def get_status(request):
id = request.GET.get("id")
file_request = ConvertRequest(id, settings.MEDIA_DIR[0])
response = {"answer": 200, "status": file_request.loadStatus()}
return JsonResponse(response)
def pdf_txt_converter(file_request: ConvertRequest):
print("pdf_txt_converter...")
filename = file_request.getPDFLocation() #'artigo.pdf'
outputFile = file_request.getTextLocation() #'article.txt'
print(filename)
print(outputFile)
# output = "output-"+"".join([chr(int(65+random.random()*25)) for i in range(0,10)])
output = "output-" + file_request.id
outputPath = file_request.path + '/' + output
os.popen("pdftoppm " + filename + " " + outputPath + " -png").read()
pages = int(os.popen("ls " + outputPath + "* | wc -l").read())
step = 30 / pages
texto = ""
for i in range(1, pages + 1):
if pages > 9 and i < 10: p = "0" + str(i)
else: p = str(i)
os.popen("tesseract " + outputPath + "-" + p + ".png " +
file_request.path + "/texto-" + output + "-" + p +
" -l por").read()
with open(file_request.path + "/texto-" + output + "-" + p + ".txt",
'r',
encoding='utf-8') as txt_file:
texto += txt_file.read()
file_request.incrementStatus(step)
os.popen("rm -rf " + outputPath + "*").read()
os.popen("rm -rf " + file_request.path + "/texto-" + output + "-*").read()
open(outputFile, "w", encoding='utf-8').write(texto)
return texto
def result(request):
id = request.GET.get("id")
file_request = ConvertRequest(id, settings.MEDIA_DIR[0])
context = {"abstract": None}
with open(file_request.getAbsPath(), 'r', encoding='utf8') as abs_file:
context["abstract"] = abs_file.read()
return render(request, 'dashboard/result.html', context)
def text_reading(file_request: ConvertRequest):
print("text_reading...")
print("File path: ", file_request.getTextLocation())
# article = open(file_request.getTextLocation(),'r')
# file = article.readlines()
# article.close()
# contiguous_string = ''
# for text in file_request.fullText:
# contiguous_string += text.strip('\n')
# abstract, body = contiguous_string.split('Introdução')
abstract, body = file_request.abs, file_request.body
#print('Abstract: \n')
#print (abstract)
#print('Body: \n')
#print (body)
sentences = body.split('.')
COLS = ['Sentences']
database = pd.DataFrame(sentences, columns=COLS)
#print ("Text Reading")
#print (database.to_string())
csvFile = open(file_request.path + "/database.csv", 'w', encoding='utf-8')
database.to_csv(csvFile,
mode='w',
columns=COLS,
index=False,
encoding="utf-8")
def process_text(request):
print("Process Text!")
response = {"answer": 200, "result": None}
if request.method == 'POST':
rid = request.POST.get("rid")
file_request = ConvertRequest(rid, settings.MEDIA_DIR[0])
file_request.portion = int(request.POST.get("percent"))
file_request.body = request.POST.get("text")
file_request.saveStatus()
file_request.saveAbs(__digest(file_request))
response["result"] = file_request.id
return JsonResponse(response)
def __handle_uploaded_file(f, rid) -> ConvertRequest:
print("Here!")
# TODO validation (.pdf) 20 pages only too
# id = "".join([chr(int(65+random.random()*25)) for i in range(0, 10)])
fileRequest = ConvertRequest(rid, settings.MEDIA_DIR[0])
fileRequest.pdfFile = f
fileRequest.isSaved = fileRequest.savePDFFile()
fileRequest.isPdf = True
return fileRequest
def __digest(file_request: ConvertRequest):
if file_request.isPdf:
file_request.fullText = pdf_txt_converter(file_request) #+10%
file_request.status = 30
if file_request.isSBC:
file_request.cleanupText() #+10%
else:
file_request.body = file_request.fullText
file_request.incrementStatus(10)
else:
file_request.status = 30
text_reading(file_request) #+10%
file_request.incrementStatus(10)
names_dict = get_name_dict(file_request.path) #+10%
file_request.incrementStatus(10)
preprocessing(names_dict, file_request.path) #+10%
file_request.incrementStatus(10)
position_weighted_metric(file_request.path) #+10%
file_request.incrementStatus(10)
graph_methodology(file_request.path) #+10%
file_request.incrementStatus(10)
# tfidf_cossine_euclidean(file_request.path) #+10%
# file_request.incrementStatus(10)
#
# brush_path(file_request.path) #+10%
# file_request.incrementStatus(10)
output = k_medoids_method(file_request) #+10%
file_request.incrementStatus(10)
return output