- 27 Jun, 2019 2 commits
-
-
Victor Yu authored
Split execute_process to force the two commands there to be executed one after another. See also: https://cmake.org/pipermail/cmake/2016-March/063074.html
- 07 Dec, 2018 3 commits
-
-
William Huhn authored
-
William Huhn authored
-
William Huhn authored
-
- 29 Nov, 2018 1 commit
-
-
Raul Laasner authored
-
- 28 Nov, 2018 2 commits
-
-
William Huhn authored
-
William Huhn authored
-
- 10 May, 2018 3 commits
-
-
William Huhn authored
-
William Huhn authored
-
William Huhn authored
Moved the file/unit check for a JSON file when writing to its own subroutine. However, based on issues observed with Intel 18.0.2, I've commented it out. (This was present in the original inlined version too.)
-
- 23 Apr, 2018 3 commits
-
-
William Huhn authored
-
William Huhn authored
-
William Huhn authored
-
- 11 Apr, 2018 1 commit
-
-
William Huhn authored
-
- 10 Apr, 2018 1 commit
-
-
William Huhn authored
-
- 09 Apr, 2018 1 commit
-
-
William Huhn authored
Added module procedure to convert values into the associated JSON formatted string, rather than doing it explicitly for each value and name/value writing.
-
- 08 Apr, 2018 7 commits
-
-
William Huhn authored
-
William Huhn authored
-
William Huhn authored
-
William Huhn authored
-
William Huhn authored
Tweaked Doxyfile to my liking (and renamed it from doxyfile). Added code fragments to README.md file; want to see how they render on GitHub.
-
William Huhn authored
-
William Huhn authored
-
- 07 Apr, 2018 7 commits
-
-
William Huhn authored
-
William Huhn authored
All output is now routed through the fjson_write subroutine. Added error checking for the file not being open to writing.
-
William Huhn authored
-
William Huhn authored
CTest now uses a script (which I shamelessly stole from the CMake mailing list) to compare the output of a test executable against a known-good/blessed output.
-
William Huhn authored
Added positive unit tests. Currently, they are always built, and no testing is done to determine if the resulting JSON actually is valid.
-
William Huhn authored
Changed some variable names which were holdovers from ELSI. Properly capitalized FortJSON throughout code.
-
William Huhn authored
Added an error case for the handle being uninitialized. If it is, the code will not output anything to it.
-
- 06 Apr, 2018 1 commit
-
-
William Huhn authored
Modified error handling. Code will no longer stop abruptly, instead it will output an error message to the JSON text (breaking the parsing) and set a dirty error bit, which is user-accessible via two functions. This allows users to determine how they want to handle errors, and it also allows them to modify the JSON by hand if the error was isolated and not overly dramatic (say, outputting a name/value pair instead of a value).
-
- 05 Apr, 2018 4 commits
-
-
William Huhn authored
-
William Huhn authored
Turns out that ECMA-404 DOES support an individual value as a JSON text. Created a test subdirectory to store testing programs and moved my test program tehre.
-
William Huhn authored
Added README file in Markdown. Removed requirement inherited from ELSI that user manually specify the Fortran compiler.
-
William Huhn authored
Added doxygen commenting to code. Needs to be added to cmake build system (as an optional feature).
-
- 04 Apr, 2018 4 commits
-
-
William Huhn authored
-
William Huhn authored
Since this code is no longer a subset of ELSI, modified the style to better suite my aethestics. Removed two legacy variables.
-
William Huhn authored
Added objects and arrays as values for name/value pairs. The main pieces for JSON output are now implemented; from here on, it's code cleanup, filling in additional value, more error handling, and testing.
-
William Huhn authored
-