codepath. The only change here is the addition of a mapping from document to repo and storing the subrepo paths per repo. The other change is the addition of tombstones. A tombstoned repository is hidden from List and Search results. This was added so we can index a new version of a repository in a compound shard without the need of recomputing the whole shard. This commit is mostly focussed on the read path. It ensures everything keeps working correctly once compound shards are introduced. However, the write path is mostly missing. We add a merge command for manual merging. However, zoekt-sourcegraph-indexserver is mostly unaware of compound shards and has no way to mutate them. This will be follow-up work. To support compound shards we had to bump the indexed format version. This is since the repoMetaData field is not backwards compatible. However, we know we plan on making other changes to the index format. So we introduced NextIndexFormatVersion. With this change we will continue to use v16, unless a common opts into v17. This will allow us to effectively feature flag the new format while we work on it. Co-authored-by: Stefan Hengl <stefan@sourcegraph.com>

4 years ago

repo17_v17.00000.zoekt

compute precise language information with go-enry for lang: queries (#220) use go-enry to compute more precise language information than ctags make lang: use filename fallback for older index versions

4 years ago

repo2_v16.00000.zoekt

ranking: add document ranks to shards (#449) We persist document ranks in the shards and sort file matches based on the rankings determined by the document ranks and match scores. Co-authored-by: Keegan Carruthers-Smith <keegan.csmith@gmail.com>

3 years ago

repo_v16.00000.zoekt

ranking: add document ranks to shards (#449) We persist document ranks in the shards and sort file matches based on the rankings determined by the document ranks and match scores. Co-authored-by: Keegan Carruthers-Smith <keegan.csmith@gmail.com>

3 years ago