'''
Created on 30 Apr 2019
@author: ostlerr
'''
import os
from imageToText.YieldBookToData import getPageScan, correctWords, removePunctuation
import configparser
import re
config = configparser.ConfigParser()
config.read('config.ini')
#experiment = config['EXPERIMENT']['name']
#outfile = open(config['EXPERIMENT']['outfile'], "w+", 1)
srcdocs = config['EXPERIMENT']['srcdocs']
fileList = os.listdir(srcdocs)
fileList.sort()
for fname in fileList:
nyear = fname[0:4]
npage = fname[4:6]
print(nyear + " - " + npage)
page = getPageScan(srcdocs + "\\" + fname)
print(page)
def loopDocs():
global year
fileList = os.listdir(srcdocs)
fileList.sort()
for fname in fileList:
nyear = fname[0:4]
#if int(nyear) >= 1992 and int(nyear) <= 2006 and fname.endswith(".jpg"):
if fname.endswith(".jpg"):
rawPage = getPageScan(srcdocs + "\\" + fname)
print("RAWPAGE: [" + rawPage + "]")
#rawPage = rawPage.replace("\n"," ") # This trick is for retaining line breaks, while allowing for testing line break joined words...
rawPage = correctWords(rawPage.split(" "), corrections)
metadata = Metadata()
hasMetadata = False
if rawPage.find("Object:") > -1:
hasMetadata = True
page = rawPage
page = trimPage(page, "Object:", "Sponsors:")
metadata.object = page.replace("\n", " ")
print("OBJECT: [" + metadata.object + "]")
metadata.field = identifyField(metadata.object)
print("FIELD: [" + metadata.field + "]")
if rawPage.find("Design:") > -1:
page = rawPage
page = trimPage(page, "Design:", "Plot dimensions")
metadata.design = page.replace("\n", " ")
print("DESIGN: [" + metadata.design + "]")
if rawPage.lower().find("plot dimensions:") > -1:
page = rawPage
page = trimPage(page, "Plot dimensions:", "Treatments")
metadata.wholeplots = page.replace("\n", " ")
print("DIMENSIONS: [" + metadata.wholeplots + "]")
#if rawPage.lower().find("Sub-plot dimensions")
if rawPage.find("Treatments:") > -1:
page = rawPage
page = trimPage(page, "Treatments:", "Experimental diary")
metadata.treatments = page.replace("\n",
"\ ") # markdown paragraph
print("TREATMENTS: [" + metadata.treatments + "]")
if hasMetadata:
metadataOutfile.write(experiment + "|" + str(nyear) + "|" +
metadata.field + "|" + metadata.object +
"|" + metadata.design + "|" +
metadata.wholeplots + "|" +
metadata.subplots + "|" +
metadata.treatments)
metadataOutfile.write("\n")
if rawPage.find("Sponsors:") > -1: # or page.find("$$Seed") > -1:
page = rawPage
#page = page.replace("$$", " ")
page = trimPage(page, "Sponsors:", "The")
page = page.replace("\n", " ")
print(page)
sponsors = getSponsors(page)
for sponsor in sponsors:
sponsorOutfile.write(experiment + "," + str(nyear) + "," +
sponsor)
sponsorOutfile.write("\n")