On 14/12/2024 4:09 pm, Ariel Otilibili wrote:
> * since 3.12 invalid escape sequences generate SyntaxWarning
> * in the future, these invalid sequences will generate SyntaxError
> * therefore changed syntax to raw string notation.
>
> Link: https://docs.python.org/3/whatsnew/3.12.html#other-language-changes
> Fixes: e45e8f69047 ("bitkeeper revision 1.803
> (4056f51d2UjBnn9uwzC9Vu3LspnUCg)")
> Fixes: d8f3a67bf98 ("pygrub: further improve grub2 support")
> Fixes: dd03048708a ("xen/pygrub: grub2/grub.cfg from RHEL 7 has new commands
> in menuentry")
> Fixes: d1b93ea2615 ("tools/pygrub: Make pygrub understand default entry in
> string format")
> Fixes: 622e368758b ("Add ZFS libfsimage support patch")
> Fixes: 02b26c02c7c ("xen/scripts: add cppcheck tool to the xen-analysis.py
> script")
> Fixes: 56c0063f4e7 ("xen/misra: xen-analysis.py: Improve the cppcheck version
> check")
>
> Cc: Anthony PERARD <[email protected]>
> Cc: Luca Fancellu <[email protected]>
> Signed-off-by: Ariel Otilibili <[email protected]>
> ---
> tools/misc/xensymoops | 4 ++--
> tools/pygrub/src/GrubConf.py | 4 ++--
> tools/pygrub/src/pygrub | 6 +++---
> xen/scripts/xen_analysis/cppcheck_analysis.py | 4 ++--
> 4 files changed, 9 insertions(+), 9 deletions(-)
>
> diff --git a/tools/misc/xensymoops b/tools/misc/xensymoops
> index 835d187e90..bec75cae93 100755
> --- a/tools/misc/xensymoops
> +++ b/tools/misc/xensymoops
> @@ -17,7 +17,7 @@ def read_oops():
> stack_addrs is a dictionary mapping potential code addresses in the stack
> to their order in the stack trace.
> """
> - stackaddr_ptn = "\[([a-z,0-9]*)\]"
> + stackaddr_ptn = r"\[([a-z,0-9]*)\]"
> stackaddr_re = re.compile(stackaddr_ptn)
>
> eip_ptn = ".*EIP:.*<([a-z,0-9]*)>.*"
Oh wow. I've not come across this script before, and it's not
referenced in the build system.
Also, it's hard-coded to 32bit Xen which was deleted in Xen 4.13 more
than a decade ago, and there are other errors in the regexes such as
including a comma in stackaddr_ptn
Worse however, it escaped the Py2->3 conversion and is still using raw
print statements.
I'll submit a patch deleting it entirely.
> diff --git a/tools/pygrub/src/GrubConf.py b/tools/pygrub/src/GrubConf.py
> index 580c9628ca..7cd2bc9aeb 100644
> --- a/tools/pygrub/src/GrubConf.py
> +++ b/tools/pygrub/src/GrubConf.py
> @@ -320,7 +320,7 @@ class GrubConfigFile(_GrubConfigFile):
> def grub2_handle_set(arg):
> (com,arg) = grub_split(arg,2)
> com="set:" + com
> - m = re.match("([\"\'])(.*)\\1", arg)
> + m = re.match(r"([\"\'])(.*)\\1", arg)
Doesn't this \\1 want to turn into just \1 now it's a raw string?
> diff --git a/tools/pygrub/src/pygrub b/tools/pygrub/src/pygrub
> index 9d51f96070..58b088d285 100755
> --- a/tools/pygrub/src/pygrub
> +++ b/tools/pygrub/src/pygrub
> @@ -1104,7 +1104,7 @@ if __name__ == "__main__":
> if chosencfg["args"]:
> zfsinfo = xenfsimage.getbootstring(fs)
> if zfsinfo is not None:
> - e = re.compile("zfs-bootfs=[\w\-\.\:@/]+" )
> + e = re.compile(r"zfs-bootfs=[\w\-\.\:@/]+" )
Related, this string looks dodgy. The \- is correct (I think, to not
have it interpreted as a range), but I'm pretty sure a literal . and :
don't need escaping inside a [], and the result here would be for a
literal \ to be included.
~Andrew
