remy/txt2bas

ZX Spectrum BASIC conversion tooling (specific support for NextBASIC)

txt2bas and bas2txt - tooling for ZX Spectrum NextBASIC

This code is inspired by the .txt2bas dot command on the ZX Spectrum Next (not that I could read the asm code!).

This project provides:

• txt2bas command line tool
• bas2txt command line tool
• NextBASIC validation
• library for BASIC and text manipulation, validation and renumbering

Installation

Node and npm (included with node) are required to install and run the code.

For the command line tooling - this installs both tools:

npm install --global txt2bas

Command line usage

Command line arguments are the same for both txt2bas and bas2txt:

txt2bas -i source.txt -o result.bas # generate a 3dos basic file
bas2txt -i source.bas -o result.txt # generates plain text

Omitting -o will print to stdout.

By default the generated file is a +3DOS format unless the output or input filename ends in .tap or using the format option -f tap:

txt2bas -i source.txt -o result.tap # generates a tap file

The command line can also read from stdin though this works best on txt2bas and not recommended for bas2txt.

Note that #autostart directive is also supported.

Command line options

• -i FILENAME - input filename
• -o FILENAME - output filename
• -t - (txt2bas only) test and validate the NextBASIC source
• -C - (txt2bas only) strip comments (to reduce final size)
• -bank - (txt2bas only) generate a BANK loadable result
• -A #n - (txt2bas only) set autostart line to #n
• -f 3dos|tap - set the output format
• -H- omit the file header (either in output or in parsing input)
• -udg - UDGs are used so encode with binary not utf8
• -tokens - (txt2bas only) show parser tokens (for debugging)
• -h - Show help options
• -v - Show current version

Library API

Problematically using the library exposes a number of paired functions:

• line2bas(String: line): Object<Uint8Array: basic, Number: lineNumber, Array: tokens, Number: length> - the byte data is contained in result.basic
• bas2line(Uint8Array: data): String - expects to include the line number, line length and the line itself as bytes
• file2bas(String: source, Object<String=3dos: format, filename=UNTITLED: String, validate=false: Boolean>): Uint8Array - results full byte array with correct format header, if validate is true, will throw on token errors
• bas2file(Uint8Array: source, String=3dos: format): String - formatted BASIC text
• formatText(String: line): String - processes the line through line2bas then bas2line to result the formatted line
• validateTxt(String: source): Array[String] - parses each line collecting and returning any token errors
• plus3DOSHeader and tapHeader - file headers for the appropriate data formats
• codes an object lookup from NextBASIC numerical value to text value, ie. 0xf5 = 'PRINT'
• statements(String: source): Array[Statement] - returns the parsed statement which include lineNumber and tokens for each line.
• renumber(String: source, Object<start: Number, end: Number, step=10: Number, base=start: Number>) - renumbers source lines and GO TO line number targets.

Development

• Currently the latest code uses node@21 (due to specific use of syntax)
• To test with another project: npm link
• For new features of language or validation changes, ensure a test is provided
• txt2bas also support different NextOS versions, specified through parser-version

Licence

• MIT