corentin/umake
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Change README

Browse source
This commit is contained in:
Corentin 2021-06-29 13:28:44 +09:00
commit 15d7304baf
2 changed files with 10 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

15
README.md
View file

@ -28,6 +28,13 @@ By default 3 targets can be used `make` (or `make all` as usual default of makef
**Be careful** : `clean` will delete the object and binary output directories defined in `make.py`. **Be careful** : `clean` will delete the object and binary output directories defined in `make.py`.
**Example :** from the root folder of a project :
```
git submodule add https://gitlab.com/corentin-pro/umake.git && cp umake/make* .
```
## How it works ## How it works
### Project layout ### Project layout
@ -35,19 +42,13 @@ By default 3 targets can be used `make` (or `make all` as usual default of makef
µmake expects all the header and source files to be place in a directories specified in the configuration (in `make.py`). µmake expects all the header and source files to be place in a directories specified in the configuration (in `make.py`).
Object files and final binaries are outputed also in directories specified in the configuration but objects files in `debug` and `release` (default) are separated. Object files and final binaries are outputed also in directories specified in the configuration but objects files in `debug` and `release` (default) are separated.
The separation of objects files is meant to switch mode quickly (only binaries are rebuilt). The separation of object/release objects files is meant to avoid rebuilding the whole project after debugging a single file.
To determine if a source file needs to be recompiled an hash is saved. The hashed are saved in a JSON file at the root of the object files directory (there are in fact 2 seperate files respectively in `debug` oand `release`). To determine if a source file needs to be recompiled an hash is saved. The hashed are saved in a JSON file at the root of the object files directory (there are in fact 2 seperate files respectively in `debug` oand `release`).
If any header file changes every source files will be recompiled to avoid any issue (hence the small or medium-size project target for this tool). This could be address in the future if needed.
### Caveats ### Caveats
* Header files dependencies is not implemented (any change triggers all compilation)
* Knowing python is recommended to understand the tool for any usage above configuration change. * Knowing python is recommended to understand the tool for any usage above configuration change.
* Targets management is not yet implemented (building all binaries), this can be easily address knowing python. A generic solution should be possible in the future.
* Template configuration is meant for C++ files (cpp/hpp) for now. * Template configuration is meant for C++ files (cpp/hpp) for now.

4
make.py
View file

@ -8,11 +8,11 @@ from umake import make
class Config: class Config:
CC = 'g++' # Compiler to call CC = 'g++' # Compiler to call
APPS = ['app_name'] # Output binaries (need to be found as .cpp directly in SOURCE_DIR) APPS = ['app_name'] # Output binaries (path to source without extension)
IGNORE_APPS = [] IGNORE_APPS = []
JOB_COUNT = int(os.cpu_count() * 0.8) # Concurent jobs (multi-processing) JOB_COUNT = int(os.cpu_count() * 0.8) # Concurent jobs (multi-processing)
BIN_DIR = Path('bin') # Output directory (binaries) BIN_DIR = Path('bin') # Output directory (binaries)
INCLUDE_DIR = Path('include') # Include directory (header files) INCLUDE_DIR = Path('include') # Include directory (header files)
OBJECT_DIR = Path('obj') # Temporary directory (object files) OBJECT_DIR = Path('obj') # Temporary directory (object files)
SOURCE_DIR = Path('src') # Source directories SOURCE_DIR = Path('src') # Source directories