Fix invalid and accidental escape sequences #2112

Avasam · 2023-08-22T22:20:52Z

Follow-up to #2045 and improved over #2100 .
The first commit was done by running ruff . --select=UP025 --exclude=Pythonwin/Scintilla --exclude=adodbapi --fix
The second one was done manually.

mhammond

Thanks, but while fixing invalid escapes are fine I don't see why we should avoid raw strings.

mhammond · 2023-09-21T15:03:49Z

setup.py

@@ -213,7 +213,7 @@ def finalize_options(self, build_ext):
            found_mfc = False
            for incl in os.environ.get("INCLUDE", "").split(os.pathsep):
                # first is a "standard" MSVC install, second is the Vista SDK.
-                for candidate in (r"..\src\occimpl.h", r"..\..\src\mfc\occimpl.h"):
+                for candidate in ("..\\src\\occimpl.h", "..\\..\\src\\mfc\\occimpl.h"):


using raw strings before was fine, right? I see no reason to avoid raw strings.

Just out of styling/consistency between using r-strings or \\ for paths strings. If you'd prefer, I can revert those specific manual changes. (where I made non-regex r-strings regular strings for consistency purpose)

I don't see a need to enforce consistency here - IMO r-strings exist purely because consistency is over-rated :)

I just remembered the other reason, it was visual (though it'll slightly differ per editor)
Looks like we're trying to escape some characters (for use in a regex)

vs

but yeah that's minor anyway, I'm reverting atm to reduce changes

huh, I see that in vscode too. That's gotta be a bug though, right?

This would be the intent:

To let you know that those characters will be escaped when used as part of a regex string.
https://docs.python.org/3/howto/regex.html#the-backslash-plague

That doesn't make sense to me. The link you pointed to basically says "if you don't want to double up your backslashes when matching them in a regex, use a raw string" - it doesn't seem to come close to saying "raw strings should only be used when you want to avoid using 4 backslashes in a regex". Eg, https://docs.python.org/3/reference/lexical_analysis.html says just "such strings are called raw strings and treat backslashes as literal characters." and makes no reference to regular expressions.

Not saying it can't be used for anything else than regex strings.
Just to answer your comment about That's gotta be a bug though, right?. It's intended semantic highlighting on microsoft's part to help with "the backslash plague" when using regex. Similarly to how it's done for JS/TS /regex/ strings

The highlight can be ignored if you use the r-string for other purposes (ie: writing raw text without escape characters)

setup.py

…id-escape-sequences

Avasam · 2023-09-21T18:33:27Z

GitHub isn't showing my new commits. But it's definitely in the branch https://github.com/Avasam/pywin32/commits/invalid-escape-sequences

Maybe a temporary GitHub issue?

mhammond · 2023-09-22T14:36:09Z

com/win32com/client/__init__.py

@@ -61,7 +61,7 @@ def GetObject(Pathname=None, Class=None, clsctx=None):
    ob = GetObject(Class = "ProgID") or GetObject(Class = clsid) will
    connect to an already running instance of the COM object.

-    ob = GetObject(r"c:\blah\blah\foo.xls") (aka the COM moniker syntax)
+    ob = GetObject(r"c:\\blah\\blah\\foo.xls") (aka the COM moniker syntax)


this change is wrong.

The docstring itself is invalid, this changes fixes it.

Before:

>>> from win32com.client import GetObject >>> print(GetObject.__doc__) Mimic VB's GetObject() function. ob = GetObject(Class = "ProgID") or GetObject(Class = clsid) will connect to an already running instance of the COM object. ob = GetObject(r"clalah♀oo.xls") (aka the COM moniker syntax) will return a ready to use Python wrapping of the required COM object. Note: You must specifiy one or the other of these arguments. I know this isn't pretty, but it is what VB does. Blech. If you don't I'll throw ValueError at you. :) This will most likely throw pythoncom.com_error if anything fails.

After:

>>> from win32com.client import GetObject >>> print(GetObject.__doc__) Mimic VB's GetObject() function. ob = GetObject(Class = "ProgID") or GetObject(Class = clsid) will connect to an already running instance of the COM object. ob = GetObject(r"c:\blah\blah\foo.xls") (aka the COM moniker syntax) will return a ready to use Python wrapping of the required COM object. Note: You must specifiy one or the other of these arguments. I know this isn't pretty, but it is what VB does. Blech. If you don't I'll throw ValueError at you. :) This will most likely throw pythoncom.com_error if anything fails.

another option is to make the docstring a raw-multiline-string:

>>> from win32com.client import GetObject >>> print(GetObject.__doc__) Mimic VB's GetObject() function. ob = GetObject(Class = "ProgID") or GetObject(Class = clsid) will connect to an already running instance of the COM object. ob = GetObject(r"c:\blah\blah\foo.xls") (aka the COM moniker syntax) will return a ready to use Python wrapping of the required COM object. Note: You must specifiy one or the other of these arguments. I know this isn't pretty, but it is what VB does. Blech. If you don't I'll throw ValueError at you. :) This will most likely throw pythoncom.com_error if anything fails.

Right - let's to the raw multiline doc string then as otherwise the code just looks wrong when read inline.

Updated for all changes. This PR is ready for a new review

mhammond · 2023-09-22T14:37:12Z

com/win32comext/authorization/demos/EditServiceSecurity.py

@@ -1,7 +1,7 @@
 """
 Implements a permissions editor for services.
 Service can be specified as plain name for local machine,
-or as a remote service of the form \\machinename\service
+or as a remote service of the form \\\\machinename\\service


I think this is wrong too - it doesn't even pretend to be a string (ie, it's not quoted), it's a textual description of how Windows represents UNC paths, which is wrong with the slashes doubled.

(same comment about docstring)
\s is an invalid escape sequence in a string (this is a multiline string)
Before:

>>> from win32comext.authorization.demos import EditServiceSecurity >>> print(EditServiceSecurity.__doc__) Implements a permissions editor for services. Service can be specified as plain name for local machine, or as a remote service of the form \machinename\service

After:

>>> from win32comext.authorization.demos import EditServiceSecurity >>> print(EditServiceSecurity.__doc__) Implements a permissions editor for services. Service can be specified as plain name for local machine, or as a remote service of the form \\machinename\service

As for the path itself, I understand it's supposed to be like network paths?

mhammond · 2023-09-22T14:39:01Z

win32/scripts/regsetup.py

@@ -506,9 +506,9 @@ def RegisterShellInfo(searchPaths):
 as well as the 2 wierd spots to locate the core Python files (eg, Python.exe,
 pythonXX.dll, the standard library and Win32 Extensions.

-"regsetup -a myappname . .\subdir"
+"regsetup -a myappname . .\\subdir"


This is an example of a command-line where the doubling is wrong too - perhaps the best fix here is to make that cleaner - eg, drop the quotes and add, say, % or c:\> as a prefix or similar?

I don't mind if you suggest a line replacement, and/or I can make this a raw-multiline-string and drop all the double slashes in this string from line 502-525

yeah, as above, let's make sure it works both as a docstring and reads correctly in the editor.

mhammond · 2023-09-22T14:39:42Z

win32/test/test_odbc.py

@@ -206,7 +206,7 @@ def testLongBinary(self):

    def testRaw(self):
        ## Test binary data
-        self._test_val("rawfield", memoryview(b"\1\2\3\4\0\5\6\7\8"))
+        self._test_val("rawfield", memoryview(b"\1\2\3\4\0\5\6\7\\8"))


why just one of the slashes here?

\8 is not a valid escape sequence, \0 to \7 is. This keeps the same behaviour:

>>> b"\0\1\2\3\4\5\6\7\8" b'\x00\x01\x02\x03\x04\x05\x06\x07\\8'

Although it seems that by the test's intention, it might be better to remove \8

huh, right - I think I'm fine with either removing the 8 or adding a comment about it.

…id-escape-sequences

Avasam added 2 commits August 22, 2023 18:16

autofixes

79d9c48

Manual fixes and adjustments

7408a02

Avasam force-pushed the invalid-escape-sequences branch from 3821e54 to 7408a02 Compare August 22, 2023 22:24

Avasam mentioned this pull request Sep 17, 2023

Demonstration of old python code using pyupgrade #2100

Closed

mhammond requested changes Sep 21, 2023

View reviewed changes

Avasam added 2 commits September 21, 2023 14:18

Merge branch 'main' of https://github.com/mhammond/pywin32 into inval…

0efd174

…id-escape-sequences

Revert manual r-string-to-double-backslash to reduce changes

10e8c5a

mhammond reviewed Sep 22, 2023

View reviewed changes

Avasam added 2 commits September 22, 2023 10:48

empty commit to try to get Github to update

d93aaec

Resolve PR comments

705cba5

Avasam requested a review from mhammond September 22, 2023 15:42

Avasam added 2 commits November 1, 2023 14:52

Merge branch 'main' of https://github.com/mhammond/pywin32 into inval…

b0209dc

…id-escape-sequences

r-docstring for GetPackageModuleName

17dbd79

mhammond approved these changes Nov 1, 2023

View reviewed changes

mhammond merged commit d26c5c8 into mhammond:main Nov 1, 2023
16 checks passed

Avasam deleted the invalid-escape-sequences branch November 1, 2023 19:42

Avasam mentioned this pull request Mar 14, 2024

DeprecationWarning over invalid escape sequences #1467

Closed

Avasam mentioned this pull request Mar 24, 2024

Prefer Ruff over isort for more efficient and complete import sorting and grouping #2220

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix invalid and accidental escape sequences #2112

Fix invalid and accidental escape sequences #2112

Avasam commented Aug 22, 2023

mhammond left a comment

mhammond Sep 21, 2023

Avasam Sep 21, 2023 •

edited

Loading

mhammond Sep 21, 2023

Avasam Sep 21, 2023 •

edited

Loading

mhammond Sep 21, 2023

Avasam Sep 21, 2023 •

edited

Loading

mhammond Sep 22, 2023

Avasam Sep 22, 2023 •

edited

Loading

Avasam commented Sep 21, 2023

mhammond Sep 22, 2023

Avasam Sep 22, 2023 •

edited

Loading

mhammond Sep 22, 2023

Avasam Nov 1, 2023

mhammond Sep 22, 2023

Avasam Sep 22, 2023 •

edited

Loading

mhammond Sep 22, 2023

Avasam Sep 22, 2023 •

edited

Loading

mhammond Sep 22, 2023

mhammond Sep 22, 2023

Avasam Sep 22, 2023 •

edited

Loading

mhammond Sep 22, 2023

Fix invalid and accidental escape sequences #2112

Fix invalid and accidental escape sequences #2112

Conversation

Avasam commented Aug 22, 2023

mhammond left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 21, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 21, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 21, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 22, 2023 • edited Loading

Choose a reason for hiding this comment

Avasam commented Sep 21, 2023

Choose a reason for hiding this comment

Avasam Sep 22, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 22, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 22, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 22, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Avasam Sep 21, 2023 •

edited

Loading

Avasam Sep 21, 2023 •

edited

Loading

Avasam Sep 21, 2023 •

edited

Loading

Avasam Sep 22, 2023 •

edited

Loading

Avasam Sep 22, 2023 •

edited

Loading

Avasam Sep 22, 2023 •

edited

Loading

Avasam Sep 22, 2023 •

edited

Loading

Avasam Sep 22, 2023 •

edited

Loading