Stefan Metzmacher
eff51fc623
r2952: add idl and torture test for DsCrackNames
...
(I need to find out what the fields mean but it works)
metze
2007-10-10 12:59:51 -05:00
Andrew Tridgell
3318cf2722
r2951: fixed the intptr_t test for discard_const()
2007-10-10 12:59:51 -05:00
Andrew Tridgell
c17d9254d4
r2950: 0x40000 is clearer than 262144
2007-10-10 12:59:51 -05:00
Andrew Tridgell
8921e34f98
r2949: added some range checks in samr.idl
2007-10-10 12:59:50 -05:00
Andrew Tridgell
dbe6430d78
r2948: added support for the [range(low,high)] attribute in pidl. This allows
...
range checking of any integer value, to help protect against denial of
service attacks (which could otherwise cause large memory allocations)
2007-10-10 12:59:50 -05:00
Andrew Tridgell
dac00ef3b0
r2941: added pvfs_flush() implementation to the posix backend
2007-10-10 12:59:50 -05:00
Jelmer Vernooij
0130713fbe
r2940: Complain when there is an attempt to register a backend for a subsystem that does
...
not (yet) exist
2007-10-10 12:59:50 -05:00
Jelmer Vernooij
c5d4fae3b1
r2939: Use /* */ style comments
2007-10-10 12:59:50 -05:00
Jelmer Vernooij
273d0049b5
r2938: Use IDL to dissect the RHS of floors in protocol towers
2007-10-10 12:59:50 -05:00
Jelmer Vernooij
edf28c42f3
r2937: Make sure all memory is initialized
2007-10-10 12:59:50 -05:00
Andrew Tridgell
1dff12fba8
r2934: - changed the unixuid module to use the nt_user_token instead of the server supplied info structure.
...
- added SID_WORLD and SID_NETWORK to the foreign sids in the
provisioning, as these are auto-added to the nt_user_token (why is
that done? Andrew?)
2007-10-10 12:59:49 -05:00
Jelmer Vernooij
309623447e
r2933: Only pull a union once when it occurs in a subcontext
2007-10-10 12:59:49 -05:00
Andrew Tridgell
2a6f59f376
r2932: character expansion in strlower_m or strupper_m is considered fatal
...
(as it could cause a overflow). Print a message giving the character
values involved in the expansion so it can be debugged if it happens.
2007-10-10 12:59:49 -05:00
Andrew Tridgell
0acf95d84c
r2931: use next_codepoint() to ensure we properly handle multi-byte characters in pvfs_unix_path()
2007-10-10 12:59:49 -05:00
Andrew Tridgell
da0481ac75
r2930: added a security context cache to the unixuid module. The module
...
doesn't actually leave us in the requested sec context between
requests yet, but it does prevent us from doing the samdb lookup on
every packet.
This change speeds up the BASE-MANGLE test against Samba4 with 5000
operations from 61 seconds to 16 seconds. For reference, Samba3 takes
27 seconds for the same test (the string and filename handling in
Samba4 is much more efficient than Samba3)
2007-10-10 12:59:49 -05:00
Andrew Tridgell
1d21c9a9d1
r2929: longhorn does not produce sorted directory listings! This is
...
excellent, as it means we don't lose any compatibility by also not
giving sorted listings. I was very much afraid that applications had
begun to rely on this (as its guaranteed by w2k NTFS).
no longer mark a non-sorting server as an error
2007-10-10 12:59:49 -05:00
Andrew Tridgell
0a475175c5
r2928: - fixed the handling of reserved names (rejecting them with ACCESS_DENIED)
...
- don't check for '.' specially in checking for legal names. Longhorn
doesn't do this any more, and its a real pain. Longhorn allows for
filenames ending in '.', and with as many '.' elements as you like.
2007-10-10 12:59:49 -05:00
Andrew Tridgell
ed52d69e8a
r2927: imported the hash2 name mangling code from Samba3 into Samba4, but
...
heavily modified to suit the Samba4 architecture.
Samba4 with posix backend now passes the BASE-MANGLE test
2007-10-10 12:59:49 -05:00
Andrew Tridgell
f85a0b725d
r2926: name->dos.attrib is not valid unless name->exists is true
2007-10-10 12:59:48 -05:00
Andrew Tridgell
9ef399a769
r2925: added the definition for a new rpc fault code I am seeing on epmapper from lhorn
2007-10-10 12:59:48 -05:00
Jelmer Vernooij
f355cd4264
r2921: Add a few more protocols and fix the numbers associated with some of
...
the current ones. It took me three hours to realise that the DCOM standard
contains false protocol numbers (apparently someone converted the protocol
numbers to hex twice, i.e. 13 -> 0c and 14 to 0d). There are no longer
duplicates in the list with protocol numbers now.
2007-10-10 12:59:48 -05:00
Jelmer Vernooij
5e0bab8911
r2920: Support passing a handle to a SAMR pipe to a RPC Binding or Select
...
Host dialog so that one can select a domain from the list of known
domains.
2007-10-10 12:59:48 -05:00
Andrew Tridgell
4a8b0302ef
r2916: longhorn client doesn't bother setting the directory bit in ntcreatex
...
when opening directories. This change allows samba4 server to work
with longhorn client.
2007-10-10 12:59:48 -05:00
Jelmer Vernooij
9560a6c8cc
r2915: Open domain first before enumerating users and groups
2007-10-10 12:59:47 -05:00
Jelmer Vernooij
b3bc055ac7
r2914: ... and support enumerating values (data is not handled yet, only the names)
2007-10-10 12:59:47 -05:00
Jelmer Vernooij
5086d6b249
r2913: - Don't print hive name if it is NULL (regtree)
...
- Initialise hive name (reg_interface)
- Fix LDB backend (enumerating keys works now!)
2007-10-10 12:59:47 -05:00
Volker Lendecke
ab52ebfdcc
r2912: Test for the sort-of weird file stamp semantics. The last_write seems to be
...
updated only by an arbitrary close() after the file was actually written to.
Volker
2007-10-10 12:59:47 -05:00
Jelmer Vernooij
e7c256a92c
r2911: Fix bug in opening relative keys
2007-10-10 12:59:47 -05:00
Andrew Tridgell
aad0e7e9d8
r2910: I noticed that the samr torture test was doing its own DOS->UNIX
...
string conversion. For RPC, all string conversions are supposed to be
done by the NDR layer, using string flags set in the IDL. The reason
this wasn't working is that I had been too lazy to do the STR_ASCII
string types properly at the NDR layer when initially writing
ndr_basic.c.
This commit fixes the ndr_basic code properly to do all ASCII
varients, by re-using the non-ascii code and a "byte_mul" local
variable. I have also removed the manual string conversion in the SAMR
torture test code.
2007-10-10 12:59:47 -05:00
Andrew Tridgell
75ca5ba509
r2909: fix some RAW-SEARCH torture mem leaks
2007-10-10 12:59:47 -05:00
Simo Sorce
f544f83063
r2908: fix typo
2007-10-10 12:59:47 -05:00
Andrew Tridgell
e3009492b8
r2907: auto destroy iconv memory handles on exit, to make valgrind leak
...
reports easier to read (less noisy)
2007-10-10 12:59:46 -05:00
Andrew Tridgell
0181fe06b4
r2906: fixed a memory leak in the smbclient -L code
2007-10-10 12:59:46 -05:00
Andrew Tridgell
657103ec62
r2904: - fixed the old style SMBsearch to return the pvfs shortname, not a truncated long name.
...
- short name can be up to 12 bytes, not 11
2007-10-10 12:59:46 -05:00
Andrew Tridgell
abcd841a85
r2903: a considerably more efficient (both in terms of CPU and memory)
...
convert_string_talloc() implementation.
the previous version used a minimum of 512 bytes, which is way above the average
of what is needed.
2007-10-10 12:59:46 -05:00
Andrew Tridgell
878f6b565f
r2902: make toupper_w() and tolower_w() slightly faster by putting the most common
...
conditions first
2007-10-10 12:59:46 -05:00
Andrew Tridgell
994392d085
r2901: if we can't load upcase.dat or lowcase.dat then don't waste 256k
...
making fake tables, instead just do the approximate upper/lower inline
with toupper() and tolower().
2007-10-10 12:59:46 -05:00
Andrew Tridgell
2132d38f9c
r2900: rusty pointed out to me that discard_const() can be done via a macro
...
on systems that have the intptr_t type, and for systems that don't
have it, they also almost certainly won't have -Wcast-qual, so we can use a
void* cast.
2007-10-10 12:59:46 -05:00
Jelmer Vernooij
4b5bf739f5
r2898: - Support enumerating remote registry values
...
- Some LDB updates
2007-10-10 12:59:45 -05:00
Jelmer Vernooij
f4860afc48
r2897: Fix double registration (of the registry subsystem) bug
2007-10-10 12:59:45 -05:00
Jelmer Vernooij
e9a7f729b3
r2896: Add torture test for EnumValue()
2007-10-10 12:59:45 -05:00
Volker Lendecke
8e26775134
r2895: Another nut to crack, not sure how hard: The statinfo on a file does obviously
...
not change during a write. The nasty thing: Excel 2003 obviosly does depend on
this.
Volker
2007-10-10 12:59:45 -05:00
Andrew Tridgell
749fa73544
r2893: added very primitive name mangling support to pvfs
2007-10-10 12:59:45 -05:00
Stefan Metzmacher
1a3b546fce
r2892: fix compiler warning
...
metze
2007-10-10 12:59:44 -05:00
Stefan Metzmacher
071c19c25d
r2891: call rootDSE only with LDAP_SEARCH_SCOPE_BASE
...
this is needed because of the global catalog
metze
2007-10-10 12:59:43 -05:00
Stefan Metzmacher
82e792a0ce
r2890: fix segfault when call is destroyed and we dereference it
...
metze
2007-10-10 12:59:43 -05:00
Stefan Metzmacher
536af87ef1
r2889: add DRSUAPI server
...
- with DsBind and DsUnbind implmented :-)
the RPC-DRSUAPI test works
metze
2007-10-10 12:59:42 -05:00
Stefan Metzmacher
de73676342
r2888: - add 'Ds' prefix to Bind and Unbind call on drsuapi
...
- rename handle -> bind_handle
- change function types to NTSTATUS
metze
2007-10-10 12:59:42 -05:00
Stefan Metzmacher
5143a9bd8a
r2887: fix comment
...
metze
2007-10-10 12:59:42 -05:00
Stefan Metzmacher
f3f2d1c676
r2886: missing stuff from last commit
...
metze
2007-10-10 12:59:42 -05:00