summaryrefslogtreecommitdiffstats
path: root/bitbake/lib/bb/codeparser.py
blob: 9a692a0c8eb0d76be6600af69ae87a65bcc992c4 (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
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
import ast
import codegen
import logging
import os.path
import bb.utils, bb.data
from itertools import chain
from pysh import pyshyacc, pyshlex, sherrors


logger = logging.getLogger('BitBake.CodeParser')
PARSERCACHE_VERSION = 2

try:
    import cPickle as pickle
except ImportError:
    import pickle
    logger.info('Importing cPickle failed.  Falling back to a very slow implementation.')


def check_indent(codestr):
    """If the code is indented, add a top level piece of code to 'remove' the indentation"""

    i = 0
    while codestr[i] in ["\n", "\t", " "]:
        i = i + 1

    if i == 0:
        return codestr

    if codestr[i-1] == "\t" or codestr[i-1] == " ":
        return "if 1:\n" + codestr

    return codestr

pythonparsecache = {}
shellparsecache = {}

def parser_cachefile(d):
    cachedir = (bb.data.getVar("PERSISTENT_DIR", d, True) or
                bb.data.getVar("CACHE", d, True))
    if cachedir in [None, '']:
        return None
    bb.utils.mkdirhier(cachedir)
    cachefile = os.path.join(cachedir, "bb_codeparser.dat")
    logger.debug(1, "Using cache in '%s' for codeparser cache", cachefile)
    return cachefile

def parser_cache_init(d):
    global pythonparsecache
    global shellparsecache

    cachefile = parser_cachefile(d)
    if not cachefile:
        return

    try:
        p = pickle.Unpickler(file(cachefile, "rb"))
        data, version = p.load()
    except:
        return

    if version != PARSERCACHE_VERSION:
        return

    pythonparsecache = data[0]
    shellparsecache = data[1]

def parser_cache_save(d):
    cachefile = parser_cachefile(d)
    if not cachefile:
        return

    glf = bb.utils.lockfile(cachefile + ".lock", shared=True)

    i = os.getpid()
    lf = None
    while not lf:
        shellcache = {}
        pythoncache = {}

        lf = bb.utils.lockfile(cachefile + ".lock." + str(i), retry=False)
        if not lf or os.path.exists(cachefile + "-" + str(i)):
            if lf:
               bb.utils.unlockfile(lf) 
               lf = None
            i = i + 1
            continue

        try:
            p = pickle.Unpickler(file(cachefile, "rb"))
            data, version = p.load()
        except (IOError, EOFError, ValueError):
            data, version = None, None

        if version != PARSERCACHE_VERSION:
            shellcache = shellparsecache
            pythoncache = pythonparsecache
        else:
            for h in pythonparsecache:
                if h not in data[0]:
                    pythoncache[h] = pythonparsecache[h]
            for h in shellparsecache:
                if h not in data[1]:
                    shellcache[h] = shellparsecache[h]

        p = pickle.Pickler(file(cachefile + "-" + str(i), "wb"), -1)
        p.dump([[pythoncache, shellcache], PARSERCACHE_VERSION])

    bb.utils.unlockfile(lf)
    bb.utils.unlockfile(glf)

def parser_cache_savemerge(d):
    cachefile = parser_cachefile(d)
    if not cachefile:
        return

    glf = bb.utils.lockfile(cachefile + ".lock")

    try:
        p = pickle.Unpickler(file(cachefile, "rb"))
        data, version = p.load()
    except (IOError, EOFError):
        data, version = None, None

    if version != PARSERCACHE_VERSION:
        data = [{}, {}]

    for f in [y for y in os.listdir(os.path.dirname(cachefile)) if y.startswith(os.path.basename(cachefile) + '-')]:
        f = os.path.join(os.path.dirname(cachefile), f)
        try:
            p = pickle.Unpickler(file(f, "rb"))
            extradata, version = p.load()
        except (IOError, EOFError):
            extradata, version = [{}, {}], None
        
        if version != PARSERCACHE_VERSION:
            continue

        for h in extradata[0]:
            if h not in data[0]:
                data[0][h] = extradata[0][h]
        for h in extradata[1]:
            if h not in data[1]:
                data[1][h] = extradata[1][h]
        os.unlink(f)

    p = pickle.Pickler(file(cachefile, "wb"), -1)
    p.dump([data, PARSERCACHE_VERSION])

    bb.utils.unlockfile(glf)


class PythonParser():
    getvars = ("d.getVar", "bb.data.getVar", "data.getVar")
    expands = ("d.expand", "bb.data.expand", "data.expand")
    execfuncs = ("bb.build.exec_func", "bb.build.exec_task")

    def warn(self, func, arg):
        """Warn about calls of bitbake APIs which pass a non-literal
        argument for the variable name, as we're not able to track such
        a reference.
        """

        try:
            funcstr = codegen.to_source(func)
            argstr = codegen.to_source(arg)
        except TypeError:
            logger.debug(2, 'Failed to convert function and argument to source form')
        else:
            logger.debug(1, self.unhandled_message % (funcstr, argstr))

    def visit_Call(self, node):
        name = self.called_node_name(node.func)
        if name in self.getvars:
            if isinstance(node.args[0], ast.Str):
                self.var_references.add(node.args[0].s)
            else:
                self.warn(node.func, node.args[0])
        elif name in self.expands:
            if isinstance(node.args[0], ast.Str):
                self.warn(node.func, node.args[0])
                self.var_expands.add(node.args[0].s)
            elif isinstance(node.args[0], ast.Call) and \
                 self.called_node_name(node.args[0].func) in self.getvars:
                pass
            else:
                self.warn(node.func, node.args[0])
        elif name in self.execfuncs:
            if isinstance(node.args[0], ast.Str):
                self.var_execs.add(node.args[0].s)
            else:
                self.warn(node.func, node.args[0])
        elif name and isinstance(node.func, (ast.Name, ast.Attribute)):
            self.execs.add(name)

    def called_node_name(self, node):
        """Given a called node, return its original string form"""
        components = []
        while node:
            if isinstance(node, ast.Attribute):
                components.append(node.attr)
                node = node.value
            elif isinstance(node, ast.Name):
                components.append(node.id)
                return '.'.join(reversed(components))
            else:
                break

    def __init__(self, name):
        self.var_references = set()
        self.var_execs = set()
        self.execs = set()
        self.var_expands = set()
        self.references = set()

        self.unhandled_message = "in call of %s, argument '%s' is not a string literal"
        self.unhandled_message = "while parsing %s, %s" % (name, self.unhandled_message)

    def parse_python(self, node):
        h = hash(str(node))

        if h in pythonparsecache:
            self.references = pythonparsecache[h]["refs"]
            self.execs = pythonparsecache[h]["execs"]
            return

        code = compile(check_indent(str(node)), "<string>", "exec",
                       ast.PyCF_ONLY_AST)

        for n in ast.walk(code):
            if n.__class__.__name__ == "Call":
                self.visit_Call(n)

        self.references.update(self.var_references)
        self.references.update(self.var_execs)

        pythonparsecache[h] = {}
        pythonparsecache[h]["refs"] = self.references
        pythonparsecache[h]["execs"] = self.execs

class ShellParser():
    def __init__(self, name):
        self.funcdefs = set()
        self.allexecs = set()
        self.execs = set()
        self.unhandled_template = "unable to handle non-literal command '%s'"
        self.unhandled_template = "while parsing %s, %s" % (name, self.unhandled_template)

    def parse_shell(self, value):
        """Parse the supplied shell code in a string, returning the external
        commands it executes.
        """

        h = hash(str(value))

        if h in shellparsecache:
            self.execs = shellparsecache[h]["execs"]
            return self.execs

        try:
            tokens, _ = pyshyacc.parse(value, eof=True, debug=False)
        except pyshlex.NeedMore:
            raise sherrors.ShellSyntaxError("Unexpected EOF")

        for token in tokens:
            self.process_tokens(token)
        self.execs = set(cmd for cmd in self.allexecs if cmd not in self.funcdefs)

        shellparsecache[h] = {}
        shellparsecache[h]["execs"] = self.execs

        return self.execs

    def process_tokens(self, tokens):
        """Process a supplied portion of the syntax tree as returned by
        pyshyacc.parse.
        """

        def function_definition(value):
            self.funcdefs.add(value.name)
            return [value.body], None

        def case_clause(value):
            # Element 0 of each item in the case is the list of patterns, and
            # Element 1 of each item in the case is the list of commands to be
            # executed when that pattern matches.
            words = chain(*[item[0] for item in value.items])
            cmds  = chain(*[item[1] for item in value.items])
            return cmds, words

        def if_clause(value):
            main = chain(value.cond, value.if_cmds)
            rest = value.else_cmds
            if isinstance(rest, tuple) and rest[0] == "elif":
                return chain(main, if_clause(rest[1]))
            else:
                return chain(main, rest)

        def simple_command(value):
            return None, chain(value.words, (assign[1] for assign in value.assigns))

        token_handlers = {
            "and_or": lambda x: ((x.left, x.right), None),
            "async": lambda x: ([x], None),
            "brace_group": lambda x: (x.cmds, None),
            "for_clause": lambda x: (x.cmds, x.items),
            "function_definition": function_definition,
            "if_clause": lambda x: (if_clause(x), None),
            "pipeline": lambda x: (x.commands, None),
            "redirect_list": lambda x: ([x.cmd], None),
            "subshell": lambda x: (x.cmds, None),
            "while_clause": lambda x: (chain(x.condition, x.cmds), None),
            "until_clause": lambda x: (chain(x.condition, x.cmds), None),
            "simple_command": simple_command,
            "case_clause": case_clause,
        }

        for token in tokens:
            name, value = token
            try:
                more_tokens, words = token_handlers[name](value)
            except KeyError:
                raise NotImplementedError("Unsupported token type " + name)

            if more_tokens:
                self.process_tokens(more_tokens)

            if words:
                self.process_words(words)

    def process_words(self, words):
        """Process a set of 'words' in pyshyacc parlance, which includes
        extraction of executed commands from $() blocks, as well as grabbing
        the command name argument.
        """

        words = list(words)
        for word in list(words):
            wtree = pyshlex.make_wordtree(word[1])
            for part in wtree:
                if not isinstance(part, list):
                    continue

                if part[0] in ('`', '$('):
                    command = pyshlex.wordtree_as_string(part[1:-1])
                    self.parse_shell(command)

                    if word[0] in ("cmd_name", "cmd_word"):
                        if word in words:
                            words.remove(word)

        usetoken = False
        for word in words:
            if word[0] in ("cmd_name", "cmd_word") or \
               (usetoken and word[0] == "TOKEN"):
                if "=" in word[1]:
                    usetoken = True
                    continue

                cmd = word[1]
                if cmd.startswith("$"):
                    logger.debug(1, self.unhandled_template % cmd)
                elif cmd == "eval":
                    command = " ".join(word for _, word in words[1:])
                    self.parse_shell(command)
                else:
                    self.allexecs.add(cmd)
                break