ZXDB API for Spectrum Next

This is an extremely simple API service for use with the ZX Spectrum Next. If offers ZXDB search and download functionality.

This service is intentionally non-SSL to allow for the ZX Spectrum Next to process the requests.

API

All requests are GET based.

If processing on a Spectrum Next, you will need to pass the HTTP response headers, you can either wait for a blank line (as per HTTP specification to represent the start of the body) or curl and count the headers you need to skip. Current this code is hosted on Heroku so the headers shouldn't change in the foreseeable future - but this is not a given.

GET /?s=SEARCHTERM

Searches for an entity based on a query using the zxinfo search engine. Returns a maximum of 10 of encoded results.

Note on spaces: if the search term includes spaces, to reduce encoding requirements, spaces are represented with a * character. So "manic miner" is manic*miner.

URL structure

http://zxdb.remysharp.com/?s=target

Example of result

^25483^Target: Renegade^Target-Renegade.tzx^116696^2^1988^
^25485^Target: Renegade^Target-Renegade128.tap^112085^1^1988^
^25486^Target: Renegade^Target-Renegade48.tap^107007^1^1988^
^25487^Target: Renegade^Target-Renegade(ErbeSoftwareS.A.).tzx^112472^2^1988^
^25489^Target: Renegade^Target-Renegade(TheHitSquad).tzx^116696^1^1990^
^25459^Renegade^Renegade.tzx^67665^2^1987^
^25461^Renegade^Renegade128.tap^71429^1^1987^
^25462^Renegade^Renegade48.tap^48977^1^1987^
^25463^Renegade^Renegade(ErbeSoftwareS.A.).tzx^59640^2^1987^
^25465^Renegade^Renegade(TheHitSquad).tzx^62903^2^1989^

Data structure explained

The fields are separated by the ^ character and rows are terminated by \n

The fields are:

1. Download id
2. Title
3. Filename and type (TAP, TZX, etc)
4. File size
5. Number of file options (default: 1)
6. Year published

The id is required to download a file.

In the case where the number of file options (field 5) to request file 2, append /2 to the /get/<id>/2 request.

Categories

By default all types of records are returned. You can narrow the results with a query string ?cat=n, where n is short for:

• g - games only
• a - apps and utilities
• d - demos including music

If a category short code isn't recognised, the filtering is disabled.

GET /get/ID

Returns a binary stream with content-length header set for the given file id. The file format will be the type associated with the particular ID.

URL structure

http://zxdb.remysharp.com/get/18840

Maintenance

1. Download the zxdb project
2. In that project directory, run $ python3 scripts/ZXDB_to_SQLite.py
3. Then create the database: $ cat ZXDB_sqlite.sql| sqlite3 zxdb.sqlite
4. Remove the unneeded/unwanted tables: $ sqllite3 zxdb.sqlite then run the following lines:

PRAGMA writable_schema = 1;
DELETE FROM sqlite_master WHERE type = 'table' AND name NOT IN ('downloads', 'entries');
PRAGMA writable_schema = 0;
VACUUM;

5. Copy across zxdb.sqlite into this project.

License

• MIT