I have an upgrade wizard (TYPO3 11) which changes the data of a table.
This is done with the querybuilder:
$queryBuilder = GeneralUtility::makeInstance(ConnectionPool::class)
->getQueryBuilderForTable('tt_content');
$queryBuilder
->update('tt_content')
->set('CType', 'newCType')
->where($queryBuilder
->expr()
->eq('CType',$queryBuilder->createNamedParameter('oldCType')))
->execute();
But I also need to rename a field in a table:
ALTER TABLE tt_content RENAME COLUMN tx_myext_old_field TO tx_myext_new_field;
I can't find any documentation or example of doing this with the querybuilder.
The normal way woult be to provide a ext_tables.sql in your extension. This is read by TYPO3 to build a virtual "database scheme" how it should look.
The database schema analyser will than provide the information, and database alteration are suggested.
You could add a database must be up to date constraint to your upgrade wizard, that way it is ensured that the field is changed.
DTL is a special task, and you have to provide the correspinng queries yourself ... which are different for different dbms systems. So using the normal way would be recommended.
The platform/driver may have some generig helper methods providing some native sql parts for doing stuffs like that. The may be possible to provide custom stuff based on SchemaMigrator or SchemaManger etc - but thats low-level stuff.
doctrine/dbal directly do not really provide these DTL as API. And the querybuilder is not meant to be used for that low level stuff at all. That's the wrong tool for such tasks.
You can also change columns of core tables that way, by providing simply the table name and the column defintion only for the field you want to change.
The official way is to handle this with ext_tables.sql and the database schema analyser.
See: https://docs.typo3.org/m/typo3/reference-coreapi/main/en-us/ExtensionArchitecture/FileStructure/ExtTablesSql.html
The concept of renaming a column could not work:
On installing the extension all new fields are generated (or should be generated if in composer mode). And as the extension should work with the new columns they are already defined.
And before the upgrade wizard could rename a column these columns are existent already which prevents a rename.
In the end I do a content copy enhancing the update query like this:
$queryBuilder = GeneralUtility::makeInstance(ConnectionPool::class)
->getQueryBuilderForTable('tt_content');
$queryBuilder
->update('tt_content')
->set('CType', 'newCType')
->set('tx_myext_newfield1',$queryBuilder->quoteIdentifier('tx_myext_oldfield1'),false)
->set('tx_myext_newfield2',$queryBuilder->quoteIdentifier('tx_myext_oldfield2'),false)
->where($queryBuilder
->expr()
->eq('CType',$queryBuilder->createNamedParameter('oldCType')))
->executeStatement();
Related
The aim is when updating the application and update the access database without altering the data so update by update only the new tables or new columns so i want to copy the exact table with it's structure to the old database vb.net and access database.
what I've tried is detecting the differences between the old database and the new one by getting in combobox1 the only missed table and in combobox2 the missed columns in the old database in exact table already there in both database and get it's data type .
so i want to copy the entire table and then create only missed columns
thank you
There is not a built in tool to do this.
But, worse yet, there is no "generate" change scripts in Access
(Like say with SQL server).
So, how do you approach this issue? What do some of the accounting systems or commercial programs that use ms-access as the database?
Well, you have to build a kind of "up-grade" system in your software.
This means two things:
To add a new column to a table (for example), you NEVER go open up the access database with access, but "add" or "write" the code to add that field in question.
In fact, I had an applcation deployed out in the field - many desktops.
So, I had a code module called upgrade. And each time I needed a new field or whatever, then I would write the code to add that new colum.
AS LONG as I always added things into that code module, I was ok. (never break the rule for adding new fields, tables or even increasing the length of some field? - use code).
And it became quite easy after I had some code written. I would in fact often cut + paste a previous bit of code to add a new column to a table.
However, after about 5 years, that messy code module had 800+ lines of code in it!!!
But, I ALSO realized that MOST things like adding a new column or whatever? Same code over and over.
So, what I did next was built a "upgrade" table. It looked like this:
Version action SQL RunCode
2.5 AddTable tblCustomers
2.5 AddField "sql here to add table"
etc. etc.
So, I had a version number, and then I compare against the up-grade table. I had "action", and the code would simple loop this table, and do whatever.
So, for example, to add a field, you can use access "DDL" command (data definition commands - most SQL systems support this, and so does Access).
so, say like this:
' any new table code goes here:
If lngVer < 1148 Then
' add event Invoice text option
ExecuteSQLNR "ALTER TABLE dbo.Events ADD InvoiceText ntext NULL"
ExecuteSQLNR "ALTER TABLE dbo.Events ADD HideEventDate bit NULL default 0"
Or, say to increase a column lengh from 50 to 55
db.Execute "ALTER TABLE tblGroupRemind ALTER COLUMN Anotes text(255)", dbFailOnError
As noted, since oh so many the commands were VERY similar, then I started putting that information into a table, and then I would execute the required upgrades in a loop.
For a whole new table? Well, I thought that was too much code, so I always included a blank empty database - and for new tables, I would place them in that upgrade.accDB table - and "transfer/copy" the table from that upgrade database to the real one. That way, I could with great ease create a whole new table, and create in Access designers, and then add/copy that table to the "upgrade.accDB" database.
As noted? The above ideas an approaches work quite well.
In fact, over time, I found it LESS hassle while coding away to add the new column or whatever LESS effort then having to open up ms-acces, and then the table, and then the designer and make the changes.
However, the BIG issue with above?
Well, you have to get all users at least upgraded to your EXISTING schema, and there is no automated tools.
in fact, before I had any automated tools? I would open up note pad, and if I added some field to some table? I would simple type into note pad that new field in such and such table is required).
Then, when on customer site, I would open up their database, and then go look at the note pad document for the list of changes I was to make. (that is what I was doing before I started automating the process - and of course it not always practical to be "on site" or have the customers database.
But, ONCE I had all of the above working?
Then during development, I would open up my "upgrade" database, add the new row and action (new table, new column, (and more).
I even had a column that defined the function to run AFTER that one command. I mean, quite often when you add a new column, or change somthing in a table, often you need to copy data, or at least process some data after you make that change.
Once you get above going?
Then you simple NEVER make changes in the data tables directly, but use your "system" for this. And that works REALLY well.
For one, a customer could open up a older data file - say one from 4 or 5 years ago. The applcation version number would be detected, and then the upgrade code would run all though the versions to update that database. (and I did this automatic on startup - so they never even knew such a upgrade had occurred).
So, you just have to make sure that for each change you make, you put that code in your upgrade system, and you are done.
But, for existing systems? You have to look at what changes you made since last deploy, and write out the "ddl" commands (the alter table SQL commands).
There is no automated way of doing this.
As FYI?
One of the BEST and more valuable free tools in Visual Tools is the SQL server compare utility. It will not only automatic detect and tell you the changes between two SQL server databases, but will also upgrade for you. (very nice).
But, such a system is not available for Access. In fact, so valuable is that utility for SQL server, you might consider upgrading from Access to SQL server for this applcation. With that utility? I can work local, add fields, columns, tables and even stored procedures to that SQL database. When I am on site (or even by VPN), then I run that compare tool - it shows the changes, and ALSO has a button to update the target schema.
I don't know of a automated "schema" checker and updater for Access.
So, what I suggest for above ONLY works if you put such a system in place, and THEN as a developer always make your schema changes to your upgrade system, and never directly in the database with ms-access.
I want to modify a column name to new name present in a table
but here problem i want to manually modify the column name present in Triggers or SP's.
Is there a any better way of doing it.
To rename a column am using this
sp_RENAME 'Tablename.old_Column', 'new_column' , 'COLUMN';
similarly how can i do it for triggers or SP's.? without opening each script?
Well, there are a bunch of 3rd party tools that are promising this type of "safe rename", some for free and some are not:
ApexSQL has a free tool for that, as MWillemse wrote in his answer,
RedGate have a commercial tool called SQLPrompt that also have a safe renaming feture, However it is far from being free.
Microsoft have a visual studio add-in called SQL Server Data Tools (or SSDT in the short version), as Dan Guzman wrote in his comment.
I have to say I've never tried any of these specific tools for that specific task, but I do have some experience with SSDT and some of RedGate's products and I consider them to be very good tools. I know nothing about ApexSQL.
Another option is to try and write the sql script yourself, However there are a couple of things to take into consideration before you start:
Can your table be accessed directly from outside the sql server? I mean, is it possible that some software is executing sql statement directly on that table? If so, you might break it when you rename that column, and no sql tool will help in this situation.
Are your sql scripting skills really that good? I consider myself to be fairly experienced with sql server, but I think writing a script like that is beyond my skills. Not that it's impossible for me, but it will probably take too much time and effort for something I can get for free.
Should you decide to write it yourself, there are a few articles that might help you in that task:
First, Microsoft official documentation of sys.sql_expression_dependencies.
Second, an article called Different Ways to Find SQL Server Object Dependencies that is written by a 13 years experience DBA,
and last but not least, a related question on StackExchange's Database Administrator's website.
You could, of course, go with the safe way Gordon Linoff suggested in his comment, or use synonyms like destination-data suggested in his answer, but then you will have to manually modify all of the columns dependencies manually, and from what I understand, that is what you want to avoid.
Renaming the Table column
Deleting the Table column
Alter Table Keys
Best way use Database Projects in Visual Studio.
Refer this links
link 1
link 2
you can do what #GorDon suggested.
Apart from this,you can also play with this query,
select o.name, sc.* from sys.syscomments sc inner join sys.objects o
on sc.id=o.object_id where sc.text like '%oldcolumnname%'
this will return list of all proc and trigger.Also you can modify filter to get exact list.then it will be very easy for you to modify,manually.
But whatever you decide,don't simply drop old column.
To be safe,even keep back up.
This suggestion relates to Oracle DB, however there may be equivalent solutions in other DBMS's.
A temporary solution to your issue is to create a pseudocolumn. This solution looks a little hacky because the syntax for a pseudocolumn requires an expression. The simplest expression I can think of is the case statement below. Let me know if you can make it more simple.
ALTER TABLE <<tablename>> ADD (
<<new_column_name>> AS (
CASE
WHEN 1=1 THEN <<tablename>>.<<old_column_name>>
END)
);
This strategy basically creates a new column on the fly by evaluating the case statement and copying the value of <<old_column_value>> to <<new_column_value>>. Because you are dynamically interpolating this column there is a performance penalty vs just selecting the original column.
The one gotcha is that this will only work if you are duplicating a column once. Multiple pseudocolumns cannot contain duplicate expressions in Oracle.
The other strategy you can consider is to create a view and you can name the columns whatever you want. You can even INSERT/UPDATE/DELETE (execute DML) against views, but this would give you a whole new table_name, not just a new column. You could however rename the old table, and name your view the same as your old table. This also has a performance penalty vs just accessing the underlying table.
You might want to replace that text in definition. However, you will be needing a dedicated administrator connection in sql server. Versions also vary in setting up a dedicated administrator connection. Setting up the startup parameter by adding ;-T7806 under advanced. And by adding Admin: before the servername upon logging in. By then, you may be able to modify the value of the definition.
I am developing one automation tool for data migration from one table to other table, here I am looking for one function or SP for which I will pass source column and destination column as input parameter. I want output parameter to return true when source column data is compatible to copy to destination column. If not then it should return false.
For example, if a source column is varchar and a destination column is integer, the script should check all the data in a source column in good enough to move to an integer column or not and return the output flag. I want a script to work like this for all types of data types. Any suggestions will be helpful.
If you're on SQL Server 2012 you have TRY_CAST(), TRY_CONVERT() and TRY_PARSE() at your disposal (see this post by Biz Nigatu of blog.dbandbi for comparison).
That said, you still need to check for truncation errors, e.g. by converting to target datatype and back, then comparing the original value with the one after conversions.
I've seen similar tools in the past, might be a good idea to see if one isn't already available online for free. Even a purchase might be less expensive than the time you put into developing and troubleshooting your own tool.
I have an SSIS solution for this that I put together using EzAPI. I've got it posted to GitHub, so feel free to look:
https://github.com/thevinnie/SyncDatabases
Now, the part of that which is relevant to you would be where I use the information schema to make sure that the source and destination are schema matches. If not, C# script task will generate the statement to create, alter or drop the necessary column(s).
The EzAPI part is cool with SSIS because it will allow you to programatically generate an SSIS package. For the project requirements, I needed to be able to load data every time and not let schema changes in the source break the process.
Comments and recommendations are welcome. Hopefully it'll help, but I think you'll be looking at the INFORMATION_SCHEMA.COLUMNS either way.
I'm trying to get work my play20 application with postgresql so I can use and later deploy to Heroku. I followed this answer.
Basically, I made connection to database (so connection from local application to Heroku postgresql database worked), but I was not able to initialise database with generated 1.sql evolution. But generated sql was not working because of postgresql is using schema (it should work without schema anyway, but apparently I'm doing something wrong or database is doing something wrong).
create table user (
id bigint not null,
email varchar(255),
gender varchar(1),
constraint pk_user primary key (id));
resulted in
ERROR: syntax error at or near "user"
Position: 14 [ERROR:0, SQLSTATE:42601]
I fixed that with adding schema to table name
create table public.user(
...
);
Ok, everything worked until I tried to read or write to database. I got again sql syntax exception and can't work with database. Seems like sql queries are somehow wrong.
Any suggestions where could be problem?
That's very common mistake while developing application with other database than in production, but fortunately there is also common solution. You can still use User model, however you have to make sure that creates database table with changed name:
#Entity
#Table(name = "users")
public class User extends Model {
...
}
In most cases in your controllers and models name-switch will be transparent for you. Only place where you have to remember the switch are RawSql queries.
BTW, that's good idea to install locally the same database for developing cause there's a lot of differences between most popular databases, like other reserved keywords, other allowed types, even other auto incrementing methods for id, so finding and fixing proper values is just easier on localhost.
Well, due to my little knowledge about postgresql, I was struggling with this all day. Here's simple solution. Don't use table called "user" on postgreqsl. This table is already used.
But why my evolution sql query worked for initialisation of database? Well if I explicitly specify in which schema I want to create table "user", that basically works.
But if schema is not specified, is used current schema. From documentation:
If a schema name is given (for example, CREATE TABLE myschema.mytable ...) then the table is created in the specified schema. Otherwise it is created in the current schema
So that explains it. But for my project, using "user" model was perfectly reasonable and for H2 file based databased it was working, so I assumed that problem was somewhere else...
Actually the problem arises since we have named a column authorization which is ok on MySQL but fails miserably on Postgres, authorization is a reserved keyword there.
We want to have all schemas in sync even on different databases, so we want to rename authorization in the MySQL catalogue(s) to something neutral.
It would be great if the renaming of the MySQL table columns would work seamlessly with older and newer versions of the applicatation at least for a couple of days so that the transition is smooth.
Does anybody know how to do this? A nice idea would be to have some sort of alias/redirection, e.g. create a new column _authorization that is actually the same column as authorization but under the new name. Queries using the new name _autorization will work as well as queries using the old name. Then we can update the application. If all servers have the latest binaries, we can drop the alias and rename the column.
Any ideas? Any help is greatly appreciated.
A nice idea would be to have some sort of alias/redirection...
No such functionality exists. The closest is to use a view based on the table, because it's easier to rename a column in a view as there isn't any underlying data to change. Otherwise:
Rename the column in MySQL using the ALTER TABLE statement:
ALTER TABLE [your table]
CHANGE authorization [new_col_name] [column_definition]
I also had to use an alias to a column (e.g. a particular app needed 'id' column instead of 'entry_id').
i was able to create a view (MySQL 5.1+ recommended for views) from the table with an extra alias to the culprit column. for the 'authorization' column from your example:
CREATE VIEW maintable_view AS
SELECT *, authorization AS authcol FROM MAINTABLE
if you run a update-query on the view, and set 'authcol' to a new value,
this will update the 'authorization' column in 'maintable'.
UPDATE maintable_view SET authcol=22 WHERE id=3
http://dev.mysql.com/doc/refman/5.1/en/create-view.html