加载操作中出现BigQuery错误：无法连接BigQuery服务器

Question

我似乎无法将任何数据从Socrata上传到BigQuery。我得到“加载操作中出现BigQuery错误：无法连接BigQuery服务器。”最初，我得到的错误限制为0个错误。现在，我已将CSV文件限制为一个数据行，因此出现了连接错误。下面是我的日志输出、代码和CSV。

我的日志输出：

INFO:root:--request-start--
INFO:root:-headers-start-
INFO:root:content-type: application/json
INFO:root:accept-encoding: gzip, deflate
INFO:root:accept: application/json
INFO:root:user-agent: google-api-python-client/1.3.1 (gzip)
INFO:root:-headers-end-
INFO:root:-path-parameters-start-
INFO:root:projectId: tokyo-unity-87516
INFO:root:-path-parameters-end-
INFO:root:body: {"configuration": {"load": {"projectionFields": [], "skipLeadingRows": 1, "destinationTable": {"projectId": "tokyo-unity-87516", "tableId": "data_seattle_gov_pu5n_trf42", "datasetId": "fromsocrata"}, "maxBadRecords": 0, "schema": {"fields": [{"type": "string", "name": "socrata__computed_region_2day_rhn5", "mode": "nullable"}, {"type": "string", "name": "socrata__computed_region_cyqu_gs94", "mode": "nullable"}, {"type": "string", "name": "socrata__computed_region_kuhn_3gp2", "mode": "nullable"}, {"type": "string", "name": "socrata__computed_region_q256_3sug", "mode": "nullable"}, {"type": "string", "name": "socrata_created_at", "mode": "nullable"}, {"type": "string", "name": "socrata_id", "mode": "nullable"}, {"type": "string", "name": "socrata_updated_at", "mode": "nullable"}, {"type": "string", "name": "socrata_version", "mode": "nullable"}, {"type": "string", "name": "at_scene_time", "mode": "nullable"}, {"type": "string", "name": "cad_cdw_id", "mode": "nullable"}, {"type": "string", "name": "cad_event_number", "mode": "nullable"}, {"type": "string", "name": "census_tract", "mode": "nullable"}, {"type": "string", "name": "district_sector", "mode": "nullable"}, {"type": "string", "name": "event_clearance_code", "mode": "nullable"}, {"type": "string", "name": "event_clearance_date", "mode": "nullable"}, {"type": "string", "name": "event_clearance_description", "mode": "nullable"}, {"type": "string", "name": "event_clearance_group", "mode": "nullable"}, {"type": "string", "name": "event_clearance_subgroup", "mode": "nullable"}, {"type": "string", "name": "general_offense_number", "mode": "nullable"}, {"type": "string", "name": "hundred_block_location", "mode": "nullable"}, {"type": "string", "name": "incident_location", "mode": "nullable"}, {"type": "string", "name": "incident_location_address", "mode": "nullable"}, {"type": "string", "name": "incident_location_city", "mode": "nullable"}, {"type": "string", "name": "incident_location_state", "mode": "nullable"}, {"type": "string", "name": "incident_location_zip", "mode": "nullable"}, {"type": "string", "name": "initial_type_description", "mode": "nullable"}, {"type": "string", "name": "initial_type_group", "mode": "nullable"}, {"type": "string", "name": "initial_type_subgroup", "mode": "nullable"}, {"type": "string", "name": "latitude", "mode": "nullable"}, {"type": "string", "name": "longitude", "mode": "nullable"}, {"type": "string", "name": "zone_beat", "mode": "nullable"}]}}}, "jobReference": {"projectId": "tokyo-unity-87516", "jobId": "bqjob_r1ad3a0e093a97474_000001545d58a62c_1"}}
INFO:root:query: ?alt=json
INFO:root:--request-end--
INFO:apiclient.discovery:URL being requested: POST https://www.googleapis.com/upload/bigquery/v2/projects/tokyo-unity-87516/jobs?uploadType=resumable&alt=json
INFO:root:--response-start--
INFO:root:status: 400
INFO:root:alternate-protocol: 443:quic
INFO:root:content-length: 37
INFO:root:server: UploadServer
INFO:root:x-guploader-uploadid: AEnB2Uo0-JZcSyKgFHlFD6MkWm1XfD6n-nhcVk4TPxgqqdjWkQzNCXzNPuTxwxvDrffWyYlxX_oS8ZvaKJjc0YKJmK0yTTFQtQ
INFO:root:date: Thu, 28 Apr 2016 14:49:31 GMT
INFO:root:alt-svc: quic=":443"; ma=2592000; v="33,32,31,30,29,28,27,26,25"
INFO:root:content-type: text/html; charset=UTF-8
INFO:root:Failed to parse Content-Range header.
INFO:root:--response-end--
BigQuery error in load operation: Could not connect with BigQuery server.
Http response status: 400
Http response content:
Failed to parse Content-Range header.

我的代码：

import rethinkdb as r
import os
r.connect( "localhost", 28015).repl()
try:
    r.db_create('queryplayground').run()
    r.db('queryplayground').table_create('socrata_datasets').run()
    r.db('queryplayground').table_create('third_party_creds').run()
    socrata_app_token = raw_input('Your Socrata app token\n')
    r.db('queryplayground').table_create('third_party_creds').insert({"id": "socrata", "app_token": socrata_app_token}).run()
    r.db('queryplayground').table('socrata_datasets').insert({"id": "data_seattle_gov_pu5n_trf4", "domain": "data.seattle.gov", "datasetid": "pu5n-trf4"}).run()
except:
    pass
import requests
r.db('queryplayground').table('socrata_datasets').replace({"id": "data_seattle_gov_pu5n_trf4", "domain": "data.seattle.gov", "datasetid": "pu5n-trf4"}).run()
for dataset in r.db('queryplayground').table('socrata_datasets').run():
    app_token = r.db('queryplayground').table('third_party_creds').get('socrata').run()['app_token']
    if not 'socrata_created_at' in dataset:
        local_filename = dataset['id']+'.csv'
        # NOTE the stream=True parameter
        url = 'https://%s/resource/%s.csv?$select=:*,*&$limit=1' % (dataset['domain'], dataset['datasetid'])
        req = requests.get(url, stream=True)
        with open(local_filename, 'wb') as f:
            for chunk in req.iter_content(chunk_size=1024): 
                if chunk: # filter out keep-alive new chunks
                    f.write(chunk)
        url = 'https://%s/resource/%s.json?' % (dataset['domain'], dataset['datasetid'])
        url += '$order=:created_at DESC&$limit=1&$select=:created_at&$$app_token=' + app_token
        print url
        dataset['socrata_created_at'] = requests.get(url).json()[0][':created_at']
        r.db('queryplayground').table('socrata_datasets').update(dataset).run()
        local_filename
        newline = os.linesep # Defines the newline based on your OS.

        source_fp = open(local_filename, 'r')
        target_fp = open('2'+local_filename, 'w')
        first_row = True
        for row in source_fp:
            if first_row:
                row = row.replace(':', 'socrata_').replace('@', '_')
                headers = row.strip().split(',')
                first_row = False
            target_fp.write(row)
        schema = []
        for col in headers:
            schema.append({"name": col.strip('"'), "type": "string", "mode": "nullable"})
        #schema = ','.join([col.strip('"')+':string:nullable' for col in headers])
        #print schema
        import json
        with open('schema.json', 'w') as f:
            f.write(json.dumps(schema))
        import json
        cmd = 'bq load --apilog=- --schema=schema.json --skip_leading_rows=1 fromsocrata.%s %s' % (dataset['id']+'2', '2'+local_filename)
        print cmd
        os.system(cmd)

我的csv文件：

"socrata__computed_region_2day_rhn5","socrata__computed_region_cyqu_gs94","socrata__computed_region_kuhn_3gp2","socrata__computed_region_q256_3sug","socrata_created_at","socrata_id","socrata_updated_at","socrata_version","at_scene_time","cad_cdw_id","cad_event_number","census_tract","district_sector","event_clearance_code","event_clearance_date","event_clearance_description","event_clearance_group","event_clearance_subgroup","general_offense_number","hundred_block_location","incident_location","incident_location_address","incident_location_city","incident_location_state","incident_location_zip","initial_type_description","initial_type_group","initial_type_subgroup","latitude","longitude","zone_beat"
,,"30","18081","2016-02-04T00:25:02.285Z","row-xkda.347u.ym5x","2016-04-04T19:09:32.730Z","rv-7pyt_iydi~nc25",,"1001512","12000437316","8100.1000","M","161","2012-12-23T13:59:00.000","TRESPASS","TRESPASS","TRESPASS","2012437316","1XX BLOCK OF PINE ST","POINT (-122.340367 47.610046)",,,,,,,,"47.610046276","-122.340367253","M3"

Answer 1

我最终发现，问题是我在尝试让BigQuery工具上传文件之前没有关闭它。