jgabaut / sprites4curses Goto Github PK

Describe the bug

For more basic details, refer to this issue.

To Reproduce
Steps to reproduce the behavior:

1. Using one of the python scripts, try parsing a .png containing more than one ? pixel on the same row. (Hint: S4C_Color_Index may suggest it's S4C_LIGHT_YELLOW. I'mma count the off-by-ones later. :) )
2. If the encoding forms a trigraph sequence, a compiler using --std=c(<23) may convert the tokens and corrupt the animation, sortening some lines for some unlucky frames.

Expected behavior
When in C generation mode, the script keeps track of how many ? chars are being put on the same row and breaks up the pattern while retaining the correct color encoding.

Screenshots

(From helapordo)

• Version >=0.3.2

Additional context

Wikipedia

All current python scripts could be wrapped by a single entry point that handles the different functions as subcommands.

As of version 0.4.3, the ./scripts directory contains some wrappers that should emulate previous behaviour.

This is sprites.py, for example:

#!/bin/python3

import sys
from s4c.core.sprites import main

if __name__ == "__main__":
    main(sys.argv)

The import works by leveraging a symlink for ./scripts/s4c:

./scripts/s4c -> ./s4c-scripts/s4c.

The old versions of the scripts are available under ./scripts-legacy.

Eventually, these will be removed. The patch needed in any Makefile using the scripts directly would be something like this:

< 	python3 ./sprites4curses/scripts/palette.py C-impl $(PALETTE_PATH) .. > ./src/palette.c
< 	python3 ./sprites4curses/scripts/palette.py C-header $(PALETTE_PATH) .. > ./src/palette.h
---
> 	python3 -m sprites4curses.s4c-scripts.s4c.core.palette C-impl $(PALETTE_PATH) .. > ./src/palette.c
> 	python3 -m sprites4curses.s4c-scripts.s4c.core.palette C-header $(PALETTE_PATH) .. > ./src/palette.h

In (#10), static void init_color_pairs_from_palette() was introduced.
It's not properly working. when called from demo.c instead of init_s4c_color_pairs().

All the needed terminal features (like, colors) should have a call to the right property macro, to check if service is possible.

Not able to generate demo executable
I cloned this repository then i installed pillow using pip. I followed the steps in README to compile for demo project.
make command is returning errors.

To Reproduce
Steps to reproduce the behavior:

1. Clone Repo
2. cd Repo
3. Clone scripts repo
4. Compile using recommended steps in README

# EDIT: I used following steps to compile from README
autoreconf; automake --add-missing; autoreconf; automake --add-missing autoreconf; ./configure make; ./demo demofile.txt

Executable file: demo
Program should compile. What am i missing?

Screenshot

I tried ./configure and make as separate commands in resulted in following error.

then, i rechecked README they are not separate commands so joined them then this error.

then, i modified the Makefile and removed -Wpedantic .. tried to make. still error

Desktop

• OS: [Arch Linux]
• Version [Latest]

I am new to all this.
Can you help me compile this?

When trying to build 0.4.3 with clang an error occurs (given we have -Werror):

./s4c-animate/animate.c:245:25: warning: a function declaration without a prototype is deprecated in all versions of C [-Wstrict-prototypes]
int s4c_check_has_colors() {
                        ^
                         void
./s4c-animate/animate.c:252:31: warning: a function declaration without a prototype is deprecated in all versions of C [-Wstrict-prototypes]
int s4c_check_can_change_color() {
                              ^
                               void
2 warnings generated.

This would make version check possible if format changes

ATM the new colors' components are weighted to a max of 680, but ncurses uses 1000 as its 255 component max.

This is due to standard colors being weighted to 680, but using 1000 could mean brighter shades are possible.