blabla
# *
# */
#
# bla_bla_bla
# bilip_bilip
#
# /* - this comment acts as a separator - */
#
# blo_blo_blo
#
#
# will only keep the first two lines of sources with
# the "blabla" block.
#
# However, the comment will be kept, with following source lines
# if it contains a starting '#' or '@' as in:
#
# /*@.....*/
# /*#.....*/
# /* @.....*/
# /* #.....*/
#
#############################################################################
#
# The DocCode class is used to store source code lines
#
# 'self.lines' contains a set of source code lines that will
# be dumped as HTML in a tag.
#
# The object is filled line by line by the parser; it strips the
# leading "margin" space from each input line before storing it
# in 'self.lines'.
#
class DocCode:
def __init__( self, margin = 0 ):
self.lines = []
self.margin = margin
def add( self, line ):
# remove margin whitespace
#
if string.strip( line[: self.margin] ) == "":
line = line[self.margin :]
self.lines.append( line )
def dump( self ):
for line in self.lines:
print "--" + line
print ""
def get_identifier( self ):
# this function should never be called
#
return "UNKNOWN_CODE_IDENTIFIER!"
def dump_html( self, identifiers = None ):
# clean the last empty lines
#
l = len( self.lines ) - 1
while l > 0 and string.strip( self.lines[l - 1] ) == "":
l = l - 1
# the code footer should be directly appended to the last code
# line to avoid an additional blank line
#
sys.stdout.write( code_header )
for line in self.lines[0 : l]:
sys.stdout.write( '\n' + line )
sys.stdout.write( code_footer )
#############################################################################
#
# The DocParagraph is used to store text paragraphs.
# 'self.words' is simply a list of words for the paragraph.
#
# The paragraph is filled line by line by the parser.
#
class DocParagraph:
def __init__( self ):
self.words = []
def add( self, line ):
# Get rid of unwanted spaces in the paragraph.
#
# The following two lines are the same as
#
# self.words.extend( string.split( line ) )
#
# but older Python versions don't have the `extend' attribute.
#
last = len( self.words )
self.words[last:last] = string.split( line )
# This function is used to retrieve the first word of a given
# paragraph.
#
def get_identifier( self ):
if self.words:
return self.words[0]
# should never happen
#
return "UNKNOWN_PARA_IDENTIFIER!"
def dump( self, identifiers = None ):
max_width = 50
cursor = 0
line = ""
extra = None
alphanum = string.lowercase + string.uppercase + string.digits + '_'
for word in self.words:
# process cross references if needed
#
if identifiers and word and word[0] == '@':
word = word[1:]
# we need to find non-alphanumeric characters
#
i = len( word )
while i > 0 and not word[i - 1] in alphanum:
i = i - 1
if i > 0:
extra = word[i:]
word = word[0:i]
block = identifiers.get( word )
if block:
word = '' + word + ''
else:
word = '?' + word
if cursor + len( word ) + 1 > max_width:
print line
cursor = 0
line = ""
line = line + word
if not extra:
line = line + " "
cursor = cursor + len( word ) + 1
# handle trailing periods, commas, etc. at the end of
# cross references.
#
if extra:
if cursor + len( extra ) + 1 > max_width:
print line
cursor = 0
line = ""
line = line + extra + " "
cursor = cursor + len( extra ) + 1
extra = None
if cursor > 0:
print line
# print "§" # for debugging only
def dump_string( self ):
s = ""
space = ""
for word in self.words:
s = s + space + word
space = " "
return s
def dump_html( self, identifiers = None ):
print para_header
self.dump( identifiers )
print para_footer
#############################################################################
#
# DocContent is used to store the content of a given marker.
#
# The "self.items" list contains (field,elements) records, where
# "field" corresponds to a given structure fields or function
# parameter (indicated by a "::"), or NULL for a normal section
# of text/code.
#
# Hence, the following example:
#
#
# This is an example of what can be put in a content section,
#
# A second line of example text.
#
# x :: A simple test field, with some contents.
# y :: Even before, this field has some code contents.
# {
# y = x+2;
# }
#
# should be stored as
#
# [ ( None, [ DocParagraph, DocParagraph] ),
# ( "x", [ DocParagraph ] ),
# ( "y", [ DocParagraph, DocCode ] ) ]
#
# in 'self.items'.
#
# The DocContent object is entirely built at creation time; you must
# pass a list of input text lines in the "lines_list" parameter.
#
class DocContent:
def __init__( self, lines_list ):
self.items = []
code_mode = 0
code_margin = 0
text = []
paragraph = None # represents the current DocParagraph
code = None # represents the current DocCode
elements = [] # the list of elements for the current field,
# contains DocParagraph or DocCode objects
field = None # the current field
for aline in lines_list:
if code_mode == 0:
line = string.lstrip( aline )
l = len( line )
margin = len( aline ) - l
# if the line is empty, this is the end of the current
# paragraph
#
if l == 0 or line == '{':
if paragraph:
elements.append( paragraph )
paragraph = None
if line == "":
continue
code_mode = 1
code_margin = margin
code = None
continue
words = string.split( line )
# test for a field delimiter on the start of the line, i.e.
# the token `::'
#
if len( words ) >= 2 and words[1] == "::":
# start a new field - complete current paragraph if any
#
if paragraph:
elements.append( paragraph )
paragraph = None
# append previous "field" to self.items
#
self.items.append( ( field, elements ) )
# start new field and elements list
#
field = words[0]
elements = []
words = words[2 :]
# append remaining words to current paragraph
#
if len( words ) > 0:
line = string.join( words )
if not paragraph:
paragraph = DocParagraph()
paragraph.add( line )
else:
# we are in code mode...
#
line = aline
# the code block ends with a line that has a single '}' on
# it that is located at the same column that the opening
# accolade...
#
if line == " " * code_margin + '}':
if code:
elements.append( code )
code = None
code_mode = 0
code_margin = 0
# otherwise, add the line to the current paragraph
#
else:
if not code:
code = DocCode()
code.add( line )
if paragraph:
elements.append( paragraph )
if code:
elements.append( code )
self.items.append( ( field, elements ) )
def get_identifier( self ):
if self.items:
item = self.items[0]
for element in item[1]:
return element.get_identifier()
# should never happen
#
return "UNKNOWN_CONTENT_IDENTIFIER!"
def get_title( self ):
if self.items:
item = self.items[0]
for element in item[1]:
return element.dump_string()
# should never happen
#
return "UNKNOWN_CONTENT_TITLE!"
def dump( self ):
for item in self.items:
field = item[0]
if field:
print ""
for element in item[1]:
element.dump()
if field:
print ""
def dump_html( self, identifiers = None ):
n = len( self.items )
in_table = 0
for i in range( n ):
item = self.items[i]
field = item[0]
if not field:
if in_table:
print ""
in_table = 0
for element in item[1]:
element.dump_html( identifiers )
else:
if not in_table:
print "| "
in_table = 1
else:
print " |
| "
print "" + field + " | "
for element in item[1]:
element.dump_html( identifiers )
if in_table:
print " |
"
def dump_html_in_table( self, identifiers = None ):
n = len( self.items )
in_table = 0
for i in range( n ):
item = self.items[i]
field = item[0]
if not field:
if item[1]:
print "| "
for element in item[1]:
element.dump_html( identifiers )
print " |
"
else:
print "| " + field + " | "
for element in item[1]:
element.dump_html( identifiers )
print " |
"
#############################################################################
#
#
# The DocBlock class is used to store a given comment block. It contains
# a list of markers, as well as a list of contents for each marker.
#
# "self.items" is a list of (marker, contents) elements, where
# 'marker' is a lowercase marker string, and 'contents' is a DocContent
# object.
#
# "self.source" is simply a list of text lines taken from the
# uncommented source itself.
#
# Finally, "self.name" is a simple identifier used to uniquely identify
# the block. It is taken from the first word of the first
# paragraph of the first marker of a given block, i.e:
#
# Goo
# Bla bla bla
#
# will have a name of "Goo"
#
class DocBlock:
def __init__( self, block_line_list = [], source_line_list = [] ):
self.items = [] # current ( marker, contents ) list
self.section = None # section this block belongs to
self.filename = "unknown" # filename defining this block
self.lineno = 0 # line number in filename
marker = None # current marker
content = [] # current content lines list
alphanum = string.letters + string.digits + "_"
self.name = None
for line in block_line_list:
line2 = string.lstrip( line )
l = len( line2 )
margin = len( line ) - l
if l > 3:
ender = None
if line2[0] == '<':
ender = '>'
elif line2[0] == '@':
ender = ':'
if ender:
i = 1
while i < l and line2[i] in alphanum:
i = i + 1
if i < l and line2[i] == ender:
if marker and content:
self.add( marker, content )
marker = line2[1 : i]
content = []
line2 = string.lstrip( line2[i + 1 :] )
l = len( line2 )
line = " " * margin + line2
content.append( line )
if marker and content:
self.add( marker, content )
self.source = []
if self.items:
self.source = source_line_list
# now retrieve block name when possible
#
if self.items:
first = self.items[0]
self.name = first[1].get_identifier()
# This function adds a new element to 'self.items'.
# 'marker' is a marker string, or None.
# 'lines' is a list of text lines used to compute a list of
# DocContent objects.
#
def add( self, marker, lines ):
# remove the first and last empty lines from the content list
#
l = len( lines )
if l > 0:
i = 0
while l > 0 and string.strip( lines[l - 1] ) == "":
l = l - 1
while i < l and string.strip( lines[i] ) == "":
i = i + 1
lines = lines[i : l]
l = len( lines )
# add a new marker only if its marker and its content list
# aren't empty
#
if l > 0 and marker:
content = DocContent( lines )
self.items.append( ( string.lower( marker ), content ) )
def find_content( self, marker ):
for item in self.items:
if ( item[0] == marker ):
return item[1]
return None
def html_address( self ):
section = self.section
if section and section.filename:
return section.filename + '#' + self.name
return "" # this block is not in a section?
def location( self ):
return self.filename + ':' + str( self.lineno )
def dump( self ):
for i in range( len( self.items ) ):
print "[" + self.items[i][0] + "]"
content = self.items[i][1]
content.dump()
def dump_html( self, identifiers = None ):
types = [ 'type', 'struct', 'functype', 'function',
'constant', 'enum', 'macro', 'structure', 'also' ]
parameters = [ 'input', 'inout', 'output', 'return' ]
if not self.items:
return
# start of a block
#
print block_header
# place html anchor if needed
#
if self.name:
print ''
print "" + self.name + "
"
print ""
# print source code
#
if not self.source:
print block_footer
return
lines = self.source
l = len( lines ) - 1
while l >= 0 and string.strip( lines[l] ) == "":
l = l - 1
print source_header
print ""
for line in lines[0 : l + 1]:
print line
print source_footer
in_table = 0
# dump each (marker,content) element
#
for element in self.items:
marker = element[0]
content = element[1]
if marker == "description":
print description_header
content.dump_html( identifiers )
print description_footer
elif not ( marker in types ):
sys.stdout.write( marker_header )
sys.stdout.write( marker )
sys.stdout.write( marker_inter + '\n' )
content.dump_html( identifiers )
print marker_footer
print ""
print block_footer
#############################################################################
#
# The DocSection class is used to store a given documentation section.
#
# Each section is made of an identifier, an abstract and a description.
#
# For example, look at:
#
# Basic_Data_Types
#
# FreeType 2 Basic Data Types
#
# <Abstract>
# Definitions of basic FreeType data types
#
# <Description>
# FreeType defines several basic data types for all its
# operations...
#
class DocSection:
def __init__( self, block ):
self.block = block
self.name = string.lower( block.name )
self.abstract = block.find_content( "abstract" )
self.description = block.find_content( "description" )
self.elements = {}
self.list = []
self.filename = self.name + ".html"
# sys.stderr.write( "new section '" + self.name + "'" )
def add_element( self, block ):
# check that we don't have a duplicate element in this
# section
#
if self.elements.has_key( block.name ):
sys.stderr.write( "ERROR - duplicate element definition for " +
"'" + block.name + "' in section '" +
self.name + "'" )
sys.exit()
self.elements[ block.name ] = block
self.list.append( block )
def dump_html( self, identifiers = None ):
"""make an HTML page from a given DocSection"""
# print HTML header
#
print html_header
# print title
#
print section_title_header
print self.title
print section_title_footer
# print description
#
print block_header
self.description.dump_html( identifiers )
print block_footer
# print elements
#
for element in self.list:
element.dump_html( identifiers )
print html_footer
class DocSectionList:
def __init__( self ):
self.sections = {} # map section names to section objects
self.list = [] # list of sections (in creation order)
self.current_section = None # current section
self.identifiers = {} # map identifiers to blocks
def append_section( self, block ):
name = string.lower( block.name )
abstract = block.find_content( "abstract" )
if self.sections.has_key( name ):
# There is already a section with this name in our
# list. We will try to complete it.
#
section = self.sections[name]
if section.abstract:
# This section already has an abstract defined;
# simply check that the new section doesn't
# provide a new one.
#
if abstract:
stderr.write( "ERROR - duplicate section definition" +
" for '" + name + "'" )
stderr.write( "previous definition in" +
" '" + section.location() )
stderr.write( "second definition in" +
" '" + block.location() )
sys.quit()
else:
# The old section didn't contain an abstract; we are
# now going to replace it.
#
section.abstract = abstract
section.description = block.find_content( "description" )
section.block = block
else:
# a new section
#
section = DocSection( block )
self.sections[name] = section
self.list.append( section )
self.current_section = section
def append_block( self, block ):
if block.name:
section = block.find_content( "section" )
if section:
self.append_section( block )
elif self.current_section:
self.current_section.add_element( block )
block.section = self.current_section
self.identifiers[block.name] = block
def prepare_files( self, file_prefix = None ):
# prepare the section list, by computing section filenames
# and the index
#
if file_prefix:
prefix = file_prefix + "-"
else:
prefix = ""
# compute section names
#
for section in self.sections.values():
title_content = section.block.find_content( "title" )
if title_content:
section.title = title_content.get_title()
else:
section.title = "UNKNOWN_SECTION_TITLE!"
# compute section filenames
#
for section in self.sections.values():
section.filename = prefix + section.name + ".html"
self.toc_filename = prefix + "toc.html"
self.index_filename = prefix + "index.html"
# compute the sorted list of identifiers for the index
#
self.index = self.identifiers.keys()
self.index.sort()
def dump_html_toc( self ):
# dump an html table of contents
#
old_stdout = sys.stdout
new_file = open( self.toc_filename, "w" )
sys.stdout = new_file
print html_header
print "<center><h1>Table of Contents</h1></center>"
print "<center><table cellpadding=5>"
for section in self.list:
if section.abstract:
print "<tr valign=top><td>"
sys.stdout.write( '<a href="' + section.filename + '">' )
sys.stdout.write( section.title )
sys.stdout.write( "</a></td><td>" + '\n' )
section.abstract.dump_html( self.identifiers )
print "</td></tr>"
print "</table></center>"
print html_footer
sys.stdout = old_stdout
def dump_html_sections( self ):
old_stdout = sys.stdout
for section in self.sections.values():
if section.filename:
new_file = open( section.filename, "w" )
sys.stdout = new_file
section.dump_html( self.identifiers )
new_file.close()
sys.stdout = old_stdout
def dump_html_index( self ):
old_stdout = sys.stdout
new_file = open( self.index_filename, "w" )
sys.stdout = new_file
num_columns = 3
total = len( self.index )
line = 0
print html_header
print "<center><h1>General Index</h1></center>"
print "<center><table cellpadding=5><tr valign=top><td>"
for ident in self.index:
block = self.identifiers[ident]
if block:
sys.stdout.write( '<a href="' + block.html_address() + '">' )
sys.stdout.write( block.name )
sys.stdout.write( '</a><br>' + '\n' )
if line * num_columns >= total:
print "</td><td>"
line = 0
else:
line = line + 1
else:
sys.stderr.write( "identifier '" + ident +
"' has no definition" + '\n' )
print "</tr></table></center>"
print html_footer
sys.stdout = old_stdout
# Filter a given list of DocBlocks. Returns a new list
# of DocBlock objects that only contains element whose
# "type" (i.e. first marker) is in the "types" parameter.
#
def filter_blocks_by_type( block_list, types ):
new_list = []
for block in block_list:
if block.items:
element = block.items[0]
marker = element[0]
if marker in types:
new_list.append( block )
return new_list
def filter_section_blocks( block ):
return block.section != None
# Perform a lexicographical comparison of two DocBlock
# objects. Returns -1, 0 or 1.
#
def block_lexicographical_compare( b1, b2 ):
if not b1.name:
return -1
if not b2.name:
return 1
id1 = string.lower( b1.name )
id2 = string.lower( b2.name )
if id1 < id2:
return -1
elif id1 == id2:
return 0
else:
return 1
# dump a list block as a single HTML page
#
def dump_html_1( block_list ):
print html_header
for block in block_list:
block.dump_html()
print html_footer
def file_exists( pathname ):
result = 1
try:
file = open( pathname, "r" )
file.close()
except:
result = None
return result
def add_new_block( list, filename, lineno, block_lines, source_lines ):
"""add a new block to the list"""
block = DocBlock( block_lines, source_lines )
block.filename = filename
block.lineno = lineno
list.append( block )
def make_block_list():
"""parse a file and extract comments blocks from it"""
file_list = []
sys.stderr.write( repr( sys.argv[1:] ) + '\n' )
for pathname in sys.argv[1:]:
if string.find( pathname, '*' ) >= 0:
newpath = glob.glob( pathname )
newpath.sort() # sort files -- this is important because
# of the order of files
else:
newpath = [pathname]
last = len( file_list )
file_list[last:last] = newpath
if len( file_list ) == 0:
file_list = None
else:
# now filter the file list to remove non-existing ones
file_list = filter( file_exists, file_list )
list = []
block = []
format = 0
lineno = 0
# We use "format" to store the state of our parser:
#
# 0 - wait for beginning of comment
# 1 - parse comment format 1
# 2 - parse comment format 2
#
# 4 - wait for beginning of source (or comment ??)
# 5 - process source
#
comment = []
source = []
state = 0
for line in fileinput.input( file_list ):
l = len( line )
if l > 0 and line[l - 1] == '\012':
line = line[0 : l - 1]
# stripped version of the line
#
line2 = string.strip( line )
l = len( line2 )
# if this line begins with a comment and we are processing some
# source, exit to state 0
#
# unless we encounter something like:
#
# /*@.....
# /*#.....
#
# /* @.....
# /* #.....
#
if format >= 4 and l > 2 and line2[0 : 2] == '/*':
if l < 4 or ( line2[3] != '@' and line2[3:4] != ' @' and
line2[3] != '#' and line2[3:4] != ' #'):
add_new_block( list, fileinput.filename(),
lineno, block, source )
format = 0
if format == 0: #### wait for beginning of comment ####
if l > 3 and line2[0 : 3] == '/**':
i = 3
while i < l and line2[i] == '*':
i = i + 1
if i == l:
# this is '/**' followed by any number of '*', the
# beginning of a Format 1 block
#
block = []
source = []
format = 1
lineno = fileinput.lineno()
elif i == l - 1 and line2[i] == '/':
# this is '/**' followed by any number of '*', followed
# by a '/', i.e. the beginning of a Format 2 or 3 block
#
block = []
source = []
format = 2
lineno = fileinput.lineno()
##############################################################
#
# FORMAT 1
#
elif format == 1:
# If the line doesn't begin with a "*", something went
# wrong, and we must exit, and forget the current block.
#
if l == 0 or line2[0] != '*':
block = []
format = 0
# Otherwise, we test for an end of block, which is an
# arbitrary number of '*', followed by '/'.
#
else:
i = 1
while i < l and line2[i] == '*':
i = i + 1
# test for the end of the block
#
if i < l and line2[i] == '/':
if block != []:
format = 4
else:
format = 0
else:
# otherwise simply append line to current block
#
block.append( line2[i:] )
continue
##############################################################
#
# FORMAT 2
#
elif format == 2:
# If the line doesn't begin with '/*' and end with '*/',
# this is the end of the format 2 format.
#
if l < 4 or line2[: 2] != '/*' or line2[-2 :] != '*/':
if block != []:
format = 4
else:
format = 0
else:
# remove the start and end comment delimiters, then
# right-strip the line
#
line2 = string.rstrip( line2[2 : -2] )
# check for end of a format2 block, i.e. a run of '*'
#
if string.count( line2, '*' ) == l - 4:
if block != []:
format = 4
else:
format = 0
else:
# otherwise, add the line to the current block
#
block.append( line2 )
continue
if format >= 4: #### source processing ####
if l > 0:
format = 5
if format == 5:
source.append( line )
if format >= 4:
add_new_block( list, fileinput.filename(), lineno, block, source )
return list
# This function is only used for debugging
#
def dump_block_list( list ):
"""dump a comment block list"""
for block in list:
print "----------------------------------------"
for line in block[0]:
print line
for line in block[1]:
print line
print "---------the end-----------------------"
def main( argv ):
"""main program loop"""
# we begin by simply building a list of DocBlock elements
#
sys.stderr.write( "extracting comment blocks from sources...\n" )
list = make_block_list()
# now, sort the blocks into sections
#
section_list = DocSectionList()
for block in list:
section_list.append_block( block )
section_list.prepare_files( "ft2" )
# dump the section list TOC and sections
#
section_list.dump_html_toc()
section_list.dump_html_sections()
section_list.dump_html_index()
# list2 = filter_blocks( list, ['type','macro','enum','constant','functype'] )
# list2 = list
# list2.sort( block_lexicographical_compare )
# dump_html_1( list2 )
# dump_doc_blocks( list )
# dump_block_lists( list )
# dump_html_1( list )
# If called from the command line
#
if __name__ == '__main__':
main( sys.argv )
# eof