Postgresql - change the size of a varchar column to lower length - sql

I have a question about the ALTER TABLE command on a really large table (almost 30 millions rows).
One of its columns is a varchar(255) and I would like to resize it to a varchar(40).
Basically, I would like to change my column by running the following command:
ALTER TABLE mytable ALTER COLUMN mycolumn TYPE varchar(40);
I have no problem if the process is very long but it seems my table is no more readable during the ALTER TABLE command.
Is there a smarter way? Maybe add a new column, copy values from the old column, drop the old column and finally rename the new one?
Note: I use PostgreSQL 9.0.

In PostgreSQL 9.1 there is an easier way
http://www.postgresql.org/message-id/162867790801110710g3c686010qcdd852e721e7a559#mail.gmail.com
CREATE TABLE foog(a varchar(10));
ALTER TABLE foog ALTER COLUMN a TYPE varchar(30);
postgres=# \d foog
Table "public.foog"
Column | Type | Modifiers
--------+-----------------------+-----------
a | character varying(30) |

There's a description of how to do this at Resize a column in a PostgreSQL table without changing data. You have to hack the database catalog data. The only way to do this officially is with ALTER TABLE, and as you've noted that change will lock and rewrite the entire table while it's running.
Make sure you read the Character Types section of the docs before changing this. All sorts of weird cases to be aware of here. The length check is done when values are stored into the rows. If you hack a lower limit in there, that will not reduce the size of existing values at all. You would be wise to do a scan over the whole table looking for rows where the length of the field is >40 characters after making the change. You'll need to figure out how to truncate those manually--so you're back some locks just on oversize ones--because if someone tries to update anything on that row it's going to reject it as too big now, at the point it goes to store the new version of the row. Hilarity ensues for the user.
VARCHAR is a terrible type that exists in PostgreSQL only to comply with its associated terrible part of the SQL standard. If you don't care about multi-database compatibility, consider storing your data as TEXT and add a constraint to limits its length. Constraints you can change around without this table lock/rewrite problem, and they can do more integrity checking than just the weak length check.

Ok, I'm probably late to the party, BUT...
THERE'S NO NEED TO RESIZE THE COLUMN IN YOUR CASE!
Postgres, unlike some other databases, is smart enough to only use just enough space to fit the string (even using compression for longer strings), so even if your column is declared as VARCHAR(255) - if you store 40-character strings in the column, the space usage will be 40 bytes + 1 byte of overhead.
The storage requirement for a short string (up to 126 bytes) is 1 byte
plus the actual string, which includes the space padding in the case
of character. Longer strings have 4 bytes of overhead instead of 1.
Long strings are compressed by the system automatically, so the
physical requirement on disk might be less. Very long values are also
stored in background tables so that they do not interfere with rapid
access to shorter column values.
(http://www.postgresql.org/docs/9.0/interactive/datatype-character.html)
The size specification in VARCHAR is only used to check the size of the values which are inserted, it does not affect the disk layout. In fact, VARCHAR and TEXT fields are stored in the same way in Postgres.

I was facing the same problem trying to truncate a VARCHAR from 32 to 8 and getting the ERROR: value too long for type character varying(8). I want to stay as close to SQL as possible because I'm using a self-made JPA-like structure that we might have to switch to different DBMS according to customer's choices (PostgreSQL being the default one). Hence, I don't want to use the trick of altering System tables.
I ended using the USING statement in the ALTER TABLE:
ALTER TABLE "MY_TABLE" ALTER COLUMN "MyColumn" TYPE varchar(8)
USING substr("MyColumn", 1, 8)
As #raylu noted, ALTER acquires an exclusive lock on the table so all other operations will be delayed until it completes.

if you put the alter into a transaction the table should not be locked:
BEGIN;
ALTER TABLE "public"."mytable" ALTER COLUMN "mycolumn" TYPE varchar(40);
COMMIT;
this worked for me blazing fast, few seconds on a table with more than 400k rows.

Adding new column and replacing new one with old worked for me, on redshift postgresql, refer this link for more details https://gist.github.com/mmasashi/7107430
BEGIN;
LOCK users;
ALTER TABLE users ADD COLUMN name_new varchar(512) DEFAULT NULL;
UPDATE users SET name_new = name;
ALTER TABLE users DROP name;
ALTER TABLE users RENAME name_new TO name;
END;

Here's the cache of the page described by Greg Smith. In case that dies as well, the alter statement looks like this:
UPDATE pg_attribute SET atttypmod = 35+4
WHERE attrelid = 'TABLE1'::regclass
AND attname = 'COL1';
Where your table is TABLE1, the column is COL1 and you want to set it to 35 characters (the +4 is needed for legacy purposes according to the link, possibly the overhead referred to by A.H. in the comments).

Try run following alter table:
ALTER TABLE public.users
ALTER COLUMN "password" TYPE varchar(300)
USING "password"::varchar;

I have found a very easy way to change the size i.e. the annotation #Size(min = 1, max = 50) which is part of "import javax.validation.constraints" i.e.
"import javax.validation.constraints.Size;"
#Size(min = 1, max = 50)
private String country;
when executing this is hibernate you get in pgAdmin III
CREATE TABLE address
(
.....
country character varying(50),
.....
)

Related

SQLiteexception database disk image is malformed

I have a weird error with a SQLite Database: You can download it here
Everytime I try to insert something in the Table "CurrencyTransactions" it fails because a new column called 7 appeared for no reason.
I tried to drop the table but
I ran PRAGMA integrity_check but I've this error then
Then I tried to export a .sql file and to import it again in a fresh new database but
1) If I import the structure only, it works fine and I don't have the 7 column anymore
2) If I import the entries then, it fails with this error:
It means something like: "Error in process #74: not an error"
To finish, I also tried this solution but the new database created is empty.
What can I do? I really need to save the entries.
What I suggest is in DB Browser.
File/Export/Database to SQL file.
Select All (for all tables)
Other options up to you other than Export Everything
Save the file.
Close the database.
Open a new database e.g nadekobotfix.db. (could be same name but different location)
Note 1-6 takes a minute or so (just under 60k).
Do the hard work according to :-
You may need to remove/ignore the first and last lines (BEGIN TRANSACTION; and the subsequent COMMIT;)
You would probably not be able to run the generated SQL directly due to constraints (tried this an failed with constraints).
You need to copy sections from the file and run according to the hierarchy as imposed by the constraints (foreign keys). If you have CHECK constraints these may need to be considered. (no Triggers to worry about).
Running SELECT * FROM sqlite_master WHERE type = 'table' AND instr(sql,'CHECK'); returns nothing so there are no CHECK constraints.
Indexes could/should be left till last (as they are in the generated SQL).
A section would consist of a table's create statement along with the insert statements.
You may wish to create a spreadsheet of the tables(sections) marking them off when they have been done.
The following query could assist as those with NA could be done first
SELECT CASE WHEN instr(sql,'FOREIGN KEY') THEN 'FK' ELSE 'NA' END AS fkey, name,sql
FROM sqlite_master
WHERE type = 'table' AND name NOT LIKE 'sqlite%' ORDER BY instr(sql,'FOREIGN KEY')
you could export individual tables from DB Browser for SQlite marking them off when done.
You may wish to do an integrity_check at regular intervals.
If this works (you might have to make adjustments to the SQL) then you can rename the old db and then rename the new (or move the old and the copy the new if using the same database name).
Note you may still have to determine how the corruption occurred.
You may wish to backup the database regularly.
You may wish to have a look at How To Corrupt An SQLite Database File
You may wish to heed :-
With few exceptions, analysis of a corrupt database does not normally
help to determine what went wrong. A better approach to avoiding
"danger", we have found, is to read and understand
https://www.sqlite.org/howtocorrupt.html
* in database main *
Page 10628: btreeInitPage() returns error code 11
This indicates that the page header is so badly corrupted that SQLite
cannot interpret this page at all. One possible reason: page 10628
has been zeroed. Can you look at a hex dump of that page? (Remember
that SQLite numbers pages beginning with 1, so the start of the page
is pgsz*10627 where pgsz is the page size.)
-- D. Richard Hipp
“btreeInitPage() returns error code 11”
Sample adjustment required
The Reminders table has a column called When, this is an SQL keyword (inadvisable column name IMO) so the generated SQL for the INSERT doesn't wrap the column name so you will get an error.
i.e. :-
CREATE TABLE IF NOT EXISTS `Reminders` (
`Id` INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT,
`ChannelId` INTEGER NOT NULL,
`IsPrivate` INTEGER NOT NULL,
`Message` TEXT,
`ServerId` INTEGER NOT NULL,
`UserId` INTEGER NOT NULL,
`When` TEXT NOT NULL,
`DateAdded` TEXT
);
INSERT INTO `Reminders` (Id,ChannelId,IsPrivate,Message,ServerId,UserId,When,DateAdded) VALUES (1270,367886754973351936,1,'Birthday Day',318127386367623170,367886754973351936,'2018-05-03 22:07:48.1860996','2018-03-18 22:07:48.186101'),
(1271,248278722656993281,1,'to remind Chanmi to remind Jayos to DeagleMomoka',318127386367623170,248278722656993281,'2018-05-05 22:08:58.4915565','2018-03-18 22:08:58.4915582'),
(1376,170240129414201344,1,'timely',318127386367623170,170240129414201344,'2018-03-29 09:00:29.4476776','2018-03-28 09:00:29.447679'),
(1377,373301201158144000,1,'timely',318127386367623170,373301201158144000,'2018-03-29 09:50:14.1631563','2018-03-28 09:50:14.1631577'),
(1378,248278722656993281,1,'timely',318127386367623170,248278722656993281,'2018-03-29 11:24:27.0250275','2018-03-28 11:24:27.025029'),
(1379,421433212716318721,1,'to timely',318127386367623170,421433212716318721,'2018-03-29 19:21:17.7465563','2018-03-28 19:21:17.7465584'),
(1380,346513954966863872,1,'t',318127386367623170,346513954966863872,'2018-03-29 19:42:23.4758798','2018-03-28 19:42:23.4758816'),
(1381,272735316002209792,1,'t!daily',318127386367623170,272735316002209792,'2018-03-29 21:01:47.5616218','2018-03-28 21:01:47.5616236'),
(1382,298272937243312132,1,'timely',318127386367623170,298272937243312132,'2018-03-29 23:18:02.8826873','2018-03-28 23:18:02.8826891'),
(1383,332340162774302720,1,'t',318127386367623170,332340162774302720,'2018-03-30 01:55:21.4704139','2018-03-29 01:55:21.4704156'),
(1384,367165474246754314,1,'tatyahaksodoeo',318127386367623170,367165474246754314,'2018-03-30 03:46:18.8805182','2018-03-29 03:46:18.8805196'),
(1385,290086674761908225,1,'timely',318127386367623170,290086674761908225,'2018-03-30 07:02:33.4115303','2018-03-29 07:02:33.4115321'),
(1386,168064128500367360,1,'timely',318127386367623170,168064128500367360,'2018-03-30 07:19:09.1915867','2018-03-29 07:19:09.1915885');
would have to be changed to use (square brackets, single or double quotes or grave accents can be used to enclose/wrap/quote the offending keyword):-
.......INSERT INTO `Reminders` (Id,ChannelId,IsPrivate,Message,ServerId,UserId,[When],DateAdded) ......
Likewise table SelfAssignableRoles has the GROUP keyword as a column name.
Likewise table Permissionv2 and table StartupCommand have the INDEX keyword as a column name.
Potential Issue
As an exercise I've tried doing the above and have managed to get 67 out of the 71 tables (66 out of 70 of your tables as sqlite_sequence is automatically created).
However, there appears to be an issue, between the Clubs table and the DiscordUser table. I believe that there is a circular reference between them. Thus as WaifuInfo and WaifuUpdates are reliant upon the DiscordUser table and as WaifuItem is realiant upon WaifInfo. The tables mentioned here have not been successfully copied.
A word of warning. If you attempt to create Clubs and or DiscordUser using the existing constraints you may end up in a situation where one always has to exist.
e.g. if DiscordUser exists but Clubs doesn't then
DROP TABLE IF EXISTS `DiscordUser`;
results in :-
no such table: main.Clubs: DROP TABLE IF EXISTS `DiscordUser`;
If you then create Clubs and try the DROP with a very basic (no Constraints) using :-
CREATE TABLE IF NOT EXISTS `Clubs` (ID INTEGER PRIMARY KEY);
DROP TABLE IF EXISTS `DiscordUser`;
The result is good as per :-
Query executed successfully: DROP TABLE IF EXISTS `DiscordUser`; (took 1ms)
Now try to DROP Clubs using :-
--CREATE TABLE IF NOT EXISTS `Clubs` (ID INTEGER PRIMARY KEY);
--DROP TABLE IF EXISTS `DiscordUser`;
DROP TABLE IF EXISTS `Clubs`;
and you can't as DiscordUser doesn't exist as per :-
no such table: main.DiscordUser: DROP TABLE IF EXISTS `Clubs`;
I've tried closing the database in case it was a caching issue but the behaviour remains.
As such, I'd strongly suggest having a good look at the constraint usage and being sure of correcting the issues before trying to copy all of the tables (I guess that there is a chance that this could be part of the cause of the corruption, however why/how is way beyond me).
P.S. The method I used was (1-6):-
Then for 7 :-
Run the sqlite_master query, from above, select all cells and copy, then drop the results into a spreadsheet (you could drop the sql column as the create gets truncated unless you try to fiddle with the delimiters).
Open the exported file (I used NotePad++) in your Editor.
Open a new DB in DB Browser (will refer to it as DBB from now) for SQLite.
In DBB in EXEC SQL tab, input PRAGMA integrity_check, run to check.
Create new tab (for next SQL).
Switch to Spreadsheet and copy the first table name that isn't marked as done.
Switch to Editor do find on EXISTS copied_table_name
Select the section (i.e the CREATE statement along to and including the last row to be inserted, note can be a pain for the larger tables so might be easier to create separate export for those tables). Copy the selection to the clipboard.
Paste into the empty tab and run.
If OK then
in DBB click to create a new tab for the next
switch to spreadsheet and mark table as DONE
goto 5.
If not OK then
If you can fix the issue by altering the SQL fix the SQL (e.g. column name needs enclosing/wrapping/quoting) and then go to 9.
If the issue is due to Constraints then go to 5 but select the table causing the constraint.
OK the issue with the DiscordUser/Club tables is that a Clubs.Ownerid requires a DiscordUser. So clubs cannot be added without the relevant Discord users (id's 1,2,7,14 and 32). Some DiscordUsers are club members so they require a club to exist.
What I have done is to load the DiscordUsers rows for the Club owners changing their ClubId to null. Load the Clubs. Update the ClubId's of the DiscordUsers so they are members of the club that they were before (i.e. undo the null) and to load the rest of the nearly 600 Discordusers (excluding those already loaded).
Here's the SQL I used for that part (note except for the Discorduser, Clubs and the 3 waifu tables, all other tables have been successfully created and loaded).
INSERT INTO `DiscordUser` (Id,AvatarId,Discriminator,UserId,DateAdded,Username,ClubId,LastLevelUp,NotifyOnLevelUp,LastXpGain,TotalXp,IsClubAdmin,CurrencyAmount) VALUES
-- ClubId was 6 changed to null
(1,'6d5212a0f5e862d57c8ffc6f254a2e85','1458',299779864045682689,'2017-10-07 18:02:04.8287878','Anubis',NULL,'2018-03-27 02:22:26.362966',0,'2017-11-17 01:19:14.0313957',7056,1,280),
-- Owns a club but not in a club
(2,'3b37e0f635706f81fdde2b6de9889283','9810',181200115539640321,'2017-10-07 18:04:39.767728','AnnaHime',NULL,'2018-01-02 02:27:38.8011863',0,'2017-11-16 01:29:49.0371488',429,0,360),
-- ClubId null was 3
(7,'612c67b6eb57d8806dcc92ed45b3a6d0','0396',177502331582021639,'2017-10-07 18:11:09.7830603','Tsuchimursu',NULL,'2018-03-28 17:45:53.7399883',0,'2017-11-17 15:53:59.084885',18156,1,4725),
-- ClubId null was 4
(14,'b2dd362171277337294de325bf92ad6a','3267',215597863441268737,'2017-10-07 18:45:54.8092675','LaLa☆Star',NULL,'2018-01-14 20:52:15.7531274',0,'2017-11-08 19:00:22.7778305',2061,1,286),
-- ClubId null was 5
(32,'667f4d802b977c4d4be974e35ae63c55','2593',251689019929395200,'2017-10-08 00:58:16.6089546','username',NULL,'2018-03-28 07:27:34.9348084',0,'2017-11-17 20:02:14.0283998',4704,1,1188),
-- ClubId was 2 changed to NULL
(91,'0adb399c9f2cd94370038e2452ab8c8d','6790',346513954966863872,'2017-10-13 05:48:51.7788964','mayoi',NULL,'2018-03-24 02:50:06.8970518',0,'2017-11-17 20:01:29.0692552',7635,1,515)
;
INSERT INTO `Clubs` (Id,DateAdded,Discrim,ImageUrl,MinimumLevelReq,Name,OwnerId,Xp,Description) VALUES
(2,'2017-11-14 07:39:57.5091592',1,'https://lh3.googleusercontent.com/_7WKFouxTx1fdFpnmmuykDAd5SoiiJOPzHdRmXKOmRRZhV5Ba4V_kZct5ooVjQ9BuzU=w300',5,'We ⤠waifus',91,40137,'Love your waifus short & tall, big & small, cute as dolls, we love ''em all!'),
(3,'2017-12-11 07:00:59.3762914',1,'',30,'Den of Faes',7,11607,NULL),
(4,'2017-12-11 07:03:59.093402',1,'',5,'Skeleton Enthusiasts',14,657,NULL),
(5,'2017-12-11 07:05:56.9111719',1,'',5,'Saki''s Juice',32,2610,NULL),
(6,'2017-12-22 04:46:24.7271709',1,'',5,'nap pile',1,24870,'For the sleeping beauties and the wandering insomniacs who enjoy a good night sleep.')
;
UPDATE `DiscordUser` SET ClubId = 6 WHERE Id=1;
UPDATE `DiscordUser` SET ClubId = 3 WHERE Id=7;
UPDATE `DiscordUser` SET ClubId = 4 WHERE Id=14;
UPDATE `DiscordUser` SET ClubId = 5 WHERE Id=32;
UPDATE `DiscordUser` SET ClubId = 2 WHERE Id=91;
-- LOAD Remaining DiscordUser rows (note incomplete)
INSERT INTO `DiscordUser` (Id,AvatarId,Discriminator,UserId,DateAdded,Username,ClubId,LastLevelUp,NotifyOnLevelUp,LastXpGain,TotalXp,IsClubAdmin,CurrencyAmount) VALUES
--(1,'6d5212a0f5e862d57c8ffc6f254a2e85','1458',299779864045682689,'2017-10-07 18:02:04.8287878','Anubis',6,'2018-03-27 02:22:26.362966',0,'2017-11-17 01:19:14.0313957',7056,1,280),
--(2,'3b37e0f635706f81fdde2b6de9889283','9810',181200115539640321,'2017-10-07 18:04:39.767728','AnnaHime',NULL,'2018-01-02 02:27:38.8011863',0,'2017-11-16 01:29:49.0371488',429,0,360),
(3,'a3cd92d397ad357834d0e6c9f10bfc59','0429',145356302347010048,'2017-10-07 18:04:49.786657','Rebel Lucy',NULL,'2018-03-26 12:55:21.1149964',0,'2017-11-17 22:21:24.0263741',6876,0,3600),
(4,'7225dccaab1c93896657a61e18595378','5286',84689434536050688,'2017-10-07 18:05:44.765554','scarletflame234',NULL,'2018-03-28 22:56:28.7427437',0,'2017-11-17 23:21:41.4446535',13368,0,288),
(5,'c1316bc0673f4a2709b3ce550ed54395','0760',303279191116480514,'2017-10-07 18:06:39.7664015','zachary',NULL,'2018-03-02 03:48:43.4817755',0,'2017-11-17 18:44:14.1082867',210,0,50),
(6,'2ed95eae7c3088c46b23e71578dacc42','8801',161369834314137601,'2017-10-07 18:07:04.7672808','Kou',NULL,'2018-03-07 06:24:32.3405246',0,'2017-11-17 23:20:00.0648699',2640,0,55),
--(7,'612c67b6eb57d8806dcc92ed45b3a6d0','0396',177502331582021639,'2017-10-07 18:11:09.7830603','Tsuchimursu',3,'2018-03-28 17:45:53.7399883',0,'2017-11-17 15:53:59.084885',18156,1,4725),
(8,'5b1d239935ab4dd6d3eee98954601d52','9859',179093512610906113,'2017-10-07 18:13:54.7939334','TheCorty',NULL,'2017-11-12 12:07:36.4752178',0,'2017-11-12 23:47:26.4744132',2460,0,205), ...........
NOTE The SQL from -- LOAD Remaining DiscordUser rows, will not work as it's only intended to show how ID's 1,2 and 7 have been commented out, as should be rows 14, 32 and 91 as they have already been loaded, the other close to 600 rows should be included.
Note I've just also loaded the outstanding 3 waifu tables so all data can be retrieved (assuming that none has been lost due to the corruption). PRAGMA integrity_check; returns OK.

is there an easy to use formula for switching the datatype on a table column from int to bit?

I'm looking at a sql server table, and one of the columns seems to have been setup to be of type INT NULL when it's really better off as a BIT NOT NULL.
Is there an easy formula or shortcut for switching the type?
My current strategy is:
add new temp column
populate with data based on old column contents
drop old column and constraints
rename new column to old column name
add constraints based on new column
this seems complicated but, really, should be automatable - so i'm wondering if anyone's automated it yet.
Update MyTable set MyColumn = 0 where MyColumn is null
Assuming you want null to be false
Alter table MyTable alter column MyColumn bit not null
Will do the job, given all non-zero values are to be treated as true.
Course you were going to backup before you did any of this weren't you. :)

Oracle SQL. How to change table field datatype CLOB-->VARCHAR2

I have database and some text fields are CLOB type, I need to change most of these into VARCHAR2.
Have tried to do that using SQL Developer tool, by clicking edit on table, but get error like this one:
The following SQL statement failed:
ALTER TABLE TBL_PEOPLE MODIFY (PERSON VARCHAR2(150) )
Want to ask, how can this change be done
You can't, directly. The code you tried will have got an `ORA-22859, presumably. (It's helpful to show the actual errors you get, of course).
You'll need to add a new varchar2 column; copy the data across - or a substring of it if it might be larger than the new column you're creating; drop the clob column. You can rename the columns so it looks fairly transparent.
As in this SQL Fiddle:
alter table tbl_people rename column person to clob_person;
alter table tbl_people add (person varchar2(150));
update tbl_people set person = clob_person;
alter table tbl_people drop column clob_person;
Obviously don't drop the old column until you're sure the data has copied without errors. Also take into account any constraints, indexes, etc. that might exist in the old column; they will need to be recreated. And anything that references the old column will have been invalidated - generally procedures will recompile themselves on next use.
So be careful, test it first, and plan some down time.
ALTER TABLE tablename ADD (FIELD_LIST_TEMP VARCHAR2);
UPDATE tablename SET FIELD_LIST_TEMP = FIELD_LIST;
ALTER TABLE tablename DROP COLUMN FIELD_LIST;
ALTER TABLE tablename RENAME COLUMN FIELD_LIST_TEMP TO FIELD_LIST;
Here FIELD_LIST existing column which is defined it as CLOB. With above query it will change from CLOB to VARCHAR2.

PostgreSQL Convert column from integer to text

I have a PostgreSQL (9.0) database with a column card_id which is currently of type integer
I need to change this to type text
What is the most best way to achieve this?
The only solution I can find involves creating a temporary column, dropping the original then renaming, I thought they might be a better method??
Have you tried what the fine manual suggests:
ALTER TABLE table ALTER COLUMN anycol TYPE anytype;
Depending on the current and the new type you may need to add USING ... to this statement.
But in your specific case that should not be necessary I believe.
ALTER TABLE table ALTER COLUMN card_id SET DATA TYPE text;

Change each record in a table with no primary key?

I have a table in a database that represents dates textually (i.e. "2008-11-09") and I would like to replace them with the UNIX timestamp. However, I don't think that MySQL is capable of doing the conversion on its own, so I'd like to write a little script to do the conversion. The way I can think to do it involves getting all the records in the table, iterating through them, and updating the database records. However, with no primary key, I can't easily get the exact record I need to update.
Is there a way to get MySQL to assign temporary IDs to records during a SELECT so that I refer back to them when doing UPDATEs?
Does this not do it?
UPDATE
MyTable
SET
MyTimeStamp = UNIX_TIMESTAMP(MyDateTime);
If for some reason you do have to iterate (the other answers cover the situation where you don't), I can think of two ways to do it (these aren't MySQL-specific):
Add a column to the table that's an auto-assigned number. Use that as the PK for your updates, then drop the column afterwards (or just keep it around for future use).
In a table with no defined PK, as long as there are no exact duplicate rows, you can use the entire row as a composite PK; just use every column in the row as your distinguishing characteristic. i.e., if the table has 3 columns, "name", "address", and "updated", do the following:
UPDATE mytable SET updated = [timestamp value] WHERE name = [name] AND address = [address] AND timestamp = [old timestamp]
Many data access frameworks use this exact strategy to implement optimistic concurrency.
No, you should be able to do this with a single update statement. If all of the dates are yyyy-mm-dd and they are just stored in some sort of text column instead of DATETIME, you can just move the data over. SQL would be like:
ALTER TABLE t ADD COLUMN dates DATETIME;
UPDATE t set t.dates=t.olddate;
This shouldn't be dependent on a PK because MySQL can scan through each row in the table. The only time PK's become an issue is if you need to update a single row, but the row may not be unique.
You can generate values during a SELECT using the MySQL user variables feature, but these values do not refer to the row; they're temporary parts of the result set only. You can't use them in UPDATE statements.
SET #v := 0;
SELECT #v:=#v+1, * FROM mytable;
Here's how I'd solve the problem. You're going to have to create another column for your UNIX timestamps anyway, so you can add it first. Then convert the values in the old datetime column to the UNIX timestamp and place it in the new column. Then drop the old textual datetime column.
ALTER TABLE mytable ADD COLUMN unix_timestamp INT UNSIGNED NOT NULL DEFAULT 0;
UPDATE mytable
SET unix_timestamp = UNIX_TIMESTAMP( STR_TO_DATE( text_timestamp, '%Y-%m-%d' ) );
ALTER TABLE mytable DROP COLUMN text_timestamp;
Of course you should confirm that the conversion has been done correctly before you drop the old column!
See UNIX_TIMESTAMP() and STR_TO_DATE()