0.1.0

• initial release

0.1.1

• fix inline definition of on-realized, which would cause (on-realized form a b) to execute form twice
• fix coercion support for java.util.Iterator
• don't automatically coerce core.async channels to deferreds (use (take! (->source chan)) instead)
• add coercion support for Java 8 BasicStream and CompletableFuture, which can be optionally disabled
• add onto method to manifold.stream to mirror the one in manifold.deferred
• add formal, configurable execution model

0.1.2

• fix lifecycle for batch and throttle when the source is a permanent stream
• fix path where manifold.stream/reduce could fail to yield any value when the reducer function throws an exception, rather than yielding that error
• add mock-clock and with-clock to manifold.time, to aid with testing timeouts and other wall-clock behavior
• add consume-async method, which expects the consume callback to return a deferred that yields a boolean, rather than simply a boolean value
• small corrections and clarifications to doc-strings

0.1.3

• Target latest Dirigiste, which is no longer compiled using JDK 8 byte code.

0.1.4

• Honor :thread-factory parameter in manifold.executor.

0.1.5

Thanks to Tsutomu Yano and Joshua Griffith

• fix bugs in finally and consume

0.1.6

Thanks to Vadim Platonov, Miikka Koskinen, Alex Engelberg, and Oleh Palianytsia

• fix bug in batch
• make reduce compatible with Clojure's reduced short-circuiting
• make sure catch can match non-Throwable errors
• allow for destructuring in loop
• add alt mechanism for choosing the first of many deferreds to be realized

0.1.7

Thanks to Ted Cushman, Vadim Platonov

• Increase stack size in the wait-pool
• Fix lifecycle bugs in throttle, partition-all, and transform
• Change let-flow to wait on all deferred values, not just the ones used by the body

0.1.8

Thanks to Paweł Stroiński

• Fix handling of non-Throwable deferred errors when dereferencing

0.1.9

• Yummly managed version
• Migrate to clj-deps
• fix bug waiting on local variables not referenced in let-flow
• let-flow doesn't resolve deferred args to alt, zip, finally, catch, let-flow, and all of their ' variants
• let-flow to use bound-fn instead of fn so dynamic variables aren't lost
• let-flow to require unique let binding names. Names are required to be unique or the results are non-deterministic
• introduce new tsasvla namespace and macro as an alternative to let-flow

0.1.10

• fix tsasvla bug causing duplicated execution after taking from an already realized value