IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Don't use nstrings to hold workgroup and netbios names. The problem with them is that MB netbios
and workgroup names in unix charset (particularly utf8) may be up to 3x bigger than the name
when represented in dos charset (ie. cp932). So go back to using fstrings for these but
translate into nstrings (ie. 16 byte length values) for transport on the wire.
Jeremy.
there are multiple "."'s in the name.
This code is protected with an #ifdef TRUNCATE_NETBIOS_NAME and this
is #define'd to 1 directly above. Should we also get rid of the #ifdef?
It appears that we pass filename through resolve_wildcards() as pstring and use fstring temporary buffer there.
As result, a long filename in unix charset (UTF-8 by default) can easily expand over 255 bytes while
Windows is able to send to us such names (e.g. Japanese name of ~190 mb chars) which we unable to process through
this small fstring buffer. Tested with W2K and smbclient, Japanese and Cyrillics.
what it was :-). Allow msdfs links to now look like UNC paths : eg.
sym_link -> msdfs://server/share/path/in/share
or :
sym_link -> msdfs:\\server\share\path\in\share
Jeremy.
boolean parameter that allows broken iconv conversions to work. Gets rid of the
nasty errno checks in mangle_hash2 and check_path_syntax and allows correct
return code checking.
Jeremy.
I know this isn't pretty, but neither was our assumption that all strings
from the directory fit inside a pstring. There was no way this worked
before will all versions of usrmgr (for example, the only version of
mine that has the TS Confic button).
character set. Because of the allowing of "broken conversions" for people
who have broken iconv libraries we can't rely on the return from convert_string()
to be valid - we must check errno instead. This is ripe for revisiting at
some stage. I prefer adding a bool parameter to all convert_string_XX varients
to specify if we will allow broken conversions or not. With "allow_broken_conversions"
set to false we could then rely on the return from convert_string rather than
checking errno.
Jeremy.
in 3.0.2a actually). We now send a correct referral back, but the client
refuses to follow it... Not sure why. Will do more tests.. Maybe unicode
character count is wrong (it looks it).
Jeremy.
Big thanks to tpot and mbp for showing how easy it can be to write a simple
unit test, and for providing the STF.
This also changes the strstr_m() code to use strstr_w() (avoiding
duplication) and fixes it so that it passes the STF.
(We now always restart before doing the unicode run, until sombody can
show me why the testsuite is wrong).
Andrew Bartlett
For example:
strstr_m("%v foo bar", "%v") would fail...
only strstr_m("foo %v", "%v") could work.
I wonder what else this broke...
Fix is to move to using strncmp() inside the strstr_m function.
Tested on ASCII only.
Andrew Bartlett
A windows DC does not reply to DCNAME\\Administrator, only to
DOMAIN\\Administrator. Fix that.
Without winbind we are wrong as domain members, we should forward the request
DOMAIN\\Username to the DC on behalf of the asking client. Winbind fixes that
nicely.
Volker
so breaks when substitution '/' and '\'). It's used by unix_clean_name(),
which is used by reduce_name, which is used by check_name() (phew!).
Now that we know all filenames passed to check_name() are in a "good"
format (no double slashes, all '\\' chars translated to '/' etc.) due
to the new check_path_syntax() we can avoid calling reduce_name unless
widelinks are denied. After this check-in I can fix all_string_sub() to
handle mb chars correctly as it won't be in the direct path in the
main path handling code.
Jeremy.
