troggle-unchained/parsers/surveys.py

311 lines
12 KiB
Python
Raw Normal View History

2020-05-24 13:30:39 +01:00
import sys
import os
import types
import stat
2011-07-11 02:10:22 +01:00
import csv
import re
import datetime
2020-05-24 13:30:39 +01:00
2020-05-24 01:57:06 +01:00
from PIL import Image
2020-05-24 13:30:39 +01:00
from functools import reduce
import settings
2021-04-26 18:18:16 +01:00
from troggle.core.models.survex import SingleScan, Wallet, DrawingFile
2021-04-13 00:43:57 +01:00
from troggle.core.models.troggle import DataIssue
2021-04-13 00:11:08 +01:00
from troggle.core.utils import save_carefully
2021-04-27 20:44:24 +01:00
'''Searches through all the :drawings: repository looking
2021-04-13 01:37:42 +01:00
for tunnel and therion files
2021-04-27 20:44:24 +01:00
Searches through all the survey scans directories in expofiles, looking for images to be referenced.
2021-04-13 01:37:42 +01:00
'''
2011-07-11 02:10:22 +01:00
def get_or_create_placeholder(year):
""" All surveys must be related to a logbookentry. We don't have a way to
automatically figure out which survey went with which logbookentry,
so we create a survey placeholder logbook entry for each year. This
function always returns such a placeholder, and creates it if it doesn't
exist yet.
"""
lookupAttribs={'date__year':int(year), 'title':"placeholder for surveys",}
nonLookupAttribs={'text':"surveys temporarily attached to this should be re-attached to their actual trips", 'date':datetime.date(int(year),1,1)}
placeholder_logbook_entry, newly_created = save_carefully(LogbookEntry, lookupAttribs, nonLookupAttribs)
return placeholder_logbook_entry
2020-05-24 01:57:06 +01:00
def listdir(*directories):
try:
return os.listdir(os.path.join(settings.SURVEYS, *directories))
except:
import urllib.request, urllib.parse, urllib.error
url = settings.SURVEYS + reduce(lambda x, y: x + "/" + y, ["listdir"] + list(directories))
folders = urllib.request.urlopen(url.replace("#", "%23")).readlines()
return [folder.rstrip(r"/") for folder in folders]
2011-07-11 02:10:22 +01:00
2021-04-27 20:44:24 +01:00
# handles url or file, so we can refer to a set of scans (not drawings) on another server
2011-07-11 02:10:22 +01:00
def GetListDir(sdir):
res = [ ]
if sdir[:7] == "http://":
2021-04-13 22:27:01 +01:00
# s = urllib.request.urlopen(sdir)
message = f"! Requesting loading from http:// NOT IMPLEMENTED. [{sdir}]"
print(message)
DataIssue.objects.create(parser='Drawings', message=message)
sdir[:7] = ""
for f in os.listdir(sdir):
if f[0] != ".":
ff = os.path.join(sdir, f)
res.append((f, ff, os.path.isdir(ff)))
2011-07-11 02:10:22 +01:00
return res
2021-04-26 19:50:03 +01:00
def LoadListScansFile(wallet):
2011-07-11 02:10:22 +01:00
gld = [ ]
2020-04-27 23:51:41 +01:00
# flatten out any directories in these wallet folders - should not be any
2021-04-26 19:50:03 +01:00
for (fyf, ffyf, fisdiryf) in GetListDir(wallet.fpath):
2011-07-11 02:10:22 +01:00
if fisdiryf:
gld.extend(GetListDir(ffyf))
else:
gld.append((fyf, ffyf, fisdiryf))
c=0
2011-07-11 02:10:22 +01:00
for (fyf, ffyf, fisdiryf) in gld:
if re.search(r"\.(?:png|jpg|jpeg|pdf|svg|gif)(?i)$", fyf):
2021-04-26 19:50:03 +01:00
singlescan = SingleScan(ffile=ffyf, name=fyf, wallet=wallet)
2020-06-24 00:18:01 +01:00
singlescan.save()
c+=1
if c>=10:
2020-06-01 00:42:48 +01:00
print(".", end='')
c = 0
2011-07-11 02:10:22 +01:00
# this iterates through the scans directories (either here or on the remote server)
# and builds up the models we can access later
2021-04-27 20:44:24 +01:00
def load_all_scans():
2015-01-19 22:41:48 +00:00
print(' - Loading Survey Scans')
2015-01-19 22:48:50 +00:00
2020-06-24 00:18:01 +01:00
SingleScan.objects.all().delete()
2021-04-26 18:18:16 +01:00
Wallet.objects.all().delete()
print(' - deleting all scansFolder and scansSingle objects')
2011-07-11 02:10:22 +01:00
# first do the smkhs (large kh survey scans) directory
2021-04-26 19:22:29 +01:00
manywallets_smkhs = Wallet(fpath=os.path.join(settings.SURVEY_SCANS, "../surveys/smkhs"), walletname="smkhs")
print("smkhs", end=' ')
2021-04-26 19:22:29 +01:00
if os.path.isdir(manywallets_smkhs.fpath):
manywallets_smkhs.save()
LoadListScansFile(manywallets_smkhs)
2011-07-11 02:10:22 +01:00
# iterate into the surveyscans directory
2020-05-24 13:30:39 +01:00
print(' - ', end=' ')
for f, ff, fisdir in GetListDir(settings.SURVEY_SCANS):
2011-07-11 02:10:22 +01:00
if not fisdir:
continue
# do the year folders
2019-02-24 14:29:14 +00:00
if re.match(r"\d\d\d\d$", f):
2020-05-24 13:30:39 +01:00
print("%s" % f, end=' ')
2011-07-11 02:10:22 +01:00
for fy, ffy, fisdiry in GetListDir(ff):
if fisdiry:
2021-04-26 19:50:03 +01:00
wallet = Wallet(fpath=ffy, walletname=fy)
wallet.save()
LoadListScansFile(wallet)
2011-07-11 02:10:22 +01:00
# do the
elif f != "thumbs":
2021-04-26 19:50:03 +01:00
wallet = Wallet(fpath=ff, walletname=f)
wallet.save()
LoadListScansFile(wallet)
2011-07-11 02:10:22 +01:00
2021-04-27 20:44:24 +01:00
print("", flush=True)
def find_tunnel_file(dwgfile, path):
2021-04-07 21:53:43 +01:00
'''Is given a line of text 'path' which may or may not contain a recognisable name of a scanned file
2021-04-13 22:27:01 +01:00
which we have already seen when we imported all the files we could find in the surveyscans direstories
2021-04-07 21:53:43 +01:00
'''
2021-04-26 19:50:03 +01:00
wallet, scansfile = None, None
2021-04-27 20:44:24 +01:00
mscansdir = re.search(r"(\d\d\d\d#X?\d+\w?|1995-96kh|92-94Surveybookkh|1991surveybook|smkhs)/(.*?(?:png|jpg|pdf|jpeg|gif))$", path)
2011-07-11 02:10:22 +01:00
if mscansdir:
2021-04-26 19:50:03 +01:00
scanswalletl = Wallet.objects.filter(walletname=mscansdir.group(1))
2021-04-27 20:44:24 +01:00
# This should be chnaged to properly detect if a list of folders is returned and do something sensible, not just pick the first.
2021-04-26 19:50:03 +01:00
if len(scanswalletl):
wallet = scanswalletl[0]
if len(scanswalletl) > 1:
2021-04-26 18:11:14 +01:00
message = "! More than one scan FOLDER matches filter query. [{}]: {} {} {} {}".format(scansfilel[0], mscansdir.group(1), mscansdir.group(2), dwgfile.dwgpath, path)
2021-04-13 22:27:01 +01:00
print(message)
DataIssue.objects.create(parser='Tunnel', message=message)
2021-04-26 19:50:03 +01:00
if wallet:
scansfilel = wallet.singlescan_set.filter(name=mscansdir.group(2))
2011-07-11 02:10:22 +01:00
if len(scansfilel):
if len(scansfilel) > 1:
2021-04-26 18:11:14 +01:00
message = "! More than one image FILENAME matches filter query. [{}]: {} {} {} {}".format(scansfilel[0], mscansdir.group(1), mscansdir.group(2), dwgfile.dwgpath, path)
print(message)
2021-04-07 21:53:43 +01:00
DataIssue.objects.create(parser='Tunnel', message=message)
2011-07-11 02:10:22 +01:00
scansfile = scansfilel[0]
2021-04-26 19:50:03 +01:00
if wallet:
dwgfile.manywallets.add(wallet)
2011-07-11 02:10:22 +01:00
if scansfile:
2021-04-26 18:08:42 +01:00
dwgfile.scans.add(scansfile)
2011-07-11 02:10:22 +01:00
2021-04-27 20:44:24 +01:00
elif path and not re.search(r"\.(?:png|jpg|pdf|jpeg|gif)$(?i)", path):
2011-07-11 02:10:22 +01:00
name = os.path.split(path)[1]
2021-04-26 18:37:59 +01:00
rdwgfilel = DrawingFile.objects.filter(dwgname=name)
2021-04-26 18:08:42 +01:00
if len(rdwgfilel):
message = "! Two paths with same name [{}]: {}".format(path, name)
print(message)
2021-04-07 21:53:43 +01:00
DataIssue.objects.create(parser='Tunnel', message=message)
2021-04-26 18:08:42 +01:00
rdwgfile = rdwgfilel[0]
2021-04-26 18:54:17 +01:00
dwgfile.dwgcontains.add(rdwgfile)
2011-07-11 02:10:22 +01:00
2021-04-26 18:08:42 +01:00
dwgfile.save()
2011-07-11 02:10:22 +01:00
2021-04-07 21:53:43 +01:00
def findimageinsert(therionfile, xth_me):
'''Tries to link the drawing file (Therion format) to the referenced image (scan) file
'''
pass
2011-07-11 02:10:22 +01:00
2021-04-07 21:53:43 +01:00
def findimportinsert(therionfile, imp):
'''Tries to link the scrap (Therion format) to the referenced therion scrap
'''
pass
rx_xth_me = re.compile(r'xth_me_image_insert.*{.*}$', re.MULTILINE)
rx_scrap = re.compile(r'^survey (\w*).*$', re.MULTILINE)
rx_input = re.compile(r'^input (\w*).*$', re.MULTILINE)
def settherionfileinfo(filetuple):
2021-04-26 18:08:42 +01:00
'''Read in the drawing file contents and sets values on the dwgfile object
2021-04-07 21:53:43 +01:00
'''
thtype, therionfile = filetuple
2021-04-26 18:42:10 +01:00
ff = os.path.join(settings.DRAWINGS_DATA, therionfile.dwgpath)
2021-04-07 21:53:43 +01:00
therionfile.filesize = os.stat(ff)[stat.ST_SIZE]
if therionfile.filesize <= 0:
message = "! Zero length therion file {}".format(ff)
print(message)
DataIssue.objects.create(parser='Therion', message=message)
return
fin = open(ff,'r')
ttext = fin.read()
fin.close()
# The equivalent for a tunnel 'path' would be a .th2 'line wall' or 'scrap'
# print(len(re.findall(r"line", ttext)))
if thtype=='th':
therionfile.npaths = len(re.findall(r"^input ", ttext, re.MULTILINE))
elif thtype=='th2':
therionfile.npaths = len(re.findall(r"^line ", ttext, re.MULTILINE))
therionfile.save()
# scan and look for survex blocks that might have been included, and image scans (as for tunnel drawings)
2021-04-26 18:08:42 +01:00
# which would populate dwgfile.survexfile
2021-04-07 21:53:43 +01:00
# in .th2 files:
# ##XTHERION## xth_me_image_insert {500 1 1.0} {1700 {}} ../../../expofiles/surveyscans/2014/01popped_elev1.jpeg 0 {}
# scrap blownout -projection plan -scale [-81.0 -42.0 216.0 -42.0 0.0 0.0 7.5438 0.0 m]
for xth_me in rx_xth_me.findall(ttext):
2021-04-26 18:37:59 +01:00
message = f'! Un-parsed image filename: {therionfile.dwgname} : {xth_me.split()[-3]} - {therionfile.dwgpath}'
2021-04-07 21:53:43 +01:00
#print(message)
DataIssue.objects.create(parser='Therion', message=message)
findimageinsert(therionfile, xth_me)
for inp in rx_input.findall(ttext):
# if this 'input' is a .th2 file we have already seen, then we can assign this as a sub-file
# but we would need to disentangle to get the current path properly
2021-04-26 18:37:59 +01:00
message = f'! Un-set Therion .th2 input: - {therionfile.dwgname} : {inp} - {therionfile.dwgpath}'
2021-04-07 21:53:43 +01:00
#print(message)
DataIssue.objects.create(parser='Therion', message=message)
findimportinsert(therionfile, inp)
therionfile.save()
rx_skpath = re.compile(rb'<skpath')
rx_pcpath = re.compile(rb'<pcarea area_signal="frame".*?sfsketch="([^"]*)" sfstyle="([^"]*)"')
2021-04-26 18:08:42 +01:00
def setdwgfileinfo(dwgfile):
'''Read in the drawing file contents and sets values on the dwgfile object
2021-04-08 01:09:06 +01:00
Should try to read the date too e.g. tunneldate="2010-08-16 22:51:57
then we could display on the master calendar per expo.
2021-04-07 21:53:43 +01:00
'''
2021-04-26 18:42:10 +01:00
ff = os.path.join(settings.DRAWINGS_DATA, dwgfile.dwgpath)
2021-04-26 18:08:42 +01:00
dwgfile.filesize = os.stat(ff)[stat.ST_SIZE]
if dwgfile.filesize <= 0:
message = "! Zero length xml file {}".format(ff)
print(message)
DataIssue.objects.create(parser='Drawings', message=message)
return
2020-05-24 13:30:39 +01:00
fin = open(ff,'rb')
2011-07-11 02:10:22 +01:00
ttext = fin.read()
fin.close()
2021-04-07 21:53:43 +01:00
2021-04-26 18:08:42 +01:00
dwgfile.npaths = len(rx_skpath.findall(ttext))
dwgfile.save()
2011-07-11 02:10:22 +01:00
2021-04-07 21:53:43 +01:00
# example drawing file in Tunnel format.
2011-07-11 02:10:22 +01:00
# <tunnelxml tunnelversion="version2009-06-21 Matienzo" tunnelproject="ireby" tunneluser="goatchurch" tunneldate="2009-06-29 23:22:17">
# <pcarea area_signal="frame" sfscaledown="12.282584" sfrotatedeg="-90.76982" sfxtrans="11.676667377221136" sfytrans="-15.677173422877454" sfsketch="204description/scans/plan(38).png" sfstyle="" nodeconnzsetrelative="0.0">
2021-04-07 21:53:43 +01:00
for path, style in rx_pcpath.findall(ttext):
2021-04-27 20:44:24 +01:00
find_tunnel_file(dwgfile, path.decode())
2021-04-07 21:53:43 +01:00
# should also scan and look for survex blocks that might have been included, and image scans
2021-04-26 18:08:42 +01:00
# which would populate dwgfile.survexfile
2021-04-26 18:08:42 +01:00
dwgfile.save()
2011-07-11 02:10:22 +01:00
2021-04-07 21:53:43 +01:00
def load_drawings_files():
'''Breadth first search of drawings directory looking for sub-directories and *.xml filesize
'''
all_xml = []
2021-04-26 18:42:10 +01:00
drawdatadir = settings.DRAWINGS_DATA
2021-04-26 18:08:42 +01:00
DrawingFile.objects.all().delete()
2021-04-07 21:53:43 +01:00
DataIssue.objects.filter(parser='Drawings').delete()
DataIssue.objects.filter(parser='Therion').delete()
DataIssue.objects.filter(parser='Tunnel').delete()
drawingsdirs = [ "" ]
while drawingsdirs:
drawdir = drawingsdirs.pop()
for f in os.listdir(os.path.join(drawdatadir, drawdir)):
2011-07-11 02:10:22 +01:00
if f[0] == "." or f[-1] == "~":
continue
lf = os.path.join(drawdir, f)
ff = os.path.join(drawdatadir, lf)
2011-07-11 02:10:22 +01:00
if os.path.isdir(ff):
2021-04-07 21:53:43 +01:00
drawingsdirs.append(lf) # lunatic! adding to list in middle of list while loop!
2011-07-11 02:10:22 +01:00
elif f[-4:] == ".xml":
2021-04-07 21:53:43 +01:00
# Always creates new
2021-04-26 18:37:59 +01:00
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
2021-04-26 18:08:42 +01:00
dwgfile.save()
all_xml.append(('xml',dwgfile))
2021-04-07 21:53:43 +01:00
elif f[-3:] == ".th":
# Always creates new
2021-04-26 18:37:59 +01:00
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
2021-04-26 18:08:42 +01:00
dwgfile.save()
all_xml.append(('th',dwgfile))
2021-04-07 21:53:43 +01:00
elif f[-4:] == ".th2":
# Always creates new
2021-04-26 18:37:59 +01:00
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
2021-04-26 18:08:42 +01:00
dwgfile.save()
all_xml.append(('th2',dwgfile))
2021-04-07 21:53:43 +01:00
print(f' - {len(all_xml)} Drawings files found')
for d in all_xml:
if d[0] == 'xml':
2021-04-26 18:08:42 +01:00
setdwgfileinfo(d[1])
2021-04-07 21:53:43 +01:00
# important to import .th2 files before .th so that we can assign them when found in .th files
if d[0] == 'th2':
settherionfileinfo(d)
if d[0] == 'th':
settherionfileinfo(d)
2021-04-26 18:08:42 +01:00
# for drawfile in DrawingFile.objects.all():
2021-04-07 21:53:43 +01:00
# SetTunnelfileInfo(drawfile)