krino: matching — scan, ignore, conditions, extraction, duplicates, explain, dry run

1 files changed, 111 insertions, 0 deletions

diff --git a/internal/cond/types.go b/internal/cond/types.go
new file mode 100644
index 0000000..81c2f45
--- /dev/null
+++ b/internal/cond/types.go
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-3.0-or-later
+
+// Package cond compiles the s-expression conditions of a rule's (when ...)
+// into a tree that Task 8's evaluator walks against one file's facts.
+package cond
+
+import (
+	"regexp"
+	"time"
+
+	"krino/internal/sexp"
+)
+
+// Options carries a rule's resolved case and fold settings into compilation.
+type Options struct {
+	IgnoreCase bool // the rule's resolved case setting is "ignore"
+	Fold       bool // the rule's resolved fold setting
+}
+
+// Cond is a compiled condition. A Cond compiled from no conditions (a rule
+// without when) is always true.
+type Cond struct {
+	root        *node      // nil: always true
+	opt         Options    // the case/fold settings conditions were compiled with; Task 8 needs them again at eval time
+	UsesContent bool       // some content test exists
+	DupDirs     [][]string // the raw directory arguments of each duplicate test, in order
+}
+
+// kind is what a compiled node tests, or how it combines its children.
+type kind int
+
+const (
+	kAnd kind = iota
+	kOr
+	kNot
+	kType
+	kName
+	kPath
+	kContent
+	kSize
+	kAge
+	kDuplicate
+	kMatched
+)
+
+// Costs, per the brief's cost order: cheapest first when sorting and/or
+// children. not takes its child's cost; and/or take the sum of theirs.
+const (
+	costCheap     = 1 // type, size, age, matched
+	costRegex     = 2 // name, path
+	costDuplicate = 5
+	costContent   = 10
+)
+
+// pattern is one name/path regex, compiled and paired with the text it was
+// written as (undecorated by (?i) or folding), for labels and reasons.
+type pattern struct {
+	re  *regexp.Regexp
+	src string
+}
+
+// keyword is one content keyword, normalised for matching and paired with
+// the text it was written as, for labels and reasons.
+type keyword struct {
+	norm string
+	src  string
+}
+
+// node is one compiled condition: a leaf test, or an and/or/not combinator
+// over other nodes. Task 8 evaluates this tree.
+type node struct {
+	kind     kind
+	pos      sexp.Pos // the position of the node as written, for diagnostics
+	label    string   // the test as written, e.g. `content "acme ltd" "0000000000"`
+	cost     int      // this node's evaluation cost; and/or sort children by it
+	children []*node  // and, or, not
+
+	// type
+	suffixes []string // leading-dot, lower-case, e.g. ".pdf"
+
+	// name, path
+	patterns []pattern
+
+	// content
+	keywords []keyword
+
+	// size, age (kind tells which is populated)
+	op      string
+	sizeVal int64
+	ageVal  time.Duration
+
+	// duplicate
+	dirs []string
+}
+
+// groups maps a (type ...) group name to the extensions it expands to,
+// spec Appendix A.
+var groups = map[string][]string{
+	"image":        {"jpg", "jpeg", "png", "gif", "webp", "bmp", "tif", "tiff", "heic", "heif", "avif", "svg", "ico", "raw", "cr2", "nef", "arw", "dng"},
+	"video":        {"mp4", "mkv", "webm", "mov", "avi", "m4v", "mpg", "mpeg", "wmv", "flv", "3gp"},
+	"audio":        {"mp3", "flac", "ogg", "opus", "m4a", "aac", "wav", "wma", "aiff"},
+	"archive":      {"zip", "tar", "gz", "tgz", "bz2", "tbz2", "xz", "txz", "zst", "7z", "rar", "lz", "lzma", "cpio"},
+	"document":     {"pdf", "doc", "docx", "odt", "rtf", "txt", "md", "tex"},
+	"spreadsheet":  {"xls", "xlsx", "ods", "csv", "tsv"},
+	"presentation": {"ppt", "pptx", "odp"},
+	"ebook":        {"epub", "mobi", "azw", "azw3", "fb2", "djvu"},
+	"code":         {"go", "c", "h", "cpp", "hpp", "py", "sh", "js", "ts", "rs", "java", "rb", "pl", "lua", "html", "css", "json", "yaml", "yml", "toml", "xml", "sql"},
+	"text":         {"txt", "md", "log", "csv", "tsv", "json", "yaml", "yml", "toml", "xml", "ini", "conf"},
+	"package":      {"deb", "rpm", "apk", "appimage", "exe", "msi", "flatpak", "snap"},
+	"font":         {"ttf", "otf", "woff", "woff2"},
+}