DbgPrint and Unicode strings?

NTFSD
1

Hi all:

Is there any format flag I can send to DbgPrint() to have it interpret a
PUNICODE_STRING? Right now, I’m using something like:

PUNICODE_STRING string;

DbgPrint(“name is %S\n”, string->Buffer);

But of course, this buffer isn’t likely to be NULL-terminated, so I get lots
of garbage in my debug output.

The DebugPrint() stuff in Chris Cant’s “Writing Windows WDM Device Driver”
has a “%T” flag that seems to work, but if I’m just using the standard
DbgPrint() API, I’m stuck.

Thanks in advance for any help you can provide.

Curt

2

%wZ is for PUNICODE_STRINGs.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]

From:
xxxxx@omnishift.com[SMTP:xxxxx@omnishift.com]
Reply To: File Systems Developers
Sent: Wednesday, September 20, 2000 23:55
To: File Systems Developers
Subject: [ntfsd] DbgPrint and Unicode strings?

Hi all:

Is there any format flag I can send to DbgPrint() to have it interpret a
PUNICODE_STRING? Right now, I’m using something like:

PUNICODE_STRING string;

DbgPrint(“name is %S\n”, string->Buffer);

But of course, this buffer isn’t likely to be NULL-terminated, so I get
lots of garbage in my debug output.

The DebugPrint() stuff in Chris Cant’s “Writing Windows WDM Device Driver”
has a “%T” flag that seems to work, but if I’m just using the standard
DbgPrint() API, I’m stuck.

Thanks in advance for any help you can provide.

Curt

3

I’m providing this explanation also for all NTFSD developers:
?
There are four types of string:

  1. NULL-terminated string (ANSI, OEM)
    ??? type PCHAR
  2. NULL-terminated UNICODE string
    ??? type PWCHAR
  3. Counted string (ANSI, OEM)
    ??? type STRING (also ANSI_STRING, OEM_STRING)
  4. Counted UNICODE string
    ??? type UNICODE_STRING
    ?
    For DbgPrint (or macro KdPrint) - which is like printf - the following
    formats should be used (expected type of argument):
    ?
  5. %s (PCHAR)
  6. %ws (PWCHAR)
  7. %Z (PSTRING, PANSI_STRING, POEM_STRING)
  8. %wZ (PUNICODE_STRING)
    ?
    In fact %Z and %wZ are not documented in the official documentation,
    but you can find them in CRT source file for _output().
    Microsoft uses them for eg. in NTOSKRNL for both debug prints and
    buffer prints ([v]s[n]printf) when some of the argument is counted
    string.
    ?
    NOTE: In the documentation of “Format Specification Fields:
    ??? printf and wprintf Functions” some fields are missing.
    ?
    type should contain Z (uppercase only) which is for counted strings.
    ???The precision?is not applied to this type.
    ?
    [{h | l | I64 | L}] should also contain w (lowecase only) which sets
    ??? the FL_WIDECHAR flag in _output(). So if the corresponding type
    ??? is some char/string type (c, s, Z), this means the UNICODE
    version
    ??? is expected as an argument.
    ?
    Paul
    ?
    PS: If you want to know more about this look into OUTPUT.C is
    reccomened.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of
xxxxx@omnishift.com
Sent: Wednesday, September 20, 2000 10:56 PM
To: File Systems Developers
Subject: [ntfsd] DbgPrint and Unicode strings?

Hi all:

Is there any format flag I can send to DbgPrint() to have it interpret a
PUNICODE_STRING?? Right now, I’m using something like:

??? PUNICODE_STRING string;

??? DbgPrint(“name is %S\n”, string->Buffer);

But of course, this buffer isn’t likely to be NULL-terminated, so I get
lots of garbage in my debug output.

The DebugPrint() stuff in Chris Cant’s “Writing Windows WDM Device
Driver” has a “%T” flag that seems to work, but if I’m just using the
standard DbgPrint() API, I’m stuck.

Thanks in advance for any help you can provide.

Curt