all: move checking REUSE annotation after all files listed

Use case: on the first scan, the file result in group unknown.
User then modify the spdxconv.cfg to add or update the match-file-pattern.
The next scan should check again the files in unknown group, in case
its match with updated config.

1 files changed, 24 insertions, 12 deletions

diff --git a/report.go b/report.go
index 64720b3..6a38f8d 100644
--- a/report.go
+++ b/report.go
@@ -109,8 +109,10 @@ func loadReport() (rep *report, err error) {
 			f.group = groupBinary
 			rep.listBinary[f.path] = f
 		case reportGroupUnknown:
-			f.group = groupUnknown
-			rep.listUnknown[f.path] = f
+			// Don't store it, we may need to rescan it later.
+			// Use case: the first scan result in unknown, user
+			// then modify the match-file-pattern and re-scan
+			// again.
 		case reportGroupDone:
 			rep.listDone[f.path] = f
 		}
@@ -124,21 +126,28 @@ func loadReport() (rep *report, err error) {
 
 func (rep *report) scan(conv *SPDXConv, listFile []string) (err error) {
 	var logp = `report.scan`
-	var ok bool
-	for _, file := range listFile {
-		if rep.hasScanned(file) {
+	for _, path := range listFile {
+		if rep.hasScanned(path) {
 			continue
 		}
-		_, ok = rep.listDone[file]
-		if ok {
+		// User may update the REUSE.toml on next scan.
+		if conv.reuse != nil && conv.reuse.isAnnotated(path) {
+			rep.listDone[path] = &file{
+				path: path,
+			}
 			continue
 		}
-
-		f, err := newFile(file, conv.cfg.MaxLineMatch)
+		f, err := newFile(path, conv.cfg.MaxLineMatch)
 		if err != nil {
 			return fmt.Errorf(`%s: %w`, logp, err)
 		}
 		if f.group == groupBinary {
+			f.licenseID = valDefault
+			f.copyrightText = valDefault
+			f.getYearFromSCM(conv)
+			if f.copyrightYear == `` {
+				f.copyrightYear = conv.cfg.CopyrightYear
+			}
 			rep.listBinary[f.path] = f
 			continue
 		}
@@ -166,8 +175,8 @@ func (rep *report) scan(conv *SPDXConv, listFile []string) (err error) {
 	return nil
 }
 
-// hasScanned return true if the file is already reported in regular or
-// binary group before.
+// hasScanned return true if the file is already reported in regular, binary,
+// or done group before.
 func (rep *report) hasScanned(path string) bool {
 	var ok bool
 	_, ok = rep.listBinary[path]
@@ -178,7 +187,10 @@ func (rep *report) hasScanned(path string) bool {
 	if ok {
 		return true
 	}
-	delete(rep.listUnknown, path)
+	_, ok = rep.listDone[path]
+	if ok {
+		return true
+	}
 	return false
 }