Quoth Rob Richardson <rdrichard...@rad-con.com>, on 2013-11-06 14:08:34 +:
> In Igor's post below, what is the meaning of the colon in front of mypid?
Parameters/placeholders.
http://sqlite.org/lang_expr.html#varparam
--->
ader dependency.
The truncation is actually a potential error: e.g., a row ID of 2^32
would be returned as 0 instead on a system with 32-bit int. It's the
sort of thing you might not see in production for a while until it
breaks everything suddenly a ways down the line.
--
you don't wish to
say). If not, then you may actually have a primary key of the whole
row, in which case I'm not sure why inventing a rowid is needed.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
nks the
same way, so there's a subjective element.)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
et as being used in the specific
technical sense involved (data is reflected automatically between
mmap and read/write), by relation to "cache coherent". So in that
sense it is both correct and not a disparagement of the OpenBSD
behavior.
---> Drake Wilson
_
is
accessible (among other ways) as pblah[0]. pblah[1] is out of bounds,
and depending on how the compiler allocates those vars it may wind up
aliasing the db pointer. This is not an SQLite problem.
---> Drake Wilson
___
sqlite-users mailing lis
esses is a speed problem and you are willing to spend more
memory to deal with it, some of the first things to try would be
fiddling with the cache_size, synchronous, and journal_mode PRAGMAs,
depending on what tradeoffs you want to make.
> - paldepind
---> Drake Wilson
two.
You probably want to do a BEGIN EXCLUSIVE before loading the schema in
most cases. (The EXCLUSIVE may not strictly be necessary, but I find
it makes things clearer.)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http:
NULL
> char._
No. I strongly suspect that's a red herring.
In summary:
- Make sure msg.num_bytes_in is actually set to what you want.
- Make sure you're handling the lifetime of the buffer correctly;
for testing purposes I'd use SQLITE_TRANSIENT rather than
SQLITE_STATIC, since that
restriction on the RHS of a MATCH.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
currently I am
> using the same basic logic for both. The problem is the second time
> around I get a SQLITE_MISUSE.
>
> What am I doing wrong?
You probably need to sqlite3_reset the statement after stepping it.
---> Drake Wilson
___
sq
g all undefined values? That doesn't
seem like a valid thing to do. Presumably you should set the result
values to indicate "no constraints used, no ordering consumed, an
arbitrary high cost estimate, and an indicator for full-scan access
(that will be recognized by the xFilter method)",
m compiled SQLite3 (for 32-bit integers; the doc is
silent re int64 type) and the default is a compile-time limit of 10.
Whether this is a problem depends on your data and application
architecture.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-u
records with the same meaning? If so, that may
indicate that a single table solution is more appropriate so that any
subset of the records can be selected at once.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.
f you want them to merely be attached to the
database connection rather than part of the permanent schema, you can
use CREATE TEMPORARY VIEW instead.
> Simon.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org
to struct pointer, then use this pointer to get data and
> insert into database, is there any faster to do this converting?
It sounds like you may be running into http://sqlite.org/faq.html#q19.
---> Drake Wilson
___
sqlite-users mailing list
sqli
ff that hasn't gotten much of anywhere yet.)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
esses anyway? You could just use one process
with both a timer and packet reception, and then a single in-memory
single-process db and db handle (or other data structures, but I
presume you've found SQLite is convenient for some other reason).
---&g
dex reads
to the compiled query when an index would not otherwise be used.
> Max
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
vant columns empirically via asking which columns would be
used in the temporary index creation, for the purpose of creating
permanent indices. This is useful as a secondary check even if one
should theoretically know which columns will be used beforehand.
---&g
/remove triggers to keep
track of that yourself in a single-row auxiliary table instead.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ed "s" instead.
Tables are not themselves sources in the strictest sense; a source
references a table (and "is" the table for many purposes).
http://archives.postgresql.org/pgsql-general/2000-08/msg00740.php
claims that SQL92 requires this behavior.
---> Drake Wilson
The time interval specified one million or more microseconds.
Note that the function is both XSI and marked obsolescent. POSIX:2008
removes usleep entirely. nanosleep is the replacement, AFAIK.
---> Drake Wilson
___
sqlite-users mailing li
e binding that bit 63 set becomes a sign bit rather than
raising an error, I suppose. I didn't see you mention that anywhere,
though.
> Thank you,
> Marco
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
writes. WAL mode would help with that, but it would increase the
underlying complexity WRT filesystem and shared memory accesses.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ome these kinds of
limitations in some cases.
> Thanks again
>
> Lynton
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
choose which only at database creation time.)
Similarly, make sure that you actually give it Unicode strings in the
target encoding; there may be some autocorrection going on if you try
to feed it Latin-1 characters, but I wouldn't rely on it.
> Thanks,
> Gr
to do this with SQL? Why not just do
it in plain application logic? In the absence of more information,
that would seem a more natural way to go about it.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
can use a WHEN
clause:
CREATE TRIGGER auto_delete_summary AFTER DELETE ON detail
FOR EACH ROW WHEN NOT EXISTS (SELECT * FROM detail WHERE key = OLD.key)
BEGIN
DELETE FROM summary WHERE key = OLD.key;
END;
---> Drake Wilson
___
sqlite-users
riggers, which run only once when triggered by a statement
that alters multiple rows. I prefer to use the explicit form, but
omitting it would change nothing.
The full CREATE TRIGGER syntax is of course part of the documentation,
at http://sqlite.org/lang_createtrigger.html.
---> Drake Wilson
a.key = b.key)
with a suitable composite equality in the subselect.
> Is my data too big for SQLIte? is it really hanging or it could be taking 3
> days?
Table A and (particularly) table B both have indices on those columns,
right?
---> Drake Wilson
_
primary keys?
A primary key declaration implicitly creates a unique index, yes.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
isk or "rewriting" sectors of a flash-based device, or similar
information for filesystems that don't do in-place writes. Pointers
would be appreciated.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
y files in both of
the cases you mention, unless (in the latter case) you actually have
per-block (or similar) crypto going on rather than purely streaming.
(You do use seekp, but some underlying streams might not support it.)
---> Drake Wilson
___
sqlite-
gardless of
any of this.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
the above takes 16 ms total; I imagine the use of GROUP BY and
EXISTS and the lack of the extra DISTINCT are the primary factors, but
I haven't checked thoroughly enough to say so confidently. I'm using
SQLite 3.7.2 on Debian GNU/Linux sid AMD64.
---> Drake Wilson
__
Quoth Drake Wilson <dr...@begriffli.ch>, on 2010-10-05 03:24:01 -0700:
> > My current task is to get the number of foods that belong to each
> > group and have at least one weight data related to them.
>
> That suggests something like:
>
> SELECT g.Z_PK AS &q
specific without knowing
what kind of application is being developed.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
using LIKE suggests that 'ian' and 'Ian' should
be treated identically, but currently your primary key allows separate
rows to exist for each of those.
Also, PRIMARY KEY UNIQUE is redundant. A primary key is always
unique.
---> Drake Wilson
__
row, right? On a
table with around 3k rows, it seems a little odd that it would take
that long, even if updating every row tends to be expensive in
general. What does your schema look like, if I might ask? Is there
significant concurrent access with that giant update?
>
e and access, because you'd be
using the SQLite components in a more natural way. The presence
of that \t suggests that you might be storing sequences of records
in those fields to start with; those could well be separate rows
in a suitable secondary t
neral case.)
See http://us2.php.net/manual/en/sqlite3stmt.bindvalue.php if you
haven't already.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
is no full procedural language.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
abases. Hmm.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
'symbol 1'), then later
('file B', 'symbol 1') and so on, and you can't trivially get the
DISTINCT out of that without sorting in temporary storage.
Using UNIQUE (symbol, file) instead would seem the obvious solution.
Is there a reason you can't do tha
Quoth Travis Orr <t...@ivl.com>, on 2010-10-12 08:17:38 -0700:
> Drake Wilson said:
> - However, it now occurs to me that it may be possible to use the
> - fts3_tokenizer() function in a trigger, which is probably a bad thing
> - when writing to untrusted databases.
>
>
page at
http://www.sqlite.org/src/wiki?name=Bug+Reports says that posting to
the list is the correct way to do this, please consider the above such
a request. (I will look into a patch if I have time, though this is
moderately unlikely.)
Additional comments are natura
IGNORE INTO Code_Units ("code", "unit")
VALUES (NEW."code", NEW."unit");
SELECT CASE WHEN (SELECT "unit" = NEW."unit" FROM Code_Units WHERE
"code" = NEW."code")
THEN 1 ELSE RAISE(ABORT, "Code associated with multiple
units")
END;
END;
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
I don't know what the SQL92 standard has to say on this, FWIW.)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
base, so if you need
to set characteristics of the file before doing the sqlite3_open(),
you can open() or CreateFile() it or whatever beforehand.
> Thanks in advance
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.
Quoth Frank Millman <fr...@chagford.com>, on 2010-10-20 11:47:06 +0200:
> Ok, thanks.
>
> Is there any chance of it being considered for a future release?
Search http://www.sqlite.org/cvstrac/wiki?p=SqliteWikiFaq for "foreign
key".
>
comparison function.
In step (d), you're doing what? Sorting the resulting rows? How
exactly would you use the index for that?
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ind is that maybe I haven't set up the
> database properly to accept UTF8 or UTF16 data, but I figured this was a
> default in SQLite3.
You have to pick one when you create the database, usually UTF-8. If
you want UTF-16 use « PRAGMA encoding = 'UTF-16' » (or 'UTF-16le' or
'UTF-16be') when you cre
specified as column types
directly. Instead, an SQL column type is used. In fact, the column
type "NONE" will be detected as NUMERIC affinity, per the rules in the
documentation. I would use a blank type to declare a column of
varying type; that would give the NONE affinit
to allow control of
within-group ordering using ORDER BY. But is this part of the public
interface, or is it an oddity that may change in future revisions?
Hipp's response upthread seems to indicate the former, but I'd rather
be sure.
---> Drake Wilson
N is guaranteed to be semantically after ORDER BY
processing and therefore allows controlling which row from a group is
selected, if one is careful.
Thanks for the corrections.
>-j
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite
nfortunately.
These aren't unsolvable, but it's a little harder than it might look.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ible to have a single query that will generate a row for each
> PattenID, COUNT(Offset_Y) combination?
Does SELECT PatternID, COUNT(DISTINCT Offset_Y) FROM Tiles GROUP BY
PatternID do what you're looking for?
---> Drake Wilson
___
sqlite-users mai
s you probably don't need it as much as you think.
http://sqlite.org/lang_altertable.html shows that modifying existing
column types in-place is not available.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sql
cit in the mykey column's index. Is
> this in fact the case?
When doing which queries?
How do you propose to look up a key value in the index without using
the collation function?
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite
/COMMIT pair.
If the condition is complicated enough and you want to save
recomputing it, you can create a temporary table to hold the _rowid_
values from the original and then use WHERE _rowid_ IN (SELECT ...)
from the temporary table to identify the rows to be moved.
---&g
propagation would work without
yielding surprising results.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
e> SELECT a,b FROM (
> ...> SELECT ID a, Price b FROM OrderTest WHERE Price < 200
> ...> UNION
> ...> SELECT ID a, Price b FROM OrderTest WHERE Price > 500
> ...> )
> ...> ORDER BY a IS 0, b;
> a|b
> 3|0.0
> 4|25.0
> 1|50.0
>
directly to do with SQLite. SQLite is only
a database engine that is used by many applications to store different
types of data. You might go search for help related to the specific
handset software in use instead.
---> Drake Wilson
___
sqlite-users m
ill gain you no
parallelism in the modes where it works. Aside from that, transaction
state is bound to a handle; you're starting a transaction and then
trying to start another one inside it.
Open two handles instead.
---> Drake Wilson
___
sqlite-users mail
IN transactions. It's
hard to tell what exactly you're doing from the description, such as
why you're doing these updates with two threads to start with, so it's
hard to give good advice. Perhaps you could show some example code?
Which threading mode do you mean? Serialized or multithreaded?
--
ry to union the two directions together, then probably
always insert non-reflexive entries in lexicographical order for
consistency (to create an invariant of one row per pairing). The
latter might be most easily done with a view of « SELECT a, b FROM t
UNION SELECT b, a FROM t » but I'm not sure how effic
Quoth Dariusz Matkowski <dmatkow...@rim.com>, on 2010-12-03 18:46:20 -0500:
> Phobic
What?
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
of the closure
structure (in this case, have each callback invocation increment the
next-pointer).
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
SQLite might change to raise errors instead of accepting the malformed
| statements covered by the exceptions above.
(I suspect the real answer is "don't do that", but I'm not entirely
confident.)
---> Drake Wilson
___
sqlite-users mailing lis
ve experience
is mostly with Access. Having additional context might allow more
useful suggestions beyond purely syntactic issues.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
>
> Note the added VALUES keyword.
Oh yes. D'oh! I think I accidentally hit kill-word before sending;
sorry about that. (The other response about the table definitions is
useful too.)
---> Drake Wilson
___
sqlite-users mailing list
sql
00.dll?".
[1]
http://kobyk.wordpress.com/2007/07/20/dynamically-linking-with-msvcrtdll-using-visual-c-2005/
[2] http://msdn.microsoft.com/en-us/library/abx4dbyh.aspx
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8
fail, stop executing
further statements and roll back, otherwise commit at the end. Don't
include the begin/end in the list. Does that not work for you?
> Thanks,
> Tom
> BareFeetWare
---> Drake Wilson
___
sqlite-users mailing list
sqlite
ax is easy to learn
It might help your queries work too
Like the oompa loompa oompa
Oompa loompa doopity do
o/`
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
e awkward unless you turn foreign keys
off first, but...
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
.. and yet you're passing a length of VARNAM_SYMBOL_MAX instead,
which I'm guessing is not 1. Pass the real length of the string (not
the size of the buffer), or -1 to treat it as a NUL-terminated C
string. Otherwise you're grabbing extra bogus bytes.
---> Drake Wilson
___
g that's the exact length in bytes, and any NUL characters within
that number of bytes will be included in the string.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ecially if the SQLite component may later be
replaced with the expectation of backwards compatibility.
> Simon.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ite3_open_v2 always takes UTF-8 and
does any filesystem-specific encoding transformations internally. (It
may still be that it does it incorrectly on some platforms, in which
case that may be a bug.)
> -- Tito
---> Drake Wilson
___
sqlite-users m
ou better index usage on the
name regardless of whether you change to a synthetic integer primary
key. Of course you have to do the normalization the same way when
writing the records to the DB in the first place.
> Thanks,
> Ian
---> Drake Wilson
__
ady know enough
to recognize exactly why their cases are different.)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
e you're actually _storing_ all the data?
Can you verify that you can get all the bytes out in any way at all?
Information about the schema in use would be helpful, in general.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ake
if you try to insert with both of them null? More generally, could
you show some example inserts with what behavior you expect? I
suspect what you're looking for is best done some other way.
---> Drake Wilson
___
sqlite-users mailing list
sqlit
ate compound UNIQUE constraint including that column, but that
doesn't allow (A, B) and (A, C) to exist simultaneously.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
er journal where
they would not also be colliding on at least one database, but I'm not
confident about that.
I would try it and see what happens, but also be rather cautious about
the design in such cases; it's hard to judge more accurately without
knowing more about the application.
>
If it's inconvenient, you
could provide your own chromium_sqlite3_openhandle(handle, ...)
function which would do the conversion and call sqlite3_open behind
the scenes.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http
Quoth Roger Binns <rog...@rogerbinns.com>, on 2011-02-28 13:03:43 -0800:
> On 02/28/2011 12:41 PM, Drake Wilson wrote:
> > Back on the original topic, I would rather think a custom VFS sounds
> > like the way to go;
>
> It is technically correct that will
Quoth Drake Wilson <dr...@begriffli.ch>, on 2011-02-28 14:44:38 -0700:
> Furthermore, another approach if the name<->FD thing is the only
> requirement would be to retrieve all the original VFS methods at init
> time (using sqlite3_vfs_find) and only alter a few of them when
of the 64-bit integer space. If
neither of those is true, you're probably looking at probing several
times to avoid collisions, and that's not something the stock "pick a
new row ID" mechanism handles AFAIK.
---> Drake Wilson
___
sqlite-users
es, it's a PRNG, but even if the total period is at least 2^64, that
doesn't guarantee no repeated 64-bit values unless the output reflects
the entire state, no? ISTR SQLite using (A)RC4.
And that doesn't help between connections.
---> Drake Wilson
_
f them will
work in some situations. Now _please_ try to understand this _before_
throwing another dozen SELECT statements at the list. Don't just ask
people to give you the magic formula or ask why it doesn't work when
you plug random things together; try to learn the underlying
principles properl
t what a string is "supposed"
to be and compare accordingly.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
g in the result?
This is not a realtime chat system. If you won't wait even ten
minutes before squawking about the same thing again, a mailing list
may not be for you. Are you understanding any of the responses at
all? Give a sign of it if so!
And "y" and "Y" are not th
ght plausibly trigger unfortunate codepaths somewhere in the kernel
in the process of failing.
I'm interested to see whether any of the above does any good, to
improve my own knowledge of NFS. :-)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-u
eed of a guaranteed response, may I suggest hiring
someone to be specifically responsible for giving you the results you
need? http://www.hwaci.com/sw/sqlite/prosupport.html is an obvious
place to start.
---> Drake Wilson
___
sqlite-users mai
a quick grep does not find currently, but that is an open
question, not a hard recommendation.)
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
important here.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
Quoth Drake Wilson <dr...@dasyatidae.net>, on 2013-04-04 10:20:44 -0500:
> So it is perfectly okay to use unprotected mmap accesses if an I/O
> error on the file will already make the entire process uncontinuable.
> The question is whether this applies to arbitrary S
n't help the
case of Unixy systems at all.
---> Drake Wilson
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
1 - 100 of 111 matches
Mail list logo
