Skip to content

Releases: korpling/graphANNIS

2.4.7

23 Oct 13:09

Choose a tag to compare

Fixed

  • When an optional node (for negation without existence) was not at the end of
    the query, find queries could give an empty output (#267).
  • Create default components for the graph type when importing GraphML files.

2.4.6

26 Jul 13:22

Choose a tag to compare

Changed

  • Compile release for macOS on version 11 (Big Sur). This is necessary, since
    GitHub actions deprecated the older macOS version.

2.4.5

26 Jul 09:57

Choose a tag to compare

Changed

  • Compile releases on Ubuntu 20.04 instead of 18.04, which means the minimal
    GLIBC version is 2.31. This is necessary, since GitHub actions deprecated this
    Ubuntu version.

Fixed

  • Update quick-xml to version 0.28 to avoid issues in future Rust versions
  • Update sstable to version 0.11 to avoid issues in future Rust versions
  • Update actix-web to version 4 to avoid issues in future Rust versions
  • Update config crate to version 0.13 to avoid issues in future Rust versions
  • Update diesel to version 2.0 due to issue in sqlite dependency

2.4.4

19 Apr 10:02

Choose a tag to compare

Fixed

  • Importing a corpus with a relative path directly under the current working
    directory would fail if the corpus has linked files.
  • Output of data items in GraphML for node/edge annotations could be unordered
    and cause test failures if comparing GraphML files.

2.4.3

15 Mar 14:50

Choose a tag to compare

Fixed

  • Update smartstring crate to version 1 to avoid issues with newer Rust
    versions.

2.4.2

06 Feb 12:16
v2.4.2

Choose a tag to compare

Fixed

  • After re-using a deleted symbol ID (used in the annotation storage), the
    retrieved value was empty.

2.4.1

30 Sep 11:16
v2.4.1

Choose a tag to compare

Fixed

  • When importing relANNIS corpora with sub-corpora, add the PartOf edge to the
    parent corpus node of the document or sub-corpora, but not automatically to
    the top-level corpus.

2.4.0

22 Sep 10:53
v2.4.0

Choose a tag to compare

Added

  • Allow to configure how spans should be interpreted in the view when the token
    layer is representing a timeline with the timeline_strategy parameter in the
    view section of the corpus configuration. This allows the view to
    reconstruct an implicit relation between spans and their segmentation nodes
    (which is not possible to represent in the legacy relANNIS data model). New
    corpora should use explicit Coverage edges between spans and their
    segmentation nodes, but in order to maintain backward compatibility with
    relANNIS, we need to support these older corpus configuration values
    (virtual_tokenization_mapping and virtual_tokenization_from_namespace),
    which only affect the display of the corpora.

2.3.0

06 Sep 14:33
v2.3.0

Choose a tag to compare

Fixed

  • Fixed wrong result order for non-token searches.
  • Estimation for negated regex was extremely off when the regex could possibly
    match all values. This caused problematic query plans including those with
    nested loop joins and long execution times.
  • Better estimation of result sizes for regular expressions with multiple
    prefixes.
  • Fix compilation issues in Rust projects that use the 2021 Rust edition.
    lalrpop/lalrpop#650
  • Faster subgraph generation for subgraph queries with context. The previous
    implementation used an AQL query that got quite complex over time and was
    difficult to execute. The new implemenation directly implements the logic
    using iterators. It also sorts the nodes in the iterator by the order of the
    node in the text.

Added

  • Add edges to the special Ordering/annis/datasource-gap between the last and
    first token of context regions in subgraph when the returned context regions
    do not overlap. This allows sorting the context regions that belong to the
    same data source but are not connected by ordinary Ordering/annis/ edges.

2.2.2

26 Jul 09:44

Choose a tag to compare

Fixed

  • Use external sorting for match results to avoid out of memory errors for large
    results.