troggle-unchained/parsers/drawings.py

325 lines
15 KiB
Python
Raw Normal View History

2011-07-11 02:10:22 +01:00
import csv
import datetime
2023-01-19 18:33:04 +00:00
import os
import re
import stat
import sys
import types
2020-05-24 13:30:39 +01:00
from functools import reduce
2023-01-19 18:33:04 +00:00
from pathlib import Path
2020-05-24 13:30:39 +01:00
import settings
2023-01-19 18:33:04 +00:00
from troggle.core.models.survex import DrawingFile, SingleScan, Wallet
2021-04-13 00:43:57 +01:00
from troggle.core.models.troggle import DataIssue
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
'''
todo='''- Rename functions more consistently between tunnel and therion variants
2022-12-07 18:22:09 +00:00
- Recode to use pathlib instead of whacky resetting of loop variable inside loop
to scan sub-folders.
- Recode rx_valid_ext to use pathlib suffix() function
- Recode load_drawings_files() to use a list of suffices not huge if-else monstrosity
2021-05-04 20:57:16 +01:00
'''
2011-07-11 02:10:22 +01:00
rx_valid_ext = re.compile(r'(?i)\.(?:png|jpg|pdf|jpeg|gif|txt)$')
def find_dwg_file(dwgfile, path):
2021-04-07 21:53:43 +01:00
'''Is given a line of text 'path' which may or may not contain a recognisable name of a scanned file
2022-12-07 18:22:09 +00:00
which we have already seen when we imported all the files we could find in the surveyscans direstories.
The purpose is to find cross-references between Tunnel drawing files. But this is not reported anywhere yet ?
What is all this really for ?! Is this data used anywhere ??
2021-04-07 21:53:43 +01:00
'''
2021-04-26 19:50:03 +01:00
wallet, scansfile = None, None
mscansdir = re.search(r"(\d\d\d\d#X?\d+\w?|1995-96kh|92-94Surveybookkh|1991surveybook|smkhs)/(.*?(?:png|jpg|pdf|jpeg|gif|txt))$", 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))
# This should be changed 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:
2022-11-23 10:41:14 +00:00
message = f"! More than one scan FOLDER matches filter query. [{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:
2021-11-06 21:37:31 +00:00
plist =[]
for sf in scansfilel:
plist.append(sf.ffile)
2022-11-23 10:41:14 +00:00
message = f"! More than one image FILENAME matches filter query. [{scansfilel[0]}]: {mscansdir.group(1)} {mscansdir.group(2)} {dwgfile.dwgpath} {path} {plist}"
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:
2022-07-27 23:48:22 +01:00
dwgfile.dwgwallets.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
2022-12-07 18:22:09 +00:00
elif path and not rx_valid_ext.search(path): # ie not recognised as a path where wallets live and not an image file type
2011-07-11 02:10:22 +01:00
name = os.path.split(path)[1]
2022-12-07 18:22:09 +00:00
rdwgfilel = DrawingFile.objects.filter(dwgname=name) # Check if it is another drawing file we have already seen
2021-04-26 18:08:42 +01:00
if len(rdwgfilel):
2021-11-06 21:37:31 +00:00
if len(rdwgfilel) > 1:
plist =[]
for df in rdwgfilel:
2022-12-07 18:22:09 +00:00
plist.append(df.dwgpath)
message = f"- Warning {len(rdwgfilel)} files named '{name}' {plist}" # should not be a problem?
2021-11-06 21:37:31 +00:00
print(message)
DataIssue.objects.create(parser='Tunnel', message=message, url=f'/dwgdataraw/{path}')
2021-11-06 21:37:31 +00:00
rdwgfile = rdwgfilel[0]
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
def findwalletimage(therionfile, foundpath):
2021-04-07 21:53:43 +01:00
'''Tries to link the drawing file (Therion format) to the referenced image (scan) file
'''
foundpath = foundpath.strip("{}")
mscansdir = re.search(r"(\d\d\d\d#\d+\w?|1995-96kh|92-94Surveybookkh|1991surveybook|smkhs)", foundpath)
if mscansdir:
scanswalletl = Wallet.objects.filter(walletname=mscansdir.group(1))
# This should be changed to properly detect if a list of folders is returned and do something sensible, not just pick the first.
if len(scanswalletl):
wallet = scanswalletl[0]
if len(scanswalletl) > 1:
message = "! More than one scan FOLDER matches filter query. [{}]: {} {} {}".format(therionfile, mscansdir.group(1), foundpath)
print(message)
DataIssue.objects.create(parser='Therion', message=message)
if wallet:
therionfile.dwgwallets.add(wallet)
scanfilename = Path(foundpath).name
scansfilel = wallet.singlescan_set.filter(name=scanfilename, wallet=wallet)
if len(scansfilel):
# message = f'! {len(scansfilel)} {scansfilel} = {scanfilename} found in the wallet specified {wallet.walletname}'
# print(message)
if len(scansfilel) > 1:
plist =[]
for sf in scansfilel:
plist.append(sf.ffile)
2022-11-23 10:41:14 +00:00
message = f"! More than one image FILENAME matches filter query. [{scansfilel[0]}]: {mscansdir.group(1)} {mscansdir.group(2)} {dwgfile.dwgpath} {path} {plist}"
print(message)
DataIssue.objects.create(parser='Therion', message=message)
scansfile = scansfilel[0]
therionfile.scans.add(scansfile)
else:
message = f'! Scanned file {scanfilename} mentioned in "{therionfile.dwgpath}" is not actually found in {wallet.walletname}'
wurl = f'/survey_scans/{wallet.walletname}/'.replace("#",":")
2022-08-14 20:52:14 +01:00
# print(message)
DataIssue.objects.create(parser='Therion', message=message, url = wurl)
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:
2022-11-23 10:41:14 +00:00
message = f"! Zero length therion file {ff}"
2021-04-07 21:53:43 +01:00
print(message)
DataIssue.objects.create(parser='Therion', message=message, url=f'/dwgdataraw/{therionfile.dwgpath}')
2021-04-07 21:53:43 +01:00
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):
# WORK IN PROGRESS. Do not clutter up the DataIssues list with this
2021-04-26 18:37:59 +01:00
message = f'! Un-parsed image filename: {therionfile.dwgname} : {xth_me.split()[-3]} - {therionfile.dwgpath}'
# print(message)
# DataIssue.objects.create(parser='xTherion', message=message, url=f'/dwgdataraw/{therionfile.dwgpath}')
# ! Un-parsed image filename: 107coldest : ../../../expofiles/surveyscans/2015/2015#20/notes.jpg - therion/plan/107coldest.th2
with open('therionrefs.log', 'a') as lg:
lg.write(message + '\n')
findwalletimage(therionfile, xth_me.split()[-3])
2021-04-07 21:53:43 +01:00
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
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='xTherion', message=message, url=f'/dwgdataraw/{therionfile.dwgpath}')
2021-04-07 21:53:43 +01:00
findimportinsert(therionfile, inp)
therionfile.save()
rx_skpath = re.compile(rb'<skpath')
rx_pcpath = re.compile(rb'<pcarea area_signal="frame".*?sfsketch="([^"]*)" sfstyle="([^"]*)"')
def settnlfileinfo(dwgfile):
2021-04-26 18:08:42 +01:00
'''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:
2022-11-23 10:41:14 +00:00
message = f"! Zero length tunnel file {ff}"
print(message)
DataIssue.objects.create(parser='Tunnel', message=message, url=f'/dwgdataraw/{dwgfile.dwgpath}')
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):
find_dwg_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
def setdrwfileinfo(dwgfile):
'''Read in the drawing file contents and sets values on the dwgfile object,
2022-03-05 22:16:03 +00:00
but these are SVGs, PDFs or .txt files, so there is no useful format to search for
This function is a placeholder in case we thnk of a way to do something
to recognise generic survex filenames.
'''
ff = Path(settings.DRAWINGS_DATA) / dwgfile.dwgpath
dwgfile.filesize = ff.stat().st_size
if dwgfile.filesize <= 0:
2022-11-23 10:41:14 +00:00
message = f"! Zero length drawing file {ff}"
print(message)
DataIssue.objects.create(parser='drawings', message=message, url=f'/dwgdataraw/{dwgfile.dwgpath}')
return
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
2022-12-07 18:22:09 +00:00
This is brain-damaged very early code. Should be replaced with proper use of pathlib.
2021-05-04 02:46:56 +01:00
Why do we have all this detection of file types/! Why not use get_mime_types ?
What is it all for ??
2022-03-05 22:16:03 +00:00
We import JPG, PNG and SVG files; which have already been put on the server,
but the upload form intentionally refuses to upload PNG and JPG (though it does allow SVG)
2021-04-07 21:53:43 +01:00
'''
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()
DataIssue.objects.filter(parser='drawings').delete()
2021-04-07 21:53:43 +01:00
DataIssue.objects.filter(parser='Therion').delete()
DataIssue.objects.filter(parser='xTherion').delete()
2021-04-07 21:53:43 +01:00
DataIssue.objects.filter(parser='Tunnel').delete()
if(os.path.isfile('therionrefs.log')):
os.remove('therionrefs.log')
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):
2022-12-07 18:22:09 +00:00
drawingsdirs.append(lf) # lunatic! adding to list in middle of list while loop! Replace with pathlib functions.
elif Path(f).suffix.lower() == ".txt":
# Always creates new
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
dwgfile.save()
all_xml.append(('txt',dwgfile))
elif Path(f).suffix.lower() == ".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))
elif Path(f).suffix.lower() == ".th":
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(('th',dwgfile))
elif Path(f).suffix.lower() == ".th2":
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(('th2',dwgfile))
elif Path(f).suffix.lower() == ".pdf":
2021-05-04 02:46:56 +01:00
# Always creates new
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
dwgfile.save()
all_xml.append(('pdf',dwgfile))
2022-03-05 22:16:03 +00:00
elif Path(f).suffix.lower() == ".png":
# Always creates new
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
dwgfile.save()
all_xml.append(('png',dwgfile))
elif Path(f).suffix.lower() == ".svg":
2021-05-04 02:46:56 +01:00
# Always creates new
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
dwgfile.save()
all_xml.append(('svg',dwgfile))
elif Path(f).suffix.lower() == ".jpg":
2021-05-04 02:46:56 +01:00
# Always creates new
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f[:-4])[1])
dwgfile.save()
all_xml.append(('jpg',dwgfile))
elif Path(f).suffix == '':
# therion file
dwgfile = DrawingFile(dwgpath=lf, dwgname=os.path.split(f)[1])
dwgfile.save()
all_xml.append(('',dwgfile))
2021-04-07 21:53:43 +01:00
print(f' - {len(all_xml)} Drawings files found')
for d in all_xml:
2022-03-05 22:16:03 +00:00
if d[0] in ['pdf', 'txt', 'svg', 'jpg', 'png', '']:
setdrwfileinfo(d[1])
2021-04-07 21:53:43 +01:00
if d[0] == 'xml':
settnlfileinfo(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)