diff options
| -rw-r--r-- | docs/markdown/Syntax.md | 3 | ||||
| -rw-r--r-- | docs/markdown/snippets/more-escape-sequences.md | 22 | ||||
| -rw-r--r-- | mesonbuild/mparser.py | 16 | ||||
| -rw-r--r-- | test cases/common/32 multiline string/meson.build | 10 | ||||
| -rw-r--r-- | test cases/common/42 string operations/meson.build | 12 |
5 files changed, 30 insertions, 33 deletions
diff --git a/docs/markdown/Syntax.md b/docs/markdown/Syntax.md index 01c8c6e..30eedf8 100644 --- a/docs/markdown/Syntax.md +++ b/docs/markdown/Syntax.md @@ -131,7 +131,8 @@ int main (int argc, char ** argv) { }''' ``` -This can also be combined with the string formatting functionality +These are raw strings that do not support the escape sequences listed above. +These strings can also be combined with the string formatting functionality described below. #### String formatting diff --git a/docs/markdown/snippets/more-escape-sequences.md b/docs/markdown/snippets/more-escape-sequences.md index 2894079..ad78ce2 100644 --- a/docs/markdown/snippets/more-escape-sequences.md +++ b/docs/markdown/snippets/more-escape-sequences.md @@ -1,17 +1,15 @@ -## String escape character update +## String escape character sequence update -The strings (both single-quoted and triple-quoted) in meson has been taught the -same set of escape sequences as in Python. It is therefore now possible to use -arbitrary bytes in strings, like for example NUL (`\0`) and other ASCII control -characters. See the chapter about *Strings* in *Syntax* for more details. +Single-quoted strings in meson have been taught the same set of escape +sequences as in Python. It is therefore now possible to use arbitrary bytes in +strings, like for example `NUL` (`\0`) and other ASCII control characters. See +the chapter about [*Strings* in *Syntax*](Syntax.md#strings) for more +details. Potential backwards compatibility issue: Any valid escape sequence according to the new rules will be interpreted as an escape sequence instead of the literal -characters. Previously only single-quote strings supported escape sequences and -the supported sequences were `\'`, `\\` and `\n`. +characters. Previously only the following escape sequences were supported in +single-quote strings: `\'`, `\\` and `\n`. -The most likely breakage is usage of backslash-n in triple-quoted strings. It -is now written in the same way as in single-quoted strings: `\\n` instead of -`\n`. In general it is now recommended to escape any usage of backslash. -However, backslash-c (`\c`), for example, is still backslash-c because it isn't -a valid escape sequence. +Note that the behaviour of triple-quoted (multiline) strings has not changed. +They behave like raw strings and do not support any escape sequences. diff --git a/mesonbuild/mparser.py b/mesonbuild/mparser.py index 9e43065..8cef377 100644 --- a/mesonbuild/mparser.py +++ b/mesonbuild/mparser.py @@ -28,18 +28,6 @@ ESCAPE_SEQUENCE_SINGLE_RE = re.compile(r''' | \\[\\'abfnrtv] # Single-character escapes )''', re.UNICODE | re.VERBOSE) -# This is the regex for the supported escape sequences of a multiline string -# literal, like '''abc\x00'''. The only difference is that single quote (') -# doesn't require escaping. -ESCAPE_SEQUENCE_MULTI_RE = re.compile(r''' - ( \\U........ # 8-digit hex escapes - | \\u.... # 4-digit hex escapes - | \\x.. # 2-digit hex escapes - | \\[0-7]{1,3} # Octal escapes - | \\N\{[^}]+\} # Unicode characters by name - | \\[\\abfnrtv] # Single-character escapes - )''', re.UNICODE | re.VERBOSE) - class MesonUnicodeDecodeError(MesonException): def __init__(self, match): super().__init__("%s" % match) @@ -187,10 +175,6 @@ This will become a hard error in a future Meson release.""", self.getline(line_s elif tid == 'multiline_string': tid = 'string' value = match_text[3:-3] - try: - value = ESCAPE_SEQUENCE_MULTI_RE.sub(decode_match, value) - except MesonUnicodeDecodeError as err: - raise MesonException("Failed to parse escape sequence: '{}' in string:\n{}".format(err.match, match_text)) lines = match_text.split('\n') if len(lines) > 1: lineno += len(lines) - 1 diff --git a/test cases/common/32 multiline string/meson.build b/test cases/common/32 multiline string/meson.build index 1f952f1..262cb15 100644 --- a/test cases/common/32 multiline string/meson.build +++ b/test cases/common/32 multiline string/meson.build @@ -23,3 +23,13 @@ quote2 = '\'' if quote1 != quote2 error('Single quote quoting is broken.') endif + +cc = meson.get_compiler('c') +prog = ''' +int main(int argc, char **argv) { + int num = 1; + printf("%d\n", num); + return 0; +}''' + +assert(cc.compiles(prog), 'multline test compile failed') diff --git a/test cases/common/42 string operations/meson.build b/test cases/common/42 string operations/meson.build index 1c289eb..6596142 100644 --- a/test cases/common/42 string operations/meson.build +++ b/test cases/common/42 string operations/meson.build @@ -78,15 +78,17 @@ assert('"1.1.20"'.strip('".') == '1.1.20', '". badly stripped') assert('"1.1.20" '.strip('" ') == '1.1.20', '". badly stripped') bs_c = '''\c''' -bs_bs_c = '''\\\c''' +bs_bs_c = '''\\c''' nl = ''' ''' -bs_n = '''\\n''' +bs_n = '''\n''' bs_nl = '''\ ''' -bs_bs_n = '''\\\\n''' -bs_bs_nl = '''\\\\ +bs_bs_n = '''\\n''' +bs_bs_nl = '''\\ ''' +bs_bs = '''\\''' +bs = '''\''' assert('\c' == bs_c, 'Single backslash broken') assert('\\c' == bs_c, 'Double backslash broken') @@ -97,3 +99,5 @@ assert('\\n' == bs_n, 'Double backslash broken before n') assert('\\\n' == bs_nl, 'Three backslash broken before n') assert('\\\\n' == bs_bs_n, 'Four backslash broken before n') assert('\\\\\n' == bs_bs_nl, 'Five backslash broken before n') +assert('\\\\' == bs_bs, 'Double-backslash broken') +assert('\\' == bs, 'Backslash broken') |