Skip to content

周伟奇 / bmw-ocr

Go to a project
Toggle navigation
Toggle navigation pinning
59cbfab2 by 周伟奇
Options

fix bug & add skip_img_sheet

1 parent 6a5899fa
Inline Side-by-side
Showing 4 changed files with 175 additions and 166 deletions
.gitignore
...@@ -33,4 +33,5 @@ data/* ...@@ -33,4 +33,5 @@ data/*
33 # 脚本 33 # 脚本
34 src/*.sh 34 src/*.sh
35 35
36 test*
...\ No newline at end of file ...\ No newline at end of file
36 test*
37 ocr_test.py
...\ No newline at end of file ...\ No newline at end of file
......
src/apps/doc/consts.py
...@@ -60,6 +60,8 @@ TRANS_MAP = { ...@@ -60,6 +60,8 @@ TRANS_MAP = {
60 } 60 }
61 TRANS = str.maketrans(TRANS_MAP) 61 TRANS = str.maketrans(TRANS_MAP)
62 ERROR_CHARS = {'.', '·', '•'} 62 ERROR_CHARS = {'.', '·', '•'}
63 SKIP_IMG_SHEET_NAME = '未处理图片'
64 SKIP_IMG_SHEET_HEADER = ('页码', '序号')
63 65
64 CARD_RATIO = 0.9 66 CARD_RATIO = 0.9
65 UNKNOWN_CARD = '未知卡号' 67 UNKNOWN_CARD = '未知卡号'
......
src/apps/doc/management/commands/doc_ocr_process.py
...@@ -80,19 +80,20 @@ class Command(BaseCommand, LoggerMixin): ...@@ -80,19 +80,20 @@ class Command(BaseCommand, LoggerMixin):
80 self.log_base, business_type, doc.id, pdf_path)) 80 self.log_base, business_type, doc.id, pdf_path))
81 return doc_data_path, excel_path, src_excel_path, pdf_path 81 return doc_data_path, excel_path, src_excel_path, pdf_path
82 82
83 @staticmethod 83 def bs_process(self, wb, ocr_data, bs_summary, unknown_summary, img_path, classify, skip_img):
84 def bs_process(wb, ocr_data, bs_summary, unknown_summary, img_path, classify):
85 sheets = ocr_data.get('data', []) 84 sheets = ocr_data.get('data', [])
86 if not sheets: 85 if not sheets:
86 skip_img.append(self.parse_img_path(img_path))
87 return 87 return
88 confidence = ocr_data.get('confidence', 1) 88 confidence = ocr_data.get('confidence', 1)
89 img_name, _ = os.path.splitext(os.path.basename(img_path)) 89 img_name, _ = os.path.splitext(os.path.basename(img_path))
90 for i, sheet in enumerate(sheets): 90 for i, sheet in enumerate(sheets):
91 sheet_name = '{0}_{1}'.format(img_name, i)
92 ws = wb.create_sheet(sheet_name)
93 cells = sheet.get('cells') 91 cells = sheet.get('cells')
94 if not cells: 92 if not cells:
93 skip_img.append(self.parse_img_path(img_path))
95 continue 94 continue
95 sheet_name = '{0}_{1}'.format(img_name, i)
96 ws = wb.create_sheet(sheet_name)
96 for cell in cells: 97 for cell in cells:
97 c1 = cell.get('start_column') 98 c1 = cell.get('start_column')
98 r1 = cell.get('start_row') 99 r1 = cell.get('start_row')
...@@ -147,9 +148,10 @@ class Command(BaseCommand, LoggerMixin): ...@@ -147,9 +148,10 @@ class Command(BaseCommand, LoggerMixin):
147 ed_list.append(summary[6]) 148 ed_list.append(summary[6])
148 149
149 @staticmethod 150 @staticmethod
150 def license1_process(ocr_data, license_summary, classify): 151 def license1_process(ocr_data, license_summary, classify, skip_img, img_path):
151 license_data = ocr_data.get('data', []) 152 license_data = ocr_data.get('data', [])
152 if not license_data: 153 if not license_data:
154 skip_img.append(img_path)
153 return 155 return
154 for license_dict in license_data: 156 for license_dict in license_data:
155 res_list = [] 157 res_list = []
...@@ -157,8 +159,7 @@ class Command(BaseCommand, LoggerMixin): ...@@ -157,8 +159,7 @@ class Command(BaseCommand, LoggerMixin):
157 res_list.append((field, value)) 159 res_list.append((field, value))
158 license_summary.setdefault(classify, []).append(res_list) 160 license_summary.setdefault(classify, []).append(res_list)
159 161
160 @staticmethod 162 def license2_process(self, ocr_res_2, license_summary, pid, classify, skip_img, img_path):
161 def license2_process(ocr_res_2, license_summary, pid, classify):
162 if ocr_res_2.get('ErrorCode') in consts.SUCCESS_CODE_SET: 163 if ocr_res_2.get('ErrorCode') in consts.SUCCESS_CODE_SET:
163 if pid == consts.BC_PID: 164 if pid == consts.BC_PID:
164 # 银行卡 165 # 银行卡
...@@ -174,113 +175,16 @@ class Command(BaseCommand, LoggerMixin): ...@@ -174,113 +175,16 @@ class Command(BaseCommand, LoggerMixin):
174 res_list.append( 175 res_list.append(
175 (field_dict.get('chn_key', ''), field_dict.get('value', ''))) 176 (field_dict.get('chn_key', ''), field_dict.get('value', '')))
176 license_summary.setdefault(classify, []).append(res_list) 177 license_summary.setdefault(classify, []).append(res_list)
177
178 async def fetch_ocr_result(self, url, json_data):
179 async with aiohttp.ClientSession(connector=aiohttp.TCPConnector(ssl=False)) as session:
180 async with session.post(url, json=json_data) as response:
181 if response.status == 200:
182 return await response.json()
183
184 # async def img_2_ocr_2_wb(self, wb, img_path, summary):
185 # res = await self.fetch_ocr_result(img_path)
186 # self.cronjob_log.info('{0} [fetch ocr result success] [img={1}] [res={2}]'.format(self.log_base, img_path, res))
187 # sheets_list = res.get('result').get('res')
188 # img_name = os.path.basename(img_path)
189 # self.append_sheet(wb, sheets_list, img_name, summary)
190
191 async def img_2_ocr_2_wb(self, wb, img_path, bs_summary, unknown_summary, license_summary):
192 with open(img_path, 'rb') as f:
193 base64_data = base64.b64encode(f.read())
194 # 获取解码后的base64值
195 file_data = base64_data.decode()
196 json_data_1 = {
197 "file": file_data
198 }
199 ocr_res_1 = await self.fetch_ocr_result(self.ocr_url_1, json_data_1)
200 if ocr_res_1 is None:
201 raise Exception('ocr 1 error, img_path={0}'.format(img_path))
202 else: 178 else:
203 self.cronjob_log.info('{0} [ocr_1 result] [img={1}] [res={2}]'.format( 179 skip_img.append(self.parse_img_path(img_path))
204 self.log_base, img_path, ocr_res_1))
205
206 if ocr_res_1.get('code') == 1:
207 ocr_data = ocr_res_1.get('data', {})
208 classify = ocr_data.get('classify')
209 if classify is None:
210 return
211 elif classify in consts.OTHER_CLASSIFY_SET: # 其他类
212 return
213 elif classify in consts.LICENSE_CLASSIFY_SET_1: # 证件1
214 self.license1_process(ocr_data, license_summary, classify)
215 elif classify in consts.LICENSE_CLASSIFY_SET_2: # 证件2
216 pid, _ = consts.LICENSE_CLASSIFY_MAPPING.get(classify)
217 json_data_2 = {
218 "pid": str(pid),
219 "key": conf.OCR_KEY,
220 "secret": conf.OCR_SECRET,
221 "file": file_data
222 }
223 ocr_res_2 = await self.fetch_ocr_result(self.ocr_url_2, json_data_2)
224 if ocr_res_2 is None:
225 raise Exception('ocr 2 error, img_path={0}'.format(img_path))
226 else:
227 # 识别结果
228 self.cronjob_log.info('{0} [ocr_2 result] [img={1}] [res={2}]'.format(
229 self.log_base, img_path, ocr_res_2))
230 self.license2_process(ocr_res_2, license_summary, pid, classify)
231 else: # 流水处理
232 self.bs_process(wb, ocr_data, bs_summary, unknown_summary, img_path, classify)
233 180
234 # def img_2_ocr_2_wb(self, wb, img_path, bs_summary, unknown_summary, license_summary): 181 # async def fetch_ocr_result(self, url, json_data):
235 # # # 流水 182 # async with aiohttp.ClientSession(connector=aiohttp.TCPConnector(ssl=False)) as session:
236 # # res = { 183 # async with session.post(url, json=json_data) as response:
237 # # 'code': 1, 184 # if response.status == 200:
238 # # 'msg': 'success', 185 # return await response.json()
239 # # 'data': { 186 #
240 # # 'classify': 0, 187 # async def img_2_ocr_2_wb(self, wb, img_path, bs_summary, unknown_summary, license_summary):
241 # # 'confidence': 0.999,
242 # # 'data': [
243 # # {
244 # # 'summary': ['户名', '卡号', '页码', '回单验证码', '打印时间', '起始时间', '终止时间'],
245 # # 'cells': []
246 # # },
247 # # {
248 # # 'summary': ['户名', '卡号', '页码', '回单验证码', '打印时间', '起始时间', '终止时间'],
249 # # 'cells': []
250 # # }
251 # # ]
252 # # }
253 # # }
254 # #
255 # # # 证件-1
256 # # res = {
257 # # 'code': 1,
258 # # 'msg': 'success',
259 # # 'data': {
260 # # 'classify': 0,
261 # # 'confidence': 0.999,
262 # # 'data': [
263 # # {
264 # # 'cn_key': 'value',
265 # # 'cn_key': 'value',
266 # # },
267 # # {
268 # # 'cn_key': 'value',
269 # # 'cn_key': 'value',
270 # # },
271 # # ]
272 # # }
273 # # }
274 # #
275 # # # 证件-2 or 其他类
276 # # res = {
277 # # 'code': 1,
278 # # 'msg': 'success',
279 # # 'data': {
280 # # 'classify': 0,
281 # # 'confidence': 0.999,
282 # # }
283 # # }
284 # with open(img_path, 'rb') as f: 188 # with open(img_path, 'rb') as f:
285 # base64_data = base64.b64encode(f.read()) 189 # base64_data = base64.b64encode(f.read())
286 # # 获取解码后的base64值 190 # # 获取解码后的base64值
...@@ -288,9 +192,10 @@ class Command(BaseCommand, LoggerMixin): ...@@ -288,9 +192,10 @@ class Command(BaseCommand, LoggerMixin):
288 # json_data_1 = { 192 # json_data_1 = {
289 # "file": file_data 193 # "file": file_data
290 # } 194 # }
291 # response_1 = requests.post(self.ocr_url_1, json=json_data_1) 195 # ocr_res_1 = await self.fetch_ocr_result(self.ocr_url_1, json_data_1)
292 # if response_1.status_code == 200: 196 # if ocr_res_1 is None:
293 # ocr_res_1 = response_1.json() 197 # raise Exception('ocr 1 error, img_path={0}'.format(img_path))
198 # else:
294 # self.cronjob_log.info('{0} [ocr_1 result] [img={1}] [res={2}]'.format( 199 # self.cronjob_log.info('{0} [ocr_1 result] [img={1}] [res={2}]'.format(
295 # self.log_base, img_path, ocr_res_1)) 200 # self.log_base, img_path, ocr_res_1))
296 # 201 #
...@@ -311,21 +216,119 @@ class Command(BaseCommand, LoggerMixin): ...@@ -311,21 +216,119 @@ class Command(BaseCommand, LoggerMixin):
311 # "secret": conf.OCR_SECRET, 216 # "secret": conf.OCR_SECRET,
312 # "file": file_data 217 # "file": file_data
313 # } 218 # }
314 # response_2 = requests.post(self.ocr_url_2, data=json_data_2) 219 # ocr_res_2 = await self.fetch_ocr_result(self.ocr_url_2, json_data_2)
315 # if response_2.status_code == 200: 220 # if ocr_res_2 is None:
221 # raise Exception('ocr 2 error, img_path={0}'.format(img_path))
222 # else:
316 # # 识别结果 223 # # 识别结果
317 # ocr_res_2 = response_2.json()
318 # self.cronjob_log.info('{0} [ocr_2 result] [img={1}] [res={2}]'.format( 224 # self.cronjob_log.info('{0} [ocr_2 result] [img={1}] [res={2}]'.format(
319 # self.log_base, img_path, ocr_res_2)) 225 # self.log_base, img_path, ocr_res_2))
320 # self.license2_process(ocr_res_2, license_summary, pid, classify) 226 # self.license2_process(ocr_res_2, license_summary, pid, classify)
321 # else:
322 # raise Exception('ocr 2 error, img_path={0}'.format(img_path))
323 # else: # 流水处理 227 # else: # 流水处理
324 # self.bs_process(wb, ocr_data, bs_summary, unknown_summary, img_path, classify) 228 # self.bs_process(wb, ocr_data, bs_summary, unknown_summary, img_path, classify)
325 # else: 229
326 # pass 230 def img_2_ocr_2_wb(self, wb, img_path, bs_summary, unknown_summary, license_summary, skip_img):
327 # else: 231 # # 流水
328 # raise Exception('ocr 1 error, img_path={0}'.format(img_path)) 232 # res = {
233 # 'code': 1,
234 # 'msg': 'success',
235 # 'data': {
236 # 'classify': 0,
237 # 'confidence': 0.999,
238 # 'data': [
239 # {
240 # 'summary': ['户名', '卡号', '页码', '回单验证码', '打印时间', '起始时间', '终止时间'],
241 # 'cells': []
242 # },
243 # {
244 # 'summary': ['户名', '卡号', '页码', '回单验证码', '打印时间', '起始时间', '终止时间'],
245 # 'cells': []
246 # }
247 # ]
248 # }
249 # }
250 #
251 # # 证件-1
252 # res = {
253 # 'code': 1,
254 # 'msg': 'success',
255 # 'data': {
256 # 'classify': 0,
257 # 'confidence': 0.999,
258 # 'data': [
259 # {
260 # 'cn_key': 'value',
261 # 'cn_key': 'value',
262 # },
263 # {
264 # 'cn_key': 'value',
265 # 'cn_key': 'value',
266 # },
267 # ]
268 # }
269 # }
270 #
271 # # 证件-2 or 其他类
272 # res = {
273 # 'code': 1,
274 # 'msg': 'success',
275 # 'data': {
276 # 'classify': 0,
277 # 'confidence': 0.999,
278 # }
279 # }
280 with open(img_path, 'rb') as f:
281 base64_data = base64.b64encode(f.read())
282 # 获取解码后的base64值
283 file_data = base64_data.decode()
284 json_data_1 = {
285 "file": file_data
286 }
287 response_1 = requests.post(self.ocr_url_1, json=json_data_1)
288 if response_1.status_code == 200:
289 ocr_res_1 = response_1.json()
290 self.cronjob_log.info('{0} [ocr_1 result] [img={1}] [res={2}]'.format(
291 self.log_base, img_path, ocr_res_1))
292
293 if ocr_res_1.get('code') == 1:
294 ocr_data = ocr_res_1.get('data', {})
295 classify = ocr_data.get('classify')
296 if classify is None:
297 skip_img.append(self.parse_img_path(img_path))
298 return
299 elif classify in consts.OTHER_CLASSIFY_SET: # 其他类
300 skip_img.append(self.parse_img_path(img_path))
301 return
302 elif classify in consts.LICENSE_CLASSIFY_SET_1: # 证件1
303 self.license1_process(ocr_data, license_summary, classify, skip_img, img_path)
304 elif classify in consts.LICENSE_CLASSIFY_SET_2: # 证件2
305 pid, _ = consts.LICENSE_CLASSIFY_MAPPING.get(classify)
306 json_data_2 = {
307 "pid": str(pid),
308 "key": conf.OCR_KEY,
309 "secret": conf.OCR_SECRET,
310 "file": file_data
311 }
312 response_2 = requests.post(self.ocr_url_2, data=json_data_2)
313 if response_2.status_code == 200:
314 # 识别结果
315 ocr_res_2 = response_2.json()
316 self.cronjob_log.info('{0} [ocr_2 result] [img={1}] [res={2}]'.format(
317 self.log_base, img_path, ocr_res_2))
318 self.license2_process(ocr_res_2, license_summary, pid, classify, skip_img, img_path)
319 else:
320 raise Exception('ocr 2 error, img_path={0}'.format(img_path))
321 else: # 流水处理
322 self.bs_process(wb, ocr_data, bs_summary, unknown_summary, img_path, classify)
323 else:
324 skip_img.append(self.parse_img_path(img_path))
325 else:
326 raise Exception('ocr 1 error, img_path={0}'.format(img_path))
327
328 @staticmethod
329 def parse_img_path(img_path):
330 img_name, _ = os.path.splitext(os.path.basename(img_path))
331 return img_name[5], img_name[11]
329 332
330 @staticmethod 333 @staticmethod
331 def get_most(value_list): 334 def get_most(value_list):
...@@ -425,8 +428,10 @@ class Command(BaseCommand, LoggerMixin): ...@@ -425,8 +428,10 @@ class Command(BaseCommand, LoggerMixin):
425 merged_bs_summary[card] = summary 428 merged_bs_summary[card] = summary
426 else: 429 else:
427 # 1卡号 430 # 1卡号
431 one_card = False
428 if len(bs_summary) == 1: 432 if len(bs_summary) == 1:
429 merged_bs_summary = self.prune_bs_summary(bs_summary) 433 merged_bs_summary = self.prune_bs_summary(bs_summary)
434 one_card = True
430 # 多卡号 435 # 多卡号
431 else: 436 else:
432 merged_bs_summary = self.merge_card(bs_summary) 437 merged_bs_summary = self.merge_card(bs_summary)
...@@ -435,7 +440,7 @@ class Command(BaseCommand, LoggerMixin): ...@@ -435,7 +440,7 @@ class Command(BaseCommand, LoggerMixin):
435 merge_role = [] 440 merge_role = []
436 classify_summary = unknown_summary.get(card_summary['classify'], {}) 441 classify_summary = unknown_summary.get(card_summary['classify'], {})
437 for role, summary in classify_summary.items(): 442 for role, summary in classify_summary.items():
438 if role in card_summary['role_set']: 443 if one_card or role in card_summary['role_set']:
439 merge_role.append(role) 444 merge_role.append(role)
440 card_summary['sheet'].extend(summary['sheet']) 445 card_summary['sheet'].extend(summary['sheet'])
441 card_summary['code'].extend(summary['code']) 446 card_summary['code'].extend(summary['code'])
...@@ -503,6 +508,7 @@ class Command(BaseCommand, LoggerMixin): ...@@ -503,6 +508,7 @@ class Command(BaseCommand, LoggerMixin):
503 bs_summary = {} 508 bs_summary = {}
504 license_summary = {} 509 license_summary = {}
505 unknown_summary = {} 510 unknown_summary = {}
511 skip_img = []
506 interest_keyword = Keywords.objects.filter( 512 interest_keyword = Keywords.objects.filter(
507 type=KeywordsType.INTEREST.value, on_off=True).values_list('keyword', flat=True) 513 type=KeywordsType.INTEREST.value, on_off=True).values_list('keyword', flat=True)
508 salary_keyword = Keywords.objects.filter( 514 salary_keyword = Keywords.objects.filter(
...@@ -515,27 +521,29 @@ class Command(BaseCommand, LoggerMixin): ...@@ -515,27 +521,29 @@ class Command(BaseCommand, LoggerMixin):
515 # wb = Workbook() 521 # wb = Workbook()
516 522
517 # 4.1 获取OCR结果 523 # 4.1 获取OCR结果
518 loop = asyncio.get_event_loop() 524 # loop = asyncio.get_event_loop()
519 tasks = [self.img_2_ocr_2_wb(wb, img_path, bs_summary, unknown_summary, license_summary) 525 # tasks = [self.img_2_ocr_2_wb(wb, img_path, bs_summary, unknown_summary, license_summary)
520 for img_path in pdf_handler.img_path_list] 526 # for img_path in pdf_handler.img_path_list]
521 loop.run_until_complete(asyncio.wait(tasks)) 527 # loop.run_until_complete(asyncio.wait(tasks))
522 # loop.close() 528 # loop.close()
523 529
524 # for img_path in pdf_handler.img_path_list: 530 for img_path in pdf_handler.img_path_list:
525 # self.img_2_ocr_2_wb(wb, img_path, bs_summary, unknown_summary, license_summary) 531 self.img_2_ocr_2_wb(wb, img_path, bs_summary, unknown_summary, license_summary, skip_img)
526 532
527 self.cronjob_log.info('{0} [bs_summary={1}] [unknown_summary={2}] [license_summary={3}]'.format( 533 self.cronjob_log.info('{0} [business_type={1}] [doc_id={2}] [bs_summary={3}] [unknown_summary={4}] '
528 self.log_base, bs_summary, unknown_summary, license_summary)) 534 '[license_summary={5}]'.format(self.log_base, business_type, doc.id, bs_summary,
535 unknown_summary, license_summary))
529 536
530 merged_bs_summary = self.rebuild_bs_summary(bs_summary, unknown_summary) 537 merged_bs_summary = self.rebuild_bs_summary(bs_summary, unknown_summary)
531 538
532 self.cronjob_log.info('{0} [merged_bs_summary={1}] [unknown_summary={2}]'.format( 539 self.cronjob_log.info('{0} [business_type={1}] [doc_id={2}] [merged_bs_summary={3}] '
533 self.log_base, merged_bs_summary, unknown_summary)) 540 '[unknown_summary={4}]'.format(self.log_base, business_type, doc.id,
541 merged_bs_summary, unknown_summary))
534 del unknown_summary 542 del unknown_summary
535 543
536 # 4.2 重构Excel文件 544 # 4.2 重构Excel文件
537 wb.save(src_excel_path) 545 wb.save(src_excel_path)
538 wb.rebuild(merged_bs_summary, license_summary) 546 wb.rebuild(merged_bs_summary, license_summary, skip_img)
539 wb.save(excel_path) 547 wb.save(excel_path)
540 except Exception as e: 548 except Exception as e:
541 doc.status = DocStatus.PROCESS_FAILED.value 549 doc.status = DocStatus.PROCESS_FAILED.value
......
src/apps/doc/ocr/wb.py
...@@ -141,32 +141,22 @@ class BSWorkbook(Workbook): ...@@ -141,32 +141,22 @@ class BSWorkbook(Workbook):
141 # month_info process 141 # month_info process
142 month_info = month_mapping.setdefault('xxxx-xx', []) 142 month_info = month_mapping.setdefault('xxxx-xx', [])
143 month_info.append((ws.title, min_row, ws.max_row, 0)) 143 month_info.append((ws.title, min_row, ws.max_row, 0))
144 elif len(month_list) == 1:
145 # reverse_trend_list process
146 reverse_trend = self.get_reverse_trend(dti.day, idx_list)
147 reverse_trend_list.append(reverse_trend)
148 # month_info process
149 month_info = month_mapping.setdefault(month_list[0], [])
150 day_mean = np.mean(dti.day.dropna())
151 if len(month_info) == 0:
152 month_info.append((ws.title, min_row, ws.max_row, day_mean))
153 else:
154 for i, item in enumerate(month_info):
155 if day_mean <= item[-1]:
156 month_info.insert(i, (ws.title, min_row, ws.max_row, day_mean))
157 break
158 else:
159 month_info.append((ws.title, min_row, ws.max_row, day_mean))
160 else: 144 else:
161 # reverse_trend_list process 145 # reverse_trend_list process
162 reverse_trend = self.get_reverse_trend(dti.day, idx_list) 146 reverse_trend = self.get_reverse_trend(dti.day, idx_list)
163 reverse_trend_list.append(reverse_trend) 147 reverse_trend_list.append(reverse_trend)
164 # month_info process 148 # month_info process
165 for i, item in enumerate(month_list[:-1]): 149 day_idx = dti.day
166 month_mapping.setdefault(item, []).append( 150 idx_list_max_idx = len(idx_list) - 1
167 (ws.title, idx_list[i] + min_row, idx_list[i + 1] + min_row - 1, self.MAX_MEAN)) 151 for i, item in enumerate(month_list):
168 month_mapping.setdefault(month_list[-1], []).insert( 152 if i == idx_list_max_idx:
169 0, (ws.title, idx_list[-1] + min_row, ws.max_row, 0)) 153 day_mean = np.mean(day_idx[idx_list[i]:].dropna())
154 month_mapping.setdefault(item, []).append(
155 (ws.title, idx_list[i] + min_row, ws.max_row, day_mean))
156 else:
157 day_mean = np.mean(day_idx[idx_list[i]: idx_list[i + 1]].dropna())
158 month_mapping.setdefault(item, []).append(
159 (ws.title, idx_list[i] + min_row, idx_list[i + 1] + min_row - 1, day_mean))
170 160
171 def build_metadata_rows(self, confidence, code, print_time, start_date, end_date): 161 def build_metadata_rows(self, confidence, code, print_time, start_date, end_date):
172 if start_date is None or end_date is None: 162 if start_date is None or end_date is None:
...@@ -259,7 +249,7 @@ class BSWorkbook(Workbook): ...@@ -259,7 +249,7 @@ class BSWorkbook(Workbook):
259 except Exception as e: 249 except Exception as e:
260 continue 250 continue
261 else: 251 else:
262 over_cell.number_format = numbers.FORMAT_NUMBER_COMMA_SEPARATED1 252 over_cell.number_format = numbers.FORMAT_GENERAL
263 253
264 # 3.4.金额转数值 254 # 3.4.金额转数值
265 try: 255 try:
...@@ -281,7 +271,7 @@ class BSWorkbook(Workbook): ...@@ -281,7 +271,7 @@ class BSWorkbook(Workbook):
281 else: 271 else:
282 if rows[consts.BORROW_IDX].value in consts.BORROW_OUTLAY_SET: 272 if rows[consts.BORROW_IDX].value in consts.BORROW_OUTLAY_SET:
283 amount_cell.value = -amount_cell.value 273 amount_cell.value = -amount_cell.value
284 amount_cell.number_format = numbers.FORMAT_NUMBER_COMMA_SEPARATED1 274 amount_cell.number_format = numbers.FORMAT_GENERAL
285 same_amount_mapping = amount_mapping.get(date_cell.value, {}) 275 same_amount_mapping = amount_mapping.get(date_cell.value, {})
286 fill_rows = same_amount_mapping.get(-amount_cell.value) 276 fill_rows = same_amount_mapping.get(-amount_cell.value)
287 if fill_rows: 277 if fill_rows:
...@@ -357,11 +347,11 @@ class BSWorkbook(Workbook): ...@@ -357,11 +347,11 @@ class BSWorkbook(Workbook):
357 end_date) 347 end_date)
358 348
359 # 3.创建月份表、提取/高亮关键行 349 # 3.创建月份表、提取/高亮关键行
360 is_reverse = False 350 # 倒序处理
361 if sum(reverse_trend_list) > 0: # 倒序处理 351 is_reverse = True if sum(reverse_trend_list) > 0 else False
362 is_reverse = True 352 for month_list in month_mapping.values():
363 for month_list in month_mapping.values(): 353 month_list.sort(key=lambda x: x[-1], reverse=is_reverse)
364 month_list.sort(key=lambda x: x[-1], reverse=True) 354
365 self.build_month_sheet(card, month_mapping, ms, is_reverse) 355 self.build_month_sheet(card, month_mapping, ms, is_reverse)
366 356
367 # 4.删除原表 357 # 4.删除原表
...@@ -379,6 +369,14 @@ class BSWorkbook(Workbook): ...@@ -379,6 +369,14 @@ class BSWorkbook(Workbook):
379 ws.append(bl_field) 369 ws.append(bl_field)
380 ws.append((None, )) 370 ws.append((None, ))
381 371
382 def rebuild(self, bs_summary, license_summary): 372 def skip_img_sheet(self, skip_img):
373 if skip_img:
374 ws = self.create_sheet(consts.SKIP_IMG_SHEET_NAME)
375 ws.append(consts.SKIP_IMG_SHEET_HEADER)
376 for img_tuple in skip_img:
377 ws.append(img_tuple)
378
379 def rebuild(self, bs_summary, license_summary, skip_img):
383 self.bs_rebuild(bs_summary) 380 self.bs_rebuild(bs_summary)
384 self.license_rebuild(license_summary) 381 self.license_rebuild(license_summary)
382 self.skip_img_sheet(skip_img)
......
Styling with Markdown is supported
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!