https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Clément Chigot changed:
What|Removed |Added
Attachment #14140|0 |1
is obsolete|
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #61 from Clément Chigot ---
Created attachment 14147
--> https://sourceware.org/bugzilla/attachment.cgi?id=14147&action=edit
0001-bfd-handle-codepage-when-opening-files-on-MinGW.patch
A better version of the previous patch
--
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Clément Chigot changed:
What|Removed |Added
CC||chigot at adacore dot com
--- Commen
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #58 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Nick Clifton :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=5f3fc928df037b21ee401d7115951b988a3ea571
commit 5f3fc928df037b21ee401d7115
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Nick Clifton changed:
What|Removed |Added
Status|REOPENED|RESOLVED
Resolution|---
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #56 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Nick Clifton :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=ba0eb22c8beb1166780d7c7e11ef4e623b8017e9
commit ba0eb22c8beb1166780d7c7e11
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #55 from Fred Eisele ---
Just to be clear, the inclusion of 'ccs=UNICODE' in the mode was an error by me
in thinking that it is used to encode the path. That is wrong, 'ccs=UNICODE'
refers only to the content. Nevertheless, it is n
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #54 from Fred Eisele ---
There is a typo in the ChangeLog 'parhs' -> 'paths'
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Fred Eisele changed:
What|Removed |Added
Attachment #13995|0 |1
is obsolete|
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #52 from Fred Eisele ---
In comment 51 Torbjörn pointed out (suggested) that 'ccs=x' is related to the
content of the file and (possibly) not the file name path.
I modified the test program and it appears he is correct.
The 'ccs=U
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #51 from Torbjörn SVENSSON ---
There is one more thing that I just started to think about.
The "ccs=UNICODE" part, when is that needed?
Is it a risk to have it compared to not have it?
According to MSDN[1], the ccs=UNICODE will do
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #50 from Torbjörn SVENSSON ---
(In reply to cvs-com...@gcc.gnu.org from comment #40)
> The master branch has been updated by Nick Clifton :
>
> https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;
> h=cb7da2a640c405e0658c135b
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #49 from Fred Eisele ---
n.b. As I added test files to https://gitlab.com/babeloff/bfdio-issue git
failed with
```
> git status
warning: could not open directory
'increasepath/increasepath/increasepath/increasepath/increasepath/in
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #48 from Fred Eisele ---
n.b. I went with the change related to changing the '/' to '\' not for the
reason suggested but because it let us get rid of one calloc.
--
You are receiving this mail because:
You are on the CC list for
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #47 from Fred Eisele ---
Created attachment 13999
--> https://sourceware.org/bugzilla/attachment.cgi?id=13999&action=edit
review the patch for discrepancies
I updated the https://gitlab.com/babeloff/bfdio-issue with the stylings
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #46 from Tom Tromey ---
(In reply to Nick Clifton from comment #43)
> (In reply to Tom Tromey from comment #42)
>
> > ../../binutils-gdb/bfd/bfdio.c:129:60: error: 'partPathOrig' undeclared
>
> *sigh* I admit that I had no way
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #45 from Nick Clifton ---
Hi Fred,
Try here:
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git
Cheers
Nick
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #44 from Fred Eisele ---
Yes, testing the patch is convoluted. That is why I supplied the test project.
https://gitlab.com/babeloff/bfdio-issue
Where exactly is the source code?
I presumed it would be here
https://github.com/bmino
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #43 from Nick Clifton ---
(In reply to Tom Tromey from comment #42)
> ../../binutils-gdb/bfd/bfdio.c:129:60: error: 'partPathOrig' undeclared
*sigh* I admit that I had no way to test the patch, so I just hoped that it was
correct
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Tom Tromey changed:
What|Removed |Added
CC||tromey at sourceware dot org
--- Comment
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #41 from Nick Clifton ---
Thanks Fred, Thanks Torbjörn,
I have now applied the patch. Do you think that it is OK to close this PR
now ?
Cheers
Nick
--
You are receiving this mail because:
You are on the CC list for the bug
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #40 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Nick Clifton :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=cb7da2a640c405e0658c135b3ab2ac5be2fdc53a
commit cb7da2a640c405e0658c135b3a
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #39 from Fred Eisele ---
I put the simple test project in gitlab https://gitlab.com/babeloff/bfdio-issue
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Fred Eisele changed:
What|Removed |Added
Attachment #13990|0 |1
is obsolete|
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #37 from Torbjörn SVENSSON ---
(In reply to Fred Eisele from comment #35)
> Created attachment 13990 [details]
> This patch handles long paths, relative paths, and paths containing '.' and
> '..' on WIN32
I think the conversion fr
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #36 from Fred Eisele ---
The patch does not include:
* handling files of the for '*.o' any differently
* does not check that the filename contains only valid characters
* there are two callocs
I presume the filename has been che
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #35 from Fred Eisele ---
Created attachment 13990
--> https://sourceware.org/bugzilla/attachment.cgi?id=13990&action=edit
This patch handles long paths, relative paths, and paths containing '.' and
'..' on WIN32
This patch handl
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #33 from Fred Eisele ---
@nick Before I make the patch could I get your impressions about the approach?
I still have some things to clear up about the approach.
* How sensitive are we to using calloc rather than allocating on the
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #34 from Fred Eisele ---
This function is called by `ar.exe` and possibly `make` which typically use
`*.o` files but not necessarily.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #32 from Fred Eisele ---
@nick Before I make the patch could I get your impressions about the approach?
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #31 from Torbjörn SVENSSON ---
I agree.
Maybe it would be sufficient to check that all bytes in the provided char* is
in the ASCII range. If a the check fails, put a clear error message to the user
that the filename is not support
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #30 from Fred Eisele ---
That makes sense.
I think that this fix is to make any long posix filenames which are
representable on NTFS do the right-thing.
i.e. ÅÄÖ and friends would still not be supported.
--
You are receiving this
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #29 from Torbjörn SVENSSON ---
(In reply to Fred Eisele from comment #28)
> However, I do not think PR 25713 is going to tackle that larger issue.
> I think it should be restricted to dealing with the filename as provided by
> the
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #28 from Fred Eisele ---
You raise a good point.
There is a problem with binutils paths generally.
POSIX path names are constrained to a fairly small set of characters.
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_c
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #27 from Torbjörn SVENSSON ---
While the overall way to do it is covered in comment 26, there are some
problems running MultiByteToWideChar.
In the example in comment 26, it's assumed that the file path is encoded in
UTF-8, or ASCI
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #26 from Fred Eisele ---
Some tweaks needed to me made to get the _fopen() to work properly.
#include
#include
#include
#include
#include
/* Mark FILE as close-on-exec. Return FILE. FILE may be NULL, in
which case nothi
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #25 from Fred Eisele ---
Use of those functions would give us something like...
size_t filelen;
/* PR 25713: Handle extra long path names possibly containing '..' and '.'.
*/
wchar_t **lpFilePart = {NULL};
const char ccs[
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #24 from Fred Eisele ---
Here is a sample program that I think demonstrates the relevant functions.
#include
#include
#include
#include
#include
#include
/**
*
https://docs.microsoft.com/en-us/windows/win32/api/fileapi/nf-
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Nick Clifton changed:
What|Removed |Added
Status|RESOLVED|REOPENED
Resolution|FIXED
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Fred Eisele changed:
What|Removed |Added
CC||fredrick.eisele at gmail dot
com
--- C
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Torbjörn SVENSSON changed:
What|Removed |Added
CC||torbjorn.svensson at st dot com
-
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Nick Clifton changed:
What|Removed |Added
Resolution|--- |FIXED
Status|REOPENED
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #19 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Nick Clifton :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=17d60030ae57fb53f5977fa6928d7d5006f42881
commit 17d60030ae57fb53f5977fa692
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #18 from Jaydeep Chauhan ---
Hi Nick,
I have tested updated patch(attachment 13095) and it is working fine.
Thanks,
Jaydeep.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Nick Clifton changed:
What|Removed |Added
Attachment #13090|0 |1
is obsolete|
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #16 from Jaydeep Chauhan ---
Hi Nick,
Thanks for this optimize patch. I have tested the same and it is working fine.
So can you please commit this patch... ?
Thanks,
Jaydeep.
--
You are receiving this mail because:
You are on t
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Nick Clifton changed:
What|Removed |Added
Attachment #13080|0 |1
is obsolete|
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #14 from Jaydeep Chauhan ---
Gentle Ping !!!
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #13 from Jaydeep Chauhan ---
Created attachment 13080
--> https://sourceware.org/bugzilla/attachment.cgi?id=13080&action=edit
Handle long relative path
Handle long relative path
--
You are receiving this mail because:
You are
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Jaydeep Chauhan changed:
What|Removed |Added
Resolution|FIXED |WONTFIX
--- Comment #12 from Jaydee
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #11 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Alan Modra :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=39a1432c09fd0242a0c832b0db04b6a71adea254
commit 39a1432c09fd0242a0c832b0db04
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #10 from Jaydeep Chauhan ---
(In reply to cvs-com...@gcc.gnu.org from comment #8)
> The master branch has been updated by Nick Clifton :
>
> https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;
> h=ca859a893931d6fad8b35cf2c20
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #9 from Nick Clifton ---
Hi Jaydeep,
Thanks for the additional patch. I have applied it with a couple
of modifications: there is no need to have two mallocs and two
copies of the path, when just one will do. Plus there alr
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #8 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Nick Clifton :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=ca859a893931d6fad8b35cf2c20afd43422a59fe
commit ca859a893931d6fad8b35cf2c20
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #7 from Jaydeep Chauhan ---
Created attachment 12544
--> https://sourceware.org/bugzilla/attachment.cgi?id=12544&action=edit
forwardslash.patch
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #6 from Jaydeep Chauhan ---
Hi Nick,
Issue is still there with ld.exe ,if we give file path with forward slash("/")
as per below.Also for normal path length(< 260 charcters) it is working fine.
ld.exe
C:/msys64/home/Jaydeep/inc
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Nick Clifton changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #4 from cvs-commit at gcc dot gnu.org ---
The master branch has been updated by Nick Clifton :
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=0b8448af68b2720d08640604355ef7d961a5acd6
commit 0b8448af68b2720d08640604355
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #3 from Jaydeep Chauhan ---
(In reply to Nick Clifton from comment #1)
Thanks Nick for reply and proposed patch.
For MAX_PATH value "windows.h" header file is required.So it is added in
proposed patch and now patch is working fine
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #2 from Jaydeep Chauhan ---
Created attachment 12402
--> https://sourceware.org/bugzilla/attachment.cgi?id=12402&action=edit
Header file added for MAX_PATH value
Added "windows.h" headerfile for MAX_PATH value in proposed patch.
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
--- Comment #1 from Nick Clifton ---
Created attachment 12401
--> https://sourceware.org/bugzilla/attachment.cgi?id=12401&action=edit
Proposed patch
Hi Jaydeep,
Unfortunately there is a small problem with your proposed patch. The
libibe
https://sourceware.org/bugzilla/show_bug.cgi?id=25713
Jaydeep Chauhan changed:
What|Removed |Added
CC||ilya.konstantinov at gmail dot
com
62 matches
Mail list logo
