From a63d76e3fc3c59c2eec2de4e538b45e41e1f8aa9 Mon Sep 17 00:00:00 2001
From: Martin Czygan <martin.czygan@gmail.com>
Date: Tue, 4 May 2021 23:59:53 +0200
Subject: tweaks; move parsing out of command

---
 skate/cmd/skate-from-unstructured/main.go | 61 +------------------------------
 1 file changed, 1 insertion(+), 60 deletions(-)

(limited to 'skate/cmd/skate-from-unstructured')

diff --git a/skate/cmd/skate-from-unstructured/main.go b/skate/cmd/skate-from-unstructured/main.go
index c2015e2..179057d 100644
--- a/skate/cmd/skate-from-unstructured/main.go
+++ b/skate/cmd/skate-from-unstructured/main.go
@@ -6,9 +6,7 @@ import (
 	"flag"
 	"log"
 	"os"
-	"regexp"
 	"runtime"
-	"strings"
 
 	"git.archive.org/martin/cgraph/skate"
 	"git.archive.org/martin/cgraph/skate/parallel"
@@ -19,11 +17,6 @@ var (
 	numWorkers   = flag.Int("w", runtime.NumCPU(), "number of workers")
 	batchSize    = flag.Int("b", 100000, "batch size")
 	bytesNewline = []byte("\n")
-
-	PatDOI         = regexp.MustCompile(`10[.][0-9]{1,8}/[^ ]*[\w]`)
-	PatDOINoHyphen = regexp.MustCompile(`10[.][0-9]{1,8}/[^ -]*[\w]`)
-	PatArxivPDF    = regexp.MustCompile(`http://arxiv.org/pdf/([0-9]{4,4}[.][0-9]{1,8})(v[0-9]{1,2})?(.pdf)?`)
-	PatArxivAbs    = regexp.MustCompile(`http://arxiv.org/abs/([0-9]{4,4}[.][0-9]{1,8})(v[0-9]{1,2})?(.pdf)?`)
 )
 
 func main() {
@@ -32,7 +25,7 @@ func main() {
 		if err := json.Unmarshal(p, &ref); err != nil {
 			return nil, err
 		}
-		if err := parseUnstructured(&ref); err != nil {
+		if err := skate.ParseUnstructured(&ref); err != nil {
 			return nil, err
 		}
 		return skate.JsonMarshalLine(&ref)
@@ -43,55 +36,3 @@ func main() {
 		log.Fatal(err)
 	}
 }
-
-// parseUnstructured will in-place augment missing DOI, arxiv id and so on.
-func parseUnstructured(ref *skate.Ref) error {
-	uns := ref.Biblio.Unstructured
-	var (
-		v  string
-		vs []string
-	)
-	// Handle things like: 10.1111/j.1550-7408.1968.tb02138.x-BIB5|cit5,
-	// 10.1111/j.1558-5646.1997.tb02431.x-BIB0008|evo02431-cit-0008, ...
-	if strings.Contains(strings.ToLower(ref.Key), "-bib") && ref.Biblio.DOI == "" {
-		parts := strings.Split(strings.ToLower(ref.Key), "-bib")
-		ref.Biblio.DOI = parts[0]
-	}
-	// DOI
-	v = PatDOI.FindString(uns)
-	if v != "" && ref.Biblio.DOI == "" {
-		ref.Biblio.DOI = v
-	}
-	// DOI in Key
-	v = PatDOINoHyphen.FindString(ref.Key)
-	if v != "" && ref.Biblio.DOI == "" {
-		ref.Biblio.DOI = v
-	}
-	// DOI in URL
-	prefixes := []string{
-		"http://doi.org/",
-		"https://doi.org/",
-		"http://dx.doi.org/",
-		"https://dx.doi.org/",
-	}
-	for _, prefix := range prefixes {
-		if ref.Biblio.DOI != "" && strings.HasPrefix(ref.Biblio.Url, prefix) {
-			ref.Biblio.DOI = strings.Replace(ref.Biblio.Url, prefix, "", -1)
-		}
-	}
-	v = PatDOINoHyphen.FindString(ref.Key)
-	if v != "" && ref.Biblio.DOI == "" {
-		ref.Biblio.DOI = v
-	}
-	// Arxiv
-	vs = PatArxivPDF.FindStringSubmatch(uns)
-	if len(vs) != 0 && ref.Biblio.ArxivId == "" {
-		ref.Biblio.ArxivId = vs[1]
-	} else {
-		vs = PatArxivAbs.FindStringSubmatch(uns)
-		if len(vs) != 0 && ref.Biblio.ArxivId == "" {
-			ref.Biblio.ArxivId = vs[1]
-		}
-	}
-	return nil
-}
-- 
cgit v1.2.3