copy_span v0.1.0

New Features

• introduce sliced! syntax for processing with anonymous ticks

Bug Fixes

• Add CHANGELOG.md

Commit Statistics

• 2 commits contributed to the release over the course of 17 calendar days.
• 2 commits were understood as conventional.
• 1 unique issue was worked on: #2256

Commit Details

view details

• #2256
  • Introduce sliced! syntax for processing with anonymous ticks (194280b)
• Uncategorized
  • Add CHANGELOG.md (bad2f84)

hydro_std v0.14.0

New Features

• Aggregate client throughput/latency
  Co-authored with @shadaj

  ---

• upgrade Stageleft to eliminate __staged compilation during development
  Before Stageleft 0.9, we always compiled the __staged module in stage
  0, which resulted in significant compilation penalties and Rust Analyzer
  thrashing since any file changes triggered a re-run of the build.rs.
  With Stageleft 0.9, we can defer compiling this module to the trybuild
  stage 1.

  Stageleft 0.9 also cleans up how paths are rewritten to use the
  __staged module, so we can simplify our logic as well. The only
  significant rewrite remaining is when running unit tests, where we have
  to regenerate __staged to access test-only module, and therefore have
  to rewrite all paths to use that module.

  Finally, in the spirit of improving compilation efficiency, we disable
  incremental builds for trybuild stage 1. We generate files with hash
  based on contents, so we were never benefitting from incremental
  compilation anyways. This reduces the disk space used significantly.

Refactor

• use async-ssh2-russh (instead of libssh2 bindings), fix #1463

New Features (BREAKING)

• add stream markers for tracking non-deterministic retries
  This introduces an additional type paramter to Stream called
  Retries, which tracks the presence (or lack) of non-determinstic
  retries in the stream. ExactlyOnce means that each element has
  deterministic order, while AtLeastOnce means that there may be
  non-deterministic duplicates.

  A TotalOrder, AtLeastOnce stream describes elements with consecutive
  duplication, but deterministic order if we ignore those immediate
  elements. A NoOrder, AtLeastOnce stream has set semantics.

  Also fixes a bug in the return type for *_keyed_*, where the output
  type was previously TotalOrder but now is NoOrder. We stream the
  results of a keyed aggregation out of a HashMap, so the order will
  indeed be non-deterministic.

Commit Statistics

• 5 commits contributed to the release over the course of 91 calendar days.
• 4 commits were understood as conventional.
• 4 unique issues were worked on: #1803, #1900, #1907, #1910

Commit Details

view details

• #1803
  • Use async-ssh2-russh (instead of libssh2 bindings), fix #1463 (5ab815f)
• #1900
  • Aggregate client throughput/latency (17f4a83)
• #1907
  • Upgrade Stageleft to eliminate __staged compilation during development (b333b45)
• #1910
  • Add stream markers for tracking non-deterministic retries (45bd6e9)
• Uncategorized
  • Release dfir_lang v0.14.0, dfir_macro v0.14.0, hydro_deploy_integration v0.14.0, lattices_macro v0.5.10, variadics_macro v0.6.1, dfir_rs v0.14.0, hydro_deploy v0.14.0, hydro_lang v0.14.0, hydro_optimize v0.13.0, hydro_std v0.14.0, safety bump 6 crates (0683595)

hydro_optimize v0.13.0

Chore

• clean up dependencies
  Using hydro_optimize as a regular dependency in hydro_test results
  in leaking many dependencies including hydro_deploy, so this moves it
  to a dev-dependency

New Features

• improve logging for profiling
• Use partitioning analysis results to partition
  Test/insta changes stem from changed implementation of
  broadcast_bincode, will change again once #1949 is implemented.
  Also added missing cases for Persists hidden behind CrossProduct,
  Difference, AntiJoin, Join, and Scan for decoupler.
• Remove commercial ilp
• Partitioning analysis
  Integrating with the partitioner next
• capture stack traces for each IR node
  Because Hydro is staged, the stack traces capture the structure of the
  program, which is helpful for profiling / visualization.
• add scan operator
• Decoupling analysis
  A Gurobi license is required to run code that uses hydro_optimize (for ILP over decoupling decisions)

Bug Fixes

• don't snapshot-test backtraces
  Backtraces aren't stable across Unix / Windows. Just have a separate
  test for them.

  Also defers resolution of backtraces until we actually need them to
  improve performance.

Refactor

• separate externals from other location kinds, clean up network operators
  First, we remove externals from LocationId, to ensure that a
  LocationId only is used for locations where we can concretely place
  compiled logic. This simplifes a lot of pattern matching where we wanted
  to disallow externals.

  Keys on network inputs / outputs (from_key and to_key) are only
  relevant to external networking. We extract the logic to instantiate
  external network collections, so that the core logic does not need to
  deal with keys.

Refactor (BREAKING)

• invert external sources and clean up locations in IR
  First, instead of creating external sources by invoking
  external.source_bincode_external(&p), we switch the API to
  p.source_bincode_external(&external) for symmetry with source_iter
  and source_stream.

  The other, much larger change is to clean up how the IR handles external
  inputs and outputs and keeps track of locations. First, we introduce
  HydroNode::ExternalInput and HydroLeaf::SendExternal as specialized
  nodes for these, so that we no longer create dummy sources / sinks.

  Then, we eliminate places where we have multiple sources of truth for
  where the output of an IR node is located, by instead referring to the
  metadata. Because it is easy in optimizer rewrites to corrupt this
  metadata, we also add a flag to transform_bottom_up that lets
  developers enable a metadata validity check. We disable it in most
  transformations for performance, but enable it in the decoupling
  rewrites since it manipulates locations in complex ways.

Commit Statistics

• 12 commits contributed to the release over the course of 12 calendar days.
• 11 commits were understood as conventional.
• 11 unique issues were worked on: #1859, #1930, #1934, #1935, #1937, #1940, #1947, #1952, #1955, #1958, #1962

Commit Details

view details

• #1859
  • Decoupling analysis (99a8f1d)
• #1930
  • Add scan operator (c4b9590)
• #1934
  • Clean up dependencies (edab6c2)
• #1935
  • Partitioning analysis (3b013ac)
• #1937
  • Capture stack traces for each IR node (d44b225)
• #1940
  • Remove commercial ilp (dfaf517)
• #1947
  • Don't snapshot-test backtraces (739b622)
• #1952
  • Use partitioning analysis results to partition (173d9c0)
• #1955
  • Improve logging for profiling (0e6403c)
• #1958
  • Invert external sources and clean up locations in IR (fb016b0)
• #1962
  • Separate externals from other location kinds, clean up network operators (22a7d0d)
• Uncategorized
  • Release dfir_lang v0.14.0, dfir_macro v0.14.0, hydro_deploy_integration v0.14.0, lattices_macro v0.5.10, variadics_macro v0.6.1, dfir_rs v0.14.0, hydro_deploy v0.14.0, hydro_lang v0.14.0, hydro_optimize v0.13.0, hydro_std v0.14.0, safety bump 6 crates (0683595)

hydro_lang v0.14.0

Chore

• enable viz feature only in dev-dependencies
  Reduces compilation burden when running hydro_test examples.
• update proc-macro-crate
• update pinned nightly to 2025-04-27, update span API usage

New Features

• Use partitioning analysis results to partition
  Test/insta changes stem from changed implementation of
  broadcast_bincode, will change again once #1949 is implemented.
  Also added missing cases for Persists hidden behind CrossProduct,
  Difference, AntiJoin, Join, and Scan for decoupler.

• make it easier to open trybuild-generated files with Rust Analyzer
  To fully compile the generated sources without error, the Stageleft
  environment variable needt to be passed to build scripts, so
  pre-configure that in workspace settings.

• graph viz for Hydro lang

• allow running generated binaries with single-threaded Tokio runtime
  Before, we had a janky architecture for establishing network connections
  which relied on blocking on futures outside an async context, which
  required a multi-threaded runtime. Now, we establish all connections
  before launching the DFIR code, so that no blocking is required there.

• capture stack traces for each IR node
  Because Hydro is staged, the stack traces capture the structure of the
  program, which is helpful for profiling / visualization.

• add scan operator

• Decoupling analysis
  A Gurobi license is required to run code that uses hydro_optimize (for ILP over decoupling decisions)

• add *_idempotent variations for (keyed) fold / reduce
  We were missing variants where the stream is totally ordered with
  consecutive retries, which show up when sampling singletons. This adds
  those in.

• upgrade Stageleft to eliminate __staged compilation during development
  Before Stageleft 0.9, we always compiled the __staged module in stage
  0, which resulted in significant compilation penalties and Rust Analyzer
  thrashing since any file changes triggered a re-run of the build.rs.
  With Stageleft 0.9, we can defer compiling this module to the trybuild
  stage 1.

  Stageleft 0.9 also cleans up how paths are rewritten to use the
  __staged module, so we can simplify our logic as well. The only
  significant rewrite remaining is when running unit tests, where we have
  to regenerate __staged to access test-only module, and therefore have
  to rewrite all paths to use that module.

  Finally, in the spirit of improving compilation efficiency, we disable
  incremental builds for trybuild stage 1. We generate files with hash
  based on contents, so we were never benefitting from incremental
  compilation anyways. This reduces the disk space used significantly.

• Assign cycles IDs that are globally unique (across clusters/processes)
  Co-authored with @shadaj

Bug Fixes

• emit appropriate Persist / Unpersist nodes for scan

• ensure that singletons and optionals always have cardinality 1 inside a tick
  Otherwise, cycling a singleton can result in a memory leak, as we found
  in PBFT implementation. Now, we strictly require that inside a tick,
  singletons/optionals are represented by a stream with 1 or fewer
  elements.

• don't snapshot-test backtraces
  Backtraces aren't stable across Unix / Windows. Just have a separate
  test for them.

  Also defers resolution of backtraces until we actually need them to
  improve performance.

• correctly enable staged-trybuild mode when cross-compiling
  RUSTFLAGS are not passed to build scripts, use a regular environment
  variable instead. Should also dramatically improve cache hit rate for
  sccache since the rustflags for non-stageleft crates are untouched.

Refactor

• separate externals from other location kinds, clean up network operators
  First, we remove externals from LocationId, to ensure that a
  LocationId only is used for locations where we can concretely place
  compiled logic. This simplifes a lot of pattern matching where we wanted
  to disallow externals.

  Keys on network inputs / outputs (from_key and to_key) are only
  relevant to external networking. We extract the logic to instantiate
  external network collections, so that the core logic does not need to
  deal with keys.

• rename ExternalProcess to External
  In preparation of supporting multi-connection source / sink pairs. There
  is no need to distinguish a single external from multiple at the
  location level, instead we will have separate APIs for declaring a
  single vs multi connection input/output.

• minimize Tokio feature flags
  Now that hydro_lang no longer needs multi-threaded runtime, we can
  eliminate it from the features used in trybuild compilation. Minimizes
  Tokio features elsewhere too.

• migrate tests from hydro_test_local

• use async-ssh2-russh (instead of libssh2 bindings), fix #1463

New Features (BREAKING)

• add stream markers for tracking non-deterministic retries
  This introduces an additional type paramter to Stream called
  Retries, which tracks the presence (or lack) of non-determinstic
  retries in the stream. ExactlyOnce means that each element has
  deterministic order, while AtLeastOnce means that there may be
  non-deterministic duplicates.

  A TotalOrder, AtLeastOnce stream describes elements with consecutive
  duplication, but deterministic order if we ignore those immediate
  elements. A NoOrder, AtLeastOnce stream has set semantics.

  Also fixes a bug in the return type for *_keyed_*, where the output
  type was previously TotalOrder but now is NoOrder. We stream the
  results of a keyed aggregation out of a HashMap, so the order will
  indeed be non-deterministic.

Refactor (BREAKING)

• invert external sources and clean up locations in IR
  First, instead of creating external sources by invoking
  external.source_bincode_external(&p), we switch the API to
  p.source_bincode_external(&external) for symmetry with source_iter
  and source_stream.

  The other, much larger change is to clean up how the IR handles external
  inputs and outputs and keeps track of locations. First, we introduce
  HydroNode::ExternalInput and HydroLeaf::SendExternal as specialized
  nodes for these, so that we no longer create dummy sources / sinks.

  Then, we eliminate places where we have multiple sources of truth for
  where the output of an IR node is located, by instead referring to the
  metadata. Because it is easy in optimizer rewrites to corrupt this
  metadata, we also add a flag to transform_bottom_up that lets
  developers enable a metadata validity check. We disable it in most
  transformations for performance, but enable it in the decoupling
  rewrites since it manipulates locations in complex ways.

• remove support for macro entrypoints

Commit Statistics

• 26 commits contributed to the release over the course of 92 calendar days.
• 25 commits were understood as conventional.
• 25 unique issues were worked on: #1803, #1843, #1859, #1902, #1905, #1907, #1909, #1910, #1916, #1930, #1936, #1937, #1938, #1939, #1941, #1944, #1945, #1947, #1948, #1951, #1952, #1956, #1958, #1959, #1962

Commit Details

view details

• #1803
  • Use async-ssh2-russh (instead of libssh2 bindings), fix #1463 (5ab815f)
• #1843
  • Update pinned nightly to 2025-04-27, update span API usage (98baec7)
• #1859
  • Decoupling analysis (99a8f1d)
• #1902
  • Assign cycles IDs that are globally unique (across clusters/processes) (863cb9e)
• #1905
  • Migrate tests from hydro_test_local (eaac1f4)
• #1907
  • Upgrade Stageleft to eliminate __staged compilation during development (b333b45)
• #1909
  • Remove support for macro entrypoints (6e29285)
• #1910
  • Add stream markers for tracking non-deterministic retries (45bd6e9)
• #1916
  • Add *_idempotent variations for (keyed) fold / reduce (6b0483d)
• #1930
  • Add scan operator (c4b9590)
• #1936
  • Graph viz for Hydro lang (4d15ff1)
• #1937
  • Capture stack traces for each IR node (d44b225)
• #1938
  • Allow running generated binaries with single-threaded Tokio runtime (bd1afdf)
• #1939
  • Minimize Tokio feature flags (59041df)
• #1941
  • Correctly enable staged-trybuild mode when cross-compiling (6699197)
• #1944
  • Update proc-macro-crate (3d40d1a)
• #1945
  • Make it easier to open trybuild-generated files with Rust Analyzer (4035cae)
• #1947
  • Don't snapshot-test backtraces (739b622)
• #1948
  • Ensure that singletons and optionals always have cardinality 1 inside a tick (8858abd)
• #1951
  • Enable viz feature only in dev-dependencies (a3280d9)
• #1952
  • Use partitioning analysis results to partition (173d9c0)
• #1956
  • Rename ExternalProcess to External (49c1918)
• #1958
  • Invert external sources and clean up locations in IR (fb016b0)
• #1959
  • Emit appropriate Persist / Unpersist nodes for scan (1fc9f0d)
• #1962
  • Separate externals from other location kinds, clean up network operators (22a7d0d)
• Uncategorized
  • Release dfir_lang v0.14.0, dfir_macro v0.14.0, hydro_deploy_integration v0.14.0, lattices_macro v0.5.10, variadi...

hydro_deploy v0.14.0

Documentation

• add basic hydro_deploy, tracing docs, fix #1205
  Also removes extensions from links, for simplicity.

New Features

• improve logging for profiling

• allow running generated binaries with single-threaded Tokio runtime
  Before, we had a janky architecture for establishing network connections
  which relied on blocking on futures outside an async context, which
  required a multi-threaded runtime. Now, we establish all connections
  before launching the DFIR code, so that no blocking is required there.

• Decoupling analysis
  A Gurobi license is required to run code that uses hydro_optimize (for ILP over decoupling decisions)

• upgrade Stageleft to eliminate __staged compilation during development
  Before Stageleft 0.9, we always compiled the __staged module in stage
  0, which resulted in significant compilation penalties and Rust Analyzer
  thrashing since any file changes triggered a re-run of the build.rs.
  With Stageleft 0.9, we can defer compiling this module to the trybuild
  stage 1.

  Stageleft 0.9 also cleans up how paths are rewritten to use the
  __staged module, so we can simplify our logic as well. The only
  significant rewrite remaining is when running unit tests, where we have
  to regenerate __staged to access test-only module, and therefore have
  to rewrite all paths to use that module.

  Finally, in the spirit of improving compilation efficiency, we disable
  incremental builds for trybuild stage 1. We generate files with hash
  based on contents, so we were never benefitting from incremental
  compilation anyways. This reduces the disk space used significantly.

• Allow VM names to be customized to ease debugging
  Co-authored with @shadaj

• update how progress is displayed, fix #1415

Bug Fixes

• VM names violating GCP's regex
• use --target-dir instead of environment variable to improve caching
  sccache includes all environment variables starting with CARGO_ in the
  cache key, so this would cause misses for all trybuild compilation.
  Along with mozilla/sccache#2424, this improves
  compilation caching.
• correctly enable staged-trybuild mode when cross-compiling
  RUSTFLAGS are not passed to build scripts, use a regular environment
  variable instead. Should also dramatically improve cache hit rate for
  sccache since the rustflags for non-stageleft crates are untouched.

Other

• remove hydro_cli to fix build on AL2

Refactor

• Encapsulate stdout/stderr handling in new PriorityBroadcast type, fix #1357
• use blake3 hash intead of random for build unique_id, fix #1337
• use async-ssh2-russh (instead of libssh2 bindings), fix #1463

New Features (BREAKING)

• re-add loop lifetimes for anti_join_multiset, tests, remove MonotonicMap, fix #1830, fix #1823
  Redo of #1835

  Also updates path of trybuild errors to allow them to be clicked in the
  IDE

  ---

  Previous commit:

  Also implements loop lifetimes for difference_multiset which uses the
  anti_join_multiset codegen.

  Updates tests for difference, difference_multiset, anti_join, and
  anti_join_multiset

Refactor (BREAKING)

• use direct &dyn Any upcasting for Rust 1.86, update pyo3, fix #1821

Commit Statistics

• 17 commits contributed to the release over the course of 91 calendar days.
• 16 commits were understood as conventional.
• 16 unique issues were worked on: #1803, #1825, #1844, #1845, #1849, #1856, #1859, #1901, #1907, #1911, #1918, #1938, #1941, #1943, #1955, #1961

Commit Details

view details

• #1803
  • Use async-ssh2-russh (instead of libssh2 bindings), fix #1463 (5ab815f)
• #1825
  • Use direct &dyn Any upcasting for Rust 1.86, update pyo3, fix #1821 (de6c8ce)
• #1844
  • Use blake3 hash intead of random for build unique_id, fix #1337 (b39aa41)
• #1845
  • Update how progress is displayed, fix #1415 (6ba3b33)
• #1849
  • Encapsulate stdout/stderr handling in new PriorityBroadcast type, fix #1357 (c983f1f)
• #1856
  • Add basic hydro_deploy, tracing docs, fix #1205 (fb7d45d)
• #1859
  • Decoupling analysis (99a8f1d)
• #1901
  • Allow VM names to be customized to ease debugging (8705f97)
• #1907
  • Upgrade Stageleft to eliminate __staged compilation during development (b333b45)
• #1911
  • Re-add loop lifetimes for anti_join_multiset, tests, remove MonotonicMap, fix #1830, fix #1823 (d6ae619)
• #1918
  • Remove hydro_cli to fix build on AL2 (555b83e)
• #1938
  • Allow running generated binaries with single-threaded Tokio runtime (bd1afdf)
• #1941
  • Correctly enable staged-trybuild mode when cross-compiling (6699197)
• #1943
  • Use --target-dir instead of environment variable to improve caching (bea805d)
• #1955
  • Improve logging for profiling (0e6403c)
• #1961
  • VM names violating GCP's regex (3a3ce7f)
• Uncategorized
  • Release dfir_lang v0.14.0, dfir_macro v0.14.0, hydro_deploy_integration v0.14.0, lattices_macro v0.5.10, variadics_macro v0.6.1, dfir_rs v0.14.0, hydro_deploy v0.14.0, hydro_lang v0.14.0, hydro_optimize v0.13.0, hydro_std v0.14.0, safety bump 6 crates (0683595)

example_test v0.0.0

Refactor

• move example testing code into separate crate
  To prep for testing of hydro_deploy #1374 #1810

Test

• test some hydro examples on localhost, fix #1374

Commit Statistics

• 2 commits contributed to the release over the course of 69 calendar days.
• 2 commits were understood as conventional.
• 2 unique issues were worked on: #1847, #1848

Commit Details

view details

• #1847
  • Move example testing code into separate crate (cb54ace)
• #1848
  • Test some hydro examples on localhost, fix #1374 (c3ccee6)

dfir_rs v0.14.0

Documentation

• add note to install nodejs

New Features

• allow running generated binaries with single-threaded Tokio runtime
  Before, we had a janky architecture for establishing network connections
  which relied on blocking on futures outside an async context, which
  required a multi-threaded runtime. Now, we establish all connections
  before launching the DFIR code, so that no blocking is required there.
• add scan operator
• Decoupling analysis
  A Gurobi license is required to run code that uses hydro_optimize (for ILP over decoupling decisions)

Bug Fixes

• Revert anti join allocation
  Added unit test for Paxos compilation and non-negative throughtput
• add type arguments to anti_join_multiset, difference_multiset to mitigate #1857
• workaround to publish example_test

Refactor

• minimize Tokio feature flags
  Now that hydro_lang no longer needs multi-threaded runtime, we can
  eliminate it from the features used in trybuild compilation. Minimizes
  Tokio features elsewhere too.
• move example testing code into separate crate
  To prep for testing of hydro_deploy #1374 #1810

Test

• test some hydro examples on localhost, fix #1374

Chore (BREAKING)

• move datalog from repo, remove datalog playground from web, #1809
  #1809

  moved to https://github.com/hydro-project/dfir-datalog

  tests moved in hydro-project/dfir-datalog#1

  Removes dedalus examples in hydro_cli_examples

  Changes pinned nightly rust version from 2024-04-05 to 2024-04-04 as the
  former did not have intel mac support.

New Features (BREAKING)

• re-add loop lifetimes for anti_join_multiset, tests, remove MonotonicMap, fix #1830, fix #1823
  Redo of #1835

  Also updates path of trybuild errors to allow them to be clicked in the
  IDE

  ---

  Previous commit:

  Also implements loop lifetimes for difference_multiset which uses the
  anti_join_multiset codegen.

  Updates tests for difference, difference_multiset, anti_join, and
  anti_join_multiset

• display loops in graph visualizations, refactor, fix #1699
  Adds loops to display, new GraphWrite.no_loops option.

  Refactors how the heirarchy of GraphWrite items is handled to be
  simpler.

Refactor (BREAKING)

• use direct &dyn Any upcasting for Rust 1.86, update pyo3, fix #1821

Commit Statistics

• 15 commits contributed to the release over the course of 93 calendar days.
• 14 commits were understood as conventional.
• 13 unique issues were worked on: #1825, #1837, #1847, #1848, #1851, #1858, #1859, #1860, #1911, #1912, #1929, #1938, #1939

Commit Details

view details

• #1825
  • Use direct &dyn Any upcasting for Rust 1.86, update pyo3, fix #1821 (de6c8ce)
• #1837
  • Move datalog from repo, remove datalog playground from web, #1809 (09fc5f3)
• #1847
  • Move example testing code into separate crate (cb54ace)
• #1848
  • Test some hydro examples on localhost, fix #1374 (c3ccee6)
• #1851
  • Display loops in graph visualizations, refactor, fix #1699 (9bb9d1f)
• #1858
  • Add type arguments to anti_join_multiset, difference_multiset to mitigate #1857 (0d841a5)
• #1859
  • Decoupling analysis (99a8f1d)
• #1860
  • Revert anti join allocation (5b5bbe5)
• #1911
  • Re-add loop lifetimes for anti_join_multiset, tests, remove MonotonicMap, fix #1830, fix #1823 (d6ae619)
• #1912
  • Add note to install nodejs (ec1d8a0)
• #1929
  • Add scan operator (b58dfc8)
• #1938
  • Allow running generated binaries with single-threaded Tokio runtime (bd1afdf)
• #1939
  • Minimize Tokio feature flags (59041df)
• Uncategorized
  • Workaround to publish example_test (96ec97a)
  • Release dfir_lang v0.14.0, dfir_macro v0.14.0, hydro_deploy_integration v0.14.0, lattices_macro v0.5.10, variadics_macro v0.6.1, dfir_rs v0.14.0, hydro_deploy v0.14.0, hydro_lang v0.14.0, hydro_optimize v0.13.0, hydro_std v0.14.0, safety bump 6 crates (0683595)

variadics_macro v0.6.1

Chore

• update proc-macro-crate

Commit Statistics

• 1 commit contributed to the release over the course of 5 calendar days.
• 1 commit was understood as conventional.
• 1 unique issue was worked on: #1944

Commit Details

view details

• #1944
  • Update proc-macro-crate (3d40d1a)

lattices_macro v0.5.10

Chore

• update proc-macro-crate

Commit Statistics

• 1 commit contributed to the release over the course of 5 calendar days.
• 1 commit was understood as conventional.
• 1 unique issue was worked on: #1944

Commit Details

view details

• #1944
  • Update proc-macro-crate (3d40d1a)

hydro_deploy_integration v0.14.0

New Features

• allow running generated binaries with single-threaded Tokio runtime
  Before, we had a janky architecture for establishing network connections
  which relied on blocking on futures outside an async context, which
  required a multi-threaded runtime. Now, we establish all connections
  before launching the DFIR code, so that no blocking is required there.

Bug Fixes

• leftover logging when setting up Unix sockets
  Oops!

Refactor

• minimize Tokio feature flags
  Now that hydro_lang no longer needs multi-threaded runtime, we can
  eliminate it from the features used in trybuild compilation. Minimizes
  Tokio features elsewhere too.
• eliminate pin-project proc macro dependency
  This was the only use of the proc-macro version along the Hydro
  dependencies, we can just use the declarative macro version.

Commit Statistics

• 4 commits contributed to the release over the course of 8 calendar days.
• 4 commits were understood as conventional.
• 4 unique issues were worked on: #1933, #1938, #1939, #1963

Commit Details

view details

• #1933
  • Eliminate pin-project proc macro dependency (903fbda)
• #1938
  • Allow running generated binaries with single-threaded Tokio runtime (bd1afdf)
• #1939
  • Minimize Tokio feature flags (59041df)
• #1963
  • Leftover logging when setting up Unix sockets (ca704e5)