add OAuth

@@ -11,6 +11,8 @@ tags:
 schemes:
 - https
 - http
+security:
+- OAuth2: []
 paths:
   /api/create/v1:
     post:
@@ -32,6 +34,43 @@ paths:
           description: ok
           schema:
             $ref: '#/definitions/ApiResponse'
+  /api/priority/v1:
+    post:
+      tags:
+      - doc
+      summary: GCAP提高申请单对应文件优先级
+      consumes:
+      - application/json
+      produces:
+      - application/json
+      parameters:
+      - in: body
+        name: body
+        required: true
+        schema:
+          $ref: '#/definitions/Application'
+      responses:
+        200:
+          description: ok
+          schema:
+            $ref: '#/definitions/ApiResponse'
+securityDefinitions:
+  OAuth2:
+    type: oauth2
+    flow: application
+    description: 'This API uses OAuth 2 with the application(clientCredentials) grant
+      flow.
+
+      client_id=sMlciTkppsMzARwHpCVarm5q7DP2Vucj3ny8JFhw
+
+      client_secret=WNoOilDx140ZLcenDKfsnikv7S2LIFs60DciYoqnrZaYLqYsKpcmt7mJIL69o9AEf84uQvRnS3K2UioxfjNyImjR4UOyXbDcF6qYgTLC4KDVByKFdVhKfrn2Lc4q4BNW
+
+      scopes=write
+
+      '
+    tokenUrl: https://staging-bmw-ocr.situdata.com/api/oauth/token/
+    scopes:
+      write: Grants write access
 responses:
   ErrorResponse:
     description: 调用异常， 具体情况请参考`HTTP`状态码和`code`字段
@@ -130,6 +169,54 @@ definitions:
             description: 元数据版本ID
             type: string
             example: '8410480'
+  Application:
+    type: object
+    required:
+    - APPLICATION_INFORMATION
+    properties:
+      APPLICATION_INFORMATION:
+        description: 申请单信息
+        type: object
+        required:
+        - SUBMIT_DATETIME
+        - STATUS
+        - ENTITY
+        - RATING
+        - APPLICATION_ID
+        - APPLICATION_VERSION
+        - INTERMEDIATE_DECISION
+        properties:
+          SUBMIT_DATETIME:
+            description: 提交时间
+            type: string
+            example: 2020-07-08 18:33:31+08:00
+          STATUS:
+            description: 状态
+            type: integer
+            example: 42
+          ENTITY:
+            description: 业务类型
+            type: string
+            example: CO00001
+            enum:
+            - CO00001
+            - CO00002
+          RATING:
+            description: 排名
+            type: integer
+            example: 4
+          APPLICATION_ID:
+            description: 申请id
+            type: string
+            example: CH-B0011010101
+          APPLICATION_VERSION:
+            description: 申请版本
+            type: integer
+            example: 1
+          INTERMEDIATE_DECISION:
+            description: ''
+            type: string
+            example: MUW
   ApiResponse:
     description: 响应对象，code字段用于表示响应的状态; data字段用于存放响应内容
     type: object

@@ -9,6 +9,7 @@ certifi==2020.6.20
 chardet==3.0.4
 defusedxml==0.6.0
 Django==2.1
+django-oauth-toolkit==1.3.2
 djangorestframework==3.9.0
 djangorestframework-jwt==1.11.0
 idna==2.9
@@ -17,6 +18,7 @@ isodate==0.6.0
 lxml==4.5.1
 marshmallow==3.6.1
 multidict==4.7.6
+oauthlib==3.1.0
 pdfminer3k==1.3.4
 Pillow==7.1.2
 ply==3.11

+from django.contrib.auth import get_user_model
+from oauth2_provider.contrib.rest_framework import OAuth2Authentication
+from oauth2_provider.oauth2_backends import get_oauthlib_core
+
+
+class OAuth2AuthenticationWithUser(OAuth2Authentication):
+    """
+    OAuth 2 authentication backend using `django-oauth-toolkit`
+    """
+    www_authenticate_realm = "api"
+    user = get_user_model().objects.first()
+
+    def authenticate(self, request):
+        """
+        Returns two-tuple of (user, token) if authentication succeeds,
+        or None otherwise.
+        """
+        oauthlib_core = get_oauthlib_core()
+        valid, r = oauthlib_core.verify_request(request, scopes=[])
+        if valid:
+            return self.user, r.access_token
+        request.oauth2_error = getattr(r, "oauth2_error", {})
+        return None

+import os
+import fitz
+import signal
+from PIL import Image
+from io import BytesIO
+
+from django.core.management import BaseCommand
+from common.mixins import LoggerMixin
+
+
+class Command(BaseCommand, LoggerMixin):
+
+    def __init__(self):
+        super().__init__()
+        self.log_base = '[pdf to img]'
+        # 处理文件开关
+        self.switch = True
+        # pdf页面转图片
+        self.zoom_x = 2.0
+        self.zoom_y = 2.0
+        self.trans = fitz.Matrix(self.zoom_x, self.zoom_y).preRotate(0)  # zoom factor 2 in each dimension
+        # 优雅退出信号：15
+        signal.signal(signal.SIGTERM, self.signal_handler)
+
+    def signal_handler(self, sig, frame):
+        self.switch = False  # 停止处理文件
+
+    @staticmethod
+    def getimage(pix):
+        if pix.colorspace.n != 4:
+            return pix
+        tpix = fitz.Pixmap(fitz.csRGB, pix)
+        return tpix
+
+    def recoverpix(self, doc, item):
+        x = item[0]  # xref of PDF image
+        s = item[1]  # xref of its /SMask
+        is_rgb = True if item[5] == 'DeviceRGB' else False
+
+        # RGB
+        if is_rgb:
+            if s == 0:
+                return doc.extractImage(x)
+            # we need to reconstruct the alpha channel with the smask
+            pix1 = fitz.Pixmap(doc, x)
+            pix2 = fitz.Pixmap(doc, s)  # create pixmap of the /SMask entry
+
+            # sanity check
+            if not (pix1.irect == pix2.irect and pix1.alpha == pix2.alpha == 0 and pix2.n == 1):
+                pix2 = None
+                return self.getimage(pix1)
+
+            pix = fitz.Pixmap(pix1)  # copy of pix1, alpha channel added
+            pix.setAlpha(pix2.samples)  # treat pix2.samples as alpha value
+            pix1 = pix2 = None  # free temp pixmaps
+            return self.getimage(pix)
+
+        # GRAY/CMYK
+        pix1 = fitz.Pixmap(doc, x)
+        pix = fitz.Pixmap(pix1)  # copy of pix1, alpha channel added
+
+        if s != 0:
+            pix2 = fitz.Pixmap(doc, s)  # create pixmap of the /SMask entry
+
+            # sanity check
+            if not (pix1.irect == pix2.irect and pix1.alpha == pix2.alpha == 0 and pix2.n == 1):
+                pix2 = None
+                return self.getimage(pix1)
+
+            pix.setAlpha(pix2.samples)  # treat pix2.samples as alpha value
+
+        pix1 = pix2 = None  # free temp pixmaps
+
+        pix = fitz.Pixmap(fitz.csRGB, pix)  # GRAY/CMYK to RGB
+        return self.getimage(pix)
+
+    @staticmethod
+    def get_img_data(pix):
+        if type(pix) is dict:  # we got a raw image
+            ext = pix["ext"]
+            img_data = pix["image"]
+        else:  # we got a pixmap
+            ext = 'png'
+            img_data = pix.getPNGData()
+        return ext, img_data
+
+    @staticmethod
+    def split_il(il):
+        img_il_list = []
+        start = 0
+        length = len(il)
+        for i in range(length):
+            if i == start:
+                if i == length - 1:
+                    img_il_list.append(il[start: length])
+                continue
+            elif i == length - 1:
+                img_il_list.append(il[start: length])
+                continue
+            if il[i][2] != il[i - 1][2]:
+                img_il_list.append(il[start: i])
+                start = i
+            elif il[i][3] != il[i - 1][3]:
+                img_il_list.append(il[start: i + 1])
+                start = i + 1
+        return img_il_list
+
+    def handle(self, *args, **kwargs):
+        pdf_dir = '/Users/clay/Desktop/普通打印-部分无线/竖版-无表格-农业银行'
+        img_dir = '/Users/clay/Desktop/普通打印-部分无线_img/竖版-无表格-农业银行'
+        os.makedirs(img_dir, exist_ok=True)
+        for d in os.listdir(pdf_dir):
+        #     if d in ['.DS_Store', 'CH-B008486764.pdf', 'CH-B008003736.pdf', 'CH-B008487476.pdf', 'CH-B006763780.pdf',
+        #              'CH-B009000564.pdf', 'CH-B009020488.pdf']:
+            if d in ['.DS_Store', '1竖版-无表格-农业银行样例.PNG']:
+                continue
+            pdf_path = os.path.join(pdf_dir, d)
+            # pdf_path = '/Users/clay/Desktop/普通打印part2/工商银行(标准版)/CH-B006754676.pdf'
+            if os.path.isfile(pdf_path):
+                img_save_path = os.path.join(img_dir, d)
+                if os.path.exists(img_save_path):
+                    continue
+                os.makedirs(img_save_path, exist_ok=True)
+                with fitz.Document(pdf_path) as pdf:
+                    self.cronjob_log.info('{0} [pdf_path={1}] [metadata={2}]'.format(
+                        self.log_base, pdf_path, pdf.metadata))
+                    # xref_list = []
+                    for pno in range(pdf.pageCount):
+                        il = pdf.getPageImageList(pno)
+                        il.sort(key=lambda x: x[0])
+                        img_il_list = self.split_il(il)
+                        del il
+
+                        print(img_il_list)
+                        if len(img_il_list) > 3:  # 单页无规律小图过多时，使用页面转图片
+                            page = pdf.loadPage(pno)
+                            pm = page.getPixmap(matrix=self.trans, alpha=False)
+                            save_path = os.path.join(img_save_path, 'page_{0}_img_0.png'.format(page.number))
+                            pm.writePNG(save_path)
+                            # img_path_list.append(save_path)
+                            # self.cronjob_log.info('{0} [page to img success] [doc_id={1}] [pdf_path={2}] '
+                            #                       '[page={3}]'.format(self.log_base, doc_id, pdf_path, page.number))
+                        else:  # 提取图片
+                            for img_index, img_il in enumerate(img_il_list):
+                                if len(img_il) == 1:  # 当只有一张图片时， 简化处理
+                                    pix = self.recoverpix(pdf, img_il[0])
+                                    ext, img_data = self.get_img_data(pix)
+                                    save_path = os.path.join(img_save_path, 'page_{0}_img_{1}.{2}'.format(
+                                        pno, img_index, ext))
+                                    with open(save_path, "wb") as f:
+                                        f.write(img_data)
+                                    # img_path_list.append(save_path)
+                                    # self.cronjob_log.info(
+                                    #     '{0} [extract img success] [doc_id={1}] [pdf_path={2}] [page={3}] '
+                                    #     '[img_index={4}]'.format(self.log_base, doc_id, pdf_path, pno, img_index))
+                                else:  # 多张图片，竖向拼接
+                                    height_sum = 0
+                                    im_list = []
+                                    width = img_il[0][2]
+                                    for img in img_il:
+                                        # xref = img[0]
+                                        # if xref in xref_list:
+                                        #     continue
+                                        height = img[3]
+                                        pix = self.recoverpix(pdf, img)
+                                        ext, img_data = self.get_img_data(pix)
+
+                                        # xref_list.append(xref)
+
+                                        im = Image.open(BytesIO(img_data))
+                                        im_list.append((height, im, ext))
+                                        height_sum += height
+
+                                    save_path = os.path.join(img_save_path, 'page_{0}_img_{1}.{2}'.format(
+                                        pno, img_index, im_list[0][2]))
+                                    res = Image.new(im_list[0][1].mode, (width, height_sum))
+                                    h_now = 0
+                                    for h, m, _ in im_list:
+                                        res.paste(m, box=(0, h_now))
+                                        h_now += h
+                                    res.save(save_path)
+            # else:
+            #     img_dir_path = os.path.join(img_dir, d)
+            #     os.makedirs(img_dir_path, exist_ok=True)

@@ -5,6 +5,7 @@ import datetime
 from django.utils import timezone
 from django.db.utils import IntegrityError
 from django.db.models import Q
+from rest_framework.permissions import IsAuthenticated
 from webargs import fields, validate
 from webargs.djangoparser import use_args, parser
 from settings import conf
@@ -15,6 +16,7 @@ from common.redis_cache import redis_handler as rh
 from .models import UploadDocRecords, DocStatus, PriorityApplication, GCAPRecords
 from .mixins import DocHandler
 from . import consts
+from apps.account.authentication import OAuth2AuthenticationWithUser
 
 
 # restframework将request.body封装至request.data, webargs从request.data中获取参数
@@ -86,7 +88,9 @@ priority_doc_args = {
 
 
 class UploadDocView(GenericView, DocHandler):
-    permission_classes = []
+    permission_classes = [IsAuthenticated]
+    authentication_classes = [OAuth2AuthenticationWithUser]
+    # required_scopes = ['write']
 
     # 上传（接收）文件接口
     @use_args(doc_upload_args, location='data')
@@ -134,8 +138,8 @@ class UploadDocView(GenericView, DocHandler):
             is_priority = PriorityApplication.objects.filter(application_id=application_id, on_off=True).exists()
             value = ['{0}_{1}'.format(prefix, doc.id)]
             redis_res = rh.enqueue(value, is_priority)
-            self.running_log.info('[doc upload success] [args={0}] [record_id={1}] [is_hil={2}] [doc_id={3}] '
-                                  '[is_priority={4}] [enqueue_res={5}]'.format(args, record.id, is_hil, doc.id,
+            self.running_log.info('[doc upload success] [args={0}] [record_id={1}] [prefix={2}] [doc_id={3}] '
+                                  '[is_priority={4}] [enqueue_res={5}]'.format(args, record.id, prefix, doc.id,
                                                                                is_priority, redis_res))
             return response.ok()
 
@@ -160,7 +164,8 @@ class UploadDocView(GenericView, DocHandler):
 
 
 class PriorityDocView(GenericView, DocHandler):
-    permission_classes = []
+    permission_classes = [IsAuthenticated]
+    authentication_classes = [OAuth2AuthenticationWithUser]
 
     # 优先级订单接口
     @use_args(priority_doc_args, location='data')
@@ -195,6 +200,25 @@ class PriorityDocView(GenericView, DocHandler):
                     args, task_str_list, enqueue_res))
         return response.ok()
 
+    post.openapi_doc = '''
+        tags: [doc]
+        summary: GCAP提高申请单对应文件优先级
+        consumes: [application/json]
+        produces: [application/json]
+        parameters:
+        - in: body
+          name: body
+          required: true
+          schema:
+            $ref: "#/definitions/Application"
+
+        responses:
+          200:
+            description: ok
+            schema:
+              $ref: '#/definitions/ApiResponse'
+        '''
+
 
 class DocView(GenericView, DocHandler):
 

@@ -22,4 +22,5 @@ urlpatterns = [
     path(r'api/create/', include('apps.doc.create_urls')),
     path(r'api/priority/', include('apps.doc.priority_urls')),
     path(r'api/doc/', include('apps.doc.internal_urls')),
+    path('api/oauth/', include('oauth2_provider.urls', namespace='oauth2_provider')),
 ]

@@ -12,10 +12,26 @@ tags:
 schemes:
 - "https"
 - "http"
+security:
+  - OAuth2: []
 '''
 
-#   scheme: bearer
+#   scheme: oauth
 security_definitions = '''
+OAuth2:
+    type: oauth2
+    flow: application
+    description: >
+        This API uses OAuth 2 with the application(clientCredentials) grant flow.
+        
+        client_id=sMlciTkppsMzARwHpCVarm5q7DP2Vucj3ny8JFhw
+        
+        client_secret=WNoOilDx140ZLcenDKfsnikv7S2LIFs60DciYoqnrZaYLqYsKpcmt7mJIL69o9AEf84uQvRnS3K2UioxfjNyImjR4UOyXbDcF6qYgTLC4KDVByKFdVhKfrn2Lc4q4BNW
+        
+        scopes=write
+    tokenUrl: https://staging-bmw-ocr.situdata.com/api/oauth/token/
+    scopes:
+      write: Grants write access
 '''
 
 responses = '''
@@ -99,6 +115,46 @@ Doc:
           type: string
           example: '8410480'
           
+Application:
+  type: object
+  required: [APPLICATION_INFORMATION]
+  properties:
+    APPLICATION_INFORMATION:
+      description: 申请单信息
+      type: object
+      required: [SUBMIT_DATETIME, STATUS, ENTITY, RATING, APPLICATION_ID, APPLICATION_VERSION, INTERMEDIATE_DECISION]
+      properties: 
+        SUBMIT_DATETIME:
+          description: 提交时间
+          type: string
+          example: 2020-07-08T18:33:31.000+08:00
+        STATUS:
+          description: 状态
+          type: integer
+          example: 42
+        ENTITY:
+          description: 业务类型
+          type: string
+          example: CO00001
+          enum: [CO00001, CO00002]
+        RATING:
+          description: 排名
+          type: integer
+          example: 4
+        APPLICATION_ID:
+          description: 申请id
+          type: string
+          example: CH-B0011010101
+        APPLICATION_VERSION:
+          description: 申请版本
+          type: integer
+          example: 1
+        INTERMEDIATE_DECISION:
+          description: ''
+          type: string
+          example: MUW
+          
+
 ApiResponse:
   description: 响应对象，code字段用于表示响应的状态; data字段用于存放响应内容
   type: object

@@ -4,6 +4,7 @@ from django.core.exceptions import PermissionDenied
 from django.utils.translation import ugettext_lazy as _
 from django.http import Http404
 from rest_framework import exceptions, status
+from rest_framework.exceptions import NotAuthenticated, PermissionDenied
 from rest_framework.response import Response
 from marshmallow.exceptions import ValidationError
 
@@ -29,7 +30,9 @@ def exception_handler(exc, context):
         if getattr(exc, 'wait', None):
             headers['Retry-After'] = '%d' % exc.wait
 
-        if exc.status_code == 403:
+        if isinstance(exc, NotAuthenticated) or isinstance(exc, PermissionDenied):
+            data = res_content(MetaStatus.NO_PERMISSION.value, MetaStatus.NO_PERMISSION.verbose_name)
+        elif exc.status_code == 403:
             data = res_content(MetaStatus.NEED_LOGIN.value, MetaStatus.NEED_LOGIN.verbose_name)
         else:
             data = res_content(MetaStatus.INTERNAL_ERROR.value, exc.detail)

@@ -48,7 +48,7 @@ class Command(BaseCommand):
             view_class = view.view_class
             url_path, path_parameters = pattern[0][0]
             url_path = unify_url_path_format(url_path)
-            if url_path != '/api/create/v1':
+            if url_path not in ['/api/create/v1', '/api/priority/v1']:
                 continue
             url_path_paramters = getattr(view, 'parameters_doc', None)
             if url_path_paramters:
@@ -82,7 +82,7 @@ class Command(BaseCommand):
                 api_doc_dct[url_path][method] = doc
         doc_dct = yaml.load(base_part)
         doc_dct['paths'] = api_doc_dct
-        # doc_dct['securityDefinitions'] = yaml.load(security_definitions)
+        doc_dct['securityDefinitions'] = yaml.load(security_definitions)
         doc_dct['responses'] = yaml.load(responses)
         doc_dct['definitions'] = yaml.load(definitions)
 

@@ -42,6 +42,7 @@ INSTALLED_APPS = [
     'django.contrib.messages',
     'django.contrib.staticfiles',
     # 'corsheaders',
+    'oauth2_provider',
     'rest_framework',
     'common',
     'apps.account',
@@ -145,6 +146,7 @@ REST_FRAMEWORK = {
     'DEFAULT_AUTHENTICATION_CLASSES': (
         'rest_framework.authentication.BasicAuthentication',
         'rest_framework_jwt.authentication.JSONWebTokenAuthentication',
+        # 'oauth2_provider.contrib.rest_framework.OAuth2Authentication',
     ),
     'EXCEPTION_HANDLER': 'common.exceptions.exception_handler'
 }
@@ -173,3 +175,8 @@ JWT_AUTH = {
 # 跨域设置
 # CORS_ORIGIN_ALLOW_ALL = True
 # CORS_ALLOW_CREDENTIALS = True
+
+OAUTH2_PROVIDER = {
+    # this is the list of available scopes
+    'SCOPES': {'read': 'Read scope', 'write': 'Write scope'}
+}