aboutsummaryrefslogtreecommitdiffstats
path: root/Tools/d_scan.py
blob: 14c6c313e9a16347f382c6237f9318556a38f071 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
#!/usr/bin/env python
# encoding: utf-8
# Thomas Nagy, 2016-2018 (ita)

"""
Provide a scanner for finding dependencies on d files
"""

import re
from waflib import Utils

def filter_comments(filename):
	"""
	:param filename: d file name
	:type filename: string
	:rtype: list
	:return: a list of characters
	"""
	txt = Utils.readf(filename)
	i = 0
	buf = []
	max = len(txt)
	begin = 0
	while i < max:
		c = txt[i]
		if c == '"' or c == "'":  # skip a string or character literal
			buf.append(txt[begin:i])
			delim = c
			i += 1
			while i < max:
				c = txt[i]
				if c == delim:
					break
				elif c == '\\':  # skip the character following backslash
					i += 1
				i += 1
			i += 1
			begin = i
		elif c == '/':  # try to replace a comment with whitespace
			buf.append(txt[begin:i])
			i += 1
			if i == max:
				break
			c = txt[i]
			if c == '+':  # eat nesting /+ +/ comment
				i += 1
				nesting = 1
				c = None
				while i < max:
					prev = c
					c = txt[i]
					if prev == '/' and c == '+':
						nesting += 1
						c = None
					elif prev == '+' and c == '/':
						nesting -= 1
						if nesting == 0:
							break
						c = None
					i += 1
			elif c == '*':  # eat /* */ comment
				i += 1
				c = None
				while i < max:
					prev = c
					c = txt[i]
					if prev == '*' and c == '/':
						break
					i += 1
			elif c == '/':  # eat // comment
				i += 1
				while i < max and txt[i] != '\n':
					i += 1
			else:  # no comment
				begin = i - 1
				continue
			i += 1
			begin = i
			buf.append(' ')
		else:
			i += 1
	buf.append(txt[begin:])
	return buf

class d_parser(object):
	"""
	Parser for d files
	"""
	def __init__(self, env, incpaths):
		#self.code = ''
		#self.module = ''
		#self.imports = []

		self.allnames = []

		self.re_module = re.compile("module\s+([^;]+)")
		self.re_import = re.compile("import\s+([^;]+)")
		self.re_import_bindings = re.compile("([^:]+):(.*)")
		self.re_import_alias = re.compile("[^=]+=(.+)")

		self.env = env

		self.nodes = []
		self.names = []

		self.incpaths = incpaths

	def tryfind(self, filename):
		"""
		Search file a file matching an module/import directive

		:param filename: file to read
		:type filename: string
		"""
		found = 0
		for n in self.incpaths:
			found = n.find_resource(filename.replace('.', '/') + '.d')
			if found:
				self.nodes.append(found)
				self.waiting.append(found)
				break
		if not found:
			if not filename in self.names:
				self.names.append(filename)

	def get_strings(self, code):
		"""
		:param code: d code to parse
		:type code: string
		:return: the modules that the code uses
		:rtype: a list of match objects
		"""
		#self.imports = []
		self.module = ''
		lst = []

		# get the module name (if present)

		mod_name = self.re_module.search(code)
		if mod_name:
			self.module = re.sub('\s+', '', mod_name.group(1)) # strip all whitespaces

		# go through the code, have a look at all import occurrences

		# first, lets look at anything beginning with "import" and ending with ";"
		import_iterator = self.re_import.finditer(code)
		if import_iterator:
			for import_match in import_iterator:
				import_match_str = re.sub('\s+', '', import_match.group(1)) # strip all whitespaces

				# does this end with an import bindings declaration?
				# (import bindings always terminate the list of imports)
				bindings_match = self.re_import_bindings.match(import_match_str)
				if bindings_match:
					import_match_str = bindings_match.group(1)
					# if so, extract the part before the ":" (since the module declaration(s) is/are located there)

				# split the matching string into a bunch of strings, separated by a comma
				matches = import_match_str.split(',')

				for match in matches:
					alias_match = self.re_import_alias.match(match)
					if alias_match:
						# is this an alias declaration? (alias = module name) if so, extract the module name
						match = alias_match.group(1)

					lst.append(match)
		return lst

	def start(self, node):
		"""
		The parsing starts here

		:param node: input file
		:type node: :py:class:`waflib.Node.Node`
		"""
		self.waiting = [node]
		# while the stack is not empty, add the dependencies
		while self.waiting:
			nd = self.waiting.pop(0)
			self.iter(nd)

	def iter(self, node):
		"""
		Find all the modules that a file depends on, uses :py:meth:`waflib.Tools.d_scan.d_parser.tryfind` to process dependent files

		:param node: input file
		:type node: :py:class:`waflib.Node.Node`
		"""
		path = node.abspath() # obtain the absolute path
		code = "".join(filter_comments(path)) # read the file and filter the comments
		names = self.get_strings(code) # obtain the import strings
		for x in names:
			# optimization
			if x in self.allnames:
				continue
			self.allnames.append(x)

			# for each name, see if it is like a node or not
			self.tryfind(x)

def scan(self):
	"look for .d/.di used by a d file"
	env = self.env
	gruik = d_parser(env, self.generator.includes_nodes)
	node = self.inputs[0]
	gruik.start(node)
	nodes = gruik.nodes
	names = gruik.names
	return (nodes, names)