CostaPy Modules - Main Process

Giving the response

Main response have 3 options:

status
desc
data

The response status only have 2 options:

success
failed

Example for the success response:

response["status"   ] = "success"
response["desc"     ] = "Product list collected"
response["data"     ] = {
    "product"   : product_list,
    "other"     : "Some random text"
}

Example for the failed response:

response["status"   ] = "failed"
response["desc"     ] = "Something went wrong"
response["data"     ] = {
    "exception"   : str(e)
}

Fetching data

fetchone

cursor.execute(f"SELECT * FROM tablename WHERE fieldname = {id} ")
row = cursor.fetchone()

token   = row['token'    ].decode()
id      = row['id'       ]
name    = row['name'     ]
email   = row['email'    ]

fetchall

cursor.execute(f"SELECT * FROM product ")
product_list = cursor.fetchall()

Nested fetchall

Variables:

l1 is mean List level 1
c2 is mean Count level 2
d3 is mean Data level 3
etc

Sample:

texture_list = []

cursor.execute(f"SELECT * FROM tableone WHERE fieldname = '{key}' ")
l1 = cursor.fetchall()
c1 = 0
for d1 in l1:

    texture_list.append({
        "id"    :d1["id"    ],
        "name"  :d1["name"  ],
        "desc"  :d1["desc"  ]
    })

    cursor.execute(f"SELECT * FROM tabletwo WHERE keyfield = '{d1['id']}' ")
    l2 = cursor.fetchone()
    texture_list[c1]["owner"] = l2

    cursor.execute(f"SELECT * FROM tablethree WHERE keyfield = '{d1['id']}' ")
    l2 = cursor.fetchall()
    texture_list[c1]["file"] = []
    c2 = 0
    for d2 in l2:
        texture_list[c1]["file"].append({
            "id"        :d2["id"        ],
            "filedir"   :d2["filedir"   ],
            "filetype"  :d2["filetype"  ]
        })

        c2 += 1

    c1 += 1

Get the last row ID from insert query

cursor.execute(f"INSERT INTO `product_files` VALUES (DEFAULT, '{webdir}', '{filename}' ) ")
product_files_lastrowid = cursor.lastrowid

Begin, Rollback, and Commit

Begin, rollback, and commit can be useful if you use more than 1 process that cannot be separate. For example: more than 1 table insertion query, inserting query while upload success, etc.

The pattern:

cursor.execute("BEGIN;")
try:
    # Process and response
except Exception as e:
    cursor.execute("ROLLBACK;")
    # Process and response when failed
cursor.execute("COMMIT;")

Sample:

cursor.execute("BEGIN;")

try:

    cursor.execute(f"INSERT INTO `files` VALUES (DEFAULT, '{webdir}', '{filename}' ) ")
    files_lastrowid = cursor.lastrowid

    cursor.execute(f"INSERT INTO `thumbnail` VALUES (DEFAULT, '{image}', '{files_lastrowid}' ) ")

    response["status"   ] = "success"
    response["desc"     ] = "insert success"

    loggorilla.prcss(APIADDR, f"insert success")

except Exception as e:

    cursor.execute("ROLLBACK;")

    response["status"   ] = "failed"
    response["desc"     ] = "There is error when processing try. See the exception for the clue."
    response["data"     ] = {
        "exception" : str(e)
    }

    loggorilla.error(APIADDR, f"{str(e)}")

cursor.execute("COMMIT;")

File management

Uploading

mediafile   = params["mediafile"]
name        = "helloworld"
ext         = pathlib.Path(mediafile.filename).suffix
dir         = f"/srv/media/material/texture"

uploading.main(mediafile, name+ext, dir)

It will be overwrite if the file already on there.

Removing

import pathlib
import glob
import os

dir     = pathlib.Path(f"/srv/media/material/texture")
name    = "helloworld.png"

for row in glob.iglob(os.path.join(dir, name)):
    os.remove(row)

It can combining with *, for example:

dir     = pathlib.Path(f"/srv/media/product/*/file")
name    = "*.zip"

4.0 KiB Raw Permalink Blame History