`lxml` examples with the `otc` data set

# Parsing example 1a:

import lxml.etree
tree = lxml.etree.parse("/anvil/projects/tdm/data/otc/valu.xml")
ingredients = tree.xpath("//ns:ingredient", namespaces = {'ns': 'urn:hl7-org:v3'})

# there are ten ingredients elements:
len(ingredients)

# get an ingredient tag
ingredients[0].tag

# get the attributes of the first ingredient (the results look like a dictionary object)
ingredients[0].attrib

# or for the last ingredient
ingredients[9].attrib

# get the attributes of each of the ingredients
for e in ingredients:
    print(e.attrib)

# get the first child of each of the ingredients, which is an ingredientSubstance for the first 9 elements
#     but is a quantity for the last element
for e in ingredients:
    print(e[0])

# Instead, we can start the search from each ingredient, and specify that we want to go next to its ingredientSubstance:
for e in ingredients:
    print(e.xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0])

# and from there, we can get their names (which happen to always be the second child of the ingredientSubstance
for e in ingredients:
    print(e.xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0][1])

# and the text inside those name elements:
for e in ingredients:
    print(e.xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0][1].text)

# Parsing example 1b:

# We can also get the ingredientSubstance element inside any specific ingredient element

# For instance, the 0th ingredient has an ingredientSubstance child, which has two children:
ingredients[0].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()

# The first is a code element with two attributes:
ingredients[0].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[0].attrib

# and the second is a name element with some text:
ingredients[0].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[1].text

# BUT the 9th ingredient has an ingredientSubstance child, which has three children:
ingredients[9].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()

# The first is a code element with two attributes:
ingredients[9].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[0].attrib

# and the second is a name element with some text:
ingredients[9].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[1].text

# and the third is an activeMoiety element with a child also called activeMoiety that has two children:
ingredients[9].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[2]

# as we can see here:
my_element = ingredients[9].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[2]
print(lxml.etree.tostring(my_element, pretty_print=True).decode('utf-8'))

# Parsing example 1c:

# It is worthwhile to walk through the elements slowly, one at a time, examining the outputs
# For instance, this is the way to extract the name of the fourth ingredient element:
ingredients[3][0][1].text

# Parsing example 1d:

# We can get all of the elements with attribute codeSystem
tree.xpath("//ns:*[@codeSystem]", namespaces = {'ns': 'urn:hl7-org:v3'})

# If you want to see the attributes of those elements, we can loop through them:
for e in tree.xpath("//ns:*[@codeSystem]", namespaces = {'ns': 'urn:hl7-org:v3'}):
    print(e.attrib)

# We can also loop through the elements that have a tag called name:
names = tree.xpath("//ns:name", namespaces = {'ns': 'urn:hl7-org:v3'})
i=1
# print the content between all name tags
for name_tag in names:
    print(i, ":", name_tag.text)
    i=i+1

# Parsing example 2a:

import lxml.etree
tree = lxml.etree.parse("/anvil/projects/tdm/data/otc/hawaii.xml")
ingredients = tree.xpath("//ns:ingredient", namespaces = {'ns': 'urn:hl7-org:v3'})

# there are fourteen ingredients elements:
len(ingredients)

# get an ingredient tag
ingredients[0].tag

# get the attributes of the first ingredient (the results look like a dictionary object)
ingredients[0].attrib

# or for the last ingredient
ingredients[13].attrib

# get the attributes of each of the ingredients
for e in ingredients:
    print(e.attrib)

# get the first child of each of the ingredients, which is an ingredientSubstance for 12 of these elements
#     but is a quantity for two of the elements
for e in ingredients:
    print(e[0])

# Instead, we can start the search from each ingredient, and specify that we want to go next to its ingredientSubstance:
for e in ingredients:
    print(e.xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0])

# and from there, we can get their names (which happen to always be the second child of the ingredientSubstance
for e in ingredients:
    print(e.xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0][1])

# and the text inside those name elements:
for e in ingredients:
    print(e.xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0][1].text)

# Parsing example 2b:

# We can also get the ingredientSubstance element inside any specific ingredient element

# For instance, the 0th ingredient has an ingredientSubstance child, which has two children:
ingredients[0].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()

# The first is a code element with two attributes:
ingredients[0].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[0].attrib

# and the second is a name element with some text:
ingredients[0].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[1].text

# BUT the 13th ingredient has an ingredientSubstance child, which has three children:
ingredients[13].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()

# The first is a code element with two attributes:
ingredients[13].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[0].attrib

# and the second is a name element with some text:
ingredients[13].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[1].text

# and the third is an activeMoiety element with a child also called activeMoiety that has two children:
ingredients[13].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[2]

# as we can see here:
my_element = ingredients[13].xpath(".//ns:ingredientSubstance", namespaces = {'ns': 'urn:hl7-org:v3'})[0].getchildren()[2]
print(lxml.etree.tostring(my_element, pretty_print=True).decode('utf-8'))

# Parsing example 2c:

# It is worthwhile to walk through the elements slowly, one at a time, examining the outputs
# For instance, this is the way to extract the name of the fourth ingredient element:
ingredients[3][0][1].text

# Parsing example 2d:

# We can get all of the elements with attribute codeSystem
tree.xpath("//ns:*[@codeSystem]", namespaces = {'ns': 'urn:hl7-org:v3'})

# If you want to see the attributes of those elements, we can loop through them:
for e in tree.xpath("//ns:*[@codeSystem]", namespaces = {'ns': 'urn:hl7-org:v3'}):
    print(e.attrib)

# We can also loop through the elements that have a tag called name:
names = tree.xpath("//ns:name", namespaces = {'ns': 'urn:hl7-org:v3'})
i=1
# print the content between all name tags
for name_tag in names:
    print(i, ":", name_tag.text)
    i=i+1

lxml examples with the otc data set

`lxml` examples with the `otc` data set