Implements URI guessing

This commit is contained in:
2016-01-14 15:07:45 +01:00
parent 21f6c21de4
commit f394ff0069
2 changed files with 74 additions and 23 deletions

69
bd.go
View File

@@ -1,6 +1,7 @@
package main
import (
"fmt"
"regexp"
"strings"
"time"
@@ -13,11 +14,16 @@ import (
type AlbumState int
const (
NEW AlbumState = iota // 0
MINT // 1
GOOD // 2
AVERAGE // 3
BAD // 4
// NEW is "État neuf" state
NEW AlbumState = iota // 0
// MINT is "Très bon état" state
MINT // 1
// GOOD is "Bon état" state
GOOD // 2
// AVERAGE is "État moyen" state
AVERAGE // 3
// BAD is "Mauvais état" state
BAD // 4
)
// An Album is the core object in our system
@@ -40,13 +46,14 @@ type Album struct {
LegalDeposit time.Time
PrintDate time.Time
PurchaseDate time.Time
}
var endDelim = regexp.MustCompile(` \(.*\)\z`)
var wordBoundaries = regexp.MustCompile(`[^[:alnum:]]+`)
var punctuation = regexp.MustCompile(`[!?\.:;,]`)
func sanitizeTitleString(title string, removeEndParent bool) string {
func sanitizeTitleString(title string) string {
// first sanitize accuented characters.
isOk := func(r rune) bool {
return r < 32 || r >= 127
@@ -57,16 +64,52 @@ func sanitizeTitleString(title string, removeEndParent bool) string {
// or io.Writer filter to automatically do such filtering when reading
// or writing data anywhere.
title, _, _ = transform.String(t, title)
//Now we remove all punctuation
if removeEndParent == true {
title = endDelim.ReplaceAllString(title, "")
}
//Now we remove all punctuat
return strings.Trim(wordBoundaries.ReplaceAllString(punctuation.ReplaceAllString(title, ""), "-"), "-")
}
func (*Album) GetBedethequeComURI() string {
return ""
// GetBedethequeComURI tries to guess the URI used by bedetheque.com to reference an album, using reverse-engineered euristics
func (a *Album) GetBedethequeComURI() string {
// we check for determinant
matches := endDelim.FindString(a.Series)
series := a.Series
titleMatch := false
if len(matches) != 0 {
series = strings.TrimSuffix(series, matches)
det := strings.Trim(matches, " ()")
if det[len(det)-1] != '\'' {
det = det + " "
}
titleCompare := det + strings.ToLower(series[:1]) + series[1:]
titleMatch = (titleCompare == a.Title)
} else {
titleMatch = (a.Series == a.Title)
}
series = sanitizeTitleString(series)
title := sanitizeTitleString(a.Title)
//first we test if we have a tome identifier
tomeIdent := ""
if a.Num < 0 {
tomeIdent = a.NumA
} else {
tomeIdent = fmt.Sprintf("Tome-%d%s", a.Num, a.NumA)
}
if titleMatch {
if len(tomeIdent) == 0 {
return fmt.Sprintf("BD-%s-%d.html", series, a.ID)
}
return fmt.Sprintf("BD-%s-%s-%d.html", series, tomeIdent, a.ID)
}
if len(tomeIdent) == 0 {
return fmt.Sprintf("BD-%s-%s-%d.html", series, title, a.ID)
}
return fmt.Sprintf("BD-%s-%s-%s-%d.html",
series,
tomeIdent,
title,
a.ID)
}
// An AlbumDescription is a more complete BD description