More python fixes
cover More python fixes
Wei Liu (4):
pygrub: fix message in grub parser pygrub/grub: always use integer for default entry
pygrub: decode string in Python 3 tools/ocaml: make python scripts 2 and 3 compatible
tools/ocaml/libs/xentoollog/genlevels.py | 5 ++++- tools/ocaml/libs/xl/genwrap.py | 17 ++++++++++------- tools/pygrub/src/GrubConf.py | 8 ++++++-- tools/pygrub/src/pygrub | 5 ++++- 4 files changed, 24 insertions(+), 11 deletions(-)
Signature · 7 lines
2.20.1 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 1/4 pygrub: fix message in grub parser
The code suggests 0 is allowed. Zero is not a positive number.
---
Backport candadiate.
--- tools/pygrub/src/GrubConf.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/pygrub/src/GrubConf.py +1 -1
@@ -236,7 +236,7 @@ class _GrubConfigFile(object): self._default = val if self._default < 0: - raise ValueError("default must be positive number") + raise ValueError("default must be non-negative number") default = property(_get_default, _set_default) def set_splash(self, val):
Signature · 7 lines
2.20.1 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 1/4] pygrub: fix message in grub parser
Quoted 3 lines · Wei Liu
The code suggests 0 is allowed. Zero is not a positive number.
Signed-off-by Wei Liu <[email protected]>
I suspect you are opening a can of number theory worms with that commit message, but the overall change is an improvement.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 2/4 pygrub/grub: always use integer for default entry
In reply to Wei LiuThe original code set the default to either a string or an integer
(0) and relies on a Python 2 specific behaviour to work (integer is allowed to be compared to string in Python 2 but not 3).
Always use integer. The caller (pygrub) already has code to handle that.
--- tools/pygrub/src/GrubConf.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
tools/pygrub/src/GrubConf.py +5 -1
@@ -233,7 +233,11 @@ class _GrubConfigFile(object): if val == "saved": self._default = 0 else: - self._default = val + try: + self._default = int(val) + except ValueError: + logging.warning("Invalid value %s, setting default to 0" %(val,)) + self._default = 0 if self._default < 0: raise ValueError("default must be non-negative number")
Signature · 7 lines
2.20.1 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] pygrub/grub: always use integer for default entry
Quoted 9 lines · Wei Liu
The original code set the default to either a string or an integer
(0) and relies on a Python 2 specific behaviour to work (integer is allowed to be compared to string in Python 2 but not 3).Always use integer. The caller (pygrub) already has code to handle that.
Reported-by M A Young <[email protected]>Signed-off-by Wei Liu <[email protected]>
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 3/4 pygrub: decode string in Python 3
In reply to Wei LiuString is unicode in 3 but bytes in 2. We need to call decode function when using Python 3.
--- tools/pygrub/src/pygrub | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
tools/pygrub/src/pygrub +4 -1
@@ -457,7 +457,10 @@ class Grub: # limit read size to avoid pathological cases buf = f.read(FS_READ_MAX) del f - self.cf.parse(buf) + if sys.version_info[0] < 3: + self.cf.parse(buf) + else: + self.cf.parse(buf.decode()) def image_index(self): if isinstance(self.cf.default, int):
Signature · 7 lines
2.20.1 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 3/4] pygrub: decode string in Python 3
Quoted 5 lines · Wei Liu
String is unicode in 3 but bytes in 2. We need to call decode function when using Python 3.
Reported-by M A Young <[email protected]>Signed-off-by Wei Liu <[email protected]>
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 4/4 tools/ocaml: make python scripts 2 and 3 compatible
In reply to Wei Liu1. Explicitly import reduce because that's required in 3.
2. Change print to function.
3. Eliminate invocations of has_key.
--- tools/ocaml/libs/xentoollog/genlevels.py | 5 ++++- tools/ocaml/libs/xl/genwrap.py | 17 ++++++++++------- 2 files changed, 14 insertions(+), 8 deletions(-)
tools/ocaml/libs/xentoollog/genlevels.py +4 -1
@@ -1,6 +1,9 @@ #!/usr/bin/python +from __future__ import print_function + import sys +from functools import reduce def read_levels(): f = open('../../../libs/toollog/include/xentoollog.h', 'r')
@@ -93,7 +96,7 @@ def autogen_header(open_comment, close_comment): if __name__ == '__main__': if len(sys.argv) < 3: - print >>sys.stderr, "Usage: genlevels.py <mli> <ml> <c-inc>" + print("Usage: genlevels.py <mli> <ml> <c-inc>", file=sys.stderr) sys.exit(1) levels, olevels = read_levels()
tools/ocaml/libs/xl/genwrap.py +10 -7
@@ -1,6 +1,9 @@ #!/usr/bin/python +from __future__ import print_function + import sys,os +from functools import reduce import idl
@@ -78,7 +81,7 @@ def ocaml_type_of(ty): elif isinstance(ty,idl.Array): return "%s array" % ocaml_type_of(ty.elem_type) elif isinstance(ty,idl.Builtin): - if not builtins.has_key(ty.typename): + if ty.typename not in builtins: raise NotImplementedError("Unknown Builtin %s (%s)" % (ty.typename, type(ty))) typename,_,_ = builtins[ty.typename] if not typename:
@@ -251,7 +254,7 @@ def gen_ocaml_ml(ty, interface, indent=""): else: s += "\texternal default : ctx -> %sunit -> t = \"stub_libxl_%s_init\"\n" % (union_args, ty.rawname) - if functions.has_key(ty.rawname): + if ty.rawname in functions: for name,args in functions[ty.rawname]: s += "\texternal %s : " % name s += " -> ".join(args)
@@ -278,7 +281,7 @@ def c_val(ty, c, o, indent="", parent = None): else: s += "%s = Int_val(%s);" % (c, o) elif isinstance(ty,idl.Builtin): - if not builtins.has_key(ty.typename): + if ty.typename not in builtins: raise NotImplementedError("Unknown Builtin %s (%s)" % (ty.typename, type(ty))) _,fn,_ = builtins[ty.typename] if not fn:
@@ -375,7 +378,7 @@ def ocaml_Val(ty, o, c, indent="", parent = None): else: s += "%s = Val_int(%s);" % (o, c) elif isinstance(ty,idl.Builtin): - if not builtins.has_key(ty.typename): + if ty.typename not in builtins: raise NotImplementedError("Unknown Builtin %s (%s)" % (ty.typename, type(ty))) _,_,fn = builtins[ty.typename] if not fn:
@@ -520,7 +523,7 @@ def autogen_header(open_comment, close_comment): if __name__ == '__main__': if len(sys.argv) < 4: - print >>sys.stderr, "Usage: genwrap.py <idl> <mli> <ml> <c-inc>" + print("Usage: genwrap.py <idl> <mli> <ml> <c-inc>", file=sys.stderr) sys.exit(1) (_,types) = idl.parse(sys.argv[1])
@@ -533,7 +536,7 @@ if __name__ == '__main__': for t in blacklist: if t not in [ty.rawname for ty in types]: - print "unknown type %s in blacklist" % t + print("unknown type %s in blacklist" % t) types = [ty for ty in types if not ty.rawname in blacklist]
@@ -564,7 +567,7 @@ if __name__ == '__main__': cinc.write("\n") cinc.write(gen_Val_ocaml(ty)) cinc.write("\n") - if functions.has_key(ty.rawname): + if ty.rawname in functions: cinc.write(gen_c_stub_prototype(ty, functions[ty.rawname])) cinc.write("\n") if ty.init_fn is not None:
Signature · 7 lines
2.20.1 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] tools/ocaml: make python scripts 2 and 3 compatible
Quoted 6 lines · Wei Liu
1. Explicitly import reduce because that's required in 3.
2. Change print to function.
3. Eliminate invocations ofhas_key.Signed-off-by M A Young <[email protected]>Signed-off-by Wei Liu <[email protected]>
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] tools/ocaml: make python scripts 2 and 3 compatible
In reply to Wei LiuQuoted 10 lines · Wei Liu
1. Explicitly import reduce because that's required in 3.
2. Change print to function.
3. Eliminate invocations ofhas_key.Signed-off-by M A Young <[email protected]>Signed-off-by Wei Liu <[email protected]>
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Quoted 11 lines · Wei Liu
Wei Liu (4):
pygrub: fix message in grub parser pygrub/grub: always use integer for default entry
pygrub: decode string in Python 3tools/ocaml: make python scripts 2 and 3 compatibletools/ocaml/libs/xentoollog/genlevels.py | 5 ++++- tools/ocaml/libs/xl/genwrap.py | 17 ++++++++++------- tools/pygrub/src/GrubConf.py | 8 ++++++-- tools/pygrub/src/pygrub | 5 ++++- 4 files changed, 24 insertions(+), 11 deletions(-)
There is a bit missing compared to my original patch which is to convert the string back to bytes for the final write out and avoid the error
Traceback (most recent call last):
File "/tmp/xencode/usr/libexec/xen/bin/pygrub", line 967, in <module>
os.write(fd, ostring)
TypeError: a bytes-like object is required, not 'str'
The attached patch works for me in a quick test on python3 though I haven't tested it on python2 yet.
Michael Young
tools/pygrub/src/pygrub renamed from tools/pygrub/src/pygrub.orig +4 -1
@@ -963,5 +963,8 @@ ostring = format_simple(bootcfg["kernel"], bootcfg["ramdisk"], args, "\0") sys.stdout.flush() - os.write(fd, ostring) + if sys.version_info[0] < 3: + os.write(fd, ostring) + else: + os.write(fd, ostring.encode())
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Quoted 30 lines · MICHAEL A.
Quoted 11 lines · Wei Liu
Wei Liu (4):
pygrub: fix message in grub parser pygrub/grub: always use integer for default entry
pygrub: decode string in Python 3tools/ocaml: make python scripts 2 and 3 compatibletools/ocaml/libs/xentoollog/genlevels.py | 5 ++++- tools/ocaml/libs/xl/genwrap.py | 17 ++++++++++------- tools/pygrub/src/GrubConf.py | 8 ++++++-- tools/pygrub/src/pygrub | 5 ++++- 4 files changed, 24 insertions(+), 11 deletions(-)There is a bit missing compared to my original patch which is to convert the string back to bytes for the final write out and avoid the error
Traceback (most recent call last):
File "/tmp/xencode/usr/libexec/xen/bin/pygrub", line 967, in <module>
os.write(fd, ostring)
TypeError: a bytes-like object is required, not 'str'The attached patch works for me in a quick test on python3 though I haven't tested it on python2 yet.
(quoted diff) +0
-0Michael Young
tools/pygrub/src/pygrub line 970
Quoted 9 lines
tools/pygrub/src/pygrub renamed from tools/pygrub/src/pygrub.orig +4
-1@@ -963,5 +963,8 @@ ostring = format_simple(bootcfg["kernel"], bootcfg["ramdisk"], args, "\0") sys.stdout.flush() - os.write(fd, ostring) + if sys.version_info[0] < 3: + os.write(fd, ostring) + else: + os.write(fd, ostring.encode())
This should work. I will fold this in to the pygrub patch.
Thanks for spotting my mistake.
Wei.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel