Commit a7c7c503 authored by ASN.1 Documenter's avatar ASN.1 Documenter
Browse files

fix Categories parse

parent 8f81c29a

asn2md.py

+47 −45
Original line number Diff line number Diff line
@@ -5,67 +5,70 @@ import argparse # parse arguments 
import os.path  # getting extension from file

import sys      # output and stuff

import re       # for regular expressions

import urllib   #



## extract doxygen-tag namespace

RE_MODULE = re.compile(

	r'^(\w[\w-]*)[\s{](?:.|\n)*DEFINITIONS(?:.|\n)*::=\s*BEGIN'

	, re.IGNORECASE | re.VERBOSE | re.DOTALL | re.MULTILINE)

RE_EXPORTS = re.compile(

	r'^\s*EXPORTS .*?;' 

	, re.VERBOSE | re.DOTALL | re.MULTILINE)

RE_MODULE   = re.compile( r'^\s*([A-Z][\w-]*)\s*({.*?})?\s*DEFINITIONS.*?::=\s*?BEGIN(.*)END', re.VERBOSE | re.MULTILINE | re.DOTALL)



RE_IMPORTS = re.compile(

	r'^\s*IMPORTS .*?;' 

	, re.VERBOSE | re.DOTALL | re.MULTILINE)

RE_SPACES   = re.compile(r'\s+')



RE_END = re.compile(

	r'^\s*END\s*$' 

	, re.VERBOSE | re.MULTILINE)

RE_COMMENTS = re.compile(r'^\s*--.*\n', re.MULTILINE)



RE_DOXY_COMMENTS = re.compile(

	r'^\s*--[-#!]\s'

	, re.VERBOSE | re.MULTILINE)

RE_BASIC_TYPES = re.compile(r'^OCTET\s+STRING|BIT\s+STRING|BOOLEAN|INTEGER|FLOAT')



RE_DOXY_C_COMMENTS = re.compile(

	r'^\s*/\*\*\s(.*?)\*/'

	, re.VERBOSE | re.MULTILINE | re.DOTALL)

RE_FIELDS = re.compile(r'^\s*([\w-]+?)\s+(OCTET\s+STRING|BIT\s+STRING|[A-Z][.\w-]+)?(.*?)(?:,|$)', re.MULTILINE | re.DOTALL)



RE_COMMENT = re.compile(

	r'--.*?(?:--|\n)'

	, re.VERBOSE | re.MULTILINE)

RE_EXPORTS = re.compile(r'^\s*EXPORTS.*?;', re.DOTALL | re.MULTILINE)



RE_TYPE = re.compile(r'(([A-Z][\w-]*)\s*::=[\w \t]+(?:{(.*?)})?.*?)\n\s*\n'

	, re.VERBOSE | re.MULTILINE | re.DOTALL)

RE_IMPORTS = re.compile(r'^\s*IMPORTS\s*(.*?);', re.DOTALL | re.MULTILINE)



extTypes = {}

RE_IMPORT_ELEMENTS = re.compile(r'^([,\s\w-]*?)FROM\s*([\w-]+)\s*({[^}]*}(?:\s+WITH\s+SUCCESSORS)?)?', re.MULTILINE)



RE_IMPORT_ELEMENT_TYPE = re.compile(r'[^,\s]+')



RE_DOXY_COMMENTS = re.compile(r'^\s*--[-!#](:?$|\s(.*))', re.MULTILINE)



RE_DOXY_C_COMMENTS = re.compile(r'^\s*/\*\*\s(.*?)\*/', re.MULTILINE | re.DOTALL)



RE_DOXY_C_COMMENTS_I = re.compile(r'\s*\*')



def indentString(s, indent):

	lines = s.splitlines()

	s = ''

	for c in lines:

		s += ''.ljust(indent, ' ') + c + '\n'

	return s

RE_DOXY_REF = re.compile(r'@ref\s+([\w-]+)')

RE_DOXY_CLASS = re.compile(r'@class\s+([\w-]+)')

RE_DOXY_BRIEF = re.compile(r'@brief\s+')



RE_TYPE = re.compile(r'(([A-Z][\w-]*)\s*::=[\w \t]+(?:{(.*?)})?.*?)\n\s*\n', re.MULTILINE | re.DOTALL)



extTypes = {}



def parseText(content, indent=None):

	# we need to keep only documenting text under the /** */ and --!

	c = re.sub(r'^\s*--[-!#](:?$|(.*))', '/** \g<1>*/', content, 0, re.MULTILINE)

	# convert --! comments to C-style ones

	c = RE_DOXY_COMMENTS.sub(r'/** \g<1>*/', content)

	if c is not None:

		content = c

	ret = ''

	for m in RE_DOXY_C_COMMENTS.finditer(content):

		lines = m.group(1).splitlines()

		for l in lines:

			l = re.sub(r'\s*\*', '', l, 1).rstrip()

			l = RE_DOXY_C_COMMENTS_I.sub('', l, 1).rstrip()

			ret += ''.ljust(indent or 0) + l + '\n'

	def repl_ref(m):

		return '[**{0}**]({1}#{0})'.format(m.group(1), extTypes.get(m.group(1),''))

	c = re.sub(r'@ref\s+([\w-]+)', repl_ref, ret, 0, re.VERBOSE | re.MULTILINE)

	c = RE_DOXY_REF.sub(repl_ref, ret)

	if c is not None:

		ret = c

	c = re.sub(r'@class\s+([\w-]+)', '', ret, 0, re.VERBOSE | re.MULTILINE)

	c = RE_DOXY_CLASS.sub('', ret)

	if c is not None:

		ret = c

	c = re.sub(r'@brief\s+', '', ret, 0, re.VERBOSE | re.MULTILINE)

	c = RE_DOXY_BRIEF.sub('', ret)

	if c is not None:

		ret = c

	def repl_category(m):

		ret = '\n&nbsp;&nbsp;&nbsp;&nbsp;Categories: '

		for l in m.group(1).split(','):

			ret += '[{0}](#{1}) '.format(l.strip(), urllib.quote_plus(l.strip()))

		return ret + '\n\n'

	c = re.sub(r'@category:\s+(.+)', repl_category, ret, re.MULTILINE)

	if c is not None:

		ret = c

	return ret
@@ -74,23 +77,23 @@ def parseModule(mname, content): 
	global cpos

	cpos = 0

	ret = ''

	m = re.search(r'^\s*IMPORTS\s*(.*?);', content, re.VERBOSE | re.DOTALL | re.MULTILINE)

	m = RE_IMPORTS.search(content)

	if m is not None:

		pos = 0

		if m.group(1) is not None:

			ret += '## Imports:\n'

			s = m.group(1)

			for fm in re.finditer(r'^([,\s\w-]*?)FROM\s*([\w-]+)\s*({[^}]*}(?:\s+WITH\s+SUCCESSORS)?)?', s, re.VERBOSE | re.MULTILINE):

			for fm in RE_IMPORT_ELEMENTS.finditer(s):

				imName = fm.group(2)

				for im in re.finditer(r'[^,\s]+', fm.group(1), 0):

				for im in RE_IMPORT_ELEMENT_TYPE.finditer(fm.group(1)):

					extTypes[im.group(0)] = imName+'.md'

				ret += ' * **{}** *{}*<br/>\n'.format(imName, re.sub(r'\s+', ' ', fm.group(3) or '', 0, 0))

				ret += ' * **{}** *{}*<br/>\n'.format(imName, RE_SPACES.sub(' ', fm.group(3) or ''))

				ret += parseText(s[pos:fm.start()], 3)+'\n'

				pos = fm.end()

			ret += parseText(s[pos:])

		cpos = m.end()



	m = re.search(r'^\s*EXPORTS.*?;', content, re.VERBOSE | re.DOTALL | re.MULTILINE)

	m = RE_EXPORTS.search(content)

	if m is not None:

		if cpos < m.end():

			cpos = m.end()
@@ -104,18 +107,18 @@ def parseModule(mname, content): 
		if m.group(3) is not None:

			ret += 'Fields:\n'

			pos = 0

			for fm in re.finditer('^\s*([\w-]+?)\s+(OCTET\s+STRING|BIT\s+STRING|[A-Z][.\w-]+)?(.*?)(?:,|$)', m.group(3), re.VERBOSE | re.MULTILINE | re.DOTALL):

			for fm in RE_FIELDS.finditer(m.group(3)):

				if fm.group(2) is not None:

					f = fm.group(1).strip()

					t = fm.group(2).strip()

					if re.match('^OCTET\s+STRING|BIT\s+STRING|BOOLEAN|INTEGER|FLOAT', t, re.VERBOSE|re.MULTILINE) is not None:

					if RE_BASIC_TYPES.match(t) is not None:

						ret += '* {0} **{1}** {2}<br>\n'.format(f, t, fm.group(3) or '')

					else:

						ret += '* {0} [**{1}**]({2}#{1}) {3}<br>\n'.format(f, t, extTypes.get(t,''), fm.group(3) or '')

				ret += parseText(fm.string[pos:fm.start()], 3)

				pos = fm.end()

		

		return ret + '```asn1\n' + re.sub(r'^\s*--.*\n', '', m.group(1).strip(), 0, re.MULTILINE) +'\n```\n\n'

		return ret + '```asn1\n' + RE_COMMENTS.sub('', m.group(1).strip()) +'\n```\n\n'



	pos = 0

	for m in RE_TYPE.finditer(content[cpos:]):
@@ -128,10 +131,9 @@ def parseModule(mname, content): 
def parseAsn(outDir, content) :

	# iterate modules in the file

	pos= 0

	RE_MODULE=re.compile(r'^\s*([A-Z][\w-]*)\s*({.*?})?\s*DEFINITIONS.*?::=\s*?BEGIN(.*)END', re.VERBOSE | re.MULTILINE | re.DOTALL)

	cnt = 0

	for m in RE_MODULE.finditer(content):

		ret = '# ASN.1 module {}\n OID: _{}_\n'.format(m.group(1), re.sub(r'\s+', ' ', m.group(2)))

		ret = '# ASN.1 module {}\n OID: _{}_\n'.format(m.group(1), RE_SPACES.sub(' ', m.group(2)))

		ret += parseText(content[pos:m.start()]) + '\n'

		if m.group(3) is not None:

			ret += parseModule(m.group(1), m.group(3))