In this commit, all the sql, non-postgres-specific, code is moved out of
the postgres package and into a new sqlbase package. This will make it
more easily reusable for future sql integrations.
In this commit, changes are made to the `kvdb/postgres` package so that
all all the non-postgres-specific code is generalised to be applicable
for all sql code. A follow up commit will move all the general sql code
into its own package.
Due to the update of the kvdb package to go1.18, the `// +build`
directives are no longer required. They are removed in this commit in
order to simplify upcoming commits. Along the way, a few typos are fixed
and an unused struct is removed.
Bump the go version to go1.18 so that we can get rid of the '// +build'
comments in the upcoming commits where some of the '//go:build'
comments will become too complex to write as '// +build' comments.
Previously the assumption existed that panic data was already an error.
If that wasn't the case for example because panic("string") was used,
the transaction wasn't unlocked.
With go 1.17 a change to the build flags was implemented:
https://go.googlesource.com/proposal/+/master/design/draft-gobuild.md
The formatter now automatically adds the forward-compatible build tag
format and the linter checks for them, so we need to include them in our
code.
This commit extends the kvdb interface in a backwards compatible way
such that we'll be able to prefetch all keys in a bucket in one go reducing the
number of roundtrips.
This commit adds more full range caching to the STM such that we'll
be able to prefetch keys and ranges on the kvdb level. This will allow
us to reduce roundtrips to etcd do fast iteration of small ranges.
In this commit we add an extra config for enabling logging to an
external file when using embedded etcd. This can be useful when running
integration tests to see more details about etcd related issues.
This commit builds on the ideas of @cfromknecht in lnd/5153. The
addition is that the design is now simpler and more robust by queueing
up everything, but allowing maximal parallelism where txns don't block.
Furthermore the commit makes CommitQueue.Done() private essentially
removing the need to understand the queue externally.
This commit fixes an issue where subsequent transaction retries may have
changed the read/write sets inside the STM which in turn left junk
references to these keys in the CommitQueue. The left keys potentially
conflicted with subsequent transactions, queueing them up causing
througput degradation.
