def get(self, request, task_id, format=None):
image_url_prefix = self.task.reel.url_prefix()
urls = []
for vol_page in range(self.task.reel.start_vol_page,
self.task.reel.end_vol_page + 1):
url = get_image_url(self.task.reel, vol_page)
urls.append(url)
pagetask_numbers = [
t.number for t in PageTask.objects.filter(
pagerect__reel=self.task.reel).order_by('pagerect__page')
]
correct_task_count = 2
if self.task.typ == Task.TYPE_CORRECT_VERIFY:
correct_task_count = Task.objects.filter(reel=self.task.reel, \
batchtask=self.task.batchtask, typ=Task.TYPE_CORRECT).count()
bar_line_count = self.task.reel.sutra.tripitaka.bar_line_count
if bar_line_count.isdigit():
bar_line_count = int(bar_line_count)
else:
bar_line_count = json.loads(bar_line_count)
response = {
'task_id': task_id,
'correct_task_count': correct_task_count,
'task_typ': self.task.typ,
'status': self.task.status,
'result': self.task.result,
'cur_focus': self.task.cur_focus,
'image_urls': urls,
'pagetask_numbers': pagetask_numbers,
'bar_line_count': bar_line_count,
}
return Response(response)
def get(self, request, s_id, format=None):
# 根据卷ID 获得页号 和页数
reel = Reel.objects.get(id=s_id)
reelPages = Page.objects.filter(reel=reel).order_by("reel_page_no")
strURLRet = ''
cut_Info_list = []
for p in reelPages:
image_url = get_image_url(reel, p.reel_page_no)
cut_Info_list.append(p.cut_info)
strURLRet = image_url + '|'
reel_ocr_text = None
reelcorrectid = -1
try:
reel_ocr_text = ReelCorrectText.objects.get(reel_id=int(s_id))
reelcorrectid = reel_ocr_text.id
text = reel_ocr_text.text
except:
reel_ocr_text = ReelOCRText.objects.get(reel_id=int(s_id))
text = reel_ocr_text.text
response = {
'sutra': text,
'pageurls': strURLRet,
'cut_Info_list': cut_Info_list,
'sutra_name': str(reel.sutra.name),
'reelcorrectid': reelcorrectid,
# 'punct_lst': punctuation,
}
return Response(response)
def cutfixed_page_detail(request, pid):
page = get_object_or_404(Page, pid=pid)
image_url = get_image_url(page.reel, page.page_no)
return render(request, 'tools/cutfixed_page_detail.html', {
'page': page,
'image_url': image_url
})
def page_url(self):
try:
pid = self.start_char_pos[0:17]
page = Page.objects.get(pid=pid)
url = get_image_url(page.reel, page.page_no)
return url
except:
pass
def get(self, request, pk, format=None):
correct_fb = CorrectFeedback.objects.get(id=pk)
correct_text = correct_fb.correct_text
text_list = []
whole_text = correct_text.text
page_txt_list = whole_text.split('p')
for p_no, i in enumerate(page_txt_list):
for line_no, k in enumerate(i.replace('b', '').split('\n')):
for char_no, l in enumerate(k):
text_list.append((p_no, 0, line_no, char_no, l))
original_txt_info = text_list[correct_fb.position]
p_no = original_txt_info[0]
b_no = original_txt_info[1]
line_no = original_txt_info[2]
char_no = original_txt_info[3]
reel = correct_text.reel
page = reel.page_set.all().order_by('reel_page_no')[p_no - 1]
image_url = get_image_url(reel, page.reel_page_no)
cut_list = []
cid = 'c{0}{1}{2:02d}n{3:02d}'.format(p_no, b_no, line_no,
char_no + 1) # 切分信息中字符从1开始计数
page_cuts = json.loads(page.cut_info)['char_data']
for c_no, cut in enumerate(page_cuts):
if cut['char_id'] == cid:
cut_list = page_cuts[c_no:c_no + len(correct_fb.original_text)]
page_lines = page_txt_list[p_no].split('\n')
fb_line = page_lines[line_no]
page_lines[
line_no] = fb_line[:
char_no] + '<span class="difftext confirmed">' + fb_line[
char_no] + '</span>' + fb_line[char_no + 1:]
if correct_fb.processor:
processor = correct_fb.processor
return Response({
'fb_id': correct_fb.id,
'origin_text': correct_fb.original_text,
'fb_text': correct_fb.fb_text,
'fb_comment': correct_fb.fb_comment,
'cut_info': cut_list,
'page_txt': page_lines,
'image_url': image_url,
'processor': processor.email,
'response': correct_fb.get_response_display()
})
else:
return Response({
'fb_id': correct_fb.id,
'origin_text': correct_fb.original_text,
'fb_text': correct_fb.fb_text,
'fb_comment': correct_fb.fb_comment,
'cut_info': cut_list,
'page_txt': page_lines,
'image_url': image_url,
'processor': '',
'response': correct_fb.get_response_display()
})
def retrieve(self, request, *args, **kwargs):
instance = self.get_object()
if (not request.user.is_admin):
if (instance.owner != request.user):
return Response({"status": -1, "msg": "No Permission!"})
serializer = self.get_serializer(instance)
if isinstance(instance, PageTask):
pagerect = instance.pagerect
page_id = pagerect.page_id
_rects = Rect.objects.filter(page_pid=page_id).all()
rects = RectSerializer(data=_rects, many=True)
rects.is_valid()
page = pagerect.page
image_url = get_image_url(page.reel, page.page_no)
return Response({
"status": instance.status,
"rects": rects.data,
"image_url": image_url,
"page_info": str(page),
"current_x": instance.current_x,
"current_y": instance.current_y,
"task_id": instance.pk
})
if isinstance(instance, PageVerifyTask):
pagerect = instance.pagerect
page_id = pagerect.page_id
_rects = Rect.objects.filter(page_pid=page_id).all()
rects = RectSerializer(data=_rects, many=True)
rects.is_valid()
page = pagerect.page
image_url = get_image_url(page.reel, page.page_no)
return Response({
"status": instance.status,
"rects": rects.data,
"image_url": image_url,
"page_info": str(page),
"current_x": instance.current_x,
"current_y": instance.current_y,
"task_id": instance.pk
})
return Response(serializer.data)
def get(self, request, rid, format=None):
reel = Reel.objects.get(id=int(rid))
img_lst = []
for page_no in range(reel.start_vol_page, reel.end_vol_page + 1):
img_lst.append({
'image_url':
encipher.get_image_url(reel, page_no),
'page_code':
"%s_%s_%d" %
(reel.sutra.tripitaka.code, reel.path_str(), page_no)
})
return Response({'results': img_lst})
def check_sutra(self, sutra):
if sutra.sid[:2] not in tcode_lst:
return
for reel in sutra.reel_set.all():
if reel.start_vol_page == 0 and reel.end_vol_page == 0:
continue
for vol_page in range(reel.start_vol_page, reel.end_vol_page + 1):
url = get_image_url(reel, vol_page)
r = requests.head(url)
if r.status_code != 200:
print(
'%s_%03d: image not exists.' %
(sutra.sid, reel.reel_no), url, r.status_code)
cut_url = url.replace('jpg', 'cut')
try:
r = requests.get(cut_url)
if r.status_code != 200:
print(
'%s_%03d: cut not exists.' %
(sutra.sid, reel.reel_no), cut_url, r.status_code)
else:
try:
cut_json = json.loads(r.content)
if not cut_json.get('char_data', []):
print(
'%s_%03d: no char data.' %
(sutra.sid, reel.reel_no), cut_url)
except:
print(
'%s_%03d: not json format.' %
(sutra.sid, reel.reel_no), cut_url)
except:
print(
'%s_%03d: check cut exception.' %
(sutra.sid, reel.reel_no), cut_url, r.status_code)
txt_url = url.replace('jpg', 'txt')
txt_exists = False
try:
r = requests.get(cut_url)
if r.status_code == 200:
txt_exists = True
except:
pass
if not txt_exists:
print(
'%s_%03d: txt not exists.' % (sutra.sid, reel.reel_no),
txt_url, r.status_code)
def sutra_page_detail(request, pid):
page = get_object_or_404(Page, pid=pid)
reel = page.reel
vol_page = reel.start_vol_page + page.reel_page_no - 1
image_url = get_image_url(reel, vol_page)
char_pos = request.GET.get('char_pos', '')
s = char_pos[-5:]
try:
line_no = int(s[0:2])
char_no = int(s[3:])
except:
return redirect('/')
context = {
'page': page,
'image_url': image_url,
'line_no': line_no,
'char_no': char_no,
}
return render(request, 'tasks/sutra_page_detail.html', context)
def obtain(self, request):
staff = request.user
task = retrieve_pagetask(staff)
if not task:
return Response({"status": -1, "msg": "All tasks have been done!"})
pagerect = task.pagerect
page_id = pagerect.page_id
_rects = Rect.objects.filter(page_pid=page_id).all()
rects = RectSerializer(data=_rects, many=True)
rects.is_valid()
page = pagerect.page
image_url = get_image_url(page.reel, page.page_no)
return Response({
"status": task.status,
"rects": rects.data,
"image_url": image_url,
"page_info": str(page),
"current_x": task.current_x,
"current_y": task.current_y,
"task_id": task.pk
})