Skip to content

Commit

Permalink
Implement digippres.org changes and min sig length (#253)
Browse files Browse the repository at this point in the history 
* Replace deprecated io.writefile with os.writefile

* Update SPARQL query to be more performant

A filter is introduced to return more relevant results, rather than
nearly everything on Wikidata. The query also introduces file format
family signatures which can be handled equally by Siegfried and fills
some gaps in identification.

Co-authored-by: Andrew Jackson <andrew.jackson@dpconline.org>

* Implement Wikidata min signature length

This reduces the query time further and tackles some of the issues
in: ffdev-info/wikidp-issues#32

The signature length can be configured but the default while we
calibrate is '6', i.e. 3 bytes.

---------

Co-authored-by: Andrew Jackson <andrew.jackson@dpconline.org>
  • Loading branch information
@ross-spencer @anjackson
ross-spencer and anjackson authored Jul 15, 2024
1 parent 5e86c03 commit 09ca73a
Show file tree
Hide file tree
Showing 4 changed files with 73 additions and 34 deletions.
3 changes: 1 addition & 2 deletions cmd/roy/harvest_wikidata.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,6 @@ package main
import (
"encoding/json"
"fmt"
"io/ioutil"
"log"
"os"
"strings"
Expand Down Expand Up @@ -119,7 +118,7 @@ func harvestWikidata() error {
)

path := config.WikidataDefinitionsPath()
err = ioutil.WriteFile(
err = os.WriteFile(
path,
[]byte(fmt.Sprintf("%s", modifiedJSON)),
config.WikidataFileMode(),
Expand Down
4 changes: 4 additions & 0 deletions cmd/roy/roy.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -163,6 +163,7 @@ var (
_, htimeout, _, _ = config.HarvestOptions()
timeout = harvest.Duration("timeout", htimeout, "set duration before timing-out harvesting requests e.g. 120s")
throttlef = harvest.Duration("throttle", 0, "set a time to wait HTTP requests e.g. 50ms")
harvestWikidataSigLen = harvest.Int("siglen", config.WikidataSigLen(), "set minimum signature length for Wikidata, e.g. 6 chars == 3 bytes")
harvestWikidataSig = harvest.Bool("wikidata", false, "harvest a static Wikidata report")
harvestWikidataLang = harvest.String("lang", config.WikidataLang(), "two-letter language-code to download Wikidata strings, e.g. \"de\"")
harvestWikidataEndpoint = harvest.String("wikidataendpoint", config.WikidataEndpoint(), "the endpoint to use to harvest Wikidata definitions from")
Expand Down Expand Up @@ -497,6 +498,9 @@ func setHarvestOptions() {
if *throttlef > 0 {
config.SetHarvestThrottle(*throttlef)
}
if *harvestWikidataSigLen != config.WikidataSigLen() {
config.SetWikidataSigLen(*harvestWikidataSigLen)
}
if *harvestWikidataLang != "" {
config.SetWikidataLang(*harvestWikidataLang)
}
Expand Down
76 changes: 51 additions & 25 deletions pkg/config/internal/wikidatasparql/sparql.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -18,9 +18,17 @@ package wikidatasparql
// the Wikidata identifier in Roy.

import (
"strconv"
"strings"
)

// sigLenTemplate gives us a field which we can replace with a
// min signature length value of our own choosing.
const sigLenTemplate = "<<siglen>>"

// Default signature length to return from Wikidata.
var wikidataSigLen = 6

// languateTemplate gives us a field which we can replace with a
// language code of our own configuration.
const languageTemplate = "<<lang>>"
Expand All @@ -35,36 +43,54 @@ var wikidataLang = "en"
// sparql represents the query required to pull all file format records
// and signatures from the Wikidata query service.
const sparql = `
# Return all file format records from Wikidata.
#
select distinct ?uri ?uriLabel ?puid ?extension ?mimetype ?encoding ?referenceLabel ?date ?relativity ?offset ?sig
where
{
?uri wdt:P31/wdt:P279* wd:Q235557. # Return records of type File Format.
optional { ?uri wdt:P2748 ?puid. } # PUID is used to map to PRONOM signatures proper.
optional { ?uri wdt:P1195 ?extension. }
optional { ?uri wdt:P1163 ?mimetype. }
optional { ?uri p:P4152 ?object; # Format identification pattern statement.
optional { ?object pq:P3294 ?encoding. } # We don't always have an encoding.
optional { ?object ps:P4152 ?sig. } # We always have a signature.
optional { ?object pq:P2210 ?relativity. } # Relativity to beginning or end of file.
optional { ?object pq:P4153 ?offset. } # Offset relative to the relativity.
optional { ?object prov:wasDerivedFrom ?provenance;
optional { ?provenance pr:P248 ?reference;
pr:P813 ?date.
}
}
}
service wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE], <<lang>>". }
}
order by ?uri
`
# Return all file format records from Wikidata.
SELECT DISTINCT ?uri ?uriLabel ?puid ?extension ?mimetype ?encoding ?referenceLabel ?date ?relativity ?offset ?sig WHERE {
{ ?uri (wdt:P31/(wdt:P279*)) wd:Q235557. }
UNION
{ ?uri (wdt:P31/(wdt:P279*)) wd:Q26085352. }
FILTER(EXISTS { ?uri (wdt:P2748|wdt:P1195|wdt:P1163|ps:P4152) _:b2. })
FILTER((STRLEN(?sig)) >= <<siglen>> )
OPTIONAL { ?uri wdt:P2748 ?puid. }
OPTIONAL { ?uri wdt:P1195 ?extension. }
OPTIONAL { ?uri wdt:P1163 ?mimetype. }
OPTIONAL {
?uri p:P4152 ?object.
OPTIONAL { ?object pq:P3294 ?encoding. }
OPTIONAL { ?object ps:P4152 ?sig. }
OPTIONAL { ?object pq:P2210 ?relativity. }
OPTIONAL { ?object pq:P4153 ?offset. }
OPTIONAL {
?object prov:wasDerivedFrom ?provenance.
OPTIONAL {
?provenance pr:P248 ?reference;
pr:P813 ?date.
}
}
}
SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE], <<lang>>". }
}
ORDER BY (?uri)
`

// WikidataSPARQL returns the SPARQL query needed to pull file-format
// signatures from Wikidata replacing various template values as we
// go.
func WikidataSPARQL() string {
return strings.Replace(sparql, languageTemplate, wikidataLang, numberReplacements)
wdSparql := strings.Replace(sparql, languageTemplate, wikidataLang, numberReplacements)
wdSparql = strings.Replace(wdSparql, sigLenTemplate, strconv.Itoa(wikidataSigLen), numberReplacements)
return wdSparql
}

// WikidataSigLen returns the minimum signature length we want the Wikidata
// SPARQL query to return.
func WikidataSigLen() int {
return wikidataSigLen
}

// SetWikidataSigLen sets the minimum signature length we want the Wikidata
// SPARQL query to return.
func SetWikidataSigLen(len int) {
wikidataSigLen = len
}

// WikidataLang will return to the caller the ISO language code
Expand Down
24 changes: 17 additions & 7 deletions pkg/config/wikidata.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -221,6 +221,18 @@ func WikidataSPARQL() string {
return wikidatasparql.WikidataSPARQL()
}

// WikidataSiglen returns the minimum signature length we want the Wikidata
// SPARQL query to return.
func WikidataSigLen() int {
return wikidatasparql.WikidataSigLen()
}

// SetWikidataLang sets the minimum signature length we want the Wikidata
// SPARQL query to return.
func SetWikidataSigLen(len int) {
wikidatasparql.SetWikidataSigLen(len)
}

// WikidataLang returns the language we want to return results in from
// Wikidata.
func WikidataLang() string {
Expand Down Expand Up @@ -262,7 +274,6 @@ func GetWikidataNoPRONOM() bool {
// returned from Wikibase, e.g. for Wikidata this URL needs to be:
//
// e.g. https://www.wikidata.org/w/index.php
//
func SetWikibaseURL(baseURL string) (func() private, error) {
_, err := url.ParseRequestURI(baseURL)
if err != nil {
Expand Down Expand Up @@ -384,12 +395,11 @@ func SetCustomWikibaseQuery() error {
//
// Example:
//
// {
// "PronomProp": "http://wikibase.example.com/entity/Q2",
// "BofProp": "http://wikibase.example.com/entity/Q3",
// "EofProp": "http://wikibase.example.com/entity/Q4"
// }
//
// {
// "PronomProp": "http://wikibase.example.com/entity/Q2",
// "BofProp": "http://wikibase.example.com/entity/Q3",
// "EofProp": "http://wikibase.example.com/entity/Q4"
// }
func WikibasePropsPath() string {
return filepath.Join(WikidataHome(), wikidata.wikibasePropsFile)
}
Expand Down

0 comments on commit 09ca73a

Please sign in to comment.