qtbase/util/cmake/qmake_parser.py
Frederik Gladhorn a5060e9f99 cmake scripts: move parser into separate file
The code is nicely separated between parsing and processing. Splitting
that into two files makes it easier to follow which function belongs to
which part.

Change-Id: I576b8613b0d05b2dae3f9c6fa65d9ed5b582a0f7
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
2019-10-10 13:35:43 +00:00

378 lines
14 KiB
Python

#!/usr/bin/env python3
#############################################################################
##
## Copyright (C) 2018 The Qt Company Ltd.
## Contact: https://www.qt.io/licensing/
##
## This file is part of the plugins of the Qt Toolkit.
##
## $QT_BEGIN_LICENSE:GPL-EXCEPT$
## Commercial License Usage
## Licensees holding valid commercial Qt licenses may use this file in
## accordance with the commercial license agreement provided with the
## Software or, alternatively, in accordance with the terms contained in
## a written agreement between you and The Qt Company. For licensing terms
## and conditions see https://www.qt.io/terms-conditions. For further
## information use the contact form at https://www.qt.io/contact-us.
##
## GNU General Public License Usage
## Alternatively, this file may be used under the terms of the GNU
## General Public License version 3 as published by the Free Software
## Foundation with exceptions as appearing in the file LICENSE.GPL3-EXCEPT
## included in the packaging of this file. Please review the following
## information to ensure the GNU General Public License requirements will
## be met: https://www.gnu.org/licenses/gpl-3.0.html.
##
## $QT_END_LICENSE$
##
#############################################################################
import collections
import os
import re
from itertools import chain
import pyparsing as pp # type: ignore
from helper import _set_up_py_parsing_nicer_debug_output
_set_up_py_parsing_nicer_debug_output(pp)
def fixup_linecontinuation(contents: str) -> str:
# Remove all line continuations, aka a backslash followed by
# a newline character with an arbitrary amount of whitespace
# between the backslash and the newline.
# This greatly simplifies the qmake parsing grammar.
contents = re.sub(r"([^\t ])\\[ \t]*\n", "\\1 ", contents)
contents = re.sub(r"\\[ \t]*\n", "", contents)
return contents
def fixup_comments(contents: str) -> str:
# Get rid of completely commented out lines.
# So any line which starts with a '#' char and ends with a new line
# will be replaced by a single new line.
#
# This is needed because qmake syntax is weird. In a multi line
# assignment (separated by backslashes and newlines aka
# # \\\n ), if any of the lines are completely commented out, in
# principle the assignment should fail.
#
# It should fail because you would have a new line separating
# the previous value from the next value, and the next value would
# not be interpreted as a value, but as a new token / operation.
# qmake is lenient though, and accepts that, so we need to take
# care of it as well, as if the commented line didn't exist in the
# first place.
contents = re.sub(r"\n#[^\n]*?\n", "\n", contents, re.DOTALL)
return contents
def flatten_list(l):
""" Flattens an irregular nested list into a simple list."""
for el in l:
if isinstance(el, collections.abc.Iterable) and not isinstance(el, (str, bytes)):
yield from flatten_list(el)
else:
yield el
def handle_function_value(group: pp.ParseResults):
function_name = group[0]
function_args = group[1]
if function_name == "qtLibraryTarget":
if len(function_args) > 1:
raise RuntimeError(
"Don't know what to with more than one function argument "
"for $$qtLibraryTarget()."
)
return str(function_args[0])
if function_name == "quote":
# Do nothing, just return a string result
return str(group)
if function_name == "files":
return str(function_args[0])
if function_name == "basename":
if len(function_args) != 1:
print(f"XXXX basename with more than one argument")
if function_args[0] == '_PRO_FILE_PWD_':
return os.path.basename(os.getcwd())
print(f"XXXX basename with value other than _PRO_FILE_PWD_")
return os.path.basename(str(function_args[0]))
if isinstance(function_args, pp.ParseResults):
function_args = list(flatten_list(function_args.asList()))
# For other functions, return the whole expression as a string.
return f"$${function_name}({' '.join(function_args)})"
class QmakeParser:
def __init__(self, *, debug: bool = False) -> None:
self.debug = debug
self._Grammar = self._generate_grammar()
def _generate_grammar(self):
# Define grammar:
pp.ParserElement.setDefaultWhitespaceChars(" \t")
def add_element(name: str, value: pp.ParserElement):
nonlocal self
if self.debug:
value.setName(name)
value.setDebug()
return value
EOL = add_element("EOL", pp.Suppress(pp.LineEnd()))
Else = add_element("Else", pp.Keyword("else"))
Identifier = add_element(
"Identifier", pp.Word(f"{pp.alphas}_", bodyChars=pp.alphanums + "_-./")
)
BracedValue = add_element(
"BracedValue",
pp.nestedExpr(
ignoreExpr=pp.quotedString
| pp.QuotedString(
quoteChar="$(", endQuoteChar=")", escQuote="\\", unquoteResults=False
)
).setParseAction(lambda s, l, t: ["(", *t[0], ")"]),
)
Substitution = add_element(
"Substitution",
pp.Combine(
pp.Literal("$")
+ (
(
(pp.Literal("$") + Identifier + pp.Optional(pp.nestedExpr()))
| (pp.Literal("(") + Identifier + pp.Literal(")"))
| (pp.Literal("{") + Identifier + pp.Literal("}"))
| (
pp.Literal("$")
+ pp.Literal("{")
+ Identifier
+ pp.Optional(pp.nestedExpr())
+ pp.Literal("}")
)
| (pp.Literal("$") + pp.Literal("[") + Identifier + pp.Literal("]"))
)
)
),
)
LiteralValuePart = add_element(
"LiteralValuePart", pp.Word(pp.printables, excludeChars="$#{}()")
)
SubstitutionValue = add_element(
"SubstitutionValue",
pp.Combine(pp.OneOrMore(Substitution | LiteralValuePart | pp.Literal("$"))),
)
FunctionValue = add_element(
"FunctionValue",
pp.Group(
pp.Suppress(pp.Literal("$") + pp.Literal("$"))
+ Identifier
+ pp.nestedExpr() # .setParseAction(lambda s, l, t: ['(', *t[0], ')'])
).setParseAction(lambda s, l, t: handle_function_value(*t)),
)
Value = add_element(
"Value",
pp.NotAny(Else | pp.Literal("}") | EOL)
+ (
pp.QuotedString(quoteChar='"', escChar="\\")
| FunctionValue
| SubstitutionValue
| BracedValue
),
)
Values = add_element("Values", pp.ZeroOrMore(Value)("value"))
Op = add_element(
"OP", pp.Literal("=") | pp.Literal("-=") | pp.Literal("+=") | pp.Literal("*=") | pp.Literal("~=")
)
Key = add_element("Key", Identifier)
Operation = add_element("Operation", Key("key") + Op("operation") + Values("value"))
CallArgs = add_element("CallArgs", pp.nestedExpr())
def parse_call_args(results):
out = ""
for item in chain(*results):
if isinstance(item, str):
out += item
else:
out += "(" + parse_call_args(item) + ")"
return out
CallArgs.setParseAction(parse_call_args)
Load = add_element("Load", pp.Keyword("load") + CallArgs("loaded"))
Include = add_element("Include", pp.Keyword("include") + CallArgs("included"))
Option = add_element("Option", pp.Keyword("option") + CallArgs("option"))
RequiresCondition = add_element("RequiresCondition", pp.originalTextFor(pp.nestedExpr()))
def parse_requires_condition(s, l, t):
# The following expression unwraps the condition via the additional info
# set by originalTextFor.
condition_without_parentheses = s[t._original_start + 1 : t._original_end - 1]
# And this replaces the colons with '&&' similar how it's done for 'Condition'.
condition_without_parentheses = (
condition_without_parentheses.strip().replace(":", " && ").strip(" && ")
)
return condition_without_parentheses
RequiresCondition.setParseAction(parse_requires_condition)
Requires = add_element(
"Requires", pp.Keyword("requires") + RequiresCondition("project_required_condition")
)
# ignore the whole thing...
DefineTestDefinition = add_element(
"DefineTestDefinition",
pp.Suppress(
pp.Keyword("defineTest")
+ CallArgs
+ pp.nestedExpr(opener="{", closer="}", ignoreExpr=pp.LineEnd())
),
)
# ignore the whole thing...
ForLoop = add_element(
"ForLoop",
pp.Suppress(
pp.Keyword("for")
+ CallArgs
+ pp.nestedExpr(opener="{", closer="}", ignoreExpr=pp.LineEnd())
),
)
# ignore the whole thing...
ForLoopSingleLine = add_element(
"ForLoopSingleLine",
pp.Suppress(pp.Keyword("for") + CallArgs + pp.Literal(":") + pp.SkipTo(EOL)),
)
# ignore the whole thing...
FunctionCall = add_element("FunctionCall", pp.Suppress(Identifier + pp.nestedExpr()))
Scope = add_element("Scope", pp.Forward())
Statement = add_element(
"Statement",
pp.Group(
Load
| Include
| Option
| Requires
| ForLoop
| ForLoopSingleLine
| DefineTestDefinition
| FunctionCall
| Operation
),
)
StatementLine = add_element("StatementLine", Statement + (EOL | pp.FollowedBy("}")))
StatementGroup = add_element(
"StatementGroup", pp.ZeroOrMore(StatementLine | Scope | pp.Suppress(EOL))
)
Block = add_element(
"Block",
pp.Suppress("{")
+ pp.Optional(EOL)
+ StatementGroup
+ pp.Optional(EOL)
+ pp.Suppress("}")
+ pp.Optional(EOL),
)
ConditionEnd = add_element(
"ConditionEnd",
pp.FollowedBy(
(pp.Optional(pp.White()) + (pp.Literal(":") | pp.Literal("{") | pp.Literal("|")))
),
)
ConditionPart1 = add_element(
"ConditionPart1", (pp.Optional("!") + Identifier + pp.Optional(BracedValue))
)
ConditionPart2 = add_element("ConditionPart2", pp.CharsNotIn("#{}|:=\\\n"))
ConditionPart = add_element(
"ConditionPart", (ConditionPart1 ^ ConditionPart2) + ConditionEnd
)
ConditionOp = add_element("ConditionOp", pp.Literal("|") ^ pp.Literal(":"))
ConditionWhiteSpace = add_element(
"ConditionWhiteSpace", pp.Suppress(pp.Optional(pp.White(" ")))
)
ConditionRepeated = add_element(
"ConditionRepeated", pp.ZeroOrMore(ConditionOp + ConditionWhiteSpace + ConditionPart)
)
Condition = add_element("Condition", pp.Combine(ConditionPart + ConditionRepeated))
Condition.setParseAction(lambda x: " ".join(x).strip().replace(":", " && ").strip(" && "))
# Weird thing like write_file(a)|error() where error() is the alternative condition
# which happens to be a function call. In this case there is no scope, but our code expects
# a scope with a list of statements, so create a fake empty statement.
ConditionEndingInFunctionCall = add_element(
"ConditionEndingInFunctionCall",
pp.Suppress(ConditionOp)
+ FunctionCall
+ pp.Empty().setParseAction(lambda x: [[]]).setResultsName("statements"),
)
SingleLineScope = add_element(
"SingleLineScope",
pp.Suppress(pp.Literal(":")) + pp.Group(Block | (Statement + EOL))("statements"),
)
MultiLineScope = add_element("MultiLineScope", Block("statements"))
SingleLineElse = add_element(
"SingleLineElse",
pp.Suppress(pp.Literal(":")) + (Scope | Block | (Statement + pp.Optional(EOL))),
)
MultiLineElse = add_element("MultiLineElse", Block)
ElseBranch = add_element("ElseBranch", pp.Suppress(Else) + (SingleLineElse | MultiLineElse))
# Scope is already add_element'ed in the forward declaration above.
Scope <<= pp.Group(
Condition("condition")
+ (SingleLineScope | MultiLineScope | ConditionEndingInFunctionCall)
+ pp.Optional(ElseBranch)("else_statements")
)
Grammar = StatementGroup("statements")
Grammar.ignore(pp.pythonStyleComment())
return Grammar
def parseFile(self, file: str):
print(f'Parsing "{file}"...')
try:
with open(file, "r") as file_fd:
contents = file_fd.read()
# old_contents = contents
contents = fixup_comments(contents)
contents = fixup_linecontinuation(contents)
result = self._Grammar.parseString(contents, parseAll=True)
except pp.ParseException as pe:
print(pe.line)
print(f"{' ' * (pe.col-1)}^")
print(pe)
raise pe
return result
def parseProFile(file: str, *, debug=False):
parser = QmakeParser(debug=debug)
return parser.parseFile(file)