Kata QA

Objetive

The objetive of this project is to create test for the different classes of the project, to cover the code and find the bugs made. The test have the objetive to use the different capabilities of the Junit 5.

The classes that should be covered has already created test clases, with some methods. This methods are orientative and the user can modify anything from inside the test part, but cannot edit nothing from the source code.

The problem does also support integration tests. All the test that include IT in their name will be execute in the integration test part.

In order to run all the test, we need to run mvn verify.

Domain

We want to create a search engine for a furniture shop. For that, we need to read the data from a .csv file and storage the information in a database. After that, we need to allow to perform some action (like retrieve data) thorough a console commands that will go against a backend search.

Index

The index must read a .csv file, analyze the data, check that the data is valid, and index the valid data and refuse the invalid data.

By default, when the program stats, it will read the file database.csv. Later, we can request to index a new .csv file to update the database. This new file can update and add new entries to the database, but not delete them (that is a separate action later explain). All the .csv files are given to us by the customer.

CSV Data Model

The .csv file has 4 columns:

• name: String, the name of the product, unique.
• type: String, the type (couch, seat, chair, table...)
• color: String, the color of the product
• size: Char, the size (s, m o l)

The order of the fields is vital, the order declares what fields are each.

Rules:

• All data is storage in minuscule, if something came in caps is must be transform in minuscule.
• The extra fields are ignored, they are not indexed.
• Every product indexed will be given automatically a unique id in ascending order, starting in the 1.
  • If a product is deleted, its id will be available to be used by a new product.
  • The id is assign to the product after it has pass successfully the analysis.
• If an entry of the .csv is not valid, is must be ignored, and it must be logged in the log a message with the reason and the number of the iteration.
• The name is mandatory and unique, there can be two products with the same name in the database.
  • If the name is duplicated in the same .csv file, from the second to the last match of a product with that ** name**, must be ignored.
  • If a name already present in the database comes in a new .csv to be indexed, the product will be updated with the new information, without changing the id.
• Type is mandatory
• Color is optional, if is not present it will be stored as an empty String.
  • If the type is seat or couch, then the color is mandatory.
• Size must be s, m o l.
• If size is empty, it will be stored as m.

Search

The search can be accessed through a command console after the program has finish of indexing the files.

• Every request that matches one of the next commands will be stored in the log.
• If one request is not correct (wrong parameters), a message will be stored in the log and the command will be ignored.

The next commands are available:

search

Retrieves all the products that match with the search criteria.

• If there is no extra params, it will retrieve all the products.
• The field name is the only field that only uses partial match, the rest of the fields uses exact match.
• For adding criteria to the search, we will introduce the field name, the symbol = or != and the term, with spaces and in minuscule.
  • The symbol = means that it must be equals (or contains for the field name) to the term.
  • The symbol != means that it must not be equals (or contains for the field name) to the term.
  • It is possible to add more search criteria using &, with spaces.

skusearch

Retrieves the product that matches the id

• An id must be sent.

delete

Deletes the product that matches the id

• An id must be sent.

update

Updates the database with the information of a file.

• By default, it will read the database_update.csv and will analyze and index the data following the indexation rules.
• The files with the same name will be updated, without changing the id.

save

Generates a file database_save.csv with the actual data stored in the database, without the id.

exit

Ends the program and generates a log.txt with all the information.