示例#1
0
def set_sig_claim(item, qid, years, permalink):
    print(f"Got {len(years)} year(s) claim for restoration")

    for year in years:
        print(f"Adding year: {year}")

        if year["type"] == "range":
            qualifiers = [
                item.get_claim(Props.START_TIME,
                               WbTime(year=int(year["value"][0]))),
                item.get_claim(Props.END_TIME,
                               WbTime(year=int(year["value"][1])))
            ]
        elif year["type"] in ["century", "decade", "year"]:
            qualifiers = [
                item.get_claim(Props.POINT_IN_TIME, convert_year(year))
            ]

            if all(["start" in year, "end" in year]):
                qualifiers.extend([
                    item.get_claim(Props.START_TIME,
                                   WbTime(year=int(year["start"]))),
                    item.get_claim(Props.END_TIME,
                                   WbTime(year=int(year["end"])))
                ])
        else:
            print("Invalid year type: " + year["type"])
            continue

        item.add_item_claim(Props.SIG_EVENT,
                            qid,
                            qualifiers=qualifiers,
                            references=get_ref(item, permalink))
示例#2
0
def convert_year(year):
    ytype = year["type"]
    val = int(year["value"])

    if year["type"] == "decade":
        return WbTime(year=val, precision="decade")
    elif year["type"] == "century":
        return WbTime(year=val, precision="century")
    else:
        return WbTime(year=val)
示例#3
0
def partial_date_to_wbtime(date):
    print(f"Parsing partial date to wbtime {date}")

    if len(date) == 4:
        return WbTime(year = int(date))
    elif len(date) == 7:
        year, month = date.split("-")
        return WbTime(year = int(year), month = int(month))
    elif len(date) == 10:
        year, month, day = date.split("-")
        return WbTime(year = int(year), month = int(month), day = int(day))
    else:
        raise TypeError(f"Invalid date to parse: {date}")
示例#4
0
def get_ref(job):
    return [
        job.item.get_url_claim(Props.REF_URL, job.data["url"]),
        job.item.get_claim(Props.RETRIEVED, WbTime(year=2021, month=11,
                                                   day=15)),
        job.item.get_item_claim(Props.LANGUAGE_WORK, Items.DUTCH)
    ]
示例#5
0
def get_refs(item, pid):
    url = f"https://pointer.kro-ncrv.nl/het-slavernijverleden-van-suriname-op-de-kaart-gezet#/plantage/{pid}"

    return [
        item.get_url_claim(Props.REF_URL, url),
        item.get_claim(Props.RETRIEVED, WbTime(year=2021, month=2, day=20)),
        item.get_item_claim(Props.LANGUAGE_WORK, Items.DUTCH)
    ]
示例#6
0
def set_year_claim(item, prop, years, permalink):
    print(f"Got {len(years)} year(s) claim for {prop}")

    for year in years:
        print(f"Adding year: {year}")

        if year["type"] == "range":
            item.add_item_claim(
                prop,
                "novalue",
                qualifiers=[
                    item.get_claim(Props.START_TIME,
                                   WbTime(year=int(year["value"][0]))),
                    item.get_claim(Props.END_TIME,
                                   WbTime(year=int(year["value"][1])))
                ],
                references=get_ref(item, permalink))
        elif year["type"] in ["century", "decade"]:
            if all(["start" in year, "end" in year]):
                qualifiers = [
                    item.get_claim(Props.START_TIME,
                                   WbTime(year=int(year["start"]))),
                    item.get_claim(Props.END_TIME,
                                   WbTime(year=int(year["end"])))
                ]
            else:
                qualifiers = None

            item.add_time_claim(prop,
                                convert_year(year),
                                qualifiers=qualifiers,
                                references=get_ref(item, permalink))
        elif year["type"] == "circa":
            item.add_time_claim(
                prop,
                convert_year(year),
                qualifiers=[item.get_item_claim(Props.PRECISION, Items.CIRCA)],
                references=get_ref(item, permalink))
        elif year["type"] == "year":
            item.add_time_claim(prop,
                                convert_year(year),
                                references=get_ref(item, permalink))
        else:
            print("Invalid year type: " + year["type"])
            continue
示例#7
0
def get_refs(item, url):
    return [
        item.get_item_claim(Props.STATED_IN, Items.PUBLIC_ART_IN_NIJMEGEN),
        item.get_url_claim(Props.REF_URL, url),
        item.get_claim(Props.RETRIEVED, WbTime(
            year = 2021, month = 1, day = 31
        )),
        item.get_item_claim(Props.LANGUAGE_WORK, Items.DUTCH)
    ]
示例#8
0
 def set_target(self, snak, value):
     if value in ('somevalue', 'novalue'):
         snak.setSnakType(value)
         return True
     if snak.type == 'wikibase-item':
         snak.setTarget(pywikibot.ItemPage(self.repo, value))
         return True
     elif snak.type == 'wikibase-property':
         snak.setTarget(pywikibot.PropertyPage(self.repo, value))
         return True
     elif snak.type == 'quantity':
         match = self.quantityR.fullmatch(value)
         if match:
             amount, error, unit = match.groups()
         else:
             match = self.quantity_oldR.fullmatch(value)
             if match:
                 amount, lower, upper, unit = match.groups()
                 error = upper, lower  # it *is* the other way around
         if match:
             if unit:
                 unit = pywikibot.ItemPage(self.repo, 'Q' + unit)
             quantity = WbQuantity(amount, unit, error, site=self.repo)
             snak.setTarget(quantity)
             return True
     elif snak.type == 'time':
         iso, _, prec = value.rpartition('/')
         if iso:
             time = WbTime.fromTimestr(iso,
                                       precision=int(prec),
                                       site=self.repo)
             snak.setTarget(time)
             return True
     elif snak.type in ('string', 'external-id', 'url', 'math'):
         if value.startswith('"') and value.endswith('"'):
             snak.setTarget(value[1:-1])
             return True
     elif snak.type == 'commonsMedia':
         if value.startswith('"') and value.endswith('"'):
             repo = self.repo.image_repository()
             snak.setTarget(pywikibot.FilePage(repo, value[1:-1]))
             return True
     #elif snak.type in ('geo-shape', 'tabular-data'):
     elif snak.type == 'monolingualtext':
         lang, _, text = value.partition(':')
         if text and text.startswith('"') and text.endswith('"'):
             monotext = WbMonolingualText(text[1:-1], lang)
             snak.setTarget(monotext)
             return True
     elif snak.type == 'globe-coordinate':
         match = self.globeR.fullmatch(value)
         if match:
             coord = Coordinate(*map(float, match.groups()), site=self.repo)
             snak.setTarget(coord)
             return True
     return False
示例#9
0
 def set_target(self, snak, value):
     if value in ('somevalue', 'novalue'):
         snak.setSnakType(value)
         return True
     if snak.type == 'wikibase-item':
         snak.setTarget(pywikibot.ItemPage(self.repo, value))
         return True
     elif snak.type == 'wikibase-property':
         snak.setTarget(pywikibot.PropertyPage(self.repo, value))
         return True
     elif snak.type == 'quantity':
         match = self.quantityR.fullmatch(value)
         if match:
             amount, error, unit = match.groups()
         else:
             match = self.quantity_oldR.fullmatch(value)
             if match:
                 amount, lower, upper, unit = match.groups()
                 error = upper, lower  # it *is* the other way around
         if match:
             if unit:
                 unit = pywikibot.ItemPage(self.repo, 'Q' + unit)
             quantity = WbQuantity(amount, unit, error, site=self.repo)
             snak.setTarget(quantity)
             return True
     elif snak.type == 'time':
         iso, _, prec = value.rpartition('/')
         if iso:
             time = WbTime.fromTimestr(
                 iso, precision=int(prec), site=self.repo)
             snak.setTarget(time)
             return True
     elif snak.type in ('string', 'external-id', 'url', 'math'):
         if value.startswith('"') and value.endswith('"'):
             snak.setTarget(value[1:-1])
             return True
     elif snak.type == 'commonsMedia':
         if value.startswith('"') and value.endswith('"'):
             repo = self.repo.image_repository()
             snak.setTarget(pywikibot.FilePage(repo, value[1:-1]))
             return True
     #elif snak.type in ('geo-shape', 'tabular-data'):
     elif snak.type == 'monolingualtext':
         lang, _, text = value.partition(':')
         if text and text.startswith('"') and text.endswith('"'):
             monotext = WbMonolingualText(text[1:-1], lang)
             snak.setTarget(monotext)
             return True
     elif snak.type == 'globe-coordinate':
         match = self.globeR.fullmatch(value)
         if match:
             coord = Coordinate(*map(float, match.groups()), site=self.repo)
             snak.setTarget(coord)
             return True
     return False
示例#10
0
def string_to_wddate(isotimestamp: str) -> WbTime:
    """
    Create a wikidata compatible wikibase date from an ISO 8601 timestamp
    """
    date = WbTime.fromTimestr(isotimestamp,
                              calendarmodel=Settings.calendarmodel)
    date.hour = 0
    date.minute = 0
    date.second = 0
    date.precision = WbTime.PRECISION["day"]
    return date
def try_get_year_from_property_timestamp(entity_dict: Dict, property_id: str,
                                         oab_type: str) -> int:
    """Function for extracting the year from an inception timestamp

    Args:
        entity_dict: JSON response from wikidata
        property_id: Wikidata property id e. g. 'P18' stands for the property image
        oab_type: openArtBrowser type e. g. 'artwork', 'motif' important for logging

    Returns:
        Year from timestamp
    """
    timestr = entity_dict[CLAIMS][property_id][0][MAINSNAK][DATAVALUE][VALUE][
        TIME]
    return WbTime.fromTimestr(timestr).year
示例#12
0
def add_party_data(row):
    print("----" * 20)
    print()
    print(row)

    title = row["title"]
    qid = row["qid"]

    if skiplist.has(qid):
        print(f"In skiplist, skipping")
        return

    item = WikidataItem(qid)

    if Props.NR_OF_SEATS in item.get_claims():
        print("Got seats already, skipping party")
        return

    for key, val in row.items():
        if not key.isdigit():
            continue

        year = int(key)

        if val == "":
            continue

        seats = int(val)

        print(f"{title} ({qid}) had {seats} seats in {year}")

        item.add_quantity_claim(
            Props.NR_OF_SEATS,
            seats,
            qualifiers=[
                item.get_item_claim(Props.LEGISLATIVE_BODY,
                                    Items.NL_LOWER_HOUSE),
                item.get_claim(Props.START_TIME, WbTime(year=year))
            ],
            references=[
                item.get_item_claim(Props.IMPORTED_FROM, Items.WIKIPEDIA_NL),
                item.get_url_claim(Props.WM_IMPORT_URL, WP_PERMALINK)
            ])

    skiplist.add(qid)
示例#13
0
def add_architect(job):
    data = job.data

    if not data["arch1_qid"]:
        print(f"No architect, skipping")
        return

    claims = job.item.get_claims()

    if Props.ARCHITECT in claims:
        print("Already has an architect, skipping")
        return

    # Add the actual data
    for index in ["1", "2", "3"]:
        prefix = f"arch{index}"

        if data[f"{prefix}_qid"] == "":
            continue

        qid = data[f"{prefix}_qid"]
        qualifiers = []

        if data[f"{prefix}_role_qid"]:
            qualifiers.append(
                job.item.get_item_claim(Props.OBJECT_HAS_ROLE,
                                        data[f"{prefix}_role_qid"]))

        if data[f"{prefix}_role_date"]:
            date = data[f"{prefix}_role_date"]
            yearstr = YEAR.match(date)

            if yearstr:
                year = int(yearstr[0])

                qualifiers.append(
                    job.item.get_claim(Props.POINT_IN_TIME, WbTime(year=year)))
            else:
                print(f"Unparsable year: {date}")

        job.item.add_item_claim(Props.ARCHITECT,
                                qid,
                                qualifiers=qualifiers,
                                references=get_refs(job.item, data["pageid"]))
示例#14
0
def wbtime_now():
    now = datetime.now()

    return WbTime(
        year = now.year, month = now.month, day = now.day
    )
示例#15
0
def add_item(data):
    print("----" * 20)
    print()
    print(data)

    title = data["title"]
    url = data["url"]
    kid = data["id"]

    # if kid != "50":
    #     return

    print(f"Handling {title}")

    if skiplist.has(kid):
        print(f"In skiplist, skipping")
        return

    creator = data["creator"]

    if creator == "":
        desc_nl = "kunstwerk in de openbare ruimte te Nijmegen"
        desc_en = "public artwork in Nijmegen, the Netherlands"
    elif creator == "Onbekend":
        desc_nl = "kunstwerk in de openbare ruimte van een onbekende maker te Nijmegen"
        desc_en = "public artwork by an unknown artist in Nijmegen, the Netherlands"
    else:
        desc_nl = f"kunstwerk van {creator} in de openbare ruimte te Nijmegen"
        desc_en = f"public artwork by {creator} in Nijmegen, the Netherlands"

    item = WikidataItem(
        summary = f"public artwork '{title}' in Nijmegen, the Netherlands",
        labels = {
            "nl" : title
        }
    )

    item.edit_descriptions({
            "de" : f"Kunst im öffentlichen Raum in Nijmegen (Niederlande)",
            "en" : desc_en,
            "es" : f"arte público en Nijmegen (Holanda)",
            "fr" : f"art public à Nimègue (Pays-Bas)",
            "nl" : desc_nl
    })

    item.edit_aliases({
        "en" : title
    })

    # Basics
    item.add_item_claim(Props.INSTANCE_OF, Items.SCULPTURE)
    item.add_item_claim(Props.COUNTRY, Items.NETHERLANDS)
    item.add_item_claim(Props.LOCATED_IN_ADMIN, Items.NIJMEGEN_MUNIP)
    item.add_item_claim(Props.GENRE, Items.PUBLIC_ART)


    # Actual data
    coord = [data["lat"], data["lon"]]
    item.add_coordinate(
        Props.COORDINATES, coord, references = get_refs(item, url)
    )

    item.add_string_claim(
        Props.INVENTORY_NR, kid,
        qualifiers = [
            item.get_item_claim(Props.COLLECTION, Items.PUBLIC_ART_IN_NIJMEGEN)
        ],
        references = get_refs(item, url)
    )

    item.add_string_claim(
        Props.DESCRIBED_AT_URL,
        url,
        qualifiers = [
            item.get_item_claim(Props.LANGUAGE_WORK, Items.DUTCH)
        ]
    )

    if data["year"] != "":
        year = int(data["year"])
        item.add_time_claim(
            Props.INCEPTION, WbTime(year = year), references = get_refs(item, url)
        )

    if data["creator_qid"] != "":
        item.add_item_claim(
            Props.CREATOR, data["creator_qid"], references = get_refs(item, url)
        )
    elif data["creator"] == "Onbekend":
        item.add_item_claim(
            Props.CREATOR, "somevalue", references = get_refs(item, url)
        )

    if data["owner"] == "gemeente":
        item.add_item_claim(
            Props.COLLECTION, Items.NIJMEGEN_MUNIP, references = get_refs(item, url)
        )
    elif data["owner"] == "particulier":
        item.add_item_claim(
            Props.COLLECTION, Items.PRIVATE_COLLECTION, references = get_refs(item, url)
        )

    if data["location_clean"] != "":
        item.add_monoling_claim(
            Props.STREET_ADDRESS, data["location_clean"], "nl", references = get_refs(item, url)
        )

    skiplist.add(kid)
def try_get_significant_events(
        result: Dict,
        oab_type: Optional[str] = SIGNIFICANT_EVENT) -> List[Dict]:
    """Maps the wikidata response for significant events to a list of dicts which is appended to an object

    Args:
        result: wikidata response
        oab_type: OpenArtBrowser type. Defaults to SIGNIFICANT_EVENT.

    Returns:
        List of JSON objects which represent significant events
    """
    significant_events = []
    qid = result[ID]
    for event in result[CLAIMS][
            PROPERTY_NAME_TO_PROPERTY_ID[SIGNIFICANT_EVENT]]:
        event_dict = {LABEL[SINGULAR]: event[MAINSNAK][DATAVALUE][VALUE][ID]}
        for qualifiers in event[QUALIFIERS].values():
            datatype = qualifiers[0][DATATYPE]
            property_id = qualifiers[0][PROPERTY]
            # Get property name from dict, if the name is not in the dict ignore it and take the id
            property = PROPERTY_ID_TO_PROPERTY_NAME.get(
                property_id, property_id)
            if datatype == TIME:
                event_dict.update({
                    property:
                    WbTime.fromTimestr(
                        qualifiers[0][DATAVALUE][VALUE][TIME]).year
                })
            elif datatype == WIKIBASE_ITEM:
                event_dict.update({
                    property:
                    list(
                        map(
                            lambda qualifier: qualifier[DATAVALUE][VALUE][ID],
                            qualifiers,
                        ))
                })
            elif datatype == QUANTITY:
                event_dict.update(
                    {property: float(qualifiers[0][DATAVALUE][VALUE][AMOUNT])})
                event_dict.update({
                    f"{property}_{UNIT}":
                    qualifiers[0][DATAVALUE][VALUE].get(UNIT, "").replace(
                        WIKIDATA_ENTITY_URL, "")
                })
            elif datatype in [STRING, URL]:
                event_dict.update({property: qualifiers[0][DATAVALUE][VALUE]})
            elif datatype == MONOLINGUALTEXT:
                event_dict.update(
                    {property: qualifiers[0][DATAVALUE][VALUE][TEXT]})
            elif datatype == COMMONS_MEDIA:
                logger.error(
                    f"commonsMedia type not supported in significant events on item {qid}"
                )
            else:
                logger.error(f"Unknown datatype: {datatype} on item {qid}")
        event_dict.update({TYPE: SIGNIFICANT_EVENT})
        significant_events.append(event_dict)

    return significant_events
示例#17
0
def create_new():
    items = Knead(PATH + "/data/reliwiki/new-churches.csv",
                  has_header=True).data()
    CITY = "Amsterdam"

    for church in items:
        print()
        print(f"Creating new church", church)
        pageid = church["pageid"]

        # Last, final check if this church doesn't exist
        if claim_exists(Props.RELIWIKI, f'"{pageid}"'):
            print(f"This Reliwiki ID exists, skipping")
            continue

        name = church["name"]

        item = WikidataItem(
            summary=f"Creating new item for Dutch church with name {name}",
            labels={
                "en": name,
                "nl": name
            },
            descriptions={
                "de": f"Kirche in {CITY} (Niederlande)",
                "en": f"church in {CITY}, the Netherlands",
                "es": f"iglesia en {CITY} (Holanda)",
                "fr": f"Église d'{CITY} (Pays-Bas)",
                "nl": f"kerk in {CITY}"
            })

        item.add_item_claim(Props.INSTANCE_OF, Items.CHURCH_BUILDING)
        item.add_string_claim(Props.RELIWIKI, pageid)
        item.add_item_claim(Props.COUNTRY, Items.NETHERLANDS)
        item.add_item_claim(Props.LOCATED_IN, church["admin_qid"])

        if church["sonneveld"] != "":
            item.add_string_claim(Props.SONNEVELD,
                                  church["sonneveld"],
                                  references=get_refs(item, pageid))

        if church["coordinates"] != "":
            coord = church["coordinates"].split(",")
            item.add_coordinate(Props.COORDINATES,
                                coord,
                                references=get_refs(item, pageid))

        if church["zipcode"] != "":
            item.add_string_claim(Props.ZIP,
                                  church["zipcode"],
                                  references=get_refs(item, pageid))

        if church["address"] != "":
            item.add_monoling_claim(Props.STREET_ADDRESS,
                                    church["address"],
                                    "nl",
                                    references=get_refs(item, pageid))

        if church["denomination_qid"] != "":
            item.add_item_claim(Props.RELIGION,
                                church["denomination_qid"],
                                references=get_refs(item, pageid))

        if church["year_use"] != "":
            if "s" in church["year_use"]:
                decade = int(church["year_use"].replace("s", ""))
                time = WbTime(year=decade, precision="decade")
            else:
                time = WbTime(year=int(church["year_use"]))

            item.add_time_claim(Props.INCEPTION,
                                time,
                                references=get_refs(item, pageid))

        print()

        break
示例#18
0
def run_bot():
    DATA_PATH = PATH / "data" / "vbvd"

    bot = CreateBot("vbvd",
                    run_once=False,
                    datapath=str(DATA_PATH / "items2.json"),
                    key="title")

    for job in bot.iterate():
        title = job.data["title"]
        creator = job.data["artist_label"]
        year = job.data["year"]

        summary = f"artwork '{title}' in Museum van Bommel van Dom in Venlo, the Netherlands",

        labels = {"nl": title}

        descriptions = {
            "de": f"Kunstwerk von {creator}",
            "en": f"work of art by {creator}",
            "es": f"obra de arte de {creator}",
            "fr": f"œuvre d’art de {creator}",
            "nl": f"kunstwerk van {creator}"
        }

        if year:
            descriptions["en"] = f"{year} " + descriptions["en"]
            descriptions["nl"] = descriptions["nl"] + f" uit {year}"

        aliases = {"en": [title]}

        job.create_item(summary, labels, descriptions, aliases)

        # Basics
        job.item.add_item_claim(Props.INSTANCE_OF, Items.WORK_OF_ART)
        job.item.add_item_claim(Props.LOCATION, Items.MUS_BOMMELVDAM)

        job.item.add_string_claim(Props.INVENTORY_NR,
                                  job.data["inventory_nr"],
                                  qualifiers=[
                                      job.item.get_item_claim(
                                          Props.COLLECTION,
                                          Items.MUS_BOMMELVDAM)
                                  ],
                                  references=get_ref(job))

        job.item.add_url_claim(Props.DESCRIBED_AT_URL,
                               job.data["url"],
                               qualifiers=[
                                   job.item.get_item_claim(
                                       Props.LANGUAGE_WORK, Items.DUTCH)
                               ])

        if job.data["artist_qid"]:
            job.item.add_item_claim(Props.CREATOR,
                                    job.data["artist_qid"],
                                    references=get_ref(job))

        if job.data["year"]:
            job.item.add_time_claim(Props.INCEPTION,
                                    WbTime(year=job.data["year"]),
                                    references=get_ref(job))

        if job.data["collection_qid"]:
            job.item.add_item_claim(Props.COLLECTION,
                                    job.data["collection_qid"],
                                    references=get_ref(job))
示例#19
0
    def _set_target(self, snak, value):
        if value in ('somevalue', 'novalue'):
            snak.setSnakType(value)
            return True

        def invalid_report():
            pywikibot.warning('Invalid value "{}" for {} datatype'.format(
                value, snak.type))

        if snak.type in self.entity_types:
            target = self.parse_entity(value)
            if target is None:
                pywikibot.warning('"LAST" magic word used without "CREATE"')
            else:
                snak.setTarget(target)
                return True
        elif snak.type == 'quantity':
            match = self.quantity_errR.fullmatch(value)
            if match:
                amount, error, unit = match.groups()
            else:
                match = self.quantity_boundsR.fullmatch(value)
                if match:
                    groups = list(match.groups())
                    unit = groups.pop()
                    amount, lower, upper = map(Decimal, groups)
                    if lower > upper:
                        error = amount - lower, upper - amount
                    else:
                        error = upper - amount, amount - lower
            if match:
                if unit:
                    unit = pywikibot.ItemPage(self.repo, 'Q' + unit)
                quantity = WbQuantity(amount, unit, error, site=self.repo)
                snak.setTarget(quantity)
                return True
            else:
                invalid_report()
        elif snak.type == 'time':
            iso, _, prec = value.rpartition('/')
            if iso:
                time = WbTime.fromTimestr(iso,
                                          precision=int(prec),
                                          site=self.repo)
                snak.setTarget(time)
                return True
            else:
                invalid_report()
        elif snak.type in ('string', 'external-id', 'url', 'math'):
            literal = self.valid_text_literal(value)
            if literal:
                snak.setTarget(literal)
                return True
            else:
                invalid_report()
        elif snak.type == 'commonsMedia':
            literal = self.valid_text_literal(value)
            if literal:
                image_repo = self.repo.image_repository()
                snak.setTarget(pywikibot.FilePage(image_repo, literal))
                return True
            else:
                invalid_report()
        # todo: elif snak.type in ('geo-shape', 'tabular-data'):
        elif snak.type == 'monolingualtext':
            lang, _, text = value.partition(':')
            literal = self.valid_text_literal(text)
            if literal:
                monotext = WbMonolingualText(literal, lang)
                snak.setTarget(monotext)
                return True
            else:
                invalid_report()
        elif snak.type == 'globe-coordinate':
            match = self.globeR.fullmatch(value)
            if match:
                coord = Coordinate(
                    *map(float, match.groups()),
                    precision=1e-4,  # hardcoded as in claimit.py
                    site=self.repo)
                snak.setTarget(coord)
                return True
            else:
                invalid_report()
        else:
            pywikibot.warning('"{}" datatype is not supported yet'.format(
                snak.type))

        return False
示例#20
0
def run_bot():
    sys.exit("No, fix your code first")
    DATA_PATH = PATH / "data" / "vbvd"

    bot = CreateBot("vbvd2",
                    datapath=str(DATA_PATH / "export-met-urls.csv"),
                    key="inventory",
                    required_fields=["inventory", "title", "url"],
                    empty_check=is_empty)

    for job in bot.iterate():
        title = job.data["title"]
        inventory_nr = job.data["inventory"]

        # If the job has a qid, check if we want to append,
        # otherwise skip because this already exists
        if not is_empty(job.data["item_qid"]) and is_empty(job.data["append"]):
            job.abort(
                f"Skipping {inventory_nr}/{title}, item already is on Wikidata"
            )
            continue

        summary = f"Adding new artwork '{title}' in Museum van Bommel van Dom in Venlo, the Netherlands"

        labels = {"nl": title}

        descriptions = {}

        # Check for empty or non-existing creator
        if is_empty(job.data["creator_name"]):
            creator_name = "een onbekende kunstenaar"
            creator_name_en = "an unknown artist"
        else:
            creator_name = job.data["creator_name"]
            creator_name_en = creator_name

        if is_empty(job.data["type"]):
            descriptions["nl"] = f"kunstwerk van {creator_name}"
        else:
            type_nl = job.data["type"]
            descriptions["nl"] = f"{type_nl} van {creator_name}"

        if is_empty(job.data["type_en"]):
            descriptions["en"] = f"work of art by {creator_name_en}"
        else:
            type_en = job.data["type_en"]
            descriptions["en"] = f"{type_en} by {creator_name_en}"

        if not is_empty(job.data["date"]):
            year = job.data["date"]
            descriptions["en"] = f"{year} " + descriptions["en"]
            descriptions["nl"] = descriptions["nl"] + f" uit {year}"

        # Check for duplicate item, and if it's there, create a description with
        # the inventory number
        if has_duplicate_item(bot):
            print(
                "Duplicate title/description, return a description with inventory suffix"
            )
            descriptions[
                "nl"] = f"{descriptions['nl']} (objectnummer {inventory_nr})"
            dd(descriptions)

        aliases = {"en": [title]}

        # If this is an append item, get the item instead of creating
        if not is_empty(job.data["item_qid"]) and not is_empty(
                job.data["append"]):
            print("Item exists, add to it instead")
            job.set_item_by_qid(job.data["item_qid"])
        else:
            # Try, if we get an exception change the description because
            # description is double
            try:
                job.create_item(summary, labels, descriptions, aliases)
            except Exception as e:
                print("Exception while creating", e)

                descriptions[
                    "nl"] = f"{descriptions['nl']} (objectnummer {inventory_nr})"
                print("Trying another description: %s" % descriptions["nl"])

                job.item.edit_descriptions(
                    descriptions,
                    f"Giving a complex description because we have a duplicate label/description"
                )

        # Save URL to Wayback Machine
        job.archive_url(job.data["url"])

        if is_empty(job.data["type_qid"]):
            job.item.add_item_claim(Props.INSTANCE_OF, Items.WORK_OF_ART)
        else:
            job.item.add_item_claim(Props.INSTANCE_OF, job.data["type_qid"])

        job.item.add_item_claim(Props.LOCATION,
                                Items.MUS_BOMMELVDAM,
                                references=get_ref(job))

        # If we have no collection, assume it's the default one
        if is_empty(job.data["collection_qid"]):
            collection_qid = "Q1994770"
        else:
            collection_qid = job.data["collection_qid"]

        job.item.add_string_claim(Props.INVENTORY_NR,
                                  inventory_nr,
                                  qualifiers=[
                                      job.item.get_item_claim(
                                          Props.COLLECTION, collection_qid)
                                  ],
                                  references=get_ref(job))

        job.item.add_url_claim(Props.DESCRIBED_AT_URL,
                               job.data["url"],
                               qualifiers=[
                                   job.item.get_item_claim(
                                       Props.LANGUAGE_WORK, Items.DUTCH)
                               ])

        if not is_empty(job.data["creator_qid"]):
            job.item.add_item_claim(Props.CREATOR,
                                    job.data["creator_qid"],
                                    references=get_ref(job))

        if not is_empty(job.data["date"]) and is_valid_year(job.data["date"]):
            job.item.add_time_claim(Props.INCEPTION,
                                    WbTime(year=int(job.data["date"])),
                                    references=get_ref(job))

        if is_empty(job.data["aquirement_qid"]):
            cause_qualifers = []
        else:
            cause_qualifers = [
                job.item.get_item_claim(Props.HAS_CAUSE,
                                        job.data["aquirement_qid"])
            ]

        job.item.add_item_claim(Props.COLLECTION,
                                collection_qid,
                                qualifiers=cause_qualifers,
                                references=get_ref(job))

        if not is_empty(job.data["material_qid"]):
            job.item.add_item_claim(Props.MATERIAL_USED,
                                    job.data["material_qid"],
                                    references=get_ref(job))

        if not is_empty(job.data["height"]):
            job.item.add_quantity_claim(
                Props.HEIGHT,
                job.data["height"],
                unit="http://www.wikidata.org/entity/Q174728",
                references=get_ref(job))