Skip to content

subset of Regular Expressions implemented in Fortran

License

Notifications You must be signed in to change notification settings

urbanjost/M_match

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

46 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

M_match

M_match

subset of Regular Expressions implemented in Fortran

A basic implementation in Fortran of a subset of regular expressions as described in "Software Tools" by Kernighan and Plauger, 1976.

The original used different metacharacters than are commonly used today.

> ORIG. NOW
> ?      .  any single character
> *         any number (including zero) of repeats of previous character
> %      ^  beginning of line
> $         end of line
> @      \  escape next metacharacter
> []        a set of characters. In a set of characters
>           -  specifies a range of characters in a set of characters.
               Must be subsets of a-z, A-Z, or 0-9
>           ^  negates the set if the first character inside a set of characters.

This code functions, but a more complete (by todays' expectations) pure-Fortran implementation is planned.

It is hoped this particular module will be useful primarily because it is a native implementation in Fortran, albeit it is not as optimized or as feature-rich as several common regex C libraries that, when available, can be accessed via an ISO_C_BINDING in modern Fortran compilers.

    git clone https://github.com/urbanjost/M_match.git
    cd M_calculator/src
    # change Makefile if not using one of the listed compilers
     
    # for gfortran
    make clean
    make F90=gfortran gfortran
     
    # for ifort
    make clean
    make F90=ifort ifort

    # for nvfortran
    make clean
    make F90=nvfortran nvfortran

    # run simple example program
    PROGRAMS/findchange '%!' <M_match.f90

A simple example that creates a grep(1)-like filter program:

program demo_m_match
use M_match,   only : getpat, match, regex_pattern
use M_match,   only : YES, NO, ERR
implicit none
character(len=1024) :: line='', argument=''
type(regex_pattern) :: p
integer             :: ios
   call get_command_argument(1,argument)
   if(argument.eq.'')stop 'missing regular expression'
   if (getpat(trim(argument), p%pat) .eq. ERR) then
      stop '*M_match* Illegal pattern.'
   endif
   INFINITE: do
      read(*,'(a)',iostat=ios)line
      if(ios.ne.0)exit INFINITE
      if (match(trim(line)//char(10), p%pat) .eq. YES) then
         write(*,'(*(g0,1x))')trim(line)
      endif
   enddo INFINITE
end program demo_m_match

SUPPORTS FPM fpm

Alternatively, download the github repository and build it with fpm ( as described at Fortran Package Manager )

     git clone https://github.com/urbanjost/M_match.git
     cd M_match
     fpm build
     fpm test

or just list it as a dependency in your fpm.toml project file.

        [dependencies]
        M_match        = { git = "https://github.com/urbanjost/M_match.git" }

DOCUMENTATION docs

USER

  • A single page that uses javascript to combine all the HTML descriptions of the man-pages is at BOOK_M_match.

  • a simple index to the man-pages in HTML form for the routines

  • There are man-pages in the repository download in the docs/ directory that may be installed on ULS (Unix-Like Systems).

  • man-pages

  • CHANGELOG provides a history of significant changes

DEVELOPER


PASSES TESTS

Sat Dec  5 01:26:24 EST 2020

GNU Fortran (GCC) 8.3.1 20191121 (Red Hat 8.3.1-5)
Copyright (C) 2018 Free Software Foundation, Inc.

ifort (IFORT) 19.1.3.304 20200925
Copyright (C) 1985-2020 Intel Corporation.  All rights reserved.

nvfortran 20.7-0 LLVM 64-bit target on x86-64 Linux -tp nehalem 
NVIDIA Compilers and Tools
Copyright (c) 2020, NVIDIA CORPORATION.  All rights reserved.