def get_outline(coords,width): left=[] right=[] uprint("Coords:\n<%s>\n, width: <%s>"%(coords,width)) for idx in xrange(len(coords)): here=coords[idx] mid=True if idx==0: prev=coords[idx] mid=False else: prev=coords[idx-1] if idx==len(coords)-1: next=coords[idx] mid=False else: next=coords[idx+1] d=(float(next[0]-prev[0]),float(next[1]-prev[1])) enwiden=1.0 if mid: d1=normalize(sub(here,prev)) d2=normalize(sub(next,here)) s=scalar(d1,d2) if s<=-1.0: s=-1.0 if s>=1.0: s=1.0 a=math.acos(s) if abs(a)>60.0/(180.0/math.pi): raise Exception("ATS route turns too sharply, at index: %d"%(idx,)) enwiden=1.0/math.cos(a) width*=enwiden d=normalize(d) left.append(add(here,mul(0.5*width,rotL(d)))) right.append(add(here,mul(0.5*width,rotR(d)))) return right+list(reversed(left))
def get_outline(coords, width): left = [] right = [] uprint("Coords:\n<%s>\n, width: <%s>" % (coords, width)) for idx in xrange(len(coords)): here = coords[idx] mid = True if idx == 0: prev = coords[idx] mid = False else: prev = coords[idx - 1] if idx == len(coords) - 1: next = coords[idx] mid = False else: next = coords[idx + 1] d = (float(next[0] - prev[0]), float(next[1] - prev[1])) enwiden = 1.0 if mid: d1 = normalize(sub(here, prev)) d2 = normalize(sub(next, here)) s = scalar(d1, d2) if s <= -1.0: s = -1.0 if s >= 1.0: s = 1.0 a = math.acos(s) if abs(a) > 60.0 / (180.0 / math.pi): raise Exception("ATS route turns too sharply, at index: %d" % (idx, )) enwiden = 1.0 / math.cos(a) width *= enwiden d = normalize(d) left.append(add(here, mul(0.5 * width, rotL(d)))) right.append(add(here, mul(0.5 * width, rotR(d)))) return right + list(reversed(left))
def getsig(coords, altspecs): poss = [] #print "Coords: <%s>"%(coords,) for y1, y2, pos in coords: poss.append(pos) uprint("poss: <%s>" % (poss, )) s = "-".join(poss) s2 = "-".join(altspecs) return md5.md5(s.encode('utf8') + s2.encode('utf8')).hexdigest()
def getsig(coords,altspecs): poss=[] #print "Coords: <%s>"%(coords,) for y1,y2,pos in coords: poss.append(pos) uprint("poss: <%s>"%(poss,)) s="-".join(poss) s2="-".join(altspecs) return md5.md5(s.encode('utf8')+s2.encode('utf8')).hexdigest()
def parse_r_areas(): p = parse.Parser("/AIP/ENR/ENR 5/ES_ENR_5_1_en.pdf", lambda x: x) res = [] for pagenr in xrange(2, p.get_num_pages()): parsed = parse_page(p, pagenr, "R") res.extend(parsed) for pa in res: uprint("pA:", repr(pa)) pretty(pa) return res
def parse_r_areas(): p=parse.Parser("/AIP/ENR/ENR 5/ES_ENR_5_1_en.pdf",lambda x: x) res=[] for pagenr in xrange(2,p.get_num_pages()): parsed=parse_page(p,pagenr,"R") res.extend(parsed) for pa in res: uprint("pA:",repr(pa)) pretty(pa) return res
def fi_parse_airfield(icao=None): spaces = [] ad = dict() assert icao != None ad['icao'] = icao sigpoints = [] #https://ais.fi/ais/eaip/pdf/aerodromes/EF_AD_2_EFET_EN.pdf #https://ais.fi/ais/eaip/aipcharts/efet/EF_AD_2_EFET_VAC.pdf #vacp=parse.Parser("/ais/eaip/aipcharts/%s/EF_AD_2_%s_VAC.pdf"%(icao.lower(),icao),lambda x: x,country="fi") def remove_italics(x): return x.replace("<i>", "").replace("</i>", "") p = parse.Parser("/ais/eaip/pdf/aerodromes/EF_AD_2_%s_EN.pdf" % (icao, ), remove_italics, country="fi") #The following doesn't actually work, since finnish VAC are bitmaps!!! :-( if 0: vacpage = vacp.parse_page_to_items(0) repp = vacpage.get_by_regex("\s*REPORTING\s*POINTS\s*") assert len(repp) > 0 for item in repp: lines = iter( page.get_lines( page.get_partially_in_rect(item.x1, item.y2 + 0.1, 100, 100))) for line in lines: uprint("Looking for reporting points:%s" % (line, )) name, lat, lon = re.match( ur"([A-ZÅÄÖ\s ]{3,})\s*([ \d]+N)\s*([ \d]+E).*", line) sigpoints.append( dict(name=icao + " " + name.strip(), kind="reporting", pos=mapper.parse_coords(lat.replace(" ", ""), lon.replace(" ", "")))) page = p.parse_page_to_items(0) nameregex = ur"%s\s+-\s+([A-ZÅÄÖ\- ]{3,})" % (icao, ) for item in page.get_by_regex(nameregex): #print "fontsize:",item.fontsize assert item.fontsize >= 14 ad['name'] = re.match(nameregex, item.text).groups()[0].strip() break for item in page.get_by_regex(ur".*ELEV\s*/\s*REF.*"): lines = page.get_lines( page.get_partially_in_rect(0, item.y1 + 0.1, 100, item.y2 - 0.1)) for line in lines: print "Line:", line ft, = re.match(".*ELEV.*([\d\.]+)\s*FT.*", line).groups() assert not 'elev' in ad ad['elev'] = float(ft)
def pretty(pa): uprint("\n\nName: %s" % (pa['name'].encode('utf8'), )) uprint( "==============================================================================" ) uprint("Floor: %s, Ceiling: %s, freqs: %s" % (pa['floor'], pa['ceiling'], pa['freqs'])) uprint("Points: %s" % (pa['points'], ))
def get_airspaces(routes): for routename,coords,altspec in routes: sig=getsig(coords,altspec) if (routename,sig) in predef: #pobj=None #for sroutename,ssig in predef: # if routename==sroutename: pobj=(routename,sig) width_nm,floor,ceiling,coordstr=predef[pobj] rawcoords=re.findall(ur"(\d{6}N)\s*(\d{7}E)",coordstr) coords=[mapper.latlon2merc(mapper.from_str(mapper.parse_coords(lats,lons)),13) for lats,lons in rawcoords] width=float(mapper.approx_scale(coords[0],13,1.25*width_nm)) try: outline=get_outline(coords,width) except Exception: uprint(u"Trouble parsing %s"%(routename,)) raise yield dict(name=routename, floor=floor, ceiling=ceiling, freqs=[], type="RNAV", points=[mapper.to_str(mapper.merc2latlon(x,13)) for x in outline]) else: uprint("Need to have predefine for route %s, with md5: %s"%(routename,sig)) uprint("Altitude, and points") raise Exception('Missing predef for route. Use: ("%s","%s") : ( 10,"FL 65","FL 95","""\n\n """),'%(routename,sig))
def low_routes(routes): for routename,coords in routes: uprint("Found route: %s, with %s cords"%(routename,coords)) assert len(coords)>=2 y1=coords[0][0] y2=coords[-1][1] foundlow=False altspec=[] for item in page.get_partially_in_rect(0,y1,100,y2): uprint("Looking for alt for %s in :"%(routename,),item.text) m=re.match(ur".*FL\s+(\d{2,3})\b(?:\s.*|$)",item.text) if m: fl,=m.groups() uprint("Matched alt:",m.groups()) fli=int(fl,10) if fli<95: uprint("Foundlow for route %s: %d"%(routename,fli)) foundlow=True altspec.append(item.text) if foundlow: yield routename,coords,altspec
def low_routes(routes): for routename, coords in routes: uprint("Found route: %s, with %s cords" % (routename, coords)) assert len(coords) >= 2 y1 = coords[0][0] y2 = coords[-1][1] foundlow = False altspec = [] for item in page.get_partially_in_rect(0, y1, 100, y2): uprint("Looking for alt for %s in :" % (routename, ), item.text) m = re.match(ur".*FL\s+(\d{2,3})\b(?:\s.*|$)", item.text) if m: fl, = m.groups() uprint("Matched alt:", m.groups()) fli = int(fl, 10) if fli < 95: uprint("Foundlow for route %s: %d" % (routename, fli)) foundlow = True altspec.append(item.text) if foundlow: yield routename, coords, altspec
def isol_routes(): cur=None accum=[] for line in page.get_lines(page.get_all_items()): uprint("Looking at ",line) m=re.match(ur"(^[A-Z]{1,2}\d+)\s+\(.*",line) if m and line.x1<25: if cur: uprint("Yielding %s: %s because of found name:%s"%(cur,accum,m.groups())) yield cur,accum accum=[] cur,=m.groups() continue m=re.match(ur"(\d{6}N)\s*(\d{7}E).*",line) if m: accum.append((line.y1,line.y2," ".join(m.groups()))) if cur: uprint("Yielding %s: %s because of end of page"%(cur,accum)) yield cur,accum accum=[]
def isol_routes(): cur = None accum = [] for line in page.get_lines(page.get_all_items()): uprint("Looking at ", line) m = re.match(ur"(^[A-Z]{1,2}\d+)\s+\(.*", line) if m and line.x1 < 25: if cur: uprint("Yielding %s: %s because of found name:%s" % (cur, accum, m.groups())) yield cur, accum accum = [] cur, = m.groups() continue m = re.match(ur"(\d{6}N)\s*(\d{7}E).*", line) if m: accum.append((line.y1, line.y2, " ".join(m.groups()))) if cur: uprint("Yielding %s: %s because of end of page" % (cur, accum)) yield cur, accum accum = []
def get_airspaces(routes): for routename, coords, altspec in routes: sig = getsig(coords, altspec) if (routename, sig) in predef: #pobj=None #for sroutename,ssig in predef: # if routename==sroutename: pobj = (routename, sig) width_nm, floor, ceiling, coordstr = predef[pobj] rawcoords = re.findall(ur"(\d{6}N)\s*(\d{7}E)", coordstr) coords = [ mapper.latlon2merc( mapper.from_str(mapper.parse_coords(lats, lons)), 13) for lats, lons in rawcoords ] width = float( mapper.approx_scale(coords[0], 13, 1.25 * width_nm)) try: outline = get_outline(coords, width) except Exception: uprint(u"Trouble parsing %s" % (routename, )) raise yield dict(name=routename, floor=floor, ceiling=ceiling, freqs=[], type="RNAV", points=[ mapper.to_str(mapper.merc2latlon(x, 13)) for x in outline ]) else: uprint( "Need to have predefine for route %s, with md5: %s" % (routename, sig)) uprint("Altitude, and points") raise Exception( 'Missing predef for route. Use: ("%s","%s") : ( 10,"FL 65","FL 95","""\n\n """),' % (routename, sig))
def replacer(args): uprint(args.groups()) y,x,w,h,font=args.groups() uprint(w,h) assert int(w)>=260 and int(w)<420 assert int(h)>=6 and int(h)<=15 f=float(w)/270.0 x1=x y1=y w1=80 h1=h x2=168*f y2=y w2=150*f h2=h did_replace[0]+=1 repl="""<text top="%s" left="%s" width="%s" height="%s" font="%s">Part of GÖTEBORG TMA</text> <text top="%s" left="%s" width="%s" height="%s" font="%s">584558N 0122951E - 584358N 0130950E - </text>"""%( y1,x1,w1,h1,font,y2,x2,w2,h2,font) uprint("\n======================================\nReplacement:\n",repl) return repl
def replacer(args): uprint(args.groups()) y, x, w, h, font = args.groups() uprint(w, h) assert int(w) >= 260 and int(w) < 420 assert int(h) >= 6 and int(h) <= 15 f = float(w) / 270.0 x1 = x y1 = y w1 = 80 h1 = h x2 = 168 * f y2 = y w2 = 150 * f h2 = h did_replace[0] += 1 repl = """<text top="%s" left="%s" width="%s" height="%s" font="%s">Part of GÖTEBORG TMA</text> <text top="%s" left="%s" width="%s" height="%s" font="%s">584558N 0122951E - 584358N 0130950E - </text>""" % ( y1, x1, w1, h1, font, y2, x2, w2, h2, font) uprint("\n======================================\nReplacement:\n", repl) return repl
def pretty(pa): uprint("\n\nName: %s"%(pa['name'].encode('utf8'),)) uprint("==============================================================================") uprint("Floor: %s, Ceiling: %s, freqs: %s"%(pa['floor'],pa['ceiling'],pa['freqs'])) uprint("Points: %s"%(pa['points'],))
def fi_parse_airfield(icao=None): spaces=[] ad=dict() assert icao!=None ad['icao']=icao sigpoints=[] #https://ais.fi/ais/eaip/pdf/aerodromes/EF_AD_2_EFET_EN.pdf #https://ais.fi/ais/eaip/aipcharts/efet/EF_AD_2_EFET_VAC.pdf #vacp=parse.Parser("/ais/eaip/aipcharts/%s/EF_AD_2_%s_VAC.pdf"%(icao.lower(),icao),lambda x: x,country="fi") def remove_italics(x): return x.replace("<i>","").replace("</i>","") p=parse.Parser("/ais/eaip/pdf/aerodromes/EF_AD_2_%s_EN.pdf"%(icao,),remove_italics,country="fi") #The following doesn't actually work, since finnish VAC are bitmaps!!! :-( if 0: vacpage=vacp.parse_page_to_items(0) repp=vacpage.get_by_regex("\s*REPORTING\s*POINTS\s*") assert len(repp)>0 for item in repp: lines=iter(page.get_lines(page.get_partially_in_rect(item.x1,item.y2+0.1,100,100))) for line in lines: uprint("Looking for reporting points:%s"%(line,)) name,lat,lon=re.match(ur"([A-ZÅÄÖ\s ]{3,})\s*([ \d]+N)\s*([ \d]+E).*",line) sigpoints.append(dict( name=icao+" "+name.strip(), kind="reporting", pos=mapper.parse_coords(lat.replace(" ",""),lon.replace(" ","")))) page=p.parse_page_to_items(0) nameregex=ur"%s\s+-\s+([A-ZÅÄÖ\- ]{3,})"%(icao,) for item in page.get_by_regex(nameregex): #print "fontsize:",item.fontsize assert item.fontsize>=14 ad['name']=re.match(nameregex,item.text).groups()[0].strip() break for item in page.get_by_regex(ur".*ELEV\s*/\s*REF.*"): lines=page.get_lines(page.get_partially_in_rect(0,item.y1+0.1,100,item.y2-0.1)) for line in lines: print "Line:",line ft,=re.match(".*ELEV.*([\d\.]+)\s*FT.*",line).groups() assert not 'elev' in ad ad['elev']=float(ft) for item in page.get_by_regex(ur"Mittapisteen.*sijainti"): lines=page.get_lines(page.get_partially_in_rect(item.x1,item.y1,100,item.y2)) for line in lines: for crd in mapper.parsecoords(line): assert not ('pos' in ad) ad['pos']=crd parse_landing_chart.help_plc(ad, "/ais/eaip/aipcharts/%s/EF_AD_2_%s_ADC.pdf"%(icao.lower(),icao.upper()), icao,ad['pos'],country='fi' ) parse_landing_chart.help_plc(ad, "/ais/eaip/aipcharts/%s/EF_AD_2_%s_VAC.pdf"%(icao.lower(),icao.upper()), icao,ad['pos'],country='fi',variant='VAC' ) parse_landing_chart.help_plc(ad, "/ais/eaip/aipcharts/%s/EF_AD_2_%s_LDG.pdf"%(icao.lower(),icao.upper()), icao,ad['pos'],country='fi',variant='landing' ) parse_landing_chart.help_plc(ad, "/ais/eaip/aipcharts/%s/EF_AD_2_%s_APDC.pdf"%(icao.lower(),icao.upper()), icao,ad['pos'],country='fi',variant='parking' ) aip_text_documents.help_parse_doc(ad,"/ais/eaip/pdf/aerodromes/EF_AD_2_%s_EN.pdf"%(icao.upper(),), icao,"fi",title="General Information",category="general") ad['runways']=[] thrs=[] freqs=[] for pagenr in xrange(p.get_num_pages()): page=p.parse_page_to_items(pagenr) if page==None: continue for item in page.get_by_regex("\s*RUNWAY\s*PHYSICAL\s*CHARACTERISTICS\s*"): lines=page.get_lines(page.get_partially_in_rect(0,item.y2+0.1,100,100)) for line in lines: if re.match(ur"AD\s+2.13",line): break m=re.match(ur".*?(RWY END)?\s*\*?(\d{6}\.\d+N)\s*(\d{6,7}\.\d+E).*",line) if not m:continue rwyend,lat,lon=m.groups() rwytxts=page.get_lines(page.get_partially_in_rect(0,line.y1,12,line.y2)) print "Rwytxts:",rwytxts rwytxt,=rwytxts uprint("rwytext:",rwytxt) rwy,=re.match(ur"\s*(\d{2}[LRCM]?)\s*[\d.]*\s*",rwytxt).groups() have_thr=False for thr in thrs: if thr['thr']==rwy: have_thr=True if rwyend!=None and have_thr: continue thrs.append(dict(pos=mapper.parse_coords(lat,lon),thr=rwy)) for item in page.get_by_regex("ATS AIRSPACE"): lines=iter(page.get_lines(page.get_partially_in_rect(0,item.y2+0.1,100,100))) spaces=[] line=lines.next() while True: while line.strip()=="": line=lines.next() print "Read line:",line if line.count("Vertical limits"): break m=re.match(ur".*?/\s+Designation and lateral limits\s*(.*\b(?:CTR|FIZ)\b.*?)\s*:?\s*$",line) if not m: m=re.match(ur"\s*(.*\b(?:CTR|FIZ)\b.*?)\s*:",line) #print "Second try:",m spacename,=m.groups() #print "Got spacename:",spacename assert spacename.strip()!="" coords=[] while True: line=lines.next() print "Further:",line if line.count("Vertical limits"): print "Breaking" break if not re.search(ur"[\d ]+N\s*[\d ]+E",line) and \ not re.search(ur"circle|cent[red]{1,5}|pitkin|point|equal\s*to",line): print "Breaking" break coords.append(line) areaspec="".join(coords) def fixup(m): lat,lon=m.groups() return lat.replace(" ","")+" "+lon.replace(" ","") areaspec=re.sub(ur"([\d ]+N)\s*([\d ]+E)",fixup,areaspec) areaspec=re.sub(ur"\(.*/\s*equal\s*to\s*Malmi\s*CTR\s*lateral\s*limits\)","",areaspec) #print "Fixed areaspec",areaspec #if icao=="EFKS": # areaspec=areaspec.replace("6615 28N","661528N") #Error! REstriction areas! spaces.append(dict( name=spacename, type="CTR", points=mapper.parse_coord_str(areaspec))) if line.count("Vertical limits"): #print "Breaking" break while not line.count("Vertical limits"): line=lines.next() #print "Matching veritcal limits--------------------------------" oldspaces=spaces spaces=[] for space in oldspaces: if space['name'].count("/"): a,b=space['name'].split("/") spaces.append(dict(space,name=a.strip())) spaces.append(dict(space,name=b.strip())) else: spaces.append(space) missing=set([space['name'] for space in spaces]) while True: for space in spaces: for it in xrange(3): cand=space['name'] if it==1: if cand.count("CTR"): cand="CTR" if cand.count("FIZ"): cand="FIZ" if it==2: if cand.count("CTR"): cand=r"CTR\s*/[\sA-Z]+" if cand.count("FIZ UPPER"): cand="FIZ UPPER" if cand.count("FIZ LOWER"): cand="FIZ LOWER" m=re.match(ur".*%s\s*:([^,:-]*)\s*-\s*([^,:-]*)"%(cand,),line) print "Matching ",cand," to ",line,"missing:",missing,m if m: break if len(spaces)==1 and not m: m=re.match(ur".*Vertical limits\s*(.*)\s*-\s*(.*)",line) if m: print "*****MATCH!!:::",m.groups() for lim in m.groups(): assert lim.count(",")==0 space['floor'],space['ceiling']=m.groups() missing.remove(space['name']) #print "Missing:" if len(missing)==0: break if len(missing)==0: break #print "Still missing:",missing line=lines.next() print "Parse f o n page",pagenr for item2 in page.get_by_regex(ur".*ATS\s*COMMUNICATION\s*FACILITIES.*"): lines=page.get_lines(page.get_partially_in_rect(0,item2.y2+0.1,100,100)) for line in lines: if line.count("RADIO NAVIGATION AND LANDING AIDS"): break print "Comm line:",line twr=re.match(ur"TWR.*(\d{3}\.\d{3})\b.*",line) if twr: freqs.append(('TWR',float(twr.groups()[0]))) atis=re.match(ur"ATIS.*(\d{3}\.\d{3})",line) if atis: freqs.append(('ATIS',float(atis.groups()[0])))
def parse_page(parser,pagenr,kind="TMA",last_sector=dict()): if kind=="TMA": thirdcols=["ATC unit","AFIS unit"] elif kind=="sector": thirdcols=["FREQ"] elif kind=="R": thirdcols=["Remarks (nature of hazard,"] else: raise Exception("Bad kind") page=parser.parse_page_to_items(pagenr) items=page.items #print "Items:",pitems #print "Possible Areas:" headings=[] for item in items: if item.text==None: continue item.text=item.text.strip() if item.text=="": continue if item.text=="Name": continue if item.y1<25 and item.text in (["Lateral limits","Vertical limits"]+thirdcols): headings.append(item) headings.sort(key=lambda x:x.x1) #print "found candidates:",zone_candidates if len(headings)==0: return [] avg_heading_y=sum(h.y1 for h in headings)/float(len(headings)) uprint("Found headings:",headings) zone_candidates=[] for item in items: if item.text==None or item.text.strip()=="": continue if item.text.strip().startswith("AMDT"): continue if item.text.strip().startswith("The LFV Group"): continue if re.match(ur"\s*LFV\s*AIRAC\s*AMDT\s*\d+/\d+\s*",item.text): continue if item.text.strip()=="LFV": continue if item.text.count('Terminal Information Areas'): continue if item.text.strip().startswith("AIRAC"): continue if kind=="R" and not is_r_or_danger_area_name(item.text.strip()): continue if item.y1>avg_heading_y+1 and item.x1<12 and not item.text in ["Name",'None',"LFV"]: if item.text.count("Established") or item.text.count(u'TROLLHÄTTAN TWR') or item.text.count(u'and/or SÅTENÄS') or item.text.count(u'TWR/TMC') or item.text.strip().endswith("TWR") or item.text.strip().endswith("TWR."): continue if item.text.count("operational hours") or item.text.count("See AIP DENMARK"): continue if item.text.count("hours of"): continue if item.text.count("Upper limit"): continue if item.text.count("that part") or item.text.count("coincides"): continue if item.text.count(u'Danger area EK D395 and') or item.text.count(u'D396 are situated within') or item.text.strip()=="TMA": continue if item.text.count(u'ÖSTGÖTA TMC is closed') or item.text.count(u'and SKAVSTA TWR is') or item.text.strip()=='open.': continue if item.text.count("SAT 0530"): continue if item.text.strip()=='OPS': continue if item.text.strip()==u'ÖSTGÖTA TMC:': continue if item.text.count(u'is open') or item.text.count('is closed'): continue if item.text.count('MON-FRI') or item.text.count('2150'): continue lines2=page.get_lines(page.get_partially_in_rect(12,item.y1+0.2,40,item.y2-0.2)) if len(lines2): zone_candidates.append(item) uprint("Found cands:",zone_candidates) zone_candidates.sort(key=lambda x:x.y1) for zone in zone_candidates: #uprint("Zone:",zone) #assert not zone.text.count("AOR") assert not zone.text.count("FIR") uprint("Headings:",headings) print "Pagenr:",pagenr assert len(headings)==3 ret=[] for i in xrange(len(zone_candidates)): d=dict() cand=zone_candidates[i] if i<len(zone_candidates)-1: nextcand=zone_candidates[i+1] else: nextcand=None y1=cand.y1-0.25 y2=100 if nextcand: y2=nextcand.y1-0.75 for j in xrange(len(headings)): head=headings[j] if j<len(headings)-1: nexthead=headings[j+1] else: nexthead=None x1=head.x1 x2=head.x2 if j==len(headings)-1: x1=headings[j-1].x2+3 x2=100 lines=page.get_lines(page.get_partially_in_rect(x1,y1,x2,y2,xsort=True,ysort=True)) #print ("Parsed %s y,%d-%d, %s: <%s>\n\n"%(cand.text,y1,y2,head.text,lines)).encode('utf8') d[head.text]=lines if kind=="R": if y2==100: y2=y1+3 d['name']=" ".join(x.strip() for x in filter_head_foot(page.get_lines(page.get_partially_in_rect(0,y1,10,y2,xsort=True,ysort=True)))) else: d['name']=cand.text.strip() ret.append(d) allow_head=2 print "Doing fixups--------------------------------------------------" tret=[] for x in ret: #print "Fixing up",x,"allow:",allow_head area="".join(x['Lateral limits']).strip() if allow_head==2 and area!="" and x['name'].strip()!="": allow_head=1 if allow_head!=1: if len(tret): tret[-1]['Lateral limits']+=x['Lateral limits'] tret[-1]['Vertical limits']+=x['Vertical limits'] else: tret.append(x) if allow_head==1: allow_head=0 if not area.endswith('-') and area!="": allow_head=2 #print " Fixed up up",x ret=tret for line in ret: print "Fixed:",line['name']," = ",line['Lateral limits'],line['Vertical limits'] out=[] for d in ret: pa=dict() curname=d['name'] if curname.count(u'Förteckning över'): continue print "D:",d arealines=[l for l in d['Lateral limits'] if l.strip()!=""] last_coord_idx=None #uprint("D:<%s> (area:%s)"%(d,arealines)) if 'FREQ' in d: freqs=[("SWEDEN CONTROL",float(x)) for x in re.findall(r"\d{3}\.\d{3}","\n".join(d['FREQ']))] #print "Parsed freqs:",freqs if freqs: last_sector['freqs']=freqs if kind=='sector': m=re.match(r"ES[A-Z]{2}\s*ACC\s*sector\s*([0-9a-zA-Z]*)",d['name']) if m: last_sector['major']=d['name'] last_sector['majorsector'],=m.groups() if len(arealines)==0: last_sector['name']=d['name'] continue if d['name'].count("Control Area and Upper Control Area"): continue if d['name'].count("SUECIA CTA"): continue if d['name'].count("SUECIA UTA"): continue m=re.match(r"([0-9a-zA-Z]*)(:.*)",d['name']) if m and 'majorsector' in last_sector: sectorname,sub=m.groups() if sectorname==last_sector['majorsector']: d['name']=last_sector['major']+sub #uprint("Fixed up name: ",d['name']) #print "Arealines:",arealines assert len(arealines) if arealines[0].strip()=="Danger area EK D395 and D396 are": arealines=arealines[1:] if arealines[0].strip()=="situated within TMA": arealines=arealines[1:] if arealines==u'Förteckning över CTA / Lists of CTA' or arealines=='Lateral limits': continue for idx in xrange(len(arealines)): if arealines[idx].lower().startswith("established"): last_coord_idx=idx pa['established']=" ".join(l for l in arealines[idx:]) break if arealines[idx].lower().startswith("danger area"): last_coord_idx=idx break if arealines[idx].strip()=="LFV": last_coord_idx=idx break if last_coord_idx==None: last_coord_idx=len(arealines) #uprint("ARealines:",arealines) #uprint("Last coord:",arealines[last_coord_idx-1]) if len(arealines)>last_coord_idx: if arealines[last_coord_idx-1:last_coord_idx+1]==[u'571324N 0161129E -', u'Established during operational hours of']: arealines[last_coord_idx-1]=arealines[last_coord_idx-1].strip("-") #uprint("Last fixed:",arealines[last_coord_idx-1]) assert not arealines[last_coord_idx-1].strip().endswith("-") #for idx in xrange(last_coord_idx-1): # print "arealine: <%s>"%(arealines[idx].strip(),) # assert arealines[idx].strip().endswith("-") or arealines[idx].strip().endswith("to") vertlim=u" ".join(d['Vertical limits']) if vertlim.strip()=="": #print "Object with no vertical limits: %s"%(repr(d['name']),) continue if d['name']=='Control Area': continue uprint("Vertlim: ",vertlim) heightst=re.findall(r"(FL\s*\d{3})|(\d+\s*ft\s*(?:\s*/\s*\d+\s*.\s*GND)?(?:\s*GND)?)|(GND)|(UNL)",vertlim) uprint("Height candidates:",heightst) heights=[] for fl,ht,gnd,unl in heightst: if fl: heights.append(fl) if ht: heights.append(ht.strip()) if gnd: heights.append(gnd.strip()) if unl: heights.append(unl.strip()) uprint("heights for ",d['name'],":",repr(heights)) if len(heights)==0 and d['name']==u'GÖTEBORG TMA': heights=['GND','FL95'] if len(heights)==1 and d['name']==u'Göteborg TMA': heights=['4500','FL95'] assert len(heights)==2 ceiling=heights[0].strip() floor=heights[1].strip() pa['name']=d['name'] pa['floor']=floor pa['ceiling']=ceiling if mapper.parse_elev(floor)>=9500: continue #uprint("Arealines:\n================\n%s\n============\n"%(arealines[:last_coord_idx])) #print pa areacoords=" ".join(arealines[:last_coord_idx]) pa['points']=parse_coord_str(areacoords) vs=[] for p in pa['points']: #print "from_str:",repr(p) x,y=mapper.latlon2merc(mapper.from_str(p),13) vs.append(Vertex(int(x),int(y))) p=Polygon(vvector(vs)) if p.calc_area()<=30*30: pass#print pa #print "Area:",p.calc_area() assert p.calc_area()>30*30 #print "Area: %f"%(p.calc_area(),) #print "Point-counts:",len(pa['points']) for p in pa['points']: assert p.count(",")==1 pa['type']=kind for thirdcol in thirdcols: if thirdcol in d: atc=d[thirdcol] break else: raise Exception("missing thirdcol") #print "ATc: <%s>"%(repr(atc),) freqs=[(y,float(x)) for x,y in re.findall(r"(\d{3}\.\d{3})\s*MHz\n(.*)","\n".join(atc))] if not freqs: freqs=last_sector.get('freqs',[]) #print repr(freqs) pa['freqs']=freqs #uprint("Cleaning up ",pa['name']) for cleaned in clean_up_polygon(list(pa['points'])): d=dict(pa) #print "cleaned",cleaned for i,tup in enumerate(cleaned): assert type(tup)==str latlon=mapper.from_str(tup) lat,lon=latlon assert lat>=-85 and lat<=85 d['points']=cleaned #uprint("cleaned:",pa['name'],len(cleaned),cleaned) #print "name:",d['name'] #print "cleaned points:",d['points'] #print "from:",areacoords #raise Exception() out.append(d) #if pa['name'].lower().count("esrange"): # print "Exit esrange" # sys.exit(1) return out
def parse_page(parser,pagenr): page=parser.parse_page_to_items(pagenr) items=page.items minx=min([item.x1 for item in items]) headings=[] majorre=ur"\s*([A-ZÅÄÖ ][A-ZÅÄÖ]{3,})\s+(?:TMA\s*\d*|MIL CTA)\s*(?:-.*)?$" minorre=ur"\s*(?:TMA|MIL CTA [SN]?)\s*[A-ZÅÄÖ ]*\s*" for item in page.get_by_regex(majorre): m,=re.match(majorre,item.text).groups() assert m!=None assert m.strip()!="" headings.append(('major',item.text.strip(),m,item)) for item in page.get_by_regex(minorre): m=re.match(minorre,item.text).group() assert m!=None assert m.strip()!="" #print "Heading %d: %s"%(item.y1,m) headings.append(('minor',item.text.strip(),m,item)) #print headings headings.sort(key=lambda x:x[3].y1) def findheadingfor(y,meta=None): minor=None major=None for (kind,full,name,item) in reversed(headings): if minor==None and kind=="minor" and item.y1<y: minor=name.strip() if meta!=None: meta['minor_y']=item.y1 if major==None and kind=="major" and item.y1<y: major=name.strip() fullname=full if meta!=None: meta['major_y']=item.y1 break assert major!=None and major.strip()!="" if minor!=None: return major+" "+minor return fullname cury=0 coordstrs=page.get_by_regex(ur".*\d{6}N \d{7}E.*") out=[] while True: found=False #print "Looking for coords, y= %d"%(cury,) for titem in coordstrs: #print "Considering coordstr: ",titem.y1 if titem.y1<=cury: continue if titem.x1<40: item=titem found=True break if not found: break cury=item.y1 headmeta=dict() name=findheadingfor(item.y1,headmeta) areaspec=[] #print "Rect: ",0,cury,minx+35,100 y1=cury lines=page.get_lines(page.get_partially_in_rect(0,cury,minx+25,100)) for idx,line in enumerate(lines): if re.search(ur"FL \d+",line) or line.count("FT MSL"): vertidx=idx break #print "Line:",line.encode('utf8') if line.strip()=="": vertidx=idx break cury=max(cury,line.y2+0.5) line=line.replace(u"–","-") if not (line.endswith("-") or line.endswith(" ")): line+=" " areaspec.append(line) verts=[] for idx in xrange(vertidx,len(lines)): #print "Looking for alt:",lines[idx],"y2:",lines[idx].y2 m=re.search(ur"(FL\s+\d+)",lines[idx].strip()) if m: verts.append((m.groups()[0],lines[idx].y1)) m=re.search(ur"(\d+ FT (?:MSL|GND|SFC))",lines[idx].strip()) if m: verts.append((m.groups()[0],lines[idx].y1)) if len(verts)>=2: break y2=verts[-1][1] freqs=[] for attempt in xrange(2): for freqcand in page.get_by_regex(ur".*\d{3}\.\d{1,3}.*"): #print "headmeta:",headmeta #print "attempt:",attempt #print "freqy1:",freqcand.y1 if freqcand.x1<30: continue if attempt==0: if freqcand.y1<y1: continue else: if 'major_y' in headmeta: if freqcand.y1<headmeta['major_y']: continue else: if freqcand.y1<y1: continue if freqcand.y1>y2: continue x,y=freqcand.x1,freqcand.y1 freq,=re.match(ur".*(\d{3}\.\d{3}).*",freqcand.text).groups() if freq=="121.500": continue lines=page.get_lines(page.get_partially_in_rect(x-10,y-1,x-0.5,y+1.5)) fname=None for line in reversed(lines): g=re.match(ur".*\b(\w{3,}\s+(?:Approach|Tower)).*",line) if g: #print "freqname Matched:",line fname,=g.groups() fname=fname.strip() break if not fname: raise Exception("Found no frequency name for freq: "+freq) freqs.append((fname,float(freq))) if len(freqs): break (ceiling,ceilingy),(floor,floory)=verts assert ceilingy<floory assert floory-ceilingy<5.0 uprint("Analyzing area for %s"%(name,)) assert "".join(areaspec).strip()!="" area=mapper.parse_coord_str("".join(areaspec),context='estonia') uprint("Done analyzing %s"%(name,)) #print area if name.count("CTA") and name.count("TMA")==0: type_="CTA" else: type_="TMA" if re.match(ur"\s*TALLINN\s*TMA\s*1\s*",name): out.append(dict( name="TALLIN TMA 2", floor='1700 ft MSL', ceiling='3500 ft MSL', freqs=freqs, type='TMA', points=mapper.parse_coord_str(""" A circle with radius 20 NM centred on 592448N 0244957E """)))
if onlyicao!=None and icao!=onlyicao: continue ad,adspace,sigpoints=fi_parse_airfield(icao) ads.append(ad) points.append(sigpoints) adspaces.extend(adspace) return ads,adspaces,points if __name__=='__main__': if len(sys.argv)==2: icaolimit=sys.argv[1] else: icaolimit=None ads,spaces,points=fi_parse_airfields(icaolimit) uprint("Spaces:") for p in points: uprint("Point: %s"%(p,)) for sp in spaces: uprint( "Name:",sp['name']) uprint( " Points:",sp['points']) uprint( " Floor:",sp['floor']) uprint( " Ceiling:",sp['ceiling']) uprint( " Freqs:",sp['freqs']) for ad in ads: uprint( "Name: ",ad['name']) uprint( " Other:",ad)
def extract_airfields(filtericao=lambda x:True,purge=True): #print getxml("/AIP/AD/AD 1/ES_AD_1_1_en.pdf") ads=[] p=Parser("/AIP/AD/AD 1/ES_AD_1_1_en.pdf") points=dict() startpage=None for pagenr in xrange(p.get_num_pages()): page=p.parse_page_to_items(pagenr) if page.count("Aerodrome directory"): startpage=pagenr break if startpage==None: raise Exception("Couldn't find aerodrome directory in file") #print "Startpage: %d"%(startpage,) #nochartf=open("nochart.txt","w") for pagenr in xrange(startpage,p.get_num_pages()): row_y=[] page=p.parse_page_to_items(pagenr) allines=[x for x in (page.get_lines(page.get_partially_in_rect(0,0,15,100))) if x.strip()] for item,next in zip(allines,allines[1:]+[""]): #print "item:",item m=re.match(ur"^\s*[A-ZÅÄÖ]{3,}(?:/.*)?\b.*",item) if m: #print "Candidate, next is:",next if re.match(r"^\s*[A-Z]{4}\b.*",next): #print "Matched:",item #print "y1:",item.y1 row_y.append(item.y1) for y1,y2 in zip(row_y,row_y[1:]+[100.0]): #print "Extacting from y-range: %f-%f"%(y1,y2) items=list(page.get_partially_in_rect(0,y1-0.25,5.0,y2+0.25,ysort=True)) if len(items)>=2: #print "Extract items",items ad=dict(name=unicode(items[0].text).strip(), icao=unicode(items[1].text).strip() ) #print "Icao:",ad['icao'] assert re.match(r"[A-Z]{4}",ad['icao']) if not filtericao(ad): continue if len(items)>=3: #print "Coord?:",items[2].text m=re.match(r".*(\d{6}N)\s*(\d{7}E).*",items[2].text) if m: lat,lon=m.groups() ad['pos']=parse_coords(lat,lon) #print "Items3:",items[3:] elev=re.findall(r"(\d{1,5})\s*ft"," ".join(t.text for t in items[3:])) #print "Elev:",elev assert len(elev)==1 ad['elev']=int(elev[0]) ads.append(ad) big_ad=set() for ad in ads: if not ad.has_key('pos'): big_ad.add(ad['icao']) for ad in ads: icao=ad['icao'] if icao in big_ad: if icao in ['ESIB','ESNY','ESCM','ESPE']: continue try: p=Parser("/AIP/AD/AD 2/%s/ES_AD_2_%s_6_1_en.pdf"%(icao,icao)) except: p=Parser("/AIP/AD/AD 2/%s/ES_AD_2_%s_6-1_en.pdf"%(icao,icao)) ad['aipvacurl']=p.get_url() for pagenr in xrange(p.get_num_pages()): page=p.parse_page_to_items(pagenr) """ for altline in exitlines: m=re.match(r"(\w+)\s+(\d+N)\s*(\d+E.*)",altline) if not m: continue name,lat,lon=m.groups() try: coord=parse_coords(lat,lon) except Exception: continue points.append(dict(name=name,pos=coord)) """ for kind in xrange(2): if kind==0: hits=page.get_by_regex(r"H[Oo][Ll][Dd][Ii][Nn][Gg]") kind="holding point" if kind==1: hits=page.get_by_regex(r"[Ee]ntry.*[Ee]xit.*point") kind="entry/exit point" if len(hits)==0: continue for holdingheading in hits: items=sorted(page.get_partially_in_rect(holdingheading.x1+2.0,holdingheading.y2+0.1,holdingheading.x1+0.5,100), key=lambda x:x.y1) items=[x for x in items if not x.text.startswith(" ")] #print "Holding items:",items for idx,item in enumerate(items): print "Holding item",item y1=item.y1 if idx==len(items)-1: y2=100 else: y2=items[idx+1].y1 items2=[x for x in page.get_partially_in_rect(item.x1+1,y1+0.3,item.x1+40,y2-0.1) if x.x1>=item.x1-0.25 and x.y1>=y1-0.05 and x.y1<y2-0.05] s=(" ".join(page.get_lines(items2))).strip() print "Holding lines:",repr(page.get_lines(items2)) #if s.startswith("ft Left/3"): #Special case for ESOK # s,=re.match("ft Left/3.*?([A-Z]{4,}.*)",s).groups() #m=re.match("ft Left/\d+.*?([A-Z]{4,}.*)",s) #if m: # s,=m.groups() if s.startswith("LjUNG"): #Really strange problem with ESCF s=s[0]+"J"+s[2:] if s.lower().startswith("holding"): sl=s.split(" ",1) if len(sl)>1: s=sl[1] s=s.strip() if kind=="entry/exit point" and s.startswith("HOLDING"): continue #reached HOLDING-part of VAC #Check for other headings #Fixup strange formatting of points in some holding items: (whitespace between coord and 'E') s=re.sub(ur"(\d+)\s*(N)\s*(\d+)\s*(E)",lambda x:"".join(x.groups()),s) m=re.match(r"([A-Z]{2,}).*?(\d+N)\s*(\d+E).*",s) if not m: m=re.match(r".*?(\d+N)\s*(\d+E).*",s) if not m: continue assert m lat,lon=m.groups() #skavsta if icao=="ESKN": if s.startswith(u"Hold north of T"): name="NORTH" elif s.startswith(u"Hold south of B"): name="SOUTH" else: assert 0 #add more specials here else: continue else: name,lat,lon=m.groups() try: coord=parse_coords(lat,lon) except Exception: print "Couldn't parse:",lat,lon continue #print name,lat,lon,mapper.format_lfv(*mapper.from_str(coord)) if name.count("REMARK") or len(name)<=2: print "Suspicious name: ",name #sys.exit(1) continue points[icao+' '+name]=dict(name=icao+' '+name,icao=icao,pos=coord,kind=kind) #for point in points.items(): # print point #sys.exit(1) def fixhex11(s): out=[] for c in s: i=ord(c) if i>=0x20: out.append(c) continue if i in [0x9,0xa,0xd]: out.append(c) continue out.append(' ') return "".join(out) for ad in ads: icao=ad['icao'] if icao in big_ad: #print "Parsing ",icao p=Parser("/AIP/AD/AD 2/%s/ES_AD_2_%s_en.pdf"%(icao,icao),loadhook=fixhex11) ad['aiptexturl']=p.get_url() firstpage=p.parse_page_to_items(0) te="\n".join(firstpage.get_all_lines()) #print te coords=re.findall(r"ARP.*(\d{6}N)\s*(\d{7}E)",te) if len(coords)>1: raise Exception("First page of airport info (%s) does not contain exactly ONE set of coordinates"%(icao,)) if len(coords)==0: print "Couldn't find coords for ",icao #print "Coords:",coords ad['pos']=parse_coords(*coords[0]) elev=re.findall(r"Elevation.*?(\d{1,5})\s*ft",te,re.DOTALL) if len(elev)>1: raise Exception("First page of airport info (%s) does not contain exactly ONE elevation in ft"%(icao,)) if len(elev)==0: print "Couldn't find elev for ",icao ad['elev']=int(elev[0]) freqs=[] found=False thrs=[] #uprint("-------------------------------------") for pagenr in xrange(p.get_num_pages()): page=p.parse_page_to_items(pagenr) #uprint("Looking on page %d"%(pagenr,)) if 0: #opening hours are no longer stored in a separate document for any airports. No need to detect which any more (since none are). for item in page.get_by_regex(".*OPERATIONAL HOURS.*"): lines=page.get_lines(page.get_partially_in_rect(0,item.y2+0.1,100,100)) for line in lines: things=["ATS","Fuelling","Operating"] if not line.count("AIP SUP"): continue for thing in things: if line.count(thing): ad['aipsup']=True for item in page.get_by_regex(".*\s*RUNWAY\s*PHYSICAL\s*CHARACTERISTICS\s*.*"): #uprint("Physical char on page") lines=page.get_lines(page.get_partially_in_rect(0,item.y2+0.1,100,100)) seen_end_rwy_text=False for line,nextline in izip(lines,lines[1:]+[None]): #uprint("MAtching: <%s>"%(line,)) if re.match(ur"AD\s+2.13",line): break if line.count("Slope of"): break if line.lower().count("end rwy:"): seen_end_rwy_text=True if line.lower().count("bgn rwy:"): seen_end_rwy_text=True m=re.match(ur".*(\d{6}\.\d+)[\s\(\)\*]*(N).*",line) if not m:continue m2=re.match(ur".*(\d{6,7}\.\d+)\s*[\s\(\)\*]*(E).*",nextline) if not m2:continue latd,n=m.groups() lond,e=m2.groups() assert n=="N" assert e=="E" lat=latd+n lon=lond+e rwytxts=page.get_lines(page.get_partially_in_rect(0,line.y1+0.05,12,nextline.y2-0.05)) uprint("Rwytxts:",rwytxts) rwy=None for rwytxt in rwytxts: #uprint("lat,lon:%s,%s"%(lat,lon)) #uprint("rwytext:",rwytxt) m=re.match(ur"\s*(\d{2}[LRCM]?)\b.*",rwytxt) if m: assert rwy==None rwy=m.groups()[0] if rwy==None and seen_end_rwy_text: continue print "Cur airport:",icao already=False assert rwy!=None seen_end_rwy_text=False for thr in thrs: if thr['thr']==rwy: raise Exception("Same runway twice on airfield:"+icao) thrs.append(dict(pos=mapper.parse_coords(lat,lon),thr=rwy)) assert len(thrs)>=2 for pagenr in xrange(0,p.get_num_pages()): page=p.parse_page_to_items(pagenr) matches=page.get_by_regex(r".*ATS\s+COMMUNICATION\s+FACILITIES.*") #print "Matches of ATS COMMUNICATION FACILITIES on page %d: %s"%(pagenr,matches) if len(matches)>0: commitem=matches[0] curname=None callsign=page.get_by_regex_in_rect(ur"Call\s*sign",0,commitem.y1,100,commitem.y2+8)[0] for idx,item in enumerate(page.get_lines(page.get_partially_in_rect(callsign.x1-0.5,commitem.y1,100,100),fudge=0.3,order_fudge=15)): if item.strip()=="": curname=None if re.match(".*RADIO\s+NAVIGATION\s+AND\s+LANDING\s+AIDS.*",item): break #print "Matching:",item m=re.match(r"(.*?)\s*(\d{3}\.\d{1,3})\s*MHz.*",item) #print "MHZ-match:",m if not m: continue #print "MHZ-match:",m.groups() who,sfreq=m.groups() freq=float(sfreq) if abs(freq-121.5)<1e-4: if who.strip(): curname=who continue #Ignore emergency frequency, it is understood if not who.strip(): if curname==None: continue else: curname=who freqs.append((curname.strip().rstrip("/"),freq)) for pagenr in xrange(0,p.get_num_pages()): page=p.parse_page_to_items(pagenr) matches=page.get_by_regex(r".*ATS\s*AIRSPACE.*") #print "Matches of ATS_AIRSPACE on page %d: %s"%(pagenr,matches) if len(matches)>0: heading=matches[0] desigitem,=page.get_by_regex("Designation and lateral limits") vertitem,=page.get_by_regex("Vertical limits") airspaceclass,=page.get_by_regex("Airspace classification") lastname=None subspacelines=dict() subspacealts=dict() for idx,item in enumerate(page.get_lines(page.get_partially_in_rect(desigitem.x2+1,desigitem.y1,100,vertitem.y1-1))): if item.count("ATS airspace not established"): assert idx==0 break if item.strip()=="": continue m=re.match(r"(.*?)(\d{6}N\s+.*)",item) if m: name,coords=m.groups() name=name.strip() else: name=item.strip() coords=None if name: lastname=name if coords: subspacelines.setdefault(lastname,[]).append(coords) assert lastname lastname=None #print "Spaces:",subspacelines #print "ICAO",ad['icao'] #altlines=page.get_lines(page.get_partially_in_rect(vertitem.x2+1,vertitem.y1,100,airspaceclass.y1-0.2)) #print "Altlines:",altlines subspacealts=dict() subspacekeys=subspacelines.keys() allaltlines=" ".join(page.get_lines(page.get_partially_in_rect(vertitem.x1+0.5,vertitem.y1+0.5,100,airspaceclass.y1-0.2))) single_vertlim=False totalts=list(mapper.parse_all_alts(allaltlines)) #print "totalts:",totalts if len(totalts)==2: single_vertlim=True for subspacename in subspacekeys: ceil=None floor=None subnames=[subspacename] if subspacename.split(" ")[-1].strip() in ["TIA","TIZ","CTR","CTR/TIZ"]: subnames.append(subspacename.split(" ")[-1].strip()) #print "Parsing alts for ",subspacename,subnames try: for nametry in subnames: if single_vertlim: #there's only one subspace, parse all of vertical limits field for this single one. items=[vertitem] else: items=page.get_by_regex_in_rect(nametry,vertitem.x2+1,vertitem.y1,100,airspaceclass.y1-0.2) for item in items: alts=[] for line in page.get_lines(page.get_partially_in_rect(item.x1+0.5,item.y1+0.5,100,airspaceclass.y1-0.2)): #print "Parsing:",line line=line.replace(nametry,"").lower().strip() parsed=list(mapper.parse_all_alts(line)) if len(parsed): alts.append(mapper.altformat(*parsed[0])) if len(alts)==2: break if alts: #print "alts:",alts ceil,floor=alts raise StopIteration except StopIteration: pass assert ceil and floor subspacealts[subspacename]=dict(ceil=ceil,floor=floor) spaces=[] for spacename in subspacelines.keys(): altspacename=spacename #print "Altspacename: %s, subspacesalts: %s"%(altspacename,subspacealts) space=dict( name=spacename, ceil=subspacealts[altspacename]['ceil'], floor=subspacealts[altspacename]['floor'], points=parse_coord_str(" ".join(subspacelines[spacename])), freqs=list(set(freqs)) ) if True: vs=[] for p in space['points']: x,y=mapper.latlon2merc(mapper.from_str(p),13) vs.append(Vertex(int(x),int(y))) p=Polygon(vvector(vs)) if p.calc_area()<=30*30: pass#print space pass#print "Area:",p.calc_area() assert p.calc_area()>30*30 #print "Area: %f"%(p.calc_area(),) spaces.append(space) #print space ad['spaces']=spaces found=True if found: break assert found ad['runways']=rwy_constructor.get_rwys(thrs) #Now find any ATS-airspace chartblobnames=[] for ad in ads: icao=ad['icao'] if icao in big_ad: parse_landing_chart.help_plc(ad,"/AIP/AD/AD 2/%s/ES_AD_2_%s_2-1_en.pdf"%(icao,icao), icao,ad['pos'],"se",variant="") parse_landing_chart.help_plc(ad,"/AIP/AD/AD 2/%s/ES_AD_2_%s_6-1_en.pdf"%(icao,icao), icao,ad['pos'],"se",variant="vac") parse_landing_chart.help_plc(ad,"/AIP/AD/AD 2/%s/ES_AD_2_%s_2-3_en.pdf"%(icao,icao), icao,ad['pos'],"se",variant="parking") #aip_text_documents.help_parse_doc(ad,"/AIP/AD/AD 2/%s/ES_AD_2_%s_6_1_en.pdf"%(icao,icao), # icao,"se",title="General Information",category="general") aip_text_documents.help_parse_doc(ad,"/AIP/AD/AD 2/%s/ES_AD_2_%s_en.pdf"%(icao,icao), icao,"se",title="General Information",category="general") #if purge: # parse_landing_chart.purge_old(chartblobnames,country="se") #sys.exit(1) for extra in extra_airfields.extra_airfields: if filtericao(extra): ads.append(extra) print print for k,v in sorted(points.items()): print k,v,mapper.format_lfv(*mapper.from_str(v['pos'])) #print "Num points:",len(points) origads=list(ads) for flygkartan_id,name,lat,lon,dummy in csv.reader(open("fplan/extract/flygkartan.csv"),delimiter=";"): found=None lat=float(lat) lon=float(lon) if type(name)==str: name=unicode(name,'utf8') mercf=mapper.latlon2merc((lat,lon),13) for a in origads: merca=mapper.latlon2merc(mapper.from_str(a['pos']),13) dist=math.sqrt((merca[0]-mercf[0])**2+(merca[1]-mercf[1])**2) if dist<120: found=a break if found: found['flygkartan_id']=flygkartan_id else: d=dict( icao='ZZZZ', name=name, pos=mapper.to_str((lat,lon)), elev=int(get_terrain_elev((lat,lon))), flygkartan_id=flygkartan_id) if filtericao(d): ads.append(d) minor_ad_charts=extra_airfields.minor_ad_charts for ad in ads: if ad['name'].count(u"Långtora"): ad['pos']=mapper.to_str(mapper.from_aviation_format("5944.83N01708.20E")) if ad['name'] in minor_ad_charts: charturl=minor_ad_charts[ad['name']] arp=ad['pos'] if 'icao' in ad and ad['icao'].upper()!='ZZZZ': icao=ad['icao'].upper() else: icao=ad['fake_icao'] parse_landing_chart.help_plc(ad,charturl,icao,arp,country='raw',variant="landing") """ assert icao!=None lc=parse_landing_chart.parse_landing_chart( charturl, icao=icao, arppos=arp,country="raw") assert lc if lc: ad['adcharturl']=lc['url'] ad['adchart']=lc """ #print ads for ad in ads: print "%s: %s - %s (%s ft) (%s)"%(ad['icao'],ad['name'],ad['pos'],ad['elev'],ad.get('flygkartan_id','inte i flygkartan')) for space in ad.get('spaces',[]): for freq in space.get('freqs',[]): print " ",freq #if 'spaces' in ad: # print " spaces: %s"%(ad['spaces'],) #if 'aiptext' in ad: # print "Aip texts:",ad['aiptext'] #else: # print "No aiptext" print "Points:" for point in sorted(points.values(),key=lambda x:x['name']): print point f=codecs.open("extract_airfields.regress.txt","w",'utf8') for ad in ads: r=repr(ad) d=md5.md5(r).hexdigest() f.write("%s - %s - %s\n"%(ad['icao'],ad['name'],d)) f.close() f=codecs.open("extract_airfields.regress-details.txt","w",'utf8') for ad in ads: r=repr(ad) f.write(u"%s - %s - %s\n"%(ad['icao'],ad['name'],r)) f.close() return ads,points.values()
for item in page.get_by_regex(ur"FREQ MHZ"): freqline=page.get_lines(page.get_partially_in_rect(item.x1,item.y2+0.05,item.x2+20,item.y2+5))[0] print "Freqline:",freqline freq,=re.match(ur"(\d{3}\.\d{3}).*",freqline).groups() ad['freq']=float(freq) for item in page.get_by_regex(ur"ELEV FT \(M\)"): elevline=page.get_lines(page.get_partially_in_rect(item.x1,item.y2+0.05,item.x2+20,item.y2+5))[0] print "elevline:",elevline elev,=re.match(ur"(\d+)\s*\(\d+\)",elevline).groups() ad['elev']=int(elev) for icao,ad in ads.items(): assert ad['icao']==icao assert 'pos' in ad return ads.values() if __name__=='__main__': if len(sys.argv)==2: icaolimit=sys.argv[1] else: icaolimit=None ads=fi_parse_small_airfields(icaolimit) for ad in ads: uprint( "Name: ",ad['name']) uprint( " Other:",ad)
adspaces = [] points = [] for icao in icaolist: if onlyicao != None and icao != onlyicao: continue ad, adspace, sigpoints = ee_parse_airfield(icao) ads.append(ad) points.append(sigpoints) adspaces.extend(adspace) return ads, adspaces, points if __name__ == "__main__": if len(sys.argv) == 2: icaolimit = sys.argv[1] else: icaolimit = None ads, spaces, points = ee_parse_airfields(icaolimit) uprint("Spaces:") for p in points: uprint("Point: %s" % (p,)) for sp in spaces: uprint("Name:", sp["name"]) uprint(" Points:", sp["points"]) uprint(" Floor:", sp["floor"]) uprint(" Ceiling:", sp["ceiling"]) uprint(" Freqs:", sp["freqs"]) for ad in ads: uprint("Name: ", ad["name"]) uprint(" Other:", ad)
type="RNAV", points=[mapper.to_str(mapper.merc2latlon(x,13)) for x in outline]) else: uprint("Need to have predefine for route %s, with md5: %s"%(routename,sig)) uprint("Altitude, and points") raise Exception('Missing predef for route. Use: ("%s","%s") : ( 10,"FL 65","FL 95","""\n\n """),'%(routename,sig)) i1=isol_routes() low1=low_routes(i1) out.extend(list(get_airspaces(low1))) out2=[] for space in out: mypoly=makepoly.poly[space['points']] for tmapoly in tmapolys: mypoly=shapemerge2d.shape_difference(mypoly,tmapoly) t=[] for mx,my in [(v.get_x(),v.get_y()) for v in mypoly.get_vertices()]: t.append(mapper.to_str(mapper.merc2latlon((mx,my),13))) space['points']=t return out2 if __name__=='__main__': import fi_parse_tma for space in fi_parse_ats_rte([makepoly.poly(x['points']) for x in fi_parse_tma.fi_parse_tma()]): uprint(space)
def extract_airfields(filtericao=lambda x: True, purge=True): # print getxml("/AIP/AD/AD 1/ES_AD_1_1_en.pdf") ads = [] p = Parser("/AIP/AD/AD 1/ES_AD_1_1_en.pdf") points = dict() startpage = None for pagenr in xrange(p.get_num_pages()): page = p.parse_page_to_items(pagenr) if page.count("Aerodrome directory"): startpage = pagenr break if startpage == None: raise Exception("Couldn't find aerodrome directory in file") # print "Startpage: %d"%(startpage,) # nochartf=open("nochart.txt","w") for pagenr in xrange(startpage, p.get_num_pages()): row_y = [] page = p.parse_page_to_items(pagenr) allines = [x for x in (page.get_lines(page.get_partially_in_rect(0, 0, 15, 100))) if x.strip()] for item, next in zip(allines, allines[1:] + [""]): # print "item:",item m = re.match(ur"^\s*[A-ZÅÄÖ]{3,}(?:/.*)?\b.*", item) if m: # print "Candidate, next is:",next if re.match(r"^\s*[A-Z]{4}\b.*", next): # print "Matched:",item # print "y1:",item.y1 row_y.append(item.y1) for y1, y2 in zip(row_y, row_y[1:] + [100.0]): # print "Extacting from y-range: %f-%f"%(y1,y2) items = list(page.get_partially_in_rect(0, y1 - 0.25, 5.0, y2 + 0.25, ysort=True)) if len(items) >= 2: # print "Extract items",items ad = dict(name=unicode(items[0].text).strip(), icao=unicode(items[1].text).strip()) # print "Icao:",ad['icao'] assert re.match(r"[A-Z]{4}", ad["icao"]) if not filtericao(ad): continue if len(items) >= 3: # print "Coord?:",items[2].text m = re.match(r".*(\d{6}N)\s*(\d{7}E).*", items[2].text) if m: lat, lon = m.groups() ad["pos"] = parse_coords(lat, lon) # print "Items3:",items[3:] elev = re.findall(r"(\d{1,5})\s*ft", " ".join(t.text for t in items[3:])) # print "Elev:",elev assert len(elev) == 1 ad["elev"] = int(elev[0]) ads.append(ad) big_ad = set() for ad in ads: if not ad.has_key("pos"): big_ad.add(ad["icao"]) for ad in ads: icao = ad["icao"] if icao in big_ad: if icao in ["ESIB", "ESNY", "ESCM", "ESPE"]: continue try: p = Parser("/AIP/AD/AD 2/%s/ES_AD_2_%s_6_1_en.pdf" % (icao, icao)) except: p = Parser("/AIP/AD/AD 2/%s/ES_AD_2_%s_6-1_en.pdf" % (icao, icao)) ad["aipvacurl"] = p.get_url() for pagenr in xrange(p.get_num_pages()): page = p.parse_page_to_items(pagenr) """ for altline in exitlines: m=re.match(r"(\w+)\s+(\d+N)\s*(\d+E.*)",altline) if not m: continue name,lat,lon=m.groups() try: coord=parse_coords(lat,lon) except Exception: continue points.append(dict(name=name,pos=coord)) """ for kind in xrange(2): if kind == 0: hits = page.get_by_regex(r"H[Oo][Ll][Dd][Ii][Nn][Gg]") kind = "holding point" if kind == 1: hits = page.get_by_regex(r"[Ee]ntry.*[Ee]xit.*point") kind = "entry/exit point" if len(hits) == 0: continue for holdingheading in hits: items = sorted( page.get_partially_in_rect( holdingheading.x1 + 2.0, holdingheading.y2 + 0.1, holdingheading.x1 + 0.5, 100 ), key=lambda x: x.y1, ) items = [x for x in items if not x.text.startswith(" ")] # print "Holding items:",items for idx, item in enumerate(items): print "Holding item", item y1 = item.y1 if idx == len(items) - 1: y2 = 100 else: y2 = items[idx + 1].y1 items2 = [ x for x in page.get_partially_in_rect(item.x1 + 1, y1 + 0.3, item.x1 + 40, y2 - 0.1) if x.x1 >= item.x1 - 0.25 and x.y1 >= y1 - 0.05 and x.y1 < y2 - 0.05 ] s = (" ".join(page.get_lines(items2))).strip() print "Holding lines:", repr(page.get_lines(items2)) # if s.startswith("ft Left/3"): #Special case for ESOK # s,=re.match("ft Left/3.*?([A-Z]{4,}.*)",s).groups() # m=re.match("ft Left/\d+.*?([A-Z]{4,}.*)",s) # if m: # s,=m.groups() if s.startswith("LjUNG"): # Really strange problem with ESCF s = s[0] + "J" + s[2:] if s.lower().startswith("holding"): sl = s.split(" ", 1) if len(sl) > 1: s = sl[1] s = s.strip() if kind == "entry/exit point" and s.startswith("HOLDING"): continue # reached HOLDING-part of VAC # Check for other headings # Fixup strange formatting of points in some holding items: (whitespace between coord and 'E') s = re.sub(ur"(\d+)\s*(N)\s*(\d+)\s*(E)", lambda x: "".join(x.groups()), s) m = re.match(r"([A-Z]{2,}).*?(\d+N)\s*(\d+E).*", s) if not m: m = re.match(r".*?(\d+N)\s*(\d+E).*", s) if not m: continue assert m lat, lon = m.groups() # skavsta if icao == "ESKN": if s.startswith(u"Hold north of T"): name = "NORTH" elif s.startswith(u"Hold south of B"): name = "SOUTH" else: assert 0 # add more specials here else: continue else: name, lat, lon = m.groups() try: coord = parse_coords(lat, lon) except Exception: print "Couldn't parse:", lat, lon continue # print name,lat,lon,mapper.format_lfv(*mapper.from_str(coord)) if name.count("REMARK") or len(name) <= 2: print "Suspicious name: ", name # sys.exit(1) continue points[icao + " " + name] = dict(name=icao + " " + name, icao=icao, pos=coord, kind=kind) # for point in points.items(): # print point # sys.exit(1) def fixhex11(s): out = [] for c in s: i = ord(c) if i >= 0x20: out.append(c) continue if i in [0x9, 0xA, 0xD]: out.append(c) continue out.append(" ") return "".join(out) for ad in ads: icao = ad["icao"] if icao in big_ad: # print "Parsing ",icao p = Parser("/AIP/AD/AD 2/%s/ES_AD_2_%s_en.pdf" % (icao, icao), loadhook=fixhex11) ad["aiptexturl"] = p.get_url() firstpage = p.parse_page_to_items(0) te = "\n".join(firstpage.get_all_lines()) # print te coords = re.findall(r"ARP.*(\d{6}N)\s*(\d{7}E)", te) if len(coords) > 1: raise Exception( "First page of airport info (%s) does not contain exactly ONE set of coordinates" % (icao,) ) if len(coords) == 0: print "Couldn't find coords for ", icao # print "Coords:",coords ad["pos"] = parse_coords(*coords[0]) elev = re.findall(r"Elevation.*?(\d{1,5})\s*ft", te, re.DOTALL) if len(elev) > 1: raise Exception( "First page of airport info (%s) does not contain exactly ONE elevation in ft" % (icao,) ) if len(elev) == 0: print "Couldn't find elev for ", icao ad["elev"] = int(elev[0]) freqs = [] found = False thrs = [] # uprint("-------------------------------------") for pagenr in xrange(p.get_num_pages()): page = p.parse_page_to_items(pagenr) # uprint("Looking on page %d"%(pagenr,)) if ( 0 ): # opening hours are no longer stored in a separate document for any airports. No need to detect which any more (since none are). for item in page.get_by_regex(".*OPERATIONAL HOURS.*"): lines = page.get_lines(page.get_partially_in_rect(0, item.y2 + 0.1, 100, 100)) for line in lines: things = ["ATS", "Fuelling", "Operating"] if not line.count("AIP SUP"): continue for thing in things: if line.count(thing): ad["aipsup"] = True for item in page.get_by_regex(".*\s*RUNWAY\s*PHYSICAL\s*CHARACTERISTICS\s*.*"): # uprint("Physical char on page") lines = page.get_lines(page.get_partially_in_rect(0, item.y2 + 0.1, 100, 100)) seen_end_rwy_text = False for line, nextline in izip(lines, lines[1:] + [None]): # uprint("MAtching: <%s>"%(line,)) if re.match(ur"AD\s+2.13", line): break if line.count("Slope of"): break if line.lower().count("end rwy:"): seen_end_rwy_text = True if line.lower().count("bgn rwy:"): seen_end_rwy_text = True m = re.match(ur".*(\d{6}\.\d+)[\s\(\)\*]*(N).*", line) if not m: continue m2 = re.match(ur".*(\d{6,7}\.\d+)\s*[\s\(\)\*]*(E).*", nextline) if not m2: continue latd, n = m.groups() lond, e = m2.groups() assert n == "N" assert e == "E" lat = latd + n lon = lond + e rwytxts = page.get_lines(page.get_partially_in_rect(0, line.y1 + 0.05, 12, nextline.y2 - 0.05)) uprint("Rwytxts:", rwytxts) rwy = None for rwytxt in rwytxts: # uprint("lat,lon:%s,%s"%(lat,lon)) # uprint("rwytext:",rwytxt) m = re.match(ur"\s*(\d{2}[LRCM]?)\b.*", rwytxt) if m: assert rwy == None rwy = m.groups()[0] if rwy == None and seen_end_rwy_text: continue print "Cur airport:", icao already = False assert rwy != None seen_end_rwy_text = False for thr in thrs: if thr["thr"] == rwy: raise Exception("Same runway twice on airfield:" + icao) thrs.append(dict(pos=mapper.parse_coords(lat, lon), thr=rwy)) assert len(thrs) >= 2 for pagenr in xrange(0, p.get_num_pages()): page = p.parse_page_to_items(pagenr) matches = page.get_by_regex(r".*ATS\s+COMMUNICATION\s+FACILITIES.*") # print "Matches of ATS COMMUNICATION FACILITIES on page %d: %s"%(pagenr,matches) if len(matches) > 0: commitem = matches[0] curname = None callsign = page.get_by_regex_in_rect(ur"Call\s*sign", 0, commitem.y1, 100, commitem.y2 + 8)[0] for idx, item in enumerate( page.get_lines( page.get_partially_in_rect(callsign.x1 - 0.5, commitem.y1, 100, 100), fudge=0.3, order_fudge=15, ) ): if item.strip() == "": curname = None if re.match(".*RADIO\s+NAVIGATION\s+AND\s+LANDING\s+AIDS.*", item): break # print "Matching:",item m = re.match(r"(.*?)\s*(\d{3}\.\d{1,3})\s*MHz.*", item) # print "MHZ-match:",m if not m: continue # print "MHZ-match:",m.groups() who, sfreq = m.groups() freq = float(sfreq) if abs(freq - 121.5) < 1e-4: if who.strip(): curname = who continue # Ignore emergency frequency, it is understood if not who.strip(): if curname == None: continue else: curname = who freqs.append((curname.strip().rstrip("/"), freq)) for pagenr in xrange(0, p.get_num_pages()): page = p.parse_page_to_items(pagenr) matches = page.get_by_regex(r".*ATS\s*AIRSPACE.*") # print "Matches of ATS_AIRSPACE on page %d: %s"%(pagenr,matches) if len(matches) > 0: heading = matches[0] desigitem, = page.get_by_regex("Designation and lateral limits") vertitem, = page.get_by_regex("Vertical limits") airspaceclass, = page.get_by_regex("Airspace classification") lastname = None subspacelines = dict() subspacealts = dict() for idx, item in enumerate( page.get_lines(page.get_partially_in_rect(desigitem.x2 + 1, desigitem.y1, 100, vertitem.y1 - 1)) ): if item.count("ATS airspace not established"): assert idx == 0 break if item.strip() == "": continue m = re.match(r"(.*?)(\d{6}N\s+.*)", item) if m: name, coords = m.groups() name = name.strip() else: name = item.strip() coords = None if name: lastname = name if coords: subspacelines.setdefault(lastname, []).append(coords) assert lastname lastname = None # print "Spaces:",subspacelines # print "ICAO",ad['icao'] # altlines=page.get_lines(page.get_partially_in_rect(vertitem.x2+1,vertitem.y1,100,airspaceclass.y1-0.2)) # print "Altlines:",altlines subspacealts = dict() subspacekeys = subspacelines.keys() allaltlines = " ".join( page.get_lines( page.get_partially_in_rect( vertitem.x1 + 0.5, vertitem.y1 + 0.5, 100, airspaceclass.y1 - 0.2 ) ) ) single_vertlim = False totalts = list(mapper.parse_all_alts(allaltlines)) # print "totalts:",totalts if len(totalts) == 2: single_vertlim = True for subspacename in subspacekeys: ceil = None floor = None subnames = [subspacename] if subspacename.split(" ")[-1].strip() in ["TIA", "TIZ", "CTR", "CTR/TIZ"]: subnames.append(subspacename.split(" ")[-1].strip()) # print "Parsing alts for ",subspacename,subnames try: for nametry in subnames: if ( single_vertlim ): # there's only one subspace, parse all of vertical limits field for this single one. items = [vertitem] else: items = page.get_by_regex_in_rect( nametry, vertitem.x2 + 1, vertitem.y1, 100, airspaceclass.y1 - 0.2 ) for item in items: alts = [] for line in page.get_lines( page.get_partially_in_rect( item.x1 + 0.5, item.y1 + 0.5, 100, airspaceclass.y1 - 0.2 ) ): # print "Parsing:",line line = line.replace(nametry, "").lower().strip() parsed = list(mapper.parse_all_alts(line)) if len(parsed): alts.append(mapper.altformat(*parsed[0])) if len(alts) == 2: break if alts: # print "alts:",alts ceil, floor = alts raise StopIteration except StopIteration: pass assert ceil and floor subspacealts[subspacename] = dict(ceil=ceil, floor=floor) spaces = [] for spacename in subspacelines.keys(): altspacename = spacename # print "Altspacename: %s, subspacesalts: %s"%(altspacename,subspacealts) space = dict( name=spacename, ceil=subspacealts[altspacename]["ceil"], floor=subspacealts[altspacename]["floor"], points=parse_coord_str(" ".join(subspacelines[spacename])), freqs=list(set(freqs)), ) if True: vs = [] for p in space["points"]: x, y = mapper.latlon2merc(mapper.from_str(p), 13) vs.append(Vertex(int(x), int(y))) p = Polygon(vvector(vs)) if p.calc_area() <= 30 * 30: pass # print space pass # print "Area:",p.calc_area() assert p.calc_area() > 30 * 30 # print "Area: %f"%(p.calc_area(),) spaces.append(space) # print space ad["spaces"] = spaces found = True if found: break assert found ad["runways"] = rwy_constructor.get_rwys(thrs)
g = re.match(ur".*\b(\w{3,}\s+(?:Approach|Tower)).*", line) if g: #print "freqname Matched:",line fname, = g.groups() fname = fname.strip() break if not fname: raise Exception("Found no frequency name for freq: " + freq) freqs.append((fname, float(freq))) if len(freqs): break (ceiling, ceilingy), (floor, floory) = verts assert ceilingy < floory assert floory - ceilingy < 5.0 uprint("Analyzing area for %s" % (name, )) assert "".join(areaspec).strip() != "" area = mapper.parse_coord_str("".join(areaspec), context='estonia') uprint("Done analyzing %s" % (name, )) #print area if name.count("CTA") and name.count("TMA") == 0: type_ = "CTA" else: type_ = "TMA" if re.match(ur"\s*TALLINN\s*TMA\s*1\s*", name): out.append( dict(name="TALLIN TMA 2", floor='1700 ft MSL', ceiling='3500 ft MSL', freqs=freqs,
freqline = page.get_lines( page.get_partially_in_rect(item.x1, item.y2 + 0.05, item.x2 + 20, item.y2 + 5))[0] print "Freqline:", freqline freq, = re.match(ur"(\d{3}\.\d{3}).*", freqline).groups() ad['freq'] = float(freq) for item in page.get_by_regex(ur"ELEV FT \(M\)"): elevline = page.get_lines( page.get_partially_in_rect(item.x1, item.y2 + 0.05, item.x2 + 20, item.y2 + 5))[0] print "elevline:", elevline elev, = re.match(ur"(\d+)\s*\(\d+\)", elevline).groups() ad['elev'] = int(elev) for icao, ad in ads.items(): assert ad['icao'] == icao assert 'pos' in ad return ads.values() if __name__ == '__main__': if len(sys.argv) == 2: icaolimit = sys.argv[1] else: icaolimit = None ads = fi_parse_small_airfields(icaolimit) for ad in ads: uprint("Name: ", ad['name']) uprint(" Other:", ad)
uprint( "Need to have predefine for route %s, with md5: %s" % (routename, sig)) uprint("Altitude, and points") raise Exception( 'Missing predef for route. Use: ("%s","%s") : ( 10,"FL 65","FL 95","""\n\n """),' % (routename, sig)) i1 = isol_routes() low1 = low_routes(i1) out.extend(list(get_airspaces(low1))) out2 = [] for space in out: mypoly = makepoly.poly[space['points']] for tmapoly in tmapolys: mypoly = shapemerge2d.shape_difference(mypoly, tmapoly) t = [] for mx, my in [(v.get_x(), v.get_y()) for v in mypoly.get_vertices()]: t.append(mapper.to_str(mapper.merc2latlon((mx, my), 13))) space['points'] = t return out2 if __name__ == '__main__': import fi_parse_tma for space in fi_parse_ats_rte( [makepoly.poly(x['points']) for x in fi_parse_tma.fi_parse_tma()]): uprint(space)
def parse_page(parser, pagenr, kind="TMA", last_sector=dict()): if kind == "TMA": thirdcols = ["ATC unit", "AFIS unit"] elif kind == "sector": thirdcols = ["FREQ"] elif kind == "R": thirdcols = ["Remarks (nature of hazard,"] else: raise Exception("Bad kind") page = parser.parse_page_to_items(pagenr) items = page.items #print "Items:",pitems #print "Possible Areas:" headings = [] for item in items: if item.text == None: continue item.text = item.text.strip() if item.text == "": continue if item.text == "Name": continue if item.y1 < 25 and item.text in ( ["Lateral limits", "Vertical limits"] + thirdcols): headings.append(item) headings.sort(key=lambda x: x.x1) #print "found candidates:",zone_candidates if len(headings) == 0: return [] avg_heading_y = sum(h.y1 for h in headings) / float(len(headings)) uprint("Found headings:", headings) zone_candidates = [] for item in items: if item.text == None or item.text.strip() == "": continue if item.text.strip().startswith("AMDT"): continue if item.text.strip().startswith("The LFV Group"): continue if re.match(ur"\s*LFV\s*AIRAC\s*AMDT\s*\d+/\d+\s*", item.text): continue if item.text.strip() == "LFV": continue if item.text.count('Terminal Information Areas'): continue if item.text.strip().startswith("AIRAC"): continue if kind == "R" and not is_r_or_danger_area_name(item.text.strip()): continue if item.y1 > avg_heading_y + 1 and item.x1 < 12 and not item.text in [ "Name", 'None', "LFV" ]: if item.text.count("Established") or item.text.count( u'TROLLHÄTTAN TWR') or item.text.count( u'and/or SÅTENÄS') or item.text.count( u'TWR/TMC') or item.text.strip().endswith( "TWR") or item.text.strip().endswith("TWR."): continue if item.text.count("operational hours") or item.text.count( "See AIP DENMARK"): continue if item.text.count("hours of"): continue if item.text.count("Upper limit"): continue if item.text.count("that part") or item.text.count("coincides"): continue if item.text.count(u'Danger area EK D395 and') or item.text.count( u'D396 are situated within') or item.text.strip() == "TMA": continue if item.text.count(u'ÖSTGÖTA TMC is closed') or item.text.count( u'and SKAVSTA TWR is') or item.text.strip() == 'open.': continue if item.text.count("SAT 0530"): continue if item.text.strip() == 'OPS': continue if item.text.strip() == u'ÖSTGÖTA TMC:': continue if item.text.count(u'is open') or item.text.count('is closed'): continue if item.text.count('MON-FRI') or item.text.count('2150'): continue lines2 = page.get_lines( page.get_partially_in_rect(12, item.y1 + 0.2, 40, item.y2 - 0.2)) if len(lines2): zone_candidates.append(item)
for item in page.get_by_regex( "\s*RUNWAY\s*PHYSICAL\s*CHARACTERISTICS\s*"): lines = page.get_lines( page.get_partially_in_rect(0, item.y2 + 0.1, 100, 100)) for line in lines: if re.match(ur"AD\s+2.13", line): break m = re.match( ur".*?(RWY END)?\s*\*?(\d{6}\.\d+N)\s*(\d{6,7}\.\d+E).*", line) if not m: continue rwyend, lat, lon = m.groups() rwytxts = page.get_lines( page.get_partially_in_rect(0, line.y1, 12, line.y2)) print "Rwytxts:", rwytxts rwytxt, = rwytxts uprint("rwytext:", rwytxt) rwy, = re.match(ur"\s*(\d{2}[LRCM]?)\s*[\d.]*\s*", rwytxt).groups() have_thr = False for thr in thrs: if thr['thr'] == rwy: have_thr = True if rwyend != None and have_thr: continue thrs.append(dict(pos=mapper.parse_coords(lat, lon), thr=rwy)) for item in page.get_by_regex("ATS AIRSPACE"): lines = iter( page.get_lines( page.get_partially_in_rect(0, item.y2 + 0.1, 100, 100))) spaces = []
continue if item.text.strip() == 'OPS': continue if item.text.strip() == u'ÖSTGÖTA TMC:': continue if item.text.count(u'is open') or item.text.count('is closed'): continue if item.text.count('MON-FRI') or item.text.count('2150'): continue lines2 = page.get_lines( page.get_partially_in_rect(12, item.y1 + 0.2, 40, item.y2 - 0.2)) if len(lines2): zone_candidates.append(item) uprint("Found cands:", zone_candidates) zone_candidates.sort(key=lambda x: x.y1) for zone in zone_candidates: #uprint("Zone:",zone) #assert not zone.text.count("AOR") assert not zone.text.count("FIR") uprint("Headings:", headings) print "Pagenr:", pagenr assert len(headings) == 3 ret = [] for i in xrange(len(zone_candidates)): d = dict() cand = zone_candidates[i]