Skip to content

Commit

Permalink
Merge branch 'develop' of https://github.com/SYM-CONTEST/Cookpad2014
Browse files Browse the repository at this point in the history
…into develop
  • Loading branch information
yabuchin committed Oct 19, 2014
2 parents 758e690 + 3bab57b commit 0eb0020
Show file tree
Hide file tree
Showing 4 changed files with 28 additions and 12 deletions.
1 change: 1 addition & 0 deletions kinennbi/crawler/anniversary.go
Original file line number Diff line number Diff line change
Expand Up @@ -105,6 +105,7 @@ func (a Anniversary) nouns() []string {
parser := Parser{}
nouns := parser.ParseToNouns(a.tweetStrings())
nouns = parser.filterNoise(nouns, a, 2)
nouns = parser.filterDuplicateNounIfCan(nouns)
return nouns
}

Expand Down
2 changes: 1 addition & 1 deletion kinennbi/crawler/main/main.go
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ import (

func main() {
// 認証した人のcrawlerを生成
c := crawler.NewCrawler("35741880-v1Oe6Ls30CzGRQfhnqTVVENyC5Ylb7INNuKkUJaW6", "jcIXMoG27zE1IW1L3JK2OiSZLn0FOmQMrNvm3uvvdDIN9")
c := crawler.NewCrawler("35741880-nEBx773oAZqU4aMq246CrO3o3YguX5fpvjSSFue0V", "GYPNfRUHKOmkIEqSf72Ilh5aqAn37L78asXevSfFopGkc")
// 認証した人のメンションを分析してそれっぽい記念日群を抽出
as := c.AnalyzeAnniversary()
// ただの確認出力なので不要
Expand Down
23 changes: 23 additions & 0 deletions kinennbi/crawler/parser.go
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,7 @@ var matchBlacklist = []string{
"なに",
"ちゃん",
"どこ",
"たん",
}

type Parser struct {
Expand Down Expand Up @@ -75,6 +76,28 @@ func (parser Parser) filterNoise(src []string, a Anniversary, minLength int) []s
return r
}

func (parser Parser) filterDuplicateNounIfCan(src []string) []string {
r := make([]string, 0, len(src))
wc := wordCount(src)
for v, c := range wc {
if c > 1 {
r = append(r, v)
}
}
if len(r) > 0 {
return r
}
return src
}

func wordCount(ss []string) map[string]int {
res := make(map[string]int)
for _, str := range ss {
res[strings.ToLower(str)]++
}
return res
}

func (parser Parser) parseToNouns(s string) []string {
p, e := mecab.Parse(s)
if e != nil {
Expand Down
14 changes: 3 additions & 11 deletions kinennbi/crawler/util.go
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ import (

func failIfNeeded(e error) {
if e != nil {
log.Fatalln(e)
log.Println(e)
}
}

Expand All @@ -29,23 +29,15 @@ func printResult(t anaconda.Tweet) {
}
}

func wordCount(ss []string) map[string]int {
res := make(map[string]int)
for _, str := range ss {
res[strings.ToLower(str)]++
}
return res
}

func containsString(ss[] string, target string) bool {
func containsString(ss []string, target string) bool {
for _, s := range ss {
if s == target {
return true
}
}
return false
}
func containsNearlyString(ss[] string, target string) bool {
func containsNearlyString(ss []string, target string) bool {
for _, s := range ss {
if strings.Contains(target, s) {
return true
Expand Down

0 comments on commit 0eb0020

Please sign in to comment.