ofort

ofort is a small tree-walking interpreter for a subset of Fortran 90/95/2003. It is extracted from the CodeBench project of yu314-coder into a standalone C command-line project.

REAL and DOUBLE PRECISION are distinguished by type tag/kind, but both are currently stored using C double internally.

Current support

ofort is intended for experimenting with small programs, examples, and interpreter development. It is not a full Fortran compiler.

Implemented pieces include scalar INTEGER, REAL, DOUBLE PRECISION, CHARACTER, LOGICAL, arrays and allocatable arrays, implicit none checking, assignment type checks, if, do, labeled do, exit, subroutines, basic functions, internal and external formatted I/O, simple unformatted stream I/O, and interactive expression evaluation.

Supported intrinsics include common numeric, conversion, character, array, and random-number routines such as abs, sqrt, sin, cos, exp, log, mod, modulo, dim, aint, nint, int, real, dble, dprod, kind, len, trim, sum, product, maxval, minval, allocated, lbound, ubound, random_number, random_seed, and system_clock.

Known missing Fortran 95 intrinsics and candidates for future work are tracked in intrinsics_to_do.txt.

Build

make

Alternative compiler targets:

make gcc
make clang

Run

.\ofort.exe examples\xtry.f90

Pass program arguments after --:

.\ofort.exe x.f90 -- alpha beta

Multiple positional source files are concatenated in command-line order before execution:

.\ofort.exe part1.f90 part2.f90

By default, undeclared variables are rejected. To enable historical implicit typing, where names beginning with I-N are integers and other names are real, use:

.\ofort.exe --implicit-typing x.f90

To start an interactive session with an existing file loaded into the edit buffer:

.\ofort.exe --load x.f90

To load a file, run it once, and stay in the interactive session:

.\ofort.exe --load-run x.f90

To compare ofort output with gfortran for a program:

.\ofort.exe --check-gfortran x.f90

To lex and parse a program without running it:

.\ofort.exe --check x.f90

To run ofort on every file matched by a glob:

python .\scripts\xofort.py "tests\cases\*.f90"
python .\scripts\xofort.py --check "tests\cases\*.f90"
python .\scripts\xofort.py --limit 5 "tests\cases\*.f90"
python .\scripts\xofort.py --timeout 30 "tests\cases\*.f90"
python .\scripts\xofort.py --max-lines 100 "tests\cases\*.f90"
python .\scripts\xofort.py --quiet "tests\cases\*.f90"
python .\scripts\xofort.py --filter "gfortran -c -std=f95" "tests\cases\*.f90"

The batch runner prints each file name and line count before processing it, uses a 30-second per-file timeout by default, and reports elapsed time at the end. --quiet suppresses output for files that ofort handles successfully. --filter runs the given command with the source file appended and skips files for which that command fails.

With no file argument, ofort starts an interactive session. Type source at the prompt and use these commands:

.       run the current source and continue
.run    run the current source; .run n repeats n times; use -- before program arguments
.runq   run the current source and quit; accepts the same count and arguments as .run
.time   run the current source and print elapsed-time statistics; .time n repeats n times
.quit   quit without running
.clear  clear the current source
.del    delete a source line or range, such as .del 3, .del 2:4, .del :3, or .del 4:
.ins    insert a source line before a line number, such as .ins 3 integer :: j
.rep    replace a source line, such as .rep 5 print *, i, j
.rename rename a variable token throughout the editable source, such as .rename i j
.list   list the current source
.decl   list declaration lines from the current source
.vars   list current variable values, or selected names such as .vars i x
.info   list declaration-style variable details, or selected names such as .info i x
.shapes list array shapes, or selected names such as .shapes x y
.sizes  list array sizes, or selected names such as .sizes x y
.stats  list numeric array statistics, or selected names such as .stats x y
.load   load a file into the current source
.load-run load a file, run it once, and keep editing

For repeated runs, command arguments can be supplied after --, for example .run 3 -- alpha beta.

When .load file.f90, .load-run file.f90, --load file.f90, or --load-run file.f90 loads a complete program, a final end... line is temporarily held aside so new input is inserted before the end of the program. The held line is restored when running or autosaving.

The shortcuts q and quit also quit, unless q or quit has been defined as a variable in the current source buffer. .quit always quits.

When an interactive session exits with a non-empty source buffer, the buffer is saved automatically as main.f90, or main1.f90, main2.f90, and so on if earlier names already exist.

In interactive mode only, a bare expression line is evaluated immediately against the current source buffer and prints its value. For example, after entering x = 3, a line containing only x displays 3 immediately. Bare expression lines are not added to the source buffer.

Test

pytest -q