https://github.com/arjenmarkus/old-programming-idioms

Explain old-style programming idioms used in FORTRAN 77 programs
https://github.com/arjenmarkus/old-programming-idioms

fortran fortran-tutorial fortran77

Last synced: about 16 hours ago
JSON representation

Explain old-style programming idioms used in FORTRAN 77 programs

• Host: GitHub
• URL: https://github.com/arjenmarkus/old-programming-idioms
• Owner: arjenmarkus
• License: mit
• Created: 2023-05-31T04:12:54.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-01-19T16:09:17.000Z (14 days ago)
• Last Synced: 2025-01-19T17:24:28.156Z (14 days ago)
• Topics: fortran, fortran-tutorial, fortran77
• Language: TeX
• Homepage:
• Size: 3.2 MB
• Stars: 23
• Watchers: 6
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
I want to write about the typical idioms you find in old-style FORTRAN

programs:

- What do they mean?

- What are modern equivalents?

This will typically be the common things from FORTRAN 77 onwards, but

perhaps also a wee bit about FORTRAN 66 (like one-pass loops?)

These are the keyword/subjects I am thinking about:

- array(*) versus array(:)

- array(10) as the starting point

- COMMON blocks and BLOCK DATA, named and blank COMMON

- history of computers:

   - hardware

   - memory management

   - tools like source code control systems

   - connections between computers, Internet

- equivalence

- constants as actual arguments

- intent

- temporary arrays - non-contiguous arrays

- implicit types

- double precision versus kind

- checking interfaces

- separate compilations, the misunderstanding of one routine per file

- fixed form and spaces

- standard input and output

- LU-numbers 5 and 6 (and 7)

- command-line arguments for file names

- real do-variables

- entry

- SAVE-attribute and SAVE as compiler property

- ASSIGN

- arithmetic IF

- computed GOTO

- F66: one-pass do-loops

- statement functions

- six characters

- numerical binary representations versus IEEE (IBM, Cray, Convex)

- big-endian and little-endian

- double complex

- unformatted versus binary files

- list-directed input and output - also: /

- narrow formats

- input formats with decimals - F5.1 and there is no decimal point in

  the input

- double precision :: x = 1.01 not the same as double precision :: x = 1.01d0

- use of d00 in input

- format statements