Generate alter syntax for column - php

I've found a way to find missing tables between two databases, and missing columns from intersect tables. I can get "create" syntax for missing tables, but how do I generate "alter" syntax for missing columns?
Is there something like SHOW CREATE TABLE $table for column?
#Edit
I get:
CREATE TABLE `table` (
`id` int(11) unsigned NOT NULL AUTO_INCREMENT,
`name` varchar(225) NOT NULL DEFAULT '',
PRIMARY KEY (`id`)
)
by SHOW CREATE TABLE $table
-> I figure out the second database's "table" misses "name" column. How do I "parse" (?) this create to generate alter to add column "name" to the second database's table?

If you want individual columns, then use ( change users to your table name )
show columns from users
If this is done on both tables, you can check the differences. This may include having to alter columns if the definition has changed.
To add a new column, use something like
alter table users add column email varchar(255) after name
The after bit allows you to position the new column at the same place as in your other table (makes it look consistent in query tools).

Related

PHP - Can't insert into mySQL (structure db problem?)

I have a problem inserting data into my MySQL database.
The structure of the db looks like this:
id | name | class | 23-02-2022 | 26-02-2022 | and so on ...
The databse is part of an attendance system. So I use dates as column names.
I use this code to open a csv file and upload some data into the db. As you can see in this part of the code I only put datas in the name and class column.
if (($handle = fopen("class.csv", "r")) !== FALSE)
{
while (($data = fgetcsv($handle, 1000, ";")) !== FALSE)
{
$query="INSERT INTO table21228 (name, class) VALUES ('$data[0]' , '$data[1]')";
if ($conn->query($query) === TRUE) {
}
else {
echo 'Error: '. $conn->error;
} fclose($handle);
}
I get this error message: Error: Field '23-02-2022' doesn't have a default value
When I use a different table, where the only columns are id, name, class it works without any problems.
So I guess the structure of my db must be the problem
Maybe all those dates columns like 23-02-2022???
Hope some might help me. Thank you!
Kind regards
Dan
The problem is that the columns of the dates dont have a DEFAULT value and since while adding a record you dont define a value for the column it is giving an error. The solution is that either you give a value for the columns while adding the records or else alter the columns and give it a default value.
But your Table structure is not at all feasible to use. You should not have columns for individual dates. Like this you will have infinite columns in your table. So instead the solution is that you insert the date of the attendance marked with the rows you add.
Could be you have a table with not null columns and you try to insert a row without a proper value for the not nullable columns .. the you have the message for field '23-02-2022' doesn't have a default value
the try insert a proper value for these columns
$query="INSERT INTO table21228 (name, class, `23-02-2022`, `26-02-2022` ) VALUES ('$data[0]' , '$data[1]', '2022-02-20', '2022-02-20')";
or try revoke the not null constranits for theese columns
alter table table21228 modify column `23-02-2022` date;
or set a default value
ALTER TABLE table21228 MODIFY column `23-02-2022` date DEFAULT '2022-02-20';
The problem is, that you try to insert a row into a table where not all columns do have a default value. You either need to give all columns a default value (using ALTER TABLE or a modified CREATE TABLE) or you have to mention all those columns in your INSERT query.
Also, your code is vulnerable to SQL injection. Read this great guide on how to prevent that:
https://phpdelusions.net/pdo
If your table looks like this:
CREATE TABLE `attendances` (
`id` INT NOT NULL AUTO_INCREMENT,
`name` VARCHAR(250) NOT NULL,
`class` VARCHAR(250) NOT NULL,
`23-02-2022` INT NOT NULL,
`26-02-2022` INT NOT NULL,
PRIMARY KEY (`id`)) ENGINE = InnoDB;
You can change it like this:
ALTER TABLE `attendances`
CHANGE `23-02-2022` `23-02-2022` INT NULL DEFAULT NULL;
or
ALTER TABLE `attendances`
CHANGE `26-02-2022` `26-02-2022` INT NOT NULL DEFAULT '0';
Here, 23-02-2022 has a default value of "NULL" and 26-02-2022 is an example with a default value of "0". Or just create the table correctly in the first place:
CREATE TABLE `attendances` (
`id` INT NOT NULL AUTO_INCREMENT,
`name` VARCHAR NOT NULL,
`class` VARCHAR NOT NULL,
`23-02-2022` INT NULL DEFAULT NULL,
`26-02-2022` INT NOT NULL DEFAULT '0',
PRIMARY KEY (`id`)) ENGINE = InnoDB;
As an alternative, you could just add all columns that have no default value to your INSERT query:
INSERT INTO `attendances` (
`id`, `name`, `class`, `23-02-2022`, `26-02-2022`
) VALUES (
NULL, 'name1', 'class1', '0', '0'
);
Make sure to protect your app from SQL injection:
<?php
$pdo->prepare("
INSERT INTO `attendances` (
`id`, `name`, `class`, `23-02-2022`, `26-02-2022`
) VALUES (
NULL, ?,?,?,?
)
")->execute(['name1', 'class1', 0, 0]);
So I use dates as column names.
...bad idea, because you theoretically have an infinite number of columns, if the system is used long term. And it will make it very difficult to write certain types of query to understand the data.
So I guess the structure of my db must be the problem
...essentially, yes.
To understand how to design your database correctly, you should learn about database normalisation.
In this scenario I'd suggest you'd have one table for the list of all people, and another for the list of all classes.
If you're running a predetermined timetable, you might then have a table which lists the class, the date and the teacher assigned to that date & class. (Or you might assign the teacher in the classes table, if one teacher normally takes the whole class.)
Then lastly you'd have a separate "attendance" table which contains columns "personID" and "attendanceDate", and "classID".
That way you will end up with multiple rows in there with the same person / class combination and different dates, to record all their attendances at each class and each date of that class. And it's completely extendable, infinitely, without you needing to modify the tables each time a new class or date is announced, or needing to dervice column names in your code when trying to generate a query.
first check your csv file has the right amount of columns as your database then set your columns default to from not NULL to null or none

Disabling unique constraint on a column in a table MYSQL [duplicate]

ColumnName is Unique (UNIQUE KEY ColumnName).
I just want to make column not unique (must be very simple, but can not understand how).
If in phpMyAdmin check at column name and at bottom click on Unique icon, get #1062 - Duplicate entry '' for key 'RegistrationNumber'. OK, see it is because, clicking on icon it ADD UNIQUE.
There is Unique icon in Structure within row. But the icon is not click-able.
As in phpMyAdmin did not found how to do it, trying with query.
Based on advices tried ALTER TABLE TableName DROP INDEX ColumnName.
Get 1091 Can't DROP 'ColumnName'; check that column/key exists
Here https://stackoverflow.com/a/4414694/2465936 found This error means that you are trying to delete a key which is being used by another table. Possibly the ColumnName is used by another table.
Please advice what need to do to make column not unique.
With SHOW CREATE TABLE get
Array
(
[0] => Array
(
[Table] => 18_6_TransactionPartners
[Create Table] => CREATE TABLE `18_6_TransactionPartners` (
`Number` int(11) NOT NULL AUTO_INCREMENT,
`CompanyName` char(255) COLLATE utf8_unicode_ci NOT NULL,
`RegistrationNumber` char(255) COLLATE utf8_unicode_ci NOT NULL,
.......
PRIMARY KEY (`Number`),
UNIQUE KEY `Number_2` (`Number`),
UNIQUE KEY `CompanyName` (`CompanyName`,`RegistrationNumber`),
KEY `Number` (`Number`)
) ENGINE=InnoDB AUTO_INCREMENT=444 DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci
)
)
Update
Based on #Bart Friederichs advice tried ALTER TABLE 18_6_TransactionPartners DROP INDEX Number and changed column RegistrationNumber not not unique. Do not understand why (possibly had some mess with unique keys). In any case can change to not unique.
Probably you have a named INDEX. By using SHOW CREATE TABLE tbl you can find out the names of the indices. Then drop them by name (e.g. some test table):
mysql> SHOW CREATE TABLE test;
CREATE TABLE `test` (
`entry_id` int(11) DEFAULT NULL,
`user_id` int(11) DEFAULT NULL,
UNIQUE KEY `k` (`entry_id`)
)
To drop the index, use this:
ALTER TABLE test DROP INDEX k;
Your key name is RegistrationNumber (as is told by the error message):
ALTER TABLE TableName DROP INDEX RegistrationNumber;
If your column was defined unique using UNIQUE clause, then you can do something like this:
ALTER TABLE mytable DROP INDEX constraint_name
For dropping the index do this:-
ALTER TABLE mytable DROP INDEX index_name;
You have to drop the index using the index name, not the column name.

Make sure every entry in a table is unique [duplicate]

I have a table:
table votes (
id,
user,
email,
address,
primary key(id),
);
Now I want to make the columns user, email, address unique (together).
How do I do this in MySql?
Of course the example is just... an example. So please don't worry about the semantics.
To add a unique constraint, you need to use two components:
ALTER TABLE - to change the table schema and,
ADD UNIQUE - to add the unique constraint.
You then can define your new unique key with the format 'name'('column1', 'column2'...)
So for your particular issue, you could use this command:
ALTER TABLE `votes` ADD UNIQUE `unique_index`(`user`, `email`, `address`);
I have a MySQL table:
CREATE TABLE `content_html` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`id_box_elements` int(11) DEFAULT NULL,
`id_router` int(11) DEFAULT NULL,
`content` mediumtext COLLATE utf8_czech_ci NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY `my_uniq_id` (`id_box_elements`,`id_router`)
);
and the UNIQUE KEY works just as expected, it allows multiple NULL rows of id_box_elements and id_router.
I am running MySQL 5.1.42, so probably there was some update on the issue discussed above. Fortunately it works and hopefully it will stay that way.
Multi column unique indexes do not work in MySQL if you have a NULL value in row as MySQL treats NULL as a unique value and at least currently has no logic to work around it in multi-column indexes. Yes the behavior is insane, because it limits a lot of legitimate applications of multi-column indexes, but it is what it is... As of yet, it is a bug that has been stamped with "will not fix" on the MySQL bug-track...
Have you tried this ?
UNIQUE KEY `thekey` (`user`,`email`,`address`)
This works for mysql version 5.5.32
ALTER TABLE `tablename` ADD UNIQUE (`column1` ,`column2`);
MySql 5 or higher behaves like this (I've just tested):
you can define unique constraints involving nullable columns. Say you define a constraint unique (A, B) where A is not nullable but B is
when evaluating such a constraint you can have (A, null) as many times you want (same A value!)
you can only have one (A, not null B) pair
Example:
PRODUCT_NAME, PRODUCT_VERSION
'glass', null
'glass', null
'wine', 1
Now if you try to insert ('wine' 1) again it will report a constraint violation
Hope this helps
You can add multiple-column unique indexes via phpMyAdmin. (I tested in version 4.0.4)
Navigate to the structure page for your target table. Add a unique index to one of the columns. Expand the Indexes list on the bottom of the structure page to see the unique index you just added. Click the edit icon, and in the following dialog you can add additional columns to that unique index.
this tutorial works for me
ALTER TABLE table_name
ADD CONSTRAINT constraint_name UNIQUE (column1, column2, ... column_n);
https://www.mysqltutorial.org/mysql-unique-constraint/
I do it like this:
CREATE UNIQUE INDEX index_name ON TableName (Column1, Column2, Column3);
My convention for a unique index_name is TableName_Column1_Column2_Column3_uindex.
If You are creating table in mysql then use following :
create table package_template_mapping (
mapping_id int(10) not null auto_increment ,
template_id int(10) NOT NULL ,
package_id int(10) NOT NULL ,
remark varchar(100),
primary key (mapping_id) ,
UNIQUE KEY template_fun_id (template_id , package_id)
);
For adding unique index following are required:
1) table_name
2) index_name
3) columns on which you want to add index
ALTER TABLE `tablename`
ADD UNIQUE index-name
(`column1` ,`column2`,`column3`,...,`columnN`);
In your case we can create unique index as follows:
ALTER TABLE `votes`ADD
UNIQUE <votesuniqueindex>;(`user` ,`email`,`address`);
If you want to avoid duplicates in future. Create another column say id2.
UPDATE tablename SET id2 = id;
Now add the unique on two columns:
alter table tablename add unique index(columnname, id2);
First get rid of existing duplicates
delete a from votes as a, votes as b where a.id < b.id
and a.user <=> b.user and a.email <=> b.email
and a.address <=> b.address;
Then add the unique constraint
ALTER TABLE votes ADD UNIQUE unique_index(user, email, address);
Verify the constraint with
SHOW CREATE TABLE votes;
Note that user, email, address will be considered unique if any of them has null value in it.
For PostgreSQL...
It didn't work for me with index; it gave me an error, so I did this:
alter table table_name
add unique(column_name_1,column_name_2);
PostgreSQL gave unique index its own name. I guess you can change the name of index in the options for the table, if it is needed to be changed...

Insert data from PHP to MySQL ONLY if multiples values don't exist in multiple columns [duplicate]

I have a table:
table votes (
id,
user,
email,
address,
primary key(id),
);
Now I want to make the columns user, email, address unique (together).
How do I do this in MySql?
Of course the example is just... an example. So please don't worry about the semantics.
To add a unique constraint, you need to use two components:
ALTER TABLE - to change the table schema and,
ADD UNIQUE - to add the unique constraint.
You then can define your new unique key with the format 'name'('column1', 'column2'...)
So for your particular issue, you could use this command:
ALTER TABLE `votes` ADD UNIQUE `unique_index`(`user`, `email`, `address`);
I have a MySQL table:
CREATE TABLE `content_html` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`id_box_elements` int(11) DEFAULT NULL,
`id_router` int(11) DEFAULT NULL,
`content` mediumtext COLLATE utf8_czech_ci NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY `my_uniq_id` (`id_box_elements`,`id_router`)
);
and the UNIQUE KEY works just as expected, it allows multiple NULL rows of id_box_elements and id_router.
I am running MySQL 5.1.42, so probably there was some update on the issue discussed above. Fortunately it works and hopefully it will stay that way.
Multi column unique indexes do not work in MySQL if you have a NULL value in row as MySQL treats NULL as a unique value and at least currently has no logic to work around it in multi-column indexes. Yes the behavior is insane, because it limits a lot of legitimate applications of multi-column indexes, but it is what it is... As of yet, it is a bug that has been stamped with "will not fix" on the MySQL bug-track...
Have you tried this ?
UNIQUE KEY `thekey` (`user`,`email`,`address`)
This works for mysql version 5.5.32
ALTER TABLE `tablename` ADD UNIQUE (`column1` ,`column2`);
MySql 5 or higher behaves like this (I've just tested):
you can define unique constraints involving nullable columns. Say you define a constraint unique (A, B) where A is not nullable but B is
when evaluating such a constraint you can have (A, null) as many times you want (same A value!)
you can only have one (A, not null B) pair
Example:
PRODUCT_NAME, PRODUCT_VERSION
'glass', null
'glass', null
'wine', 1
Now if you try to insert ('wine' 1) again it will report a constraint violation
Hope this helps
You can add multiple-column unique indexes via phpMyAdmin. (I tested in version 4.0.4)
Navigate to the structure page for your target table. Add a unique index to one of the columns. Expand the Indexes list on the bottom of the structure page to see the unique index you just added. Click the edit icon, and in the following dialog you can add additional columns to that unique index.
this tutorial works for me
ALTER TABLE table_name
ADD CONSTRAINT constraint_name UNIQUE (column1, column2, ... column_n);
https://www.mysqltutorial.org/mysql-unique-constraint/
I do it like this:
CREATE UNIQUE INDEX index_name ON TableName (Column1, Column2, Column3);
My convention for a unique index_name is TableName_Column1_Column2_Column3_uindex.
If You are creating table in mysql then use following :
create table package_template_mapping (
mapping_id int(10) not null auto_increment ,
template_id int(10) NOT NULL ,
package_id int(10) NOT NULL ,
remark varchar(100),
primary key (mapping_id) ,
UNIQUE KEY template_fun_id (template_id , package_id)
);
For adding unique index following are required:
1) table_name
2) index_name
3) columns on which you want to add index
ALTER TABLE `tablename`
ADD UNIQUE index-name
(`column1` ,`column2`,`column3`,...,`columnN`);
In your case we can create unique index as follows:
ALTER TABLE `votes`ADD
UNIQUE <votesuniqueindex>;(`user` ,`email`,`address`);
If you want to avoid duplicates in future. Create another column say id2.
UPDATE tablename SET id2 = id;
Now add the unique on two columns:
alter table tablename add unique index(columnname, id2);
First get rid of existing duplicates
delete a from votes as a, votes as b where a.id < b.id
and a.user <=> b.user and a.email <=> b.email
and a.address <=> b.address;
Then add the unique constraint
ALTER TABLE votes ADD UNIQUE unique_index(user, email, address);
Verify the constraint with
SHOW CREATE TABLE votes;
Note that user, email, address will be considered unique if any of them has null value in it.
For PostgreSQL...
It didn't work for me with index; it gave me an error, so I did this:
alter table table_name
add unique(column_name_1,column_name_2);
PostgreSQL gave unique index its own name. I guess you can change the name of index in the options for the table, if it is needed to be changed...

php SQL DATABASE

I have a create database and have about 8 tables in Database also created Primary keys and foreign-keys in appropriate tables. But when I insert data in primary-table, my other table doesn't show updated data.
I mean, say I have a table which has data for names like ;
N is (name)
N1 = George, N2 = Ross, N3 = Rim ...etc now that means i have Primary key N1,N2,N3 etc..
Now, when I insert this primary keys in others table it should shows me name like George, ross and rim instead of primary-key number it self(N1,N2,N3).
How can I get names instead PK itself?
You are misunderstanding the concept of keys in relational databases. Keys are there, not to copy data from similar tables but to show the relations between data in different tables. They help to understand how the data between different tables is related - that is where the name "relational database" comes from. They also speed up querying of that data if indexed.
You can read more about the usage of keys here: Keys and normalization
I am still unclear on what exactly you want to do with the database. but let me demonstrate you on the basic of database and how you should be using it. Consider a table users where you will be storing the data related to user.
CREATE TABLE `users` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`name` varchar(50) NOT NULL,
`email` varchar(50) DEFAULT NULL,
`phone` varchar(30) DEFAULT NULL,
PRIMARY KEY (`id`),
);
the column id holds the primary key and have an attribute called auto_increment, now what this means is every time you insert a record to this table the id attribute gets incremented and you don't have to worry about inserting any value in this column because your database will take care of that. for example take a look at insert query below.
INSERT INTO users(name,email,phone) VALUES('First Name', 'first#domain.com', '9999999999');
INSERT INTO users(name,email,phone) VALUES('Second Name', 'second#domain.com', '8888888888');
INSERT INTO users(name,email,phone) VALUES('Third Name', 'third#domain.com', '2222222222');
INSERT INTO users(name,email,phone) VALUES('Fourth Name', 'fourth#domain.com', '3333333333');
did you see you did not insert any id here. this is because it is database who will handle the logic. now the first record will hold the value 1 the second will have 2 the third one 3 and the fourth one 4 and so on.
hope this helps you.

Categories