[cifs-protocol] GptTmpl.inf syntax documentation issues in MS-GPSB [REG:118052218237829]

Tom Jebo tomjebo at microsoft.com
Tue May 22 23:05:54 UTC 2018


[dochelp to bcc]
[casemail to cc]

Hi Garming, 

Thank you for reaching out to dochelp about this issue. One of the Open Specifications team members will be responding shortly to assist you with this. 
In the meantime, I've created a case and added the number (118052218237829) to the subject. Please keep the number in the subject and refer to this when corresponding to us about this issue. 

Best regards,
Tom Jebo
Sr Escalation Engineer
Microsoft Open Specifications

-----Original Message-----
From: Garming Sam <garming at catalyst.net.nz> 
Sent: Tuesday, May 22, 2018 3:38 PM
To: Interoperability Documentation Help <dochelp at microsoft.com>
Cc: cifs-protocol at lists.samba.org
Subject: GptTmpl.inf syntax documentation issues in MS-GPSB

Hi,

In section 2.2 of MS-GPSB, it lists "Service General Settings" as a possible HeaderValue, when upon looking at 2.2.8 Service General Settings, it should be "Service General Setting" without the additional 's'.

Looking more closely at section 2.2.8, you also find that the ABNF description of the .inf file in 2.2 appears incomplete. In section 2.2, it would appear that all settings are in the form 'key = value'. Section
2.2.8 (2.2.7 and also 2.2.9) describe an alternate format which is stored 'name,mode,ACL' where there is no '=' in the line at all.

Examples generated from Windows:

[Service General Setting]
"Dhcp",3,""

[Registry Keys]
"MACHINE\SOFTWARE\Microsoft\TelnetServer\Defaults",0,"D:PAR(A;CI;KA;;;BA)(A;CIIO;KA;;;CO)(A;CI;KA;;;SY)(A;CI;KR;;;BU)(A;CI;KR;;;S-1-15-2-1)"

For the top-level description to be generally correct, I would've expected a syntax more like the following:

Settings = Settings / Setting Settings
Setting = Key Wsp "=" Wsp ValueList LineBreak / Name "," Mode ","
AclString LineBreak

Name = String / QuotedString
Mode = [0-9]+
AclString = SDDL / DQUOTE SDDL DQUOTE


Cheers,

Garming



More information about the cifs-protocol mailing list