refactored & audited, real analizer added
This commit is contained in:
@@ -0,0 +1,103 @@
|
||||
package moonblood
|
||||
|
||||
import (
|
||||
"bufio"
|
||||
"os"
|
||||
"sort"
|
||||
"strings"
|
||||
|
||||
"git.kplus.net.ua/yevhen/resource-scraper/internal/releaseparser"
|
||||
)
|
||||
|
||||
type GroupStats struct {
|
||||
Total int
|
||||
YearDistribution map[int]int
|
||||
SourceDistribution map[string]int
|
||||
BitDepthDistribution map[string]int
|
||||
ResolutionDistribution map[string]int
|
||||
LanguageDistribution map[string]int
|
||||
EditionDistribution map[string]int
|
||||
UnknownTokens []releaseparser.UnknownTokenStats
|
||||
AvgConfidence float64
|
||||
}
|
||||
|
||||
func AnalyzeGroup() (GroupStats, error) {
|
||||
stats := GroupStats{
|
||||
YearDistribution: make(map[int]int),
|
||||
SourceDistribution: make(map[string]int),
|
||||
BitDepthDistribution: make(map[string]int),
|
||||
ResolutionDistribution: make(map[string]int),
|
||||
LanguageDistribution: make(map[string]int),
|
||||
EditionDistribution: make(map[string]int),
|
||||
UnknownTokens: []releaseparser.UnknownTokenStats{},
|
||||
}
|
||||
|
||||
file, err := os.Open("internal/releaseparser/resources/moonblood.txt")
|
||||
if err != nil {
|
||||
return stats, err
|
||||
}
|
||||
defer file.Close()
|
||||
|
||||
scanner := bufio.NewScanner(file)
|
||||
totalConfidence := 0.0
|
||||
|
||||
unknownMap := make(map[string]*releaseparser.UnknownTokenStats)
|
||||
|
||||
for scanner.Scan() {
|
||||
line := strings.TrimSpace(scanner.Text())
|
||||
if line == "" {
|
||||
continue
|
||||
}
|
||||
|
||||
result := Parse(line)
|
||||
stats.Total++
|
||||
|
||||
if result.Year != 0 {
|
||||
stats.YearDistribution[result.Year]++
|
||||
}
|
||||
if result.Source != "" {
|
||||
stats.SourceDistribution[result.Source]++
|
||||
}
|
||||
if result.BitDepth != "" {
|
||||
stats.BitDepthDistribution[result.BitDepth]++
|
||||
}
|
||||
if result.Resolution != "" {
|
||||
stats.ResolutionDistribution[result.Resolution]++
|
||||
}
|
||||
if result.Language != "" {
|
||||
stats.LanguageDistribution[result.Language]++
|
||||
}
|
||||
if result.Edition != "" {
|
||||
stats.EditionDistribution[result.Edition]++
|
||||
}
|
||||
|
||||
for _, unknown := range result.UnknownTokens {
|
||||
if _, exists := unknownMap[unknown]; !exists {
|
||||
unknownMap[unknown] = &releaseparser.UnknownTokenStats{
|
||||
Token: unknown,
|
||||
Count: 0,
|
||||
Releases: []string{},
|
||||
}
|
||||
}
|
||||
unknownMap[unknown].Count++
|
||||
unknownMap[unknown].Releases = append(unknownMap[unknown].Releases, line)
|
||||
}
|
||||
|
||||
totalConfidence += result.Confidence
|
||||
}
|
||||
|
||||
// Convert map to slice and sort by count
|
||||
for _, tokenStats := range unknownMap {
|
||||
stats.UnknownTokens = append(stats.UnknownTokens, *tokenStats)
|
||||
}
|
||||
|
||||
sort.Slice(stats.UnknownTokens, func(i, j int) bool {
|
||||
return stats.UnknownTokens[i].Count > stats.UnknownTokens[j].Count
|
||||
})
|
||||
|
||||
if stats.Total > 0 {
|
||||
stats.AvgConfidence = totalConfidence / float64(stats.Total)
|
||||
}
|
||||
|
||||
return stats, nil
|
||||
}
|
||||
Reference in New Issue
Block a user