View Category
XML
Parsing

Process an XML document

Given the XML Document:

<shopping>
  <item name="bread" quantity="3" price="2.50"/>
  <item name="milk" quantity="2" price="3.50"/>
</shopping>

Print out the total cost of the items, e.g. $14.50
python
from xml.dom.minidom import parseString
document = parseString(
"""<shopping>
<item name="bread" quantity="3" price="2.50"/>
<item name="milk" quantity="2" price="3.50"/>
</shopping>""").documentElement
total = sum([float(item.getAttribute('price')) *
int(item.getAttribute('quantity'))
for item in document.getElementsByTagName('item')])
print '$%.2f' % total
erlang
-include_lib("xmerl/include/xmerl.hrl").
-export([get_total/1]).

get_total(ShoppingList) ->
{XmlElt, _} = xmerl_scan:string(ShoppingList),
Items = xmerl_xpath:string("/shopping/item", XmlElt),
Total = lists:foldl(fun(Item, Tot) ->
[#xmlAttribute{value = PriceString}] = xmerl_xpath:string("/item/@price", Item),
{Price, _} = string:to_float(PriceString),
[#xmlAttribute{value = QuantityString}] = xmerl_xpath:string("/item/@quantity", Item),
{Quantity, _} = string:to_integer(QuantityString),
Tot + Price*Quantity
end,
0, Items),
io:format("$~.2f~n", [Total]).
cpp
char input[] =
"<shopping>"
" <item name=\"bread\" quantity=\"3\" price=\"2.50\"/>"
" <item name=\"milk\" quantity=\"2\" price=\"3.50\"/>"
"</shopping>";

xml_document<> doc;
doc.parse<0>(input);
xml_node<> *shopping = doc.first_node();

float total_price = 0;
for (xml_node<> *item = shopping->first_node(); item != NULL; item = item->next_sibling())
{
float item_sum = 0;
float val;

if (string(item->name()) != "item")
continue;

for (xml_attribute<> *attr = item->first_attribute(); attr != NULL; attr = attr->next_attribute())
{
string name(attr->name());
if (name == "quantity" || name == "price")
{
stringstream v(attr->value());
v >> val;
if (item_sum)
item_sum *= val;
else
item_sum = val;
}
}
total_price += item_sum;
}

cout.setf(ios::fixed, ios::floatfield);
cout << "Total price is $" << setprecision(2) << total_price << endl;
Creation

create some XML programmatically

Given the following CSV:

bread,3,2.50
milk,2,3.50

Produce the equivalent information in XML, e.g.:

<shopping>
  <item name="bread" quantity="3" price="2.50" />
  <item name="milk" quantity="2" price="3.50" />
</shopping>
python
from xml.dom import minidom

csv = """bread,3,2.50
milk,2,3.50"""

doc = minidom.Document()
shopping = doc.createElement("shopping")

for line in csv.split("\n"):
name, quantity, price = line.split(",")
el = doc.createElement("item")
el.setAttribute("name", name)
el.setAttribute("quantity", quantity)
el.setAttribute("price", price)
shopping.appendChild(el)

print shopping.toprettyxml()
from xml.etree.ElementTree import Element, SubElement, tostring

csv = """bread,3,2.50
milk,2,3.50"""

root = Element('shopping')

for line in csv.split("\n"):
name, quantity, price = line.split(",")
SubElement(root,'item', {'name' : name,
'quantity' : quantity,
'price' : price })

print tostring(root)
erlang
to_xml(ShoppingList) ->
Items = lists:map(fun(L) ->
[Name, Quantity, Price] = string:tokens(L, ","),
{item, [{name, Name}, {quantity, Quantity}, {price, Price}], []}
end, string:tokens(ShoppingList, "\n")),
xmerl:export_simple([{shopping, [], Items}], xmerl_xml).
cpp
string input("bread,3,2.50\nmilk,2,3.50\n");

tokenizer<char_separator<char> > tokens(input, char_separator<char>(", \n"));
tokenizer<char_separator<char> >::iterator it = tokens.begin();

xml_document<> doc;
xml_node<> *shopping = doc.allocate_node(node_element, "shopping");
doc.append_node(shopping);

while (it != tokens.end()) {
xml_node<> *item = doc.allocate_node(node_element, "item");
shopping->append_node(item);

item->append_attribute(doc.allocate_attribute("name", doc.allocate_string((*it++).c_str())));
item->append_attribute(doc.allocate_attribute("quantity", doc.allocate_string((*it++).c_str())));
item->append_attribute(doc.allocate_attribute("price", doc.allocate_string((*it++).c_str())));
}

cout << doc << endl;