Generating REST APIs for a MySql database which does not follow conventions of frameworks such as rails, django, laravel etc is a small adventure that one like to avoid ..
Hence this.
xmysql requires node >= 7.6.0
npm install -g xmysql
xmysql -h localhost -u mysqlUsername -p mysqlPassword -d databaseName
http://localhost:3000
That is it! Simple and minimalistic!
Happy hackery!
Example : Generate REST APIs for Magento
Powered by popular node packages : (express, mysql) => { xmysql }
- Generates API for ANY MySql database 🔥🔥
- Serves APIs irrespective of naming conventions of primary keys, foreign keys, tables etc 🔥🔥
- Support for composite primary keys 🔥🔥
- REST API Usual suspects : CRUD, List, FindOne, Count, Exists, Distinct
- Bulk insert, Bulk delete, Bulk read 🔥
- Relations
- Pagination
- Sorting
- Column filtering - Fields 🔥
- Row filtering - Where 🔥
- Aggregate functions
- Group By, Having (as query params) 🔥🔥
- Group By, Having (as a separate API) 🔥🔥
- Multiple group by in one API 🔥🔥🔥🔥
- Chart API for numeric column 🔥🔥🔥🔥🔥🔥
- Auto Chart API - (a gift for lazy while prototyping) 🔥🔥🔥🔥🔥🔥
- XJOIN - (Supports any number of JOINS) 🔥🔥🔥🔥🔥🔥🔥🔥🔥
- Supports views
- Prototyping (features available when using local MySql server only)
- Run dynamic queries 🔥🔥🔥
- Upload single file
- Upload multiple files
- Download file
- Health and version apis
- Use more than one CPU Cores
- Docker support and Nginx reverse proxy config 🔥🔥🔥 - Thanks to @markuman
- AWS Lambda Example - Thanks to @bertyhell 🔥🔥🔥
Use HTTP clients like Postman or similar tools to invoke REST API calls
Download node, mysql (setup mysql), sample database - if you haven't on your system.
| HTTP Type | API URL | Comments |
|---|---|---|
| GET | / | Gets all REST APIs |
| GET | /api/tableName | Lists rows of table |
| POST | /api/tableName | Create a new row |
| PUT | /api/tableName | Replaces existing row with new row |
| POST 🔥 | /api/tableName/bulk | Create multiple rows - send object array in request body |
| GET 🔥 | /api/tableName/bulk | Lists multiple rows - /api/tableName/bulk?_ids=1,2,3 |
| DELETE 🔥 | /api/tableName/bulk | Deletes multiple rows - /api/tableName/bulk?_ids=1,2,3 |
| GET | /api/tableName/:id | Retrieves a row by primary key |
| PATCH | /api/tableName/:id | Updates row element by primary key |
| DELETE | /api/tableName/:id | Delete a row by primary key |
| GET | /api/tableName/findOne | Works as list but gets single record matching criteria |
| GET | /api/tableName/count | Count number of rows in a table |
| GET | /api/tableName/distinct | Distinct row(s) in table - /api/tableName/distinct?_fields=col1 |
| GET | /api/tableName/:id/exists | True or false whether a row exists or not |
| GET | /api/parentTable/:id/childTable | Get list of child table rows with parent table foreign key |
| GET 🔥 | /api/tableName/aggregate | Aggregate results of numeric column(s) |
| GET 🔥 | /api/tableName/groupby | Group by results of column(s) |
| GET 🔥 | /api/tableName/ugroupby | Multiple group by results using one call |
| GET 🔥 | /api/tableName/chart | Numeric column distribution based on (min,max,step) or(step array) or (automagic) |
| GET 🔥 | /api/tableName/autochart | Same as Chart but identifies which are numeric column automatically - gift for lazy while prototyping |
| GET 🔥 | /api/xjoin | handles join |
| GET 🔥 | /dynamic | execute dynamic mysql statements with params |
| GET 🔥 | /upload | upload single file |
| GET 🔥 | /uploads | upload multiple files |
| GET 🔥 | /download | download a file |
| GET | /api/tableName/describe | describe each table for its columns |
| GET | /api/tables | get all tables in database |
| GET | /_health | gets health of process and mysql -- details query params for more details |
| GET | /_version | gets version of Xmysql, mysql, node |
xmysql identifies foreign key relations automatically and provides GET api.
/api/blogs/103/comments
eg: blogs is parent table and comments is child table. API invocation will result in all comments for blog primary key 103.
/api/payments/103___JM555205
___ : If there are multiple primary keys - separate them by three underscores as shown
_p indicates page and _size indicates size of response rows
By default 20 records and max of 100 are returned per GET request on a table.
/api/payments?_size=50
/api/payments?_p=2
/api/payments?_p=2&_size=50
When _size is greater than 100 - number of records defaults to 100 (i.e maximum)
When _size is less than or equal to 0 - number of records defaults to 20 (i.e minimum)
/api/payments?_sort=column1
eg: sorts ascending by column1
/api/payments?_sort=-column1
eg: sorts descending by column1
/api/payments?_sort=column1,-column2
eg: sorts ascending by column1 and descending by column2
/api/payments?_fields=customerNumber,checkNumber
eg: gets only customerNumber and checkNumber in response of each record
/api/payments?_fields=-checkNumber
eg: gets all fields in table row but not checkNumber
eq - '=' - (colName,eq,colValue)
ne - '!=' - (colName,ne,colValue)
gt - '>' - (colName,gt,colValue)
gte - '>=' - (colName,gte,colValue)
lt - '<' - (colName,lt,colValue)
lte - '<=' - (colName,lte,colValue)
is - 'is' - (colName,is,true/false/null)
in - 'in' - (colName,in,val1,val2,val3,val4)
bw - 'between' - (colName,bw,val1,val2)
like - 'like' - (colName,like,~name) note: use ~ in place of %
nlike - 'not like' - (colName,nlike,~name) note: use ~ in place of %
/api/payments?_where=(checkNumber,eq,JM555205)~or((amount,gt,200)~and(amount,lt,2000))
~or - 'or'
~and - 'and'
~xor - 'xor'
eg: simple logical expression
/api/payments?_where=(checkNumber,eq,JM555205)~or(checkNumber,eq,OM314933)
eg: complex logical expression
/api/payments?_where=((checkNumber,eq,JM555205)~or(checkNumber,eq,OM314933))~and(amount,gt,100)
eg: logical expression with sorting(_sort), pagination(_p), column filtering (_fields)
/api/payments?_where=(amount,gte,1000)&_sort=-amount&p=2&_fields=customerNumber
eg: filter of rows using _where is available for relational route URLs too.
/api/offices/1/employees?_where=(jobTitle,eq,Sales%20Rep)
/api/tableName/findOne?_where=(id,eq,1)
Works similar to list but only returns top/one result. Used in conjunction with _where
/api/tableName/count
Returns number of rows in table
/api/tableName/1/exists
Returns true or false depending on whether record exists
/api/offices?_groupby=country
eg: SELECT country,count(*) FROM offices GROUP BY country
/api/offices?_groupby=country&_having=(_count,gt,1)
eg: SELECT country,count(1) as _count FROM offices GROUP BY country having _count > 1
/api/offices/groupby?_fields=country
eg: SELECT country,count(*) FROM offices GROUP BY country
/api/offices/groupby?_fields=country,city
eg: SELECT country,city,count(*) FROM offices GROUP BY country,city
/api/offices/groupby?_fields=country,city&_having=(_count,gt,1)
eg: SELECT country,city,count(*) as _count FROM offices GROUP BY country,city having _count > 1
/api/offices/groupby?_fields=country,city&_sort=city
eg: SELECT country,city,count(*) FROM offices GROUP BY country,city ORDER BY city ASC
/api/offices/groupby?_fields=country,city&_sort=city,country
eg: SELECT country,city,count(*) FROM offices GROUP BY country,city ORDER BY city ASC, country ASC
/api/offices/groupby?_fields=country,city&_sort=city,-country
eg: SELECT country,city,count(*) FROM offices GROUP BY country,city ORDER BY city ASC, country DESC
http://localhost:3000/api/payments/aggregate?_fields=amount
response body
[
{
"min_of_amount": 615.45,
"max_of_amount": 120166.58,
"avg_of_amount": 32431.645531,
"sum_of_amount": 8853839.23,
"stddev_of_amount": 20958.625377426568,
"variance_of_amount": 439263977.71130896
}
]
eg: retrieves all numeric aggregate of a column in a table
http://localhost:3000/api/orderDetails/aggregate?_fields=priceEach,quantityOrdered
response body
[
{
"min_of_priceEach": 26.55,
"max_of_priceEach": 214.3,
"avg_of_priceEach": 90.769499,
"sum_of_priceEach": 271945.42,
"stddev_of_priceEach": 36.576811252187795,
"variance_of_priceEach": 1337.8631213781719,
"min_of_quantityOrdered": 6,
"max_of_quantityOrdered": 97,
"avg_of_quantityOrdered": 35.219,
"sum_of_quantityOrdered": 105516,
"stddev_of_quantityOrdered": 9.832243813502942,
"variance_of_quantityOrdered": 96.67301840816688
}
]
eg: retrieves numeric aggregate can be done for multiple columns too
🔥🔥[ HOTNESS ALERT ]
Group by multiple columns in one API call using _fields query params - comes really handy
http://localhost:3000/api/employees/ugroupby?_fields=jobTitle,reportsTo
response body
{
"jobTitle":[
{
"Sales Rep":17
},
{
"President":1
},
{
"Sale Manager (EMEA)":1
},
{
"Sales Manager (APAC)":1
},
{
"Sales Manager (NA)":1
},
{
"VP Marketing":1
},
{
"VP Sales":1
}
],
"reportsTo":[
{
"1002":2
},
{
"1056":4
},
{
"1088":3
},
{
"1102":6
},
{
"1143":6
},
{
"1621":1
}
{
"":1
},
]
}
🔥🔥🔥🔥🔥🔥 [ HOTNESS ALERT ]
Chart API returns distribution of a numeric column in a table
It comes in SEVEN powerful flavours
- Chart : With min, max, step in query params 🔥🔥
⤴️
This API returns the number of rows where amount is between (0,25000), (25001,50000) ...
/api/payments/chart?_fields=amount&min=0&max=131000&step=25000
Response
[
{
"amount": "0 to 25000",
"_count": 107
},
{
"amount": "25001 to 50000",
"_count": 124
},
{
"amount": "50001 to 75000",
"_count": 30
},
{
"amount": "75001 to 100000",
"_count": 7
},
{
"amount": "100001 to 125000",
"_count": 5
},
{
"amount": "125001 to 150000",
"_count": 0
}
]
- Chart : With step array in params 🔥🔥
⤴️
This API returns distribution between the step array specified
/api/payments/chart?_fields=amount&steparray=0,10000,20000,70000,140000
Response
[
{
"amount": "0 to 10000",
"_count": 42
},
{
"amount": "10001 to 20000",
"_count": 36
},
{
"amount": "20001 to 70000",
"_count": 183
},
{
"amount": "70001 to 140000",
"_count": 12
}
]
- Chart : With step pairs in params 🔥🔥
⤴️
This API returns distribution between each step pair
/api/payments/chart?_fields=amount&steppair=0,50000,40000,100000
Response
[
{"amount":"0 to 50000","_count":231},
{"amount":"40000 to 100000","_count":80}
]
- Chart : with no params 🔥🔥
⤴️
This API figures out even distribution of a numeric column in table and returns the data
/api/payments/chart?_fields=amount
Response
[
{
"amount": "-9860 to 11100",
"_count": 45
},
{
"amount": "11101 to 32060",
"_count": 91
},
{
"amount": "32061 to 53020",
"_count": 109
},
{
"amount": "53021 to 73980",
"_count": 16
},
{
"amount": "73981 to 94940",
"_count": 7
},
{
"amount": "94941 to 115900",
"_count": 3
},
{
"amount": "115901 to 130650",
"_count": 2
}
]
- Chart : range, min, max, step in query params 🔥🔥
⤴️
This API returns the number of rows where amount is between (0,25000), (0,50000) ... (0,maxValue)
Number of records for amount is counted from min value to extended Range instead of incremental steps
/api/payments/chart?_fields=amount&min=0&max=131000&step=25000&range=1
Response
[
{
"amount": "0 to 25000",
"_count": 107
},
{
"amount": "0 to 50000",
"_count": 231
},
{
"amount": "0 to 75000",
"_count": 261
},
{
"amount": "0 to 100000",
"_count": 268
},
{
"amount": "0 to 125000",
"_count": 273
}
]
- Range can be specified with step array like below
/api/payments/chart?_fields=amount&steparray=0,10000,20000,70000,140000&range=1
[
{
"amount": "0 to 10000",
"_count": 42
},
{
"amount": "0 to 20000",
"_count": 78
},
{
"amount": "0 to 70000",
"_count": 261
},
{
"amount": "0 to 140000",
"_count": 273
}
]
- Range can be specified without any step params like below
/api/payments/chart?_fields=amount&range=1
[
{
"amount": "-9860 to 11100",
"_count": 45
},
{
"amount": "-9860 to 32060",
"_count": 136
},
...
]
Please Note: _fields in Chart API can only take numeric column as its argument.
Identifies numeric columns in a table which are not any sort of key and applies chart API as before - feels like magic when there are multiple numeric columns in table while hacking/prototyping and you invoke this API.
http://localhost:3000/api/payments/autochart
[
{
"column": "amount",
"chart": [
{
"amount": "-9860 to 11100",
"_count": 45
},
{
"amount": "11101 to 32060",
"_count": 91
},
{
"amount": "32061 to 53020",
"_count": 109
},
{
"amount": "53021 to 73980",
"_count": 16
},
{
"amount": "73981 to 94940",
"_count": 7
},
{
"amount": "94941 to 115900",
"_count": 3
},
{
"amount": "115901 to 130650",
"_count": 2
}
]
}
]
_join : List of tableNames alternated by type of join to be made (_j, _ij,_ lj, _rj)
alias.tableName : TableName as alias
_j : Join [ _j => join, _ij => ij, _lj => left join , _rj => right join)
_onNumber : Number 'n' indicates condition to be applied for 'n'th join between (n-1) and 'n'th table in list
Sql join query:
SELECT pl.field1, pr.field2
FROM productlines as pl
JOIN products as pr
ON pl.productline = pr.productline
Equivalent xjoin query API:
/api/xjoin?_join=pl.productlines,_j,pr.products&_on1=(pl.productline,eq,pr.productline)&_fields=pl.field1,pr.field2
Sql join query:
SELECT pl.field1, pr.field2, ord.field3
FROM productlines as pl
JOIN products as pr
ON pl.productline = pr.productline
JOIN orderdetails as ord
ON pr.productcode = ord.productcodeEquivalent xjoin query API:
/api/xjoin?_join=pl.productlines,_j,pr.products,_j,ord.orderDetails&_on1=(pl.productline,eq,pr.productline)&_on2=(pr.productcode,eq,ord.productcode)&_fields=pl.field1,pr.field2,ord.field3
Explanation:
pl.productlines => productlines as pl
_j => join
pr.products => products as pl
_on1 => join condition between productlines and products => (pl.productline,eq,pr.productline)
_on2 => join condition between products and orderdetails => (pr.productcode,eq,ord.productcode)
Example to use : _fields, _where, _p, _size in query params
/api/xjoin?_join=pl.productlines,_j,pr.products&_on1=(pl.productline,eq,pr.productline)&_fields=pl.productline,pr.productName&_size=2&_where=(productName,like,1972~)
Response:
[{"pl_productline":"Classic Cars","pr_productName":"1972 Alfa Romeo GTA"}]
Please note :
Xjoin response has aliases for fields like below aliasTableName + '_' + columnName.
eg: pl.productline in _fields query params - returns as pl_productline in response.
Dynamic queries on a database can be run by POST method to URL localhost:3000/dynamic
This is enabled ONLY when using local mysql server i.e -h localhost or -h 127.0.0.1 option.
Post body takes two fields : query and params.
query: SQL query or SQL prepared query (ones with ?? and ?)
params : parameters for SQL prepared query
POST /dynamic
{
"query": "select * from ?? limit 1,20",
"params": ["customers"]
}
POST /dynamic URL can have any suffix to it - which can be helpful in prototyping
eg:
POST /dynamic/weeklyReport
POST /dynamic/user/update
POST /upload
Do POST operation on /upload url with multiform 'field' assigned to local file to be uploaded
eg: curl --form file=@/Users/me/Desktop/a.png http://localhost:3000/upload
returns uploaded file name else 'upload failed'
(Note: POSTMAN has issues with file uploading hence examples with curl)
POST /uploads
Do POST operation on /uploads url with multiform 'fields' assigned to local files to be uploaded
Notice 's' near /api/uploads and files in below example
eg: curl --form files=@/Users/me/Desktop/a.png --form files=@/Users/me/Desktop/b.png http://localhost:3000/uploads
returns uploaded file names as string
http://localhost:3000/download?name=fileName
For upload and download of files -> you can specify storage folder using -s option Upload and download apis are available only with local mysql server
{"process_uptime":3.858,"mysql_uptime":"2595"}
Shows up time of Xmysql process and mysql server
http://localhost:3000/_health?details=1
{"process_uptime":1.151,"mysql_uptime":"2798",
"os_total_memory":17179869184,
"os_free_memory":2516357120,
"os_load_average":[2.29931640625,2.1845703125,2.13818359375],
"v8_heap_statistics":{"total_heap_size":24735744,
"total_heap_size_executable":5242880,
"total_physical_size":23521048,
"total_available_size":1475503064,
"used_heap_size":18149064,
"heap_size_limit":1501560832,
"malloced_memory":8192,
"peak_malloced_memory":11065664,
"does_zap_garbage":0}}
Provides more details on process.
Infact passing any query param gives detailed health output: example below
http://localhost:3000/_health?voila
{"process_uptime":107.793,"mysql_uptime":"2905","os_total_memory":17179869184,"os_free_memory":2573848576,"os_load_average":[2.052734375,2.12890625,2.11767578125],"v8_heap_statistics":{"total_heap_size":24735744,"total_heap_size_executable":5242880,"total_physical_size":23735016,"total_available_size":1475411128,"used_heap_size":18454968,"heap_size_limit":1501560832,"malloced_memory":8192,"peak_malloced_memory":11065664,"does_zap_garbage":0}}
http://localhost:3000/_version
{"Xmysql":"0.4.1","mysql":"5.7.15","node":"8.2.1"}
- You need just REST APIs for (ANY) MySql database at blink of an eye (literally).
- You are learning new frontend frameworks and need REST APIs for your MySql database.
- You are working on a demo, hacks etc
- If you are in need of a full blown MVC framework, ACL, Validations, Authorisation etc - its early days please watch/star this repo to keep a tab on progress.
Options:
-V, --version Output the version number
-h, --host <n> Hostname of database -> localhost by default
-u, --user <n> Username of database -> root by default
-p, --password <n> Password of database -> empty by default
-d, --database <n> database schema name
-r, --ipAddress <n> IP interface of your server / localhost by default
-n, --portNumber <n> Port number for app -> 3000 by default
-o, --port <n> Port number of mysql -> 3306 by default
-a, --apiPrefix <n> Api url prefix -> /api/ by default
-s, --storageFolder <n> Storage folder -> current working dir by default (available only with local)
-i, --ignoreTables <n> Comma separated table names to ignore
-c, --useCpuCores <n> Specify number of cpu cores to use / 1 by default / 0 to use max
-y, --readOnly readonly apis -> false by default
-h, --help Output usage information
Examples:
$ xmysql -u username -p password -d databaseSchema
Simply run with docker run -p 3000:80 -d markuman/xmysql:0.4.2
The best way for testing is to run mysql in a docker container too and create a docker network, so that xmysql can access the mysql container with a name from docker network.
- Create network
docker network create mynet
- Start mysql with docker name
some-mysqland bind to docker networkmynetdocker run --name some-mysql -p 3306:3306 --net mynet -e MYSQL_ROOT_PASSWORD=password -d markuman/mysql
- run xmysql and set env variable for
some-mysqlfrom step 2docker run -p 3000:80 -d -e DATABASE_HOST=some-mysql --net mynet markuman/xmysql
You can also pass the environment variables to a file and use them as an option with docker like docker run --env-file ./env.list -p 3000:80 --net mynet -d markuman/xmysql
environment variables which can be used:
ENV DATABASE_HOST 127.0.0.1
ENV DATABASE_USER root
ENV DATABASE_PASSWORD password
ENV DATABASE_NAME sakila
Furthermore, the docker container of xmysql is listen on port 80. You can than request it just with http://xmysql/api/ in other services running in the same docker network.
Given you've deployed your xmysql docker container like
docker run -d \
--network local_dev \
--name xmysql \
-p 3000:80 \
-e DATABASE_HOST=mysql_host \
-e DATABASE_USER=root \
-e DATABASE_PASSWORD=password \
-e DATABASE_NAME=sys \
markuman/xmysql:0.4.2but the response is just
["http://127.0.0.1:3000/api/tables","http://127.0.0.1:3000/api/xjoin"]then obviously the connection to your mysql database failed.
- attache to the xmysql image
docker exec -ti xmysql
- install mysql cli client
apk --update --no-cache add mysql-client
- try to access your mysql database
mysql-client -h mysql_host
- profit from the
mysql-clienterror output and improve the environment variables for mysql
This is a config example when you use nginx as reverse proxy
events {
worker_connections 1024;
}
http {
server {
server_name 127.0.0.1;
listen 80 ;
location / {
rewrite ^/(.*) /$1 break;
proxy_redirect off;
proxy_set_header Host $host;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_pass http://127.0.0.1:3000;
}
}
}
e.g.
- create a docker network
docker network create local_dev - start a mysql server
docker run -d --name mysql -p 3306:3306 --network local_dev -e MYSQL_ROOT_PASSWORD=password mysql - start xmysql
docker run -d --network local_dev --name xmyxql -e DATABASE_NAME=sys -e DATABASE_HOST=mysql -p 3000:80 markuman/xmysql:0.4.2 - start nginx on host system with the config above
sudo nginx -g 'daemon off;' -c /tmp/nginx.conf - profit
curl http://127.0.0.1/api/host_summary_by_file_io_type/describe
When you start your nginx proxy in a docker container too, use as proxy_pass the --name value of xmysql. E.g. proxy_pass http://xmysql (remember, xmysql runs in it's docker container already on port 80).
docker-compose run test
- Requires
docker-composeto be installed on your machine.