13 years agor9443: We can now move windows around anywhere on our SWAT "desktop".
Deryck Hodge [Sun, 21 Aug 2005 05:41:48 +0000 (05:41 +0000)]
r9443: We can now move windows around anywhere on our SWAT "desktop".
QxWidget doesn't accept percentages, so I feed the toplevel
widget the actual brower height/width onload and onresize.

(This used to be commit dadc89f8839b5c481ba47eeadec9f7bf4d3e14a9)

13 years agor9442: Support some more primitive types. Fix charset support.
Jelmer Vernooij [Sun, 21 Aug 2005 02:12:25 +0000 (02:12 +0000)]
r9442: Support some more primitive types. Fix charset support.
(This used to be commit eeced29eb3dd2cbc4540555bdb642b2685eb9c5c)

13 years agor9441: Use "const char *" for fixed-size arrays with charset() because
Jelmer Vernooij [Sun, 21 Aug 2005 01:59:41 +0000 (01:59 +0000)]
r9441: Use "const char *" for fixed-size arrays with charset() because
these can require more elements in the local charset (usually UTF8) then
in the wire one.
(This used to be commit a0e63c2691f596cdacbc2e15404829ebca075429)

13 years agor9440: Fix bug introduced by new zero-padding code. Caught by the buildfarm.
Jelmer Vernooij [Sun, 21 Aug 2005 00:07:51 +0000 (00:07 +0000)]
r9440: Fix bug introduced by new zero-padding code. Caught by the buildfarm.
(This used to be commit 9be03c057e229e9cf7fe8b1db04adb9d2f1efc64)

13 years agor9439: Make sure the remainder of the array is always initialized if the buffer is
Jelmer Vernooij [Sat, 20 Aug 2005 23:36:43 +0000 (23:36 +0000)]
r9439: Make sure the remainder of the array is always initialized if the buffer is
larger then the string to be pushed.
(This used to be commit 70b52e26f31b00637ed7f90f77ff0b2794dad729)

13 years agor9438: Add initial files for ldb_map
Jelmer Vernooij [Sat, 20 Aug 2005 23:30:10 +0000 (23:30 +0000)]
r9438: Add initial files for ldb_map
(This used to be commit a66eeadecdcc5f67dd34c4ae82f5ae9191163499)

13 years agor9437: Update PLAN and README for Samba3 compatibility layer
Jelmer Vernooij [Sat, 20 Aug 2005 23:28:14 +0000 (23:28 +0000)]
r9437: Update PLAN and README for Samba3 compatibility layer
Add support for reading tdbsam files
(This used to be commit 75ac972909ac601fb876e208a992eeebafaf6417)

13 years agor9434: moved the registry editor into a common js library. Deryck, does this
Andrew Tridgell [Sat, 20 Aug 2005 22:45:48 +0000 (22:45 +0000)]
r9434: moved the registry editor into a common js library. Deryck, does this
look ok? Apologies if this treads on your toes.
(This used to be commit 8295eb01c0e1eec4f7c5bd211eac80b92ff9549e)

13 years agor9433: - Update TODO & README
Jelmer Vernooij [Sat, 20 Aug 2005 21:54:45 +0000 (21:54 +0000)]
r9433: - Update TODO & README
- Extend compatibility module (work towards being able to generate
  Midl-compatible IDL files, pretty much just removing attributes
  midl doesn't know such as value() and charset())
- Support charset() on fixed-size arrays of characters
(This used to be commit 43817eb0bd46aab10d1595ce008d8380565246f0)

13 years agor9431: Check an error code in BASE-DELETE.
Volker Lendecke [Sat, 20 Aug 2005 21:28:30 +0000 (21:28 +0000)]
r9431: Check an error code in BASE-DELETE.

(This used to be commit 71571fffc0493a5658c5980e6ebe4d8f9ada4699)

13 years agor9430: Remove obsolete file.
Jelmer Vernooij [Sat, 20 Aug 2005 19:39:19 +0000 (19:39 +0000)]
r9430: Remove obsolete file.
Update pidl documentation, including description on ethereal conformance files
Nicer --help ouput
(This used to be commit 6c36864daee75b64fef86ce99cf980106624e0f7)

13 years agor9429: Update link to DCOM spec. Fix HRESULT values.
Jelmer Vernooij [Sat, 20 Aug 2005 18:32:44 +0000 (18:32 +0000)]
r9429: Update link to DCOM spec. Fix HRESULT values.
(This used to be commit fad3413de5655eb6b1a1c4172b02acd80ae24cda)

13 years agor9424: Remove duplicate interface name in DCERPC subdissector strings. Also
Tim Potter [Sat, 20 Aug 2005 11:42:11 +0000 (11:42 +0000)]
r9424: Remove duplicate interface name in DCERPC subdissector strings.  Also
shorten some of the autogenerated function names.
(This used to be commit f380e86806aca0aa5159fc3ad1f4bdf633375c86)

13 years agor9423: Fix a typo.
Tim Potter [Sat, 20 Aug 2005 09:54:56 +0000 (09:54 +0000)]
r9423: Fix a typo.
(This used to be commit d971da4b1d1f21925f95b9c000f32a88f55b5da0)

13 years agor9422: Include crypto.h header.
Andrew Bartlett [Sat, 20 Aug 2005 08:30:41 +0000 (08:30 +0000)]
r9422: Include crypto.h header.

Andrew Bartlett
(This used to be commit 340955220443533cc70dcb0ef9bae4815f30e6c5)

13 years agor9421: Move arcfour code into it's own file, in lib/crypto.
Andrew Bartlett [Sat, 20 Aug 2005 07:59:00 +0000 (07:59 +0000)]
r9421: Move arcfour code into it's own file, in lib/crypto.

Andrew Bartlett
(This used to be commit ca6cf462708810637544d4b4bef0f404fb89a002)

13 years agor9420: Fix the SPNEGO system again: Update the state position after
Andrew Bartlett [Sat, 20 Aug 2005 07:31:29 +0000 (07:31 +0000)]
r9420: Fix the SPNEGO system again: Update the state position after
processing the state.

Andrew Bartlett
(This used to be commit c3a8080a189e084c5774bdd1a9f3ea62daed715e)

13 years agor9419: Silly, silly, untested mistake...
Andrew Bartlett [Sat, 20 Aug 2005 07:04:13 +0000 (07:04 +0000)]
r9419: Silly, silly, untested mistake...

Andrew Bartlett
(This used to be commit 66cdd4dbd7f4024b49bcc9fe1c7ff2a0de7e82e8)

13 years agor9418: SPNEGO fixes:
Andrew Bartlett [Sat, 20 Aug 2005 06:36:35 +0000 (06:36 +0000)]
r9418: SPNEGO fixes:

- Fix mixing of code and data
- send mechListMic again in SPENGO server
- only send optomistic first packet in the client.
(This used to be commit 9941da8081ef5a669b0946265860d2f20d3718d3)

13 years agor9417: Ask for the ASYNC_REPLIES feature, as will want that.
Andrew Bartlett [Sat, 20 Aug 2005 06:14:46 +0000 (06:14 +0000)]
r9417: Ask for the ASYNC_REPLIES feature, as will want that.

Andrew Bartlett
(This used to be commit 8cb8f99ae602f3bca0235e5e695ecb1ba55b0e46)

13 years agor9416: Cleanups inspired by jra's work to migrate Samba4's NTLMSSP code back
Andrew Bartlett [Sat, 20 Aug 2005 06:14:14 +0000 (06:14 +0000)]
r9416: Cleanups inspired by jra's work to migrate Samba4's NTLMSSP code back
into Samba3.

The NTLMSSP sign/seal code now assumes that GENSEC has already checked
to see if SIGN or SEAL should be permitted.  This simplfies the code
ensures that no matter what the mech, the correct code paths have been
set in place.

Also remove duplication caused by the NTLMv2 code's history, and
document why some of the things a bit funny.

In SPNEGO, create a new routine to handle the negTokenInit creation.
We no longer send an OID for a mech we can't start (like kerberos on
the server without a valid trust account).

Andrew Bartlett
(This used to be commit fe45ef608f961a6950d4d19b4cb5e7c27b38ba5f)

13 years agor9415: Remove old kerberos code (including salt guessing code) that has only
Andrew Bartlett [Sat, 20 Aug 2005 06:08:52 +0000 (06:08 +0000)]
r9415: Remove old kerberos code (including salt guessing code) that has only
caused me pain (and covourty warnings).

Simply gensec_gssapi to assume the properties of lorikeet-heimdal,
rather than having #ifdef around critical features.  This simplifies
the code rather a lot.

Andrew Bartlett
(This used to be commit 11156f556db678c3d325fe5ced5e41a76ed6a3f1)

13 years agor9414: Fix failure to find own domain info due to recent ldb_dn upgrade - we
Andrew Bartlett [Sat, 20 Aug 2005 06:01:49 +0000 (06:01 +0000)]
r9414: Fix failure to find own domain info due to recent ldb_dn upgrade - we
don't use a simple char* here any more.

Andrew Bartlett
(This used to be commit f21c31c7e151e23caa2699b1727d1ae310fbed4e)

13 years agor9413: Bring Samba4 back up to date with lorikeet-heimdal.
Andrew Bartlett [Sat, 20 Aug 2005 06:00:50 +0000 (06:00 +0000)]
r9413: Bring Samba4 back up to date with lorikeet-heimdal.

Delete test_crypto_wrapping.c, previously included but unbuilt.

Andrew Bartlett
(This used to be commit d5fb30fb0cef330e0947969f0c9afc1f58fc4c7d)

13 years agor9412: Simplfy this NTLM authentication code by requiring the caller to
Andrew Bartlett [Sat, 20 Aug 2005 05:59:27 +0000 (05:59 +0000)]
r9412: Simplfy this NTLM authentication code by requiring the caller to
supply the user_sess_key and lm_sess_key parameters.  Inspired by
coverty complaining about inconsistant checking.

Also factor out some of this code, where we deal with just NT and LM
hashes, or embedded plaintext passwords.

Andrew Bartlett
(This used to be commit ceec35564f44c8043888c8ffa776f137bd1171c8)

13 years agor9411: Ensure we don't send a challenge without first getting a negotiate in
Andrew Bartlett [Sat, 20 Aug 2005 04:42:19 +0000 (04:42 +0000)]
r9411: Ensure we don't send a challenge without first getting a negotiate in
NTLMSSP, unless we are in datagram mode (not fully implemented yet).

Andrew Bartlett
(This used to be commit 727f5109421e9414a335e42e3ad3dd3ff19776bd)

13 years agor9410: - a winreg_CloseKey() should return a zero key on success (zeroing the
Andrew Tridgell [Sat, 20 Aug 2005 04:40:08 +0000 (04:40 +0000)]
r9410: - a winreg_CloseKey() should return a zero key on success (zeroing the
callers key). This is the normal pattern with rpc handles.

- fixed reference to undefined error variable in winreg_DeleteKey()
(This used to be commit 6757d51a280d6c393648a384f1ef3ca403517352)

13 years agor9409: fix a problem that volker noticed with web page timeouts causing smbd
Andrew Tridgell [Sat, 20 Aug 2005 04:38:35 +0000 (04:38 +0000)]
r9409: fix a problem that volker noticed with web page timeouts causing smbd
to crash. This is one of the downsides of the fact that the ejs engine
is not event driven, resulting in the rendering of each web page being
'semi-async'. We need to protect the web context from the timeout
processing until we have unwound the stack back to the point that the
'web' variable representing the page rendering logic won't be used any
(This used to be commit 97e3c9eaf1a917eb504a2c3414c5d2b64cd4539a)

13 years agor9406: Add const.
Andrew Bartlett [Sat, 20 Aug 2005 00:10:03 +0000 (00:10 +0000)]
r9406: Add const.

Andrew Bartlett
(This used to be commit 4c6052ccfd6ff01eb00d8f953ae8912ea6dedba3)

13 years agor9397: Playing with qooxdoo a bit and saving my work, so I
Deryck Hodge [Fri, 19 Aug 2005 12:02:30 +0000 (12:02 +0000)]
r9397: Playing with qooxdoo a bit and saving my work, so I
can get to it later today at work.  Not much to see yet,
and not linked to from SWAT yet.

Playing with the idea of a web-based desktop, and just
seeing how widgets can be used.

(This used to be commit 435467f90d809899c3d0efeaf991fb659d97202d)

13 years agor9396: ntlm_auth updates, including again support for the NTLMSSP client
Andrew Bartlett [Thu, 18 Aug 2005 22:36:12 +0000 (22:36 +0000)]
r9396: ntlm_auth updates, including again support for the NTLMSSP client
mode, and specification of the workstation.

Andrew Bartlett
(This used to be commit 8f53d995c63b718b40fb52ec102fa74f94719b65)

13 years agor9394: avoid to use BOOL in ldb
Simo Sorce [Thu, 18 Aug 2005 16:41:27 +0000 (16:41 +0000)]
r9394: avoid to use BOOL in ldb
(This used to be commit d2055849fba56e8620403621f3fb9684f24e853f)

13 years agor9393: Fix ldb standalone build
Simo Sorce [Thu, 18 Aug 2005 16:27:09 +0000 (16:27 +0000)]
r9393: Fix ldb standalone build
(This used to be commit 796d0ea2fcf5132b157a397cc1a54aa26c042691)

13 years agor9392: Fix ldb_dn_compose to make build farm happy
Simo Sorce [Thu, 18 Aug 2005 16:18:48 +0000 (16:18 +0000)]
r9392: Fix ldb_dn_compose to make build farm happy

Add ldb_dn_string_compose so that you can build a dn starting from a
struct ldb_dn base and a set of parameters to be composed in a format
string with the same syntax of printf
(This used to be commit 31c69d0655752cc8ea3bc5b7ea87792291302091)

13 years agor9391: Convert all the code to use struct ldb_dn to ohandle ldap like distinguished...
Simo Sorce [Thu, 18 Aug 2005 15:02:01 +0000 (15:02 +0000)]
r9391: Convert all the code to use struct ldb_dn to ohandle ldap like distinguished names
Provide more functions to handle DNs in this form
(This used to be commit 692e35b7797e39533dd2a1c4b63d9da30f1eb5ba)

13 years agor9390: fixed mixing of code and data
Andrew Tridgell [Thu, 18 Aug 2005 12:21:42 +0000 (12:21 +0000)]
r9390: fixed mixing of code and data
(This used to be commit ae2122e76a7e50435dafcc412ee425b522c0c766)

13 years agor9389: handle errors reading from files in web server
Andrew Tridgell [Thu, 18 Aug 2005 12:19:16 +0000 (12:19 +0000)]
r9389: handle errors reading from files in web server

bug found by coverity
(This used to be commit 1a3229777adf12f8148267af7b33689ebe880ed8)

13 years agor9388: we should fault bad handles given to winreg_GetVersion()
Andrew Tridgell [Thu, 18 Aug 2005 11:16:32 +0000 (11:16 +0000)]
r9388: we should fault bad handles given to winreg_GetVersion()
(This used to be commit 00e55be9efc6050e3bda13603270b028dbc19c3f)

13 years agor9387: regedit uses "New Key #nn" for newly created keys, which conflicts with the...
Andrew Tridgell [Thu, 18 Aug 2005 11:15:15 +0000 (11:15 +0000)]
r9387: regedit uses "New Key #nn" for newly created keys, which conflicts with the stricter
DN rules in ldb. Escape the DN components to cope.

Simo, sorry for making a change in ldb_dn.c while you have changes
pending. Please feel free to revert these and switch reg_backend_ldb.c
to use the new dn construction code.
(This used to be commit 136ecf5cb23758558b4119b08047fc273be8b0f8)

13 years agor9386: OpenKey with a bad name must return WERR_BADFILE (w2k3 regedit relies on this)
Andrew Tridgell [Thu, 18 Aug 2005 10:58:05 +0000 (10:58 +0000)]
r9386: OpenKey with a bad name must return WERR_BADFILE (w2k3 regedit relies on this)
(This used to be commit ae42976590b3c532790d854be45fff49074aef68)

13 years agor9385: Remove unused functions
Simo Sorce [Thu, 18 Aug 2005 10:23:53 +0000 (10:23 +0000)]
r9385: Remove unused functions
(This used to be commit fac8ff623778250acd830f358fcd34b85f7983b6)

13 years agor9384: added a debug to show the dcerpc fault code for any calls we fault
Andrew Tridgell [Thu, 18 Aug 2005 10:12:55 +0000 (10:12 +0000)]
r9384: added a debug to show the dcerpc fault code for any calls we fault
(This used to be commit 92ed69695b2a5459e79f3edb37790a17de5a66c6)

13 years agor9383: remove unused file
Andrew Tridgell [Thu, 18 Aug 2005 09:33:03 +0000 (09:33 +0000)]
r9383: remove unused file
(This used to be commit dae90b6959508a19d35f711caf4c3f4bb9b3796a)

13 years agor9382: Add a dummy registry for HKEY_USERS to get rid of the annoying dialog
Tim Potter [Thu, 18 Aug 2005 07:06:19 +0000 (07:06 +0000)]
r9382: Add a dummy registry for HKEY_USERS to get rid of the annoying dialog
presented by regedt32.  I think this hive is dynamically generated
from SAM information.
(This used to be commit feb341969e59540bc22e78df0d44b4d42c336d31)

13 years agor9381: Line wrapping.
Tim Potter [Thu, 18 Aug 2005 07:00:37 +0000 (07:00 +0000)]
r9381: Line wrapping.
(This used to be commit 2cebdd59387786c8aeee4c6757e38ed5303f7315)

13 years agor9379: the valgrind test box is now just going past the max 30 minute smbd
Andrew Tridgell [Thu, 18 Aug 2005 06:19:32 +0000 (06:19 +0000)]
r9379: the valgrind test box is now just going past the max 30 minute smbd
runtime for testing. Increased to 45 minutes.
(This used to be commit 4fd436260901785e808de23763a6e127bc742142)

13 years agor9378: initialise the last_mod attribute in the ldb backend. Better to return
Andrew Tridgell [Thu, 18 Aug 2005 06:14:10 +0000 (06:14 +0000)]
r9378: initialise the last_mod attribute in the ldb backend. Better to return
0 than an uninitialised value, but we should put proper last_modified
time support into the ldb winreg backend in the future
(This used to be commit 899bf07908d4a04cb3d1cd4034dca0c91e6a912b)

13 years agor9377: made winreg a user tool (I find it quite useful). I expect it to get the ability
Andrew Tridgell [Thu, 18 Aug 2005 05:09:26 +0000 (05:09 +0000)]
r9377: made winreg a user tool (I find it quite useful). I expect it to get the ability
to add/remove keys and values in the future.

add it to the standard tests, ensuring that we know if winreg
breaks. This is particularly important as winreg uses such unusual IDL
(This used to be commit e4ca36bda34cf5e6fecaef5fe60e5dd397ebee3c)

13 years agor9374: HPUX is also missing setegid()
Andrew Tridgell [Thu, 18 Aug 2005 01:57:43 +0000 (01:57 +0000)]
r9374: HPUX is also missing setegid()
(This used to be commit 57e6bd61395e82064c72510dcc326b11b7bdf7fd)

13 years agor9373: - create a hierachical memory tree with recursiv ndr_pull_* functions
Stefan Metzmacher [Thu, 18 Aug 2005 01:24:08 +0000 (01:24 +0000)]
r9373: - create a hierachical memory tree with recursiv ndr_pull_* functions
- with this it's also possible to talloc_free() the ndr_pull structure
  and talloc_steal(ndr->current_mem_ctx); to fetch the whole data of the hierachical tree
- if the toplevel struct is a valid talloc pointer it's also possible to use
  NDR_PULL_SET_MEM_CTX(ndr, mem_ctx); to the the toplevel pointer with the struct pointer

(NOTE: no callers are using this yet, but they shortly will)

(This used to be commit 1a2b8369586642cc9bc15d015c1e4256c3a92732)

13 years agor9372: - make the subcontext handling autogenerated code look nicer,
Stefan Metzmacher [Thu, 18 Aug 2005 00:45:19 +0000 (00:45 +0000)]
r9372: - make the subcontext handling autogenerated code look nicer,
- unify the handling of subcontext, compression and obfucation

(This used to be commit 09de7e0af7f9f7539cf63791baf90ac202536176)

13 years agor9371: add the usefull trick hack for NT_STATUS_NO_MEMORY too
Stefan Metzmacher [Thu, 18 Aug 2005 00:43:00 +0000 (00:43 +0000)]
r9371: add the usefull trick hack for NT_STATUS_NO_MEMORY too

(This used to be commit 9b9635833f44e63f8acfd619ae2ecc06f6b2312a)

13 years agor9370: need a configure test for setresuid()
Andrew Tridgell [Thu, 18 Aug 2005 00:31:40 +0000 (00:31 +0000)]
r9370: need a configure test for setresuid()
(This used to be commit 5a38b9a10b3c5363cddc7acaaea2e1e9aceb254c)

13 years agor9369: an attempt to fix the build on HPUX. This is based on work by Don
Andrew Tridgell [Thu, 18 Aug 2005 00:20:40 +0000 (00:20 +0000)]
r9369: an attempt to fix the build on HPUX. This is based on work by Don
McCall, but takes a slightly different approach that I hope will be
more generic
(This used to be commit e8260a81cf99be2ccae64135ca0572c8a6ae62ad)

13 years agor9360: fixed the IDL for winreg_SetValue()
Andrew Tridgell [Wed, 17 Aug 2005 13:14:40 +0000 (13:14 +0000)]
r9360: fixed the IDL for winreg_SetValue()
(This used to be commit 8aff6a0bd808358162b646514d996ad432bfb70d)

13 years agor9359: don't check for size overflow if value is NULL
Andrew Tridgell [Wed, 17 Aug 2005 13:09:42 +0000 (13:09 +0000)]
r9359: don't check for size overflow if value is NULL

this gets viewing the product options in smbd HKLM from w2k3 regedit
(This used to be commit 17f3250dd61bc9aad5cf3804e271383e219752f6)

13 years agor9358: - opening a winreg key of "" is the same as re-opening the hive. The
Andrew Tridgell [Wed, 17 Aug 2005 13:06:05 +0000 (13:06 +0000)]
r9358: - opening a winreg key of "" is the same as re-opening the hive. The
  w2k3 regedit does this.

- w2k3 expects null termination is string lengths
(This used to be commit bcef9e592acd96a2fd7028b299d3d93968d044ae)

13 years agor9357: Remove DBGC_CLASS cruft copied over from Samba 3. I would like to
Tim Potter [Wed, 17 Aug 2005 12:51:07 +0000 (12:51 +0000)]
r9357: Remove DBGC_CLASS cruft copied over from Samba 3.  I would like to
replace this with something funkier.
(This used to be commit 8d376d56c78894b9bbd27ed7fa70da415c0cd038)

13 years agor9356: a better way of coping with NULL arrays in the array bounds checking. This...
Andrew Tridgell [Wed, 17 Aug 2005 12:28:47 +0000 (12:28 +0000)]
r9356: a better way of coping with NULL arrays in the array bounds checking. This copes with the
case of size_is(*size) where size is NULL, and the array is NULL
(This used to be commit 56769b4b1d900cce60cd35298b642a85e4eddfee)

13 years agor9355: return the EnumKey and EnumValue list we have so far when we get a rpc fault
Andrew Tridgell [Wed, 17 Aug 2005 12:27:28 +0000 (12:27 +0000)]
r9355: return the EnumKey and EnumValue list we have so far when we get a rpc fault
(This used to be commit 3a9c63923fbb21dfb8e5fc549dde8fad8cb8f354)

13 years agor9347: this array bounds checking is harder than it looks ...
Andrew Tridgell [Wed, 17 Aug 2005 03:30:45 +0000 (03:30 +0000)]
r9347: this array bounds checking is harder than it looks ...

this copes with 2 more situations:

1) where the array is NULL, which would previously be coped with by a
   if (ptr) check, but now in the deferred array bounds checking needs
   to look at the array variable in the ndr code. Not nice.

2) nest the array checking along with the SCALARS vs BUFFERS checks, ensuring we don't
   do array bounds checking for a buffer when in scalars only mode
(This used to be commit ad1b9867a5a14bc9ed2e1a5eb8f05bb2046bc645)

13 years agor9346: allow test_ldap.sh to be called when $CONFFILE is not set
Andrew Tridgell [Wed, 17 Aug 2005 02:56:39 +0000 (02:56 +0000)]
r9346: allow test_ldap.sh to be called when $CONFFILE is not set
(This used to be commit 3ad02091f621e3ff51aa082ca7a47b9da238aa40)

13 years agor9345: used the data blob functions in the echo.js test code
Andrew Tridgell [Wed, 17 Aug 2005 02:56:07 +0000 (02:56 +0000)]
r9345: used the data blob functions in the echo.js test code
(This used to be commit f217f08b679ec3ed4784ed0fe74e9cbb935cdce9)

13 years agor9344: started adding calls for manipulation of data blobs in ejs
Andrew Tridgell [Wed, 17 Aug 2005 02:55:31 +0000 (02:55 +0000)]
r9344: started adding calls for manipulation of data blobs in ejs


(This used to be commit cc5a3e9cc3b44f418eb5dfcd4e855a36c5ac8327)

13 years agor9342: removed extra libinclude of base.js
Andrew Tridgell [Wed, 17 Aug 2005 02:21:40 +0000 (02:21 +0000)]
r9342: removed extra libinclude of base.js
(This used to be commit bda6943c311e005cff0b46722355e9aa693db67a)

13 years agor9341: updated the winreg test program to take advantage of the new EnumValue
Andrew Tridgell [Wed, 17 Aug 2005 01:40:14 +0000 (01:40 +0000)]
r9341: updated the winreg test program to take advantage of the new EnumValue
functionality. It now completely enumerates a winreg tree.

This would make a good basis for a js version of regshell if someone
felt like taking that on.
(This used to be commit 75d0cf78ec8c01dee2a6475d4c64417349d9ec3a)

13 years agor9340: print the [in] contents when debugging even if the marshalling
Andrew Tridgell [Wed, 17 Aug 2005 01:30:47 +0000 (01:30 +0000)]
r9340: print the [in] contents when debugging even if the marshalling
failed. This makes it easier to track down marshalling bugs.
(This used to be commit a7a1752ca47a18b325a0ad0984cf7d02da1e2afe)

13 years agor9339: treat arrays of uint8 values as a special DATA_BLOB type in the ejs
Andrew Tridgell [Wed, 17 Aug 2005 01:29:35 +0000 (01:29 +0000)]
r9339: treat arrays of uint8 values as a special DATA_BLOB type in the ejs
interfaces to RPC. This makes large blobs of data much saner. Tim, you
will probably want to do the same for the smb_interfaces.h generated

Next we will need ways of extracting different data types from these
blobs, for example asking for the blob to be interpreted as a utf16
string, or as a little-endian integer. That will allow for registry
scripting to be quite sane.
(This used to be commit a8bca2e8e27c953c0413693326ec3b5ecf17ba41)

13 years agor9338: fixed the winreg IDL to be correct for the EnumKey and EnumValue
Andrew Tridgell [Wed, 17 Aug 2005 01:25:58 +0000 (01:25 +0000)]
r9338: fixed the winreg IDL to be correct for the EnumKey and EnumValue
calls. The previous IDL was just a workaround for the limitations of
our older rpc infrastructure. Now that Jelmer has added much improved
string support using the charset keyword we can correctly implemenent
the unusual winreg string buffers.

Jelmer, note the little comment I put on winreg_StringBuf() about why
I couldn't use [value()] for the length field.

This also fixes EnumKey() and EnumValue() to use NTTIME fields for the
last_changed_time. I don't know why we were using a pair of uint32's,
as it is just a NTTIME.
(This used to be commit 8354b016122cc4f3cff042b3ada1de07e1614eb7)

13 years agor9337: defer the checking of array sizes until the end of the
Andrew Tridgell [Wed, 17 Aug 2005 01:20:22 +0000 (01:20 +0000)]
r9337: defer the checking of array sizes until the end of the
function/structure, as the checks might involve variables that have
not yet been unmarshalled. This is needed to cope with the correct IDL
for the winreg pipe.

Jelmer, can you look at this when you get a chance and see if you
think this is a reasonable approach? Sorry its wrecks the nice
indentation in the generated code.
(This used to be commit 127ed0cfe2d210f5e927d3768d5372042d1c070c)

13 years agor9335: only copy the in side of an array to the out side of an array when the
Andrew Tridgell [Tue, 16 Aug 2005 23:19:17 +0000 (23:19 +0000)]
r9335: only copy the in side of an array to the out side of an array when the
array is a [ref] pointer. For non-ref arrays it is quite valid for a
server to return a larger response array then the client gave (as can
happen with winreg) in which case this memcpy() will fault.
(This used to be commit 6cf20e7adb9891119fdc9a0b208c3f94fe823334)

13 years agor9328: Remove background:ThreeDFace from qooxdoo styles to
Deryck Hodge [Tue, 16 Aug 2005 20:00:24 +0000 (20:00 +0000)]
r9328: Remove background:ThreeDFace from qooxdoo styles to
get our samba.org background bacl.

(This used to be commit 71291f0d566eaa493a5687a6bce7c5ec9373a7de)

13 years agor9327: Rename parent to fParent to prevent confusion with window.parent object.
Deryck Hodge [Tue, 16 Aug 2005 19:47:27 +0000 (19:47 +0000)]
r9327: Rename parent to fParent to prevent confusion with window.parent object.
(Thanks, Derrell, for the reminder.)
Rename child to fChild, to be consistent and maintain readability.

(This used to be commit 6040f9401eedb195aef160b3b58462a1a6728af7)

13 years agor9321: Fix potential bug found by Coverity. src_len has to be int but later we do...
Alexander Bokovoy [Tue, 16 Aug 2005 13:22:01 +0000 (13:22 +0000)]
r9321: Fix potential bug found by Coverity. src_len has to be int but later we do pass it as size_t. In case src_len is negative, we need to register a failure and return to the caller
(This used to be commit 95d96c79a538814bb524d7905e1e8f64df6341ca)

13 years agor9320: Fix premature dereference bug found by Coverty and also get rid of non-used...
Alexander Bokovoy [Tue, 16 Aug 2005 10:57:21 +0000 (10:57 +0000)]
r9320: Fix premature dereference bug found by Coverty and also get rid of non-used memory context
(This used to be commit 127e06492a545940443c93e9aec66eebefa26dc2)

13 years agor9319: updated newuser script for new OO style for nss object
Andrew Tridgell [Tue, 16 Aug 2005 10:52:02 +0000 (10:52 +0000)]
r9319: updated newuser script for new OO style for nss object

thanks to Hotarut for spotting this
(This used to be commit 3f30c6118ba22fbf52068630f48bcde82182b8a6)

13 years agor9318: fix searches with scope ONE and SUB,
Stefan Metzmacher [Tue, 16 Aug 2005 06:55:40 +0000 (06:55 +0000)]
r9318: fix searches with scope ONE and SUB,

the problem was that ldb_dn_compare_base() just looked at if
both dn's mtach some how, and the following happens:

basedn: CN=bar,DC=foo,DC=com
dn: DC=foo,DC=com

and dn: DC=foo,DC=com was return as result of a sub and base search

and also the ONE search with
basedn: DC=foo,DC=com

returned this

dn: CN=bla,CN=bar,DC=foo,DC=com

(This used to be commit 2a107472c373e425013050e28b944c830f653a87)

13 years agor9311: Missing null pointer check in ejsLib
Steve French [Tue, 16 Aug 2005 00:01:25 +0000 (00:01 +0000)]
r9311: Missing  null pointer check in ejsLib

Found by coverity
(This used to be commit ae7b073af3d690275dbfa11c8406963702c57d3b)

13 years agor9310: Add two missing return statements after null pointer checks to fix potential...
Steve French [Mon, 15 Aug 2005 23:52:15 +0000 (23:52 +0000)]
r9310: Add two missing return statements after null pointer checks to fix potential oops in kdc code.
Found by coverity.
(This used to be commit 3b707b928969c87ac7e9948a567a3ebbc754f28c)

13 years agor9309: Incorrect null pointer check in return from talloc.
Steve French [Mon, 15 Aug 2005 23:15:59 +0000 (23:15 +0000)]
r9309: Incorrect null pointer check in return from talloc.

Found by coverity.
(This used to be commit bafd3afbef9b8d3a8baa55d4b31cc62bfeeed459)

13 years agor9305: Use the check-var.m4 from roken to really, really detect h_errno correctly.
Andrew Bartlett [Mon, 15 Aug 2005 01:38:21 +0000 (01:38 +0000)]
r9305: Use the check-var.m4 from roken to really, really detect h_errno correctly.

This fixes the build on Fedora Core 4.

Andrew Bartlett
(This used to be commit aa36b2adad0d66b8171aa3ccca3057298361bec5)

13 years agor9300: cope with zero length in ndr_pull_charset()
Andrew Tridgell [Sat, 13 Aug 2005 03:17:18 +0000 (03:17 +0000)]
r9300: cope with zero length in ndr_pull_charset()
(This used to be commit 1ae255aba44f4444486ae5bc634c8ab1a6328c87)

13 years agor9299: fixed the evaluation of pointer expressions that evaluate to boolean
Andrew Tridgell [Sat, 13 Aug 2005 03:16:03 +0000 (03:16 +0000)]
r9299: fixed the evaluation of pointer expressions that evaluate to boolean
(This used to be commit 17679676d4629ac81116ae1b102f1a311b5d4eed)

13 years agor9298: pull non-array elements before array elements to overcome the problem
Andrew Tridgell [Sat, 13 Aug 2005 03:14:47 +0000 (03:14 +0000)]
r9298: pull non-array elements before array elements to overcome the problem
of arrays that have length_is() or size_is() properties that refer to
later elements
(This used to be commit 9526f2a5104b50f4b6ab575cfa6270abb0377e51)

13 years agor9297: Add a NDR_CHECK() around each ejs push/pull.
Tim Potter [Sat, 13 Aug 2005 02:59:56 +0000 (02:59 +0000)]
r9297: Add a NDR_CHECK() around each ejs push/pull.

Some experimentation with the top level function calls.
(This used to be commit e8d53d15e4371bc266a691c3c3c82675373ba7f4)

13 years agor9296: Reinstate generation of push/pull functions for each field in a struct
Tim Potter [Sat, 13 Aug 2005 02:47:14 +0000 (02:47 +0000)]
r9296: Reinstate generation of push/pull functions for each field in a struct
or union.
(This used to be commit 72420f24e654661c8792bd7d0eae610ad182d529)

13 years agor9295: Tridge had some good suggestions for simplifying the smb_interfaces ejs
Tim Potter [Sat, 13 Aug 2005 02:40:02 +0000 (02:40 +0000)]
r9295: Tridge had some good suggestions for simplifying the smb_interfaces ejs
code.  I will try and get just the push/pull functions working rather than
getting in a mess with trying to do everything at once.
(This used to be commit a7504ebf9557796bd50907620115f8367ea586f0)

13 years agor9292: More compiles fixes for autogenerated smb_interfaces ejs wrappers.
Tim Potter [Sat, 13 Aug 2005 01:42:14 +0000 (01:42 +0000)]
r9292: More compiles fixes for autogenerated smb_interfaces ejs wrappers.

Add some code to try and work out whether a smb_interface requires a
TALLOC_CTX in the function signature.  I'm not sure whether this is
going to work as the raw api is a bit inconsistent I think.
(This used to be commit 13a101653ea7bba3b8b1f74637d981eb004484cc)

13 years agor9291: Add a stub for smb_raw_lpq() which is in smb_interfaces.h but doesn't
Tim Potter [Sat, 13 Aug 2005 01:15:32 +0000 (01:15 +0000)]
r9291: Add a stub for smb_raw_lpq() which is in smb_interfaces.h but doesn't
seem to be used anywhere yet.
(This used to be commit 9e5ce3a28892241e2b080c0fa187ee99042c2330)

13 years agor9284: Whitespace. (-:
Tim Potter [Fri, 12 Aug 2005 23:01:57 +0000 (23:01 +0000)]
r9284: Whitespace.  (-:
(This used to be commit 6a7922a02c19b5ea2b77011ff593c6d535d0b212)

13 years agor9279: In Samba3 debuglevel 10 used to print RPC debugs. I think this should
Tim Potter [Fri, 12 Aug 2005 22:55:53 +0000 (22:55 +0000)]
r9279: In Samba3 debuglevel 10 used to print RPC debugs.  I think this should
apply to Samba4 as well so change the DEBUGLEVEL > 10 to DEBUGLEVEL >= 10.
(This used to be commit 2215cc08b7b94df3fff9cfff6c839bbf2cb6e664)

13 years agor9260: only overwrite destination files when the temp files was created correct
Stefan Metzmacher [Fri, 12 Aug 2005 09:51:40 +0000 (09:51 +0000)]
r9260: only overwrite destination files when the temp files was created correct

(This used to be commit b5626c978fe3349f0972e1aeea51b05e66c6ce33)

13 years agor9258: return is only supported in function, not in plain shell scripts
Stefan Metzmacher [Fri, 12 Aug 2005 09:20:32 +0000 (09:20 +0000)]
r9258: return is only supported in function, not in plain shell scripts

(This used to be commit 3b99c3685734e398ab007e93dd7f468027c1847f)

13 years agor9257: - I assume we don't need unistd.h in this file,
Stefan Metzmacher [Fri, 12 Aug 2005 09:05:09 +0000 (09:05 +0000)]
r9257: - I assume we don't need unistd.h in this file,
  that sould fixes the build on aix
- use "" instead of <> for includes in our tree

(This used to be commit 8c8fd4cfc6bbad9e8ec2ae9998a6cf4bd90b4bc6)

13 years agor9240: - move struct security_token to the idl file, with this we can
Stefan Metzmacher [Thu, 11 Aug 2005 13:12:45 +0000 (13:12 +0000)]
r9240: - move struct security_token to the idl file, with this we can
  the ndr_pull/push/print functions for it in the ntacl-lsm module

- fix compiler warnings in the ldap_encode_ndr_* code

(This used to be commit 83d65d0d7ed9c240ad44aa2c881c1f07212bfda4)

13 years agor9239: I think this is the right IDL for the eventlog record. Pidl is
Tim Potter [Thu, 11 Aug 2005 04:05:21 +0000 (04:05 +0000)]
r9239: I think this is the right IDL for the eventlog record.  Pidl is
generating incorrect code for arrays of strings here.
(This used to be commit 3b2476e0a00dbd3b552ccde736147e93655732f1)

13 years agor9238: Some test code to do user-unmarshalling of eventlog_Record data.
Tim Potter [Thu, 11 Aug 2005 04:04:16 +0000 (04:04 +0000)]
r9238: Some test code to do user-unmarshalling of eventlog_Record data.
(This used to be commit d0225f10797eaeeb6d10cf753578703e144ee8dd)

13 years agor9237: Fix eventlog_Record typedef so it decodes records source and computer
Tim Potter [Thu, 11 Aug 2005 03:31:35 +0000 (03:31 +0000)]
r9237: Fix eventlog_Record typedef so it decodes records source and computer
names.  Need to get working with SIDs and extra data.
(This used to be commit 2543f78df61b76295acf6fe4837adefbe08ca5c4)

13 years agor9235: Remove attribute search we no longer reference.
Andrew Bartlett [Wed, 10 Aug 2005 22:28:37 +0000 (22:28 +0000)]
r9235: Remove attribute search we no longer reference.

Andrew Bartlett
(This used to be commit cf8bf1e9f3d771fb3ea14949cd6963c3e9ac6c2d)

13 years agor9234: Ensure we always change the end of the PAC, no matter what it is. Fix
Andrew Bartlett [Wed, 10 Aug 2005 22:27:55 +0000 (22:27 +0000)]
r9234: Ensure we always change the end of the PAC, no matter what it is.  Fix
typo in comment.

Andrew Bartlett
(This used to be commit c96f8594b16c6a5310e2e8727bb6975f2a193231)