diff options
author | Martin Czygan <martin.czygan@gmail.com> | 2021-09-23 21:52:20 +0200 |
---|---|---|
committer | Martin Czygan <martin.czygan@gmail.com> | 2021-09-23 21:52:20 +0200 |
commit | ed353fc849b1c6014b9cc91c1936729d1f79d8fb (patch) | |
tree | fbe86ed0f60e054e6ef444a9464a78ba6b4c4697 | |
parent | 3e2c21a396a0e1ce98a2fdb2de5018de0a380f71 (diff) | |
download | refcat-ed353fc849b1c6014b9cc91c1936729d1f79d8fb.tar.gz refcat-ed353fc849b1c6014b9cc91c1936729d1f79d8fb.zip |
misc: fix and improve comments
-rw-r--r-- | skate/cleanup.go | 7 | ||||
-rw-r--r-- | skate/reduce.go | 15 | ||||
-rw-r--r-- | skate/verify.go | 7 | ||||
-rw-r--r-- | skate/xio/util.go | 4 |
4 files changed, 28 insertions, 5 deletions
diff --git a/skate/cleanup.go b/skate/cleanup.go index c8ce6d1..4a10063 100644 --- a/skate/cleanup.go +++ b/skate/cleanup.go @@ -9,6 +9,8 @@ import ( "mvdan.cc/xurls/v2" ) +// TODO: This should be revisited entirely. + var rxRelaxed = xurls.Relaxed() // URLFilter is a line oriented URL filter. @@ -21,6 +23,7 @@ type FilterURL struct { AllowedSchemas []string } +// Run executes the filter on a blob of data, most likely a line. func (f *FilterURL) Run(p []byte) ([]byte, error) { parts := strings.Split(string(p), f.Delimiter) if len(parts) < f.Index { @@ -59,6 +62,7 @@ type FilterDOI struct { AllowedSchema []string } +// Run executes the filter on a blob of data, most likely a line. func (f *FilterDOI) Run(p []byte) ([]byte, error) { parts := strings.Split(string(p), f.Delimiter) if len(parts) < f.Index { @@ -78,8 +82,11 @@ func (f *FilterDOI) Run(p []byte) ([]byte, error) { return []byte(strings.Join(parts, f.Delimiter)), nil } +// FilterRawRef is an ad-hoc filter. type FilterRawRef struct{} +// Run executes the filter. TODO: Gather cleanup functions together and make +// them more easily shared. func (f *FilterRawRef) Run(p []byte) ([]byte, error) { var ref Ref if err := json.Unmarshal(p, &ref); err != nil { diff --git a/skate/reduce.go b/skate/reduce.go index 1100e2f..c68b2dd 100644 --- a/skate/reduce.go +++ b/skate/reduce.go @@ -17,6 +17,21 @@ // * [ ] switch to faster logging, e.g. zerolog, https://github.com/rs/zerolog#benchmarks // * [x] batch, parallelize // * [ ] unify flags to "-a", "-b" +// +// A couple more ideas to improve. +// +// * each reducer could be its own type instead of a function in order to allow +// for customizations, options, e.g. +// +// type ReduceExact struct { +// ReleasesReader io.Reader +// RefsReader io.Reader +// W io.Writer +// MatchResult MatchResult +// } +// +// func (r *ReduceExact) Run() error { ... } +// package skate import ( diff --git a/skate/verify.go b/skate/verify.go index 5df0a7e..c5fbeff 100644 --- a/skate/verify.go +++ b/skate/verify.go @@ -112,8 +112,7 @@ type MatchResult struct { Reason Reason } -// VerificationPair groups two identifiers and their match status and -// match reason. +// MatchPair groups two identifiers and their match status and match reason. type MatchPair struct { A string B string @@ -135,7 +134,9 @@ func JsonMarshalNewline(v interface{}) ([]byte, error) { return b, nil } -// Verify verifies two releases and will ignore short titles. +// Verify verifies two releases and will ignore short titles. TODO: We want +// this to be a type with configuration options and maybe some pluggable +// pieces, e.g. for author matching, etc. func Verify(a, b *Release) MatchResult { return VerifyMinTitleLength(a, b, 5) } diff --git a/skate/xio/util.go b/skate/xio/util.go index 10225a1..4416c9f 100644 --- a/skate/xio/util.go +++ b/skate/xio/util.go @@ -9,13 +9,13 @@ import ( "sync" ) -// SyncWriter makes any writer thread safe, using a lock. +// SyncWriter makes any writer thread-safe using a lock. type SyncWriter struct { sync.Mutex w io.Writer } -// NewSyncWriter returns an io.Writer that can be safely accessed by multiple +// NewSyncWriter returns an io.Writer that can be safely written to by multiple // goroutines. func NewSyncWriter(w io.Writer) *SyncWriter { return &SyncWriter{w: w} |