Python Article.load_from_uuid方法代码示例

# 需要导入模块: from article import Article [as 别名]
# 或者: from article.Article import load_from_uuid [as 别名]
def reparse_api(version=1):
    """ Reparse an already parsed and stored article with a given UUID """
    if not (1 <= version <= 1):
        return better_jsonify(valid="False", reason="Unsupported version")

    uuid = request.form.get("id", "").strip()[0:_MAX_UUID_LENGTH]
    tokens = None
    register = {}
    stats = {}

    with SessionContext(commit=True) as session:
        # Load the article
        a = ArticleProxy.load_from_uuid(uuid, session)
        if a is not None:
            # Found: Parse it (with a fresh parser) and store the updated version
            a.parse(session, verbose=True, reload_parser=True)
            # Save the tokens
            tokens = a.tokens
            # Build register of person names
            register = a.create_register(session)
            stats = dict(
                num_tokens=a.num_tokens,
                num_sentences=a.num_sentences,
                num_parsed=a.num_parsed,
                ambiguity=a.ambiguity,
            )

    # Return the tokens as a JSON structure to the client,
    # along with a name register and article statistics
    return better_jsonify(valid=True, result=tokens, register=register, stats=stats)

# 需要导入模块: from article import Article [as 别名]
# 或者: from article.Article import load_from_uuid [as 别名]
def article_api(version=1):
    """ Obtain information about an article, given its URL or id """

    if not (1 <= version <= 1):
        return better_jsonify(valid=False, reason="Unsupported version")

    if request.method == "GET":
        url = request.args.get("url")
        uuid = request.args.get("id")
    else:
        url = request.form.get("url")
        uuid = request.form.get("id")
    if url:
        url = url.strip()[0:_MAX_URL_LENGTH]
    if uuid:
        uuid = uuid.strip()[0:_MAX_UUID_LENGTH]
    if url:
        # URL has priority, if both are specified
        uuid = None
    if not url and not uuid:
        return better_jsonify(valid=False, reason="No url or id specified in query")

    with SessionContext(commit=True) as session:

        if uuid:
            a = ArticleProxy.load_from_uuid(uuid, session)
        elif url.startswith("http:") or url.startswith("https:"):
            a = ArticleProxy.load_from_url(url, session)
        else:
            a = None

        if a is None:
            return better_jsonify(valid=False, reason="Article not found")

        if a.html is None:
            return better_jsonify(valid=False, reason="Unable to fetch article")

        # Prepare the article for display
        a.prepare(session)
        register = a.create_register(session, all_names=True)
        # Fetch names of article topics, if any
        topics = (
            session.query(ArticleTopic).filter(ArticleTopic.article_id == a.uuid).all()
        )
        topics = [dict(name=t.topic.name, id=t.topic.identifier) for t in topics]

    return better_jsonify(
        valid=True,
        url=a.url,
        id=a.uuid,
        heading=a.heading,
        author=a.author,
        ts=a.timestamp.isoformat()[0:19],
        num_sentences=a.num_sentences,
        num_parsed=a.num_parsed,
        ambiguity=a.ambiguity,
        register=register,
        topics=topics,
    )

# 需要导入模块: from article import Article [as 别名]
# 或者: from article.Article import load_from_uuid [as 别名]
def speed_test(uuid):
    try:
        print("Starting speed test")
        t0 = time.time()
        with SessionContext(commit = True) as session:
            # Load the article
            a = Article.load_from_uuid(uuid, session)
            if a is not None:
                # Parse it and store the updated version
                a.parse(session, verbose = True)
        t1 = time.time()
        print("Parsing finished in {0:.2f} seconds".format(t1 - t0))
    finally:
        Article.cleanup()

# 需要导入模块: from article import Article [as 别名]
# 或者: from article.Article import load_from_uuid [as 别名]
def page():
    """ Handler for a page displaying the parse of an arbitrary web
        page by URL or an already scraped article by UUID """
    url = request.args.get("url", None)
    uuid = request.args.get("id", None)
    if url:
        url = url.strip()[0:_MAX_URL_LENGTH]
    if uuid:
        uuid = uuid.strip()[0:_MAX_UUID_LENGTH]
    if url:
        # URL has priority, if both are specified
        uuid = None
    if not url and not uuid:
        # !!! TODO: Separate error page
        return redirect(url_for("routes.main"))

    with SessionContext(commit=True) as session:

        if uuid:
            a = ArticleProxy.load_from_uuid(uuid, session)
        elif url.startswith("http:") or url.startswith("https:"):
            a = ArticleProxy.scrape_from_url(url, session)  # Forces a new scrape
        else:
            a = None

        if a is None:
            # !!! TODO: Separate error page
            return redirect(url_for("routes.main"))

        # Prepare the article for display (may cause it to be parsed and stored)
        a.prepare(session, verbose=True, reload_parser=True)
        register = a.create_register(session, all_names=True)

        # Fetch names of article topics, if any
        topics = (
            session.query(ArticleTopic).filter(ArticleTopic.article_id == a.uuid).all()
        )
        topics = [dict(name=t.topic.name, id=t.topic.identifier) for t in topics]

        # Fetch similar (related) articles, if any
        DISPLAY = 10  # Display at most 10 matches
        similar = Search.list_similar_to_article(session, a.uuid, n=DISPLAY)

        return render_template(
            "page.html", article=a, register=register, topics=topics, similar=similar
        )