comparison v1

@@ -1019,15 +1019,108 @@ BASE_XML_TEXT = """<?xml version="1.0" encoding="utf-8"?>
 CDATA_TEXT = """<![CDATA[<Exec xmlns="http://tempuri.org/"><strXMLParm>&lt;Request&gt;&lt;Framework&gt;&lt;UserName&gt;SFCHINA\qqcout0&lt;/UserName&gt;&lt;GUID&gt;70d0efcb-3bc2-4018-ac4e-681c8f3131b6&lt;/GUID&gt;&lt;DetailedTracingEnabled&gt;False&lt;/DetailedTracingEnabled&gt;&lt;ServiceName&gt;AMSWebService&lt;/ServiceName&gt;&lt;SupportsRedirection&gt;true&lt;/SupportsRedirection&gt;&lt;ServiceType&gt;Service&lt;/ServiceType&gt;&lt;/Framework&gt;&lt;Parms&gt;&lt;InputXML type="string"&gt;&amp;lt;?xml version="1.0" encoding="utf-16"?&amp;gt;&amp;lt;InputXML&amp;gt; &amp;lt;Result&amp;gt; {0} &amp;lt;/Result&amp;gt;&amp;lt;AuthorizationData&amp;gt;&amp;lt;ServiceComponent&amp;gt;OCR&amp;lt;/ServiceComponent&amp;gt;&amp;lt;RoleId/&amp;gt;&amp;lt;CompanyId/&amp;gt;&amp;lt;/AuthorizationData&amp;gt;&amp;lt;/InputXML&amp;gt;&lt;/InputXML&gt;&lt;/Parms&gt;&lt;/Request&gt;</strXMLParm></Exec>]]>"""
 
 RESULT_MAPPING = {
-    MVI_CLASSIFY: 'mvi_ocr',
+    # MVI_CLASSIFY: 'mvi_ocr',
     IC_CLASSIFY: 'ic_ocr',
     RP_CLASSIFY: 'rp_ocr',
-    BC_CLASSIFY: 'bc_ocr',
+    # BC_CLASSIFY: 'bc_ocr',
     BL_CLASSIFY: 'bl_ocr',
-    UCI_CLASSIFY: 'uci_ocr',
+    # UCI_CLASSIFY: 'uci_ocr',
     EEP_CLASSIFY: 'eep_ocr',
     DL_CLASSIFY: 'dl_ocr',
     PP_CLASSIFY: 'pp_ocr',
     MVC_CLASSIFY: 'mvc_ocr',
-    VAT_CLASSIFY: 'vat_ocr',
+    # VAT_CLASSIFY: 'vat_ocr',
 }
+
+COMPARE_FIELDS = ('ic_ocr', 'rp_ocr', 'bl_ocr', 'eep_ocr', 'dl_ocr', 'pp_ocr', 'mvc_ocr')
+
+# 身份证
+ITPRC = {
+    'customerChineseName': ('姓名', 'common_compare', {}),
+    'idNum': ('公民身份号码', 'common_compare', {}),
+    # 20200410-20250410 OCR识别为长期，向GCAP发送：2099-12-31 00:00:00.0
+    'idExpiryDate': ('有效期限', 'date_compare', {'long': True, 'ocr_split': True, 'input_replace': ''}),
+}
+
+# 护照
+ITPSP = {
+    'customerChineseName': ('英文姓名', 'common_compare', {}),
+    'idNum': ('护照号码', 'common_compare', {}),
+    'idExpiryDate': ('有效期至', 'date_compare', {'input_replace': ''}),  # 20250410
+    'dateOfBirth': ('出生日期', 'date_compare', {'input_replace': ''}),  # 20250410
+}
+
+# 港澳台通行证
+ITHKM_ITTID = {
+    'customerChineseName': ('中文名', 'common_compare', {}),
+    'idNum': ('证件号码', 'common_compare', {}),
+    'idExpiryDate': ('有效期限', 'date_compare', {'ocr_split': True, 'input_replace': '.'}),  # 2013.10.24-2023.10.23
+    'dateOfBirth': ('出生日期', 'date_compare', {'input_replace': '.'}),  # 2023.10.23
+    # 'secondIdNum': ''
+}
+
+# 居住证
+ITRES = {
+    'customerChineseName': ('姓名', 'common_compare', {}),
+    'idNum': ('公民身份号码', 'common_compare', {}),
+    'idExpiryDate': ('有效期限', 'date_compare', {'ocr_split': True, 'input_replace': ''}),  # 20200410-20250410
+    'secondIdNum': ('通行证号码', 'common_compare', {})
+}
+
+ID_TYPE_COMPARE = {
+    'ITPRC': {'model_field': 'ic_ocr', 'compare_field': ITPRC},
+    'ITPSP': {'model_field': 'pp_ocr', 'compare_field': ITPSP},
+    'ITHKM': {'model_field': 'eep_ocr', 'compare_field': ITHKM_ITTID},
+    'ITTID': {'model_field': 'eep_ocr', 'compare_field': ITHKM_ITTID},
+    'ITRES': {'model_field': 'rp_ocr', 'compare_field': ITRES},
+}
+
+# 1. 分别对比 POS车架号 vs 车辆登记正，POS车架号 vs 行驶证
+#     a）两种比对 均一致：Y
+#     b）其中一个 不一致：N，向GCAP发送：不一致的OCR识别结果
+#     c）两中比对 均不一致：N，向GCAP发送：车辆登记证 & 行驶证识别结果
+# 车辆登记证
+PCUSD_MVC = {
+    'vinNo': ('9.车辆识别代号/车架号', 'common_compare', {}),
+    'manufactureDate': ('32.车辆出厂日期', 'common_compare', {}),
+    'firstRegistrationDate': ('3.登记日期', 'common_compare', {}),
+}
+
+# 行驶证
+PCUSD_DL = {
+    'vinNo': ('车辆识别代码', 'common_compare', {}),
+    # 'manufactureDate': '',
+    # 'firstRegistrationDate': '',
+}
+
+# 营业执照
+TCCOR = {
+    'customerChineseName': ('企业名称', 'common_compare', {}),
+    'legalRepName': ('经营者姓名', 'common_compare', {}),
+    'idNum': ('注册号', 'common_compare', {}),
+    'businessLicenseNo': ('注册号', 'common_compare', {}),
+    'taxRegistrationCode': ('注册号', 'common_compare', {}),
+    'incorporationDate': ('成立日期', 'date_compare', {'ocr_replace': True}),  # 2017年07月11日
+    # 2017年07月11日至长期  1. OCR识别为长期，向GCAP发送：2099-12-31 00:00:00.0
+    'businessLicenseDueDate': ('营业期限', 'date_compare', {'long': True, 'ocr_replace': True}),
+    'capitalRegAmount': ('注册资本', 'rmb_compare', {}),  # 壹拾万元整 将OCR识别结果（人民币大写）转化为数字
+}
+
+TCSEP = {
+    'companyName': ('企业名称', 'common_compare', {}),
+    'registeredCapital': ('注册资本', 'rmb_compare', {}),  # 壹拾万元整 将OCR识别结果（人民币大写）转化为数字
+    'selfEmployedSubType': ('企业类型', 'type_compare', {}),  # 有限责任公司
+}
+
+# 1. POS数据OCR识别结果对应关系如下：
+#     a）Individual Businessman CSIBM => 个体工商户
+#     b）Small and Micro Enterprise Owners CSSME => 个人独资企业、有限合伙企业、股份合作制、有限责任公司（***）【只需比对“有限责任公司”即可】
+#     c）Others CSOTH => 识别结果不一致时，向GCAP发送 OCR的识别结果
+
+RESULT_Y = 'Y'
+RESULT_N = 'N'
+RESULT_NA = 'NA'
+
+
+
+

@@ -12,3 +12,7 @@ class OCR2Exception(Exception):
 
 class OCR4Exception(Exception):
     pass
+
+
+class GCAPException(Exception):
+    pass

@@ -888,6 +888,7 @@ class Command(BaseCommand, LoggerMixin):
                                     os.remove(excel_path)
                         finally:
                             # TODO 识别结果存一张表，方便跑报表
+                            if doc.document_scheme == consts.DOC_SCHEME_LIST[0]:
                                 try:
                                     # 更新OCR累计识别结果表
                                     result_class = HILOCRResult if business_type == consts.HIL_PREFIX else AFCOCRResult

@@ -3,6 +3,7 @@ from requests.auth import HTTPBasicAuth
 from settings import conf
 from common.tools.dict_to_xml import dicttoxml, escape_xml
 from apps.doc import consts
+from apps.doc.exceptions import GCAPException
 
 
 class GCAP:
@@ -19,11 +20,10 @@ class GCAP:
         comparison_xml = dicttoxml(comparison_res, root=False, attr_type=False)
         return consts.BASE_XML_TEXT.format(consts.CDATA_TEXT.format(escape_xml(comparison_xml))).encode('utf-8')
 
-    def send(self, comparison_res):
-        data = self.dict_to_xml(comparison_res)
+    def send(self, data):
         response = requests.post(self.url, headers=self.headers, data=data, verify=False, auth=self.auth)
-        return response
-
-
+        if response.status_code != 200:
+            raise GCAPException('GCAP response with code: {0}'.format(response.status_code))
 
 
+gcap = GCAP()

+import json
 import logging
+import traceback
 from . import app
-from apps.doc.models import AFCDoc
+from apps.doc.models import AFCOCRResult, HILOCRResult, AFCComparisonInfo, HILComparisonInfo
+from apps.doc import consts
+from apps.doc.ocr.gcap import gcap
+from apps.doc.exceptions import GCAPException
+from common.tools.comparison import cp
 
 compare_log = logging.getLogger('compare')
+log_base = '[CA Compare]'
 
 
 @app.task
@@ -10,8 +17,190 @@ def compare(application_id, application_entity, uniq_seq, ocr_res_id):
     # POS: application_id, application_entity, uniq_seq, None
     # OCR: application_id, business_type(application_entity), None, ocr_res_id
 
+    compare_log.info('{0} [receive task] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}]'.format(
+        log_base, application_entity, application_id, uniq_seq, ocr_res_id))
+
     # 根据application_id查找最新的比对信息，如果没有，结束
-    # 分析比对信息，需要比对的license
+    comparison_class = HILComparisonInfo if application_entity == consts.HIL_PREFIX else AFCComparisonInfo
+    last_obj = comparison_class.objects.filter(application_id=application_id).last()
+    if last_obj is None:
+        compare_log.info('{0} [comparison info empty] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}]'.format(
+            log_base, application_entity, application_id, uniq_seq, ocr_res_id
+        ))
+        return
+
     # 根据application_id查找OCR累计结果指定license字段，如果没有，结束
-    # 比对信息，将比对结果发送GCAP
-    pass
+    result_class = HILOCRResult if application_entity == consts.HIL_PREFIX else AFCOCRResult
+    if ocr_res_id is None:
+        ocr_res_dict = result_class.objects.filter(application_id=application_id).values(*consts.COMPARE_FIELDS).first()
+    else:
+        ocr_res_dict = result_class.objects.filter(id=ocr_res_id).values(*consts.COMPARE_FIELDS).first()
+    if ocr_res_dict is None:
+        compare_log.info('{0} [ocr info empty] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}]'.format(
+            log_base, application_entity, application_id, uniq_seq, ocr_res_id
+        ))
+        return
+
+    # 比对信息
+    comparison_res = {
+        'OCR_Input': {
+            'uniqSeq': last_obj.uniq_seq,
+            'applicationId': application_id,
+            'applicationEntity': application_entity,
+            'applicationVersion': last_obj.application_version,
+            'vehicleStatus': last_obj.vehicle_status,
+            # 'wholeResult': 'Y',
+            'wholeResultMessage': '',
+            'applicationLink': '',
+        }
+    }
+
+    res_set = set()
+
+    is_sep = True if last_obj.customer_type == consts.CUSTOMER_TYPE[5] else False
+
+    individual_cus_info_list = json.loads(last_obj.individual_cus_info)
+    for individual_cus_info in individual_cus_info_list:
+        individual_cus_info['customerType'] = last_obj.customer_type
+        # sep营业执照
+        if is_sep and individual_cus_info.get('companyName') is not None:
+            sep_is_find = False
+            sep_ocr_res_str = ocr_res_dict.get('bl_ocr')
+            if sep_ocr_res_str is not None:
+                sep_ocr_list = json.loads(sep_ocr_res_str)
+                for sep_ocr in sep_ocr_list:
+                    company_name = sep_ocr.get(consts.TCSEP.get('companyName')[0])
+                    if company_name is None or company_name != individual_cus_info.get('companyName'):
+                        continue
+                    sep_is_find = True
+                    for sep_field, sep_tuple in consts.TCSEP.items():
+                        sep_res = getattr(cp, sep_tuple[1])(
+                            individual_cus_info.get(sep_field), sep_ocr.get(sep_tuple[0]), sep_tuple[2])
+                        individual_cus_info[sep_field + 'Result'] = sep_res
+                        res_set.add(sep_res)
+                    break
+            if not sep_is_find:
+                res_set.add(consts.RESULT_N)
+                for field in consts.TCSEP.keys():
+                    individual_cus_info[field + 'Result'] = consts.RESULT_N
+
+        # 个人信息证件
+        id_type = individual_cus_info.get('idType')
+        compare_target = consts.ID_TYPE_COMPARE.get(id_type)
+        if compare_target is None:
+            continue
+        is_find = False
+        ocr_res_str = ocr_res_dict.get(compare_target.get('model_field'))
+        if ocr_res_str is not None:
+            ocr_res_list = json.loads(ocr_res_str)
+            compare_target_dict = compare_target.get('compare_field')
+            for ocr_res in ocr_res_list:
+                base_name = ocr_res.get(compare_target_dict.get('customerChineseName')[0])
+                if base_name is None or base_name != individual_cus_info.get('customerChineseName'):  # TODO 特殊姓名比对
+                    continue
+                is_find = True
+                for compare_field, compare_tuple in compare_target.get('compare_field').items():
+                    compare_res = getattr(cp, compare_tuple[1])(
+                        individual_cus_info.get(compare_field), ocr_res.get(compare_tuple[0]), compare_tuple[2])
+                    individual_cus_info[compare_field + 'Result'] = compare_res
+                    res_set.add(compare_res)
+                break
+        if not is_find:
+            res_set.add(consts.RESULT_N)
+            for field in compare_target.get('compare_field').keys():
+                individual_cus_info[field + 'Result'] = consts.RESULT_N
+
+    comparison_res['individualCusInfo'] = individual_cus_info_list
+
+    if last_obj.corporate_cus_info is not None:
+        corporate_cus_info = json.loads(last_obj.corporate_cus_info)
+        corporate_cus_info['customerType'] = last_obj.customer_type
+
+        is_bl_find = False
+        bl_ocr_res_str = ocr_res_dict.get('bl_ocr')
+        if bl_ocr_res_str is not None:
+            bl_ocr_list = json.loads(bl_ocr_res_str)
+            for bl_ocr in bl_ocr_list:
+                company_name = bl_ocr.get(consts.TCCOR.get('customerChineseName')[0])
+                if company_name is None or company_name != corporate_cus_info.get('customerChineseName'):
+                    continue
+                is_bl_find = True
+                for bl_field, bl_tuple in consts.TCCOR.items():
+                    bl_res = getattr(cp, bl_tuple[1])(
+                        corporate_cus_info.get(bl_field), bl_ocr.get(bl_tuple[0]), bl_tuple[2])
+                    corporate_cus_info[bl_field + 'Result'] = bl_res
+                    res_set.add(bl_res)
+                break
+        if not is_bl_find:
+            res_set.add(consts.RESULT_N)
+            for field in consts.TCCOR.keys():
+                corporate_cus_info[field + 'Result'] = consts.RESULT_N
+
+        comparison_res['corporateCusInfo'] = corporate_cus_info
+
+    if last_obj.vehicle_status == consts.VEHICLE_STATUS[0] and last_obj.usedcar_info is not None:
+        usedcar_info = json.loads(last_obj.usedcar_info)
+
+        is_usedcar_find = False
+        mvc_ocr_res_str = ocr_res_dict.get('mvc_ocr')
+        if mvc_ocr_res_str is not None:
+            mvc_ocr_list = json.loads(mvc_ocr_res_str)
+            for mvc_ocr in mvc_ocr_list:
+                vin_no = mvc_ocr.get(consts.PCUSD_MVC.get('vinNo')[0])
+                if vin_no is None or vin_no != usedcar_info.get('vinNo'):
+                    continue
+                is_usedcar_find = True
+                for mvc_field, mvc_tuple in consts.PCUSD_MVC.items():
+                    mvc_res = getattr(cp, mvc_tuple[1])(
+                        usedcar_info.get(mvc_field), mvc_ocr.get(mvc_tuple[0]), mvc_tuple[2])
+                    usedcar_info[mvc_field + 'Result'] = mvc_res
+                    res_set.add(mvc_res)
+
+                dl_find = False
+                dl_ocr_res_str = ocr_res_dict.get('dl_ocr')
+                if dl_ocr_res_str is not None:
+                    dl_ocr_list = json.loads(dl_ocr_res_str)
+                    for dl_ocr in dl_ocr_list:
+                        dl_vin_no = dl_ocr.get(consts.PCUSD_DL.get('vinNo')[0])
+                        if dl_vin_no is None or dl_vin_no != usedcar_info.get('vinNo'):
+                            continue
+                        dl_find = True
+                        break
+                if not dl_find:
+                    res_set.add(consts.RESULT_N)
+                    usedcar_info['vinNo' + 'Result'] = consts.RESULT_N
+                break
+        if not is_usedcar_find:
+            res_set.add(consts.RESULT_N)
+            for field in consts.PCUSD_MVC.keys():
+                usedcar_info[field + 'Result'] = consts.RESULT_N
+        comparison_res['usedCarInfo'] = usedcar_info
+
+    comparison_res['wholeResult'] = consts.RESULT_N if consts.RESULT_N in res_set else consts.RESULT_Y
+    print(comparison_res)
+
+    # 将比对结果发送GCAP
+    # try:
+    #     data = gcap.dict_to_xml(comparison_res)
+    # except Exception as e:
+    #     compare_log.error('{0} [dict to xml failed] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}] '
+    #                       '[error={5}]'.format(log_base, application_entity, application_id, uniq_seq, ocr_res_id,
+    #                                            traceback.format_exc()))
+    # else:
+    #     try:
+    #         for times in range(consts.RETRY_TIMES):
+    #             try:
+    #                 gcap.send(data)
+    #             except Exception as e:
+    #                 gcap_exc = str(e)
+    #             else:
+    #                 break
+    #         else:
+    #             raise GCAPException(gcap_exc)
+    #     except Exception as e:
+    #         compare_log.error('{0} [gcap failed] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}] '
+    #                           '[error={5}]'.format(log_base, application_entity, application_id, uniq_seq,
+    #                                                ocr_res_id, traceback.format_exc()))
+    #     else:
+    #         compare_log.info('{0} [task success] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}]'.format(
+    #             log_base, application_entity, application_id, uniq_seq, ocr_res_id))

+import re
+from .rmb_upper import to_rmb_upper
+
+
+class Comparison:
+
+    def __init__(self):
+        self.CSIBM = 'CSIBM'
+        self.CSSME = 'CSSME'
+        self.CSOTH = 'CSOTH'
+
+        self.TYPE_MAPPING = (
+            (r'个体工商户', self.CSIBM),
+            (r'有限责任公司', self.CSSME),
+            (r'个人独资企业', self.CSSME),
+            (r'有限合伙企业', self.CSSME),
+            (r'股份合作制', self.CSSME),
+        )
+
+        self.RESULT_Y = 'Y'
+        self.RESULT_N = 'N'
+        self.RESULT_NA = 'NA'  # TODO NA情况
+
+    def build_res(self, result):
+        if result:
+            return self.RESULT_Y
+        else:
+            return self.RESULT_N
+
+    def common_compare(self, input_str, ocr_str, **kwargs):
+        return self.build_res(input_str == ocr_str)
+
+    def date_compare(self, input_str, ocr_str, **kwargs):
+        if kwargs.get('long', False) and '长期' in ocr_str:
+            return '2099-12-31'
+        if kwargs.get('ocr_split', False):
+            ocr_str = ocr_str.split('-')[-1]
+        if kwargs.get('ocr_replace', False):
+            ocr_str = ocr_str.replace('年', '-').replace('月', '-').replace('日', '')
+        if kwargs.get('input_replace') is not None:
+            input_str = input_str.replace('-', kwargs.get('replace'))
+        return self.build_res(input_str == ocr_str)
+
+    def rmb_compare(self, input_str, ocr_str, **kwargs):
+        input_rmb_upper = to_rmb_upper(float(input_str))
+        return self.build_res(input_rmb_upper == ocr_str)
+
+    def type_compare(self, input_str, ocr_str, **kwargs):
+        for map_tuple in self.TYPE_MAPPING:
+            if re.search(map_tuple[0], ocr_str) is not None:
+                compare_str = map_tuple[1]
+                break
+        else:
+            compare_str = self.CSOTH
+
+        return self.build_res(input_str == compare_str)
+
+
+cp = Comparison()
+
+

+from io import StringIO
+import math
+
+_RMB_DIGITS = ['零', '壹', '贰', '叁', '肆', '伍', '陆', '柒', '捌', '玖' ]
+_SECTION_CHARS = ['', '拾', '佰', '仟', '万' ]
+
+
+def to_rmb_upper(price):
+    price = round(price, 2)
+    integer_part = int(price)
+    wanyi_part = integer_part // 1000000000000
+    yi_part = integer_part % 1000000000000 // 100000000
+    wan_part = integer_part % 100000000 // 10000
+    qian_part = integer_part % 10000
+    dec_part = int(round(price * 100 % 100))
+
+    strio = StringIO()
+
+    zero_count = 0
+    #处理万亿以上的部分
+    if integer_part >= 1000000000000 and wanyi_part > 0:
+        zero_count = _parse_integer(strio, wanyi_part, zero_count, True)
+        strio.write('万')
+
+    #处理亿到千亿的部分
+    if integer_part >= 100000000 and yi_part > 0:
+        is_first_section = integer_part >= 100000000 and integer_part < 1000000000000
+        zero_count = _parse_integer(strio, yi_part, zero_count, is_first_section)
+        strio.write('亿')
+
+    #处理万的部分
+    if integer_part >= 10000 and wan_part > 0:
+        is_first_section = integer_part >= 1000 and integer_part < 10000000
+        zero_count = _parse_integer(strio, wan_part, zero_count, is_first_section)
+        strio.write('万')
+
+    #处理千及以后的部分
+    if qian_part > 0:
+        is_first_section = integer_part < 1000
+        zero_count = _parse_integer(strio, qian_part, zero_count, is_first_section)
+    else:
+        zero_count += 1
+    if integer_part > 0:
+        strio.write('元')
+
+    #处理小数
+    if dec_part > 0:
+        _parse_decimal(strio, integer_part, dec_part, zero_count)
+    elif dec_part == 0 and integer_part > 0:
+        strio.write('整')
+    else:
+        strio.write('零元整')
+
+    return strio.getvalue()
+
+
+def _parse_integer(strio, value, zero_count = 0, is_first_section = False):
+    assert value > 0 and value <= 9999
+    ndigits = int(math.floor(math.log10(value))) + 1
+    if value < 1000 and not is_first_section:
+        zero_count += 1
+    for i in range(0, ndigits):
+        factor = int(pow(10, ndigits - 1 - i))
+        digit = int(value / factor)
+        if digit != 0:
+            if zero_count > 0:
+                strio.write('零')
+            strio.write(_RMB_DIGITS[digit])
+            strio.write(_SECTION_CHARS[ndigits - i - 1])
+            zero_count = 0
+        else:
+            zero_count += 1
+        value -= value // factor * factor
+    return zero_count
+
+
+def _parse_decimal(strio, integer_part, value, zero_count):
+    assert value > 0 and value <= 99
+    jiao = value // 10
+    fen = value % 10
+    if zero_count > 0 and (jiao > 0 or fen > 0) and integer_part > 0:
+        strio.write('零')
+    if jiao > 0:
+        strio.write(_RMB_DIGITS[jiao])
+        strio.write('角')
+    if zero_count == 0 and jiao == 0 and fen > 0 and integer_part > 0:
+        strio.write('零')
+    if fen > 0:
+        strio.write(_RMB_DIGITS[fen])
+        strio.write('分')
+    else:
+        strio.write('整')
\ No newline at end of file