Eksempler >Data om land

Land i verden

Lese CVS | Lese XML | Lese og parse XML med SAX | Lese CSV til objekter | CSV eller XML til objekter

Hva

Land og hovedsteder i verden

Denne modulen demonstrerer alternative datastrukturer i Python. Vi bruker to datasett som beskriver alle land i verden, en CSV-fil og en XML-fil.

Datagrunnlaget er hentet fra geonames [1] . Datasettene som er brukt er kopiert lokat og er beskrevet i modulen: Noen datasett .

Datasettene kan kopieres herfra: CSV-fila: land.txt og XML-fila: land.xml

Vi skal gjøre følgende øvelser på disse dataene:

Lese CSV-fila og rapportere land fordelt på verdensdel
Lese XML-fila og rapportere land fordelt på verdensdel
Lese XML-fila parse med SAX og rapportere land fordelt på verdensdel
Lese CSV-fila og opprette en liste av land-objekter og rapportere derfra
Lese XML-fila og opprette en liste av land-objekter og rapportere derfra
Lage et fleksibelt Land objekt

Vi skal altså ikke gjøre noe særlig "nyttig". Hensikten er bare å illustrerer noen Python mekanismer.

Lese CVS

Vi leser inn CVS-fila, og rapporterer land fordelt på verdensdel.

"""
Read land.txt
Report countries sorted on continent
"""
#------------------------
# helper to read a file
def getTextFile(filename):
    try:
        file=open(filename,'r')
        res=file.read()
        file.close()
        return res
    except:
        print 'Trouble reading: '+filename
        return None
#-------------------------------
# continets [code,fullname,collected countries]
continents=[['EU','Europa',[]],
            ['AS','Asia',[]],
            ['NA', 'North America',[]],
            ['AN','Antarctica',[]],
            ['SA','South america',[]],
            ['OC','Oceania',[]],
            ['','No region',[]]]
#--------------------------------
# wherever you place the file:
T=getTextFile('..\\..\\commondata\\geonames\\land.txt')
if T!=None:
    lines=T.split('\n')
    for line in lines:
        parts=line.split('\t')
        if len(parts)!=12:
            continue
        (iso2,iso3,isonum,fips,country,capital,
         area,population,continent,language,money,gid)=parts
        for c in continents:
            if c[0]==continent:
                c[2].append( country)
    #--------------------
    # report, collected in a string, and ready for filewrite
    result=''
    for c in continents:
        result+= c[1]+'\n'
        for co in sorted(c[2]):
            result+='\t'+co+'\n'
    print result

Lese XML

Vi leser inn XML-fila, og rapporterer land fordelt på verdensdel.

import xml.dom.minidom
"""
Read land.xml
Report countries sorted on continent
"""
#------------------------
# helper to read a file
def getTextFile(filename):
    try:
        file=open(filename,'r')
        res=file.read()
        file.close()
        return res
    except:
        print 'Trouble reading: '+filename
        return None
#-----------------------
# helper to collect all text in a node
def getText(nodelist):
    rc = ''
    for node in nodelist:
        if node.nodeType == node.TEXT_NODE:
            t=node.data.encode('ISO-8859-1')
            rc += t
    return rc
#-------------------------------
# continets [code,fullname,collected countries]
continents=[['EU','Europa',[]],
            ['AS','Asia',[]],
            ['NA', 'North America',[]],
            ['AN','Antarctica',[]],
            ['SA','South america',[]],
            ['OC','Oceania',[]],
            ['','No region',[]]]
#--------------------------------
# wherever you place the file:
T=getTextFile('..\\..\\commondata\\geonames\\land.xml')
if T!=None:
    dom=xml.dom.minidom.parseString(T)
    countries=dom.getElementsByTagName('country')
    for co in countries:
        country=getText(co.getElementsByTagName('countryName')[0].childNodes)
        continent=getText(co.getElementsByTagName('continent')[0].childNodes)
        for c in continents:
            if c[0]==continent:
                c[2].append( country)
    #--------------------
    # report, collected in a string, and ready for filewrite
    result=''
    for c in continents:
        result+= c[1]+'\n'
        for co in sorted(c[2]):
            result+='\t'+co+'\n'
    print result

Lese og parse XML med SAX

Vi leser inn XML-fila, og rapporterer land fordelt på verdensdel.

import xml.parsers.expat
"""
Read land.xml
Parse with SAX
Report countries sorted on continent
"""
#------------------------
# helper to read a file
def getTextFile(filename):
    try:
        file=open(filename,'r')
        res=file.read()
        file.close()
        return res
    except:
        print 'Trouble reading: '+filename
        return None
#-------------------------------
# continets [code,fullname,collected countries]
continents=[['EU','Europa',[]],
            ['AS','Asia',[]],
            ['NA', 'North America',[]],
            ['AN','Antarctica',[]],
            ['SA','South america',[]],
            ['OC','Oceania',[]],
            ['','No region',[]]]
#-------------------------------------------------------
# 2 flags to remember type of element
# 2 globals to store wanted data
# 3 primitive handler functions used by saxparser
country=''
continent=''
countryFlag=False
continentFlag=False
#---------------------
# opening an element
def start_element(name, attrs):
    global countryFlag
    global continentFlag
    if name=='countryName':
        countryFlag=True
    elif name=='continent':
        continentFlag=True
#---------------------
# closing an element
def end_element(name):
    global countryFlag
    global continentFlag
    global country
    global continent
    global continents
    countryFlag=False
    continentFlag=False
    if name=='country':
        for c in continents:
            if c[0]==continent:
                c[2].append( country)
#---------------------
# content of an element
def char_data(data):
    global country
    global continent
    global continents
    if continentFlag:
        continent=data
    elif countryFlag:
        country=data
#--------------------------------
# wherever you place the file:
p = xml.parsers.expat.ParserCreate()
p.returns_unicode = 0
# tell parser about our callbacks
p.StartElementHandler = start_element
p.EndElementHandler = end_element
p.CharacterDataHandler = char_data
T=getTextFile('..\\..\\commondata\\geonames\\land.xml')
if T!=None:
    p.Parse(T)
    #--------------------
    # report, collected in a string, and ready for filewrite
    result=''
    for c in continents:
        result+= c[1]+'\n'
        for co in sorted(c[2]):
            result+='\t'+co+'\n'
    print result

Lese CSV til objekter

Vi leser inn csv-fila, ordner i objektliste og rapporterer land fordelt på verdensdel.

"""
Read land.txt
build list of objects
Report countries sorted on continent
"""
#------------------------
# defining a class
class Country:
    def __init__(self, name,continent):
        self.name=name
        self.continent=continent
#------------------------
# helper to read a file
def getTextFile(filename):
    try:
        file=open(filename,'r')
        res=file.read()
        file.close()
        return res
    except:
        print 'Trouble reading: '+filename
        return None
#-------------------------------
# continets [code,fullname,collected countries]
continents=[['EU','Europa',[]],
            ['AS','Asia',[]],
            ['NA', 'North America',[]],
            ['AN','Antarctica',[]],
            ['SA','South america',[]],
            ['OC','Oceania',[]],
            ['','No region',[]]]
#--------------------------------
# wherever you place the file:
T=getTextFile('..\\..\\commondata\\geonames\\land.txt')
allCountries=[]
if T!=None:
    lines=T.split('\n')
    for line in lines:
        parts=line.split('\t')
        if len(parts)!=12:
            continue
        allCountries.append(Country(parts[4],parts[8]))
    #--------------------
    # report, collected in a string, and ready for filewrite
    result=''
    for c in continents:
        result+= c[1]+'\n'
        for co in allCountries:
            if co.continent==c[0]:
                result+='\t'+co.name+'\n'
    print result

CSV eller XML til objekter

Skisse av en løsning som skal kunne handtere begge formatene , ordne i objektliste og rapporterer land fordelt på verdensdel.

import xml.dom.minidom
"""
Read land.txt or land.xml
build list of objects
Report countries sorted on continent
"""
#------------------------
# defining a class
class Country:
    def __init__(self, clist=None,cnode=None):
        if clist!=None:
            self.name=clist[4]
            self.continent=clist[8]
        elif cnode!=None:
            cnn=cnode.getElementsByTagName('countryName')[0]
            self.name=getText(cnn.childNodes)
            ccc=cnode.getElementsByTagName('continent')[0]
            self.continent=getText(ccc.childNodes)
        else:
            self.name='langvekkistan'
            self.continent=''
            
#------------------------
# helper to read a file
def getTextFile(filename):
    try:
        file=open(filename,'r')
        res=file.read()
        file.close()
        return res
    except:
        print 'Trouble reading: '+filename
        return None
#-----------------------
# helper to collect all text in a node
def getText(nodelist):
    rc = ''
    for node in nodelist:
        if node.nodeType == node.TEXT_NODE:
            t=node.data.encode('ISO-8859-1')
            rc += t
    return rc
#-------------------------------
# continets [code,fullname,collected countries]
continents=[['EU','Europa',[]],
            ['AS','Asia',[]],
            ['NA', 'North America',[]],
            ['AN','Antarctica',[]],
            ['SA','South america',[]],
            ['OC','Oceania',[]],
            ['','No region',[]]]
   
#--------------------------------
allCountries=[]
#------ this section-----
##T=getTextFile('..\\..\\commondata\\geonames\\land.txt')
##if T!=None:
##    lines=T.split('\n')
##    for line in lines:
##        parts=line.split('\t')
##        if len(parts)!=12:
##            continue
##        allCountries.append(Country(clist=parts))
#------ or this section-----
T=getTextFile('..\\..\\commondata\\geonames\\land.xml')
if T!=None:
    dom=xml.dom.minidom.parseString(T)
    countries=dom.getElementsByTagName('country')
    for co in countries:
        allCountries.append(Country(cnode=co))
#--------------------
# report, collected in a string, and ready for filewrite
result=''
for c in continents:
    result+= c[1]+'\n'
    for co in allCountries:
        if co.continent==c[0]:
            result+='\t'+co.name+'\n'
print result

Referanser

Geonames GeoNames www.geonames.org/ 14-03-2010

Vedlikehold

B.Stenseth, september 2007

( Velkommen ) Eksempler >Data om land ( Presentasjon )