TableParser Namespace Reference

Functions
def	extractPages
def	extractPagesForPackage
def	generateTab

Variables
list	allRows = []
string	destTabClassName = 'tabs'
	fileNameTemplate = None
	htmlFileName = None
	htmlFilePath = None
	htmlFullPath = None
	htmlPage = None
tuple	newTab = generateTab(pageNames, page)
tuple	oldTab = temp.find('div', {'class' : destTabClassName})
tuple	page = page.replace(' ', '_')
tuple	pageNames = pages.keys()
tuple	pages = extractPagesForPackage()
tuple	table = htmlPage.find('table', {'class' : tableClassName})
string	tableClassName = 'directory'
tuple	temp = BeautifulSoup(str(htmlPage))

Function Documentation

def TableParser.extractPages

(

configFileFlag = False

)

Definition at line 19 of file TableParser.py.

References bitset_utilities.append().

def extractPages(configFileFlag = False):
    # initial page, A
    pages = {'A':[]}
    # find all class/namespace talbe rows.
    table = htmlPage.find('table', {'class' : tableClassName})
    for row in table.findAll('tr'):
        # please see the related html file (annotated.html) to understand the
        # approach here. you will see that, only hidden rows have style
        # attribute and these hidden rows must be added to pages of their
        # parents. This is why we need to check whether row has a style
        # attribute or not.
        styleFlag = False
        if row.has_key('style'): styleFlag = True
        # change the first letter if row is not hidden (child) one
        if not styleFlag: firstLetter = row.findAll('td')[0].text[0].upper()
        # if pages dict doesn't have the page yet..
        if not pages.has_key(firstLetter):
            pages[firstLetter] = []
        # insert the row into the related page
        if configFileFlag:
            url = row.find('a')['href']
            if '_cff' in url or '_cfi' in url or '_cfg' in url:
                pages[firstLetter].append(row)
        else:
            pages[firstLetter].append(row)
    return pages

# load rows from the package documentation page. output structure: