#!/usr/bin/env python3

"""

Creates a clean, distributable .zip of the CppMode folder for a release.

Includes everything in CppMode/ EXCEPT the contents of cache/ subfolders

(cache/linux-x64, cache/windows-x64, etc.) -- those are machine-specific

compiled build artifacts (Processing.o and similar), not something that

belongs in a release archive; they get rebuilt fresh on whatever machine

actually uses the release.

Usage:

create-release.py [version]

If no version is given, the zip is named CppMode.zip; otherwise it's

named CppMode-<version>.zip (e.g. CppMode-0.2.0.zip).

"""

import os

import sys

import zipfile

# This script lives in CppMode/scripts/, so CppMode's own root is one

# directory up from wherever this script actually is.

CPPMODE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))

OUTPUT_DIR = os.path.dirname(CPPMODE_DIR) # one level above CppMode/ itself

def should_skip_file(path):

"""

Returns True for .o files inside cache/<platform>/, or anything

inside .git/ or .github/. The cache/<platform>/ DIRECTORIES

themselves are kept (so the release has the right folder layout

ready for a fresh rebuild), just not the compiled .o artifacts

sitting inside them -- those are machine-specific and get rebuilt

fresh wherever the release is actually used.

"""

rel = os.path.relpath(path, CPPMODE_DIR)

parts = rel.split(os.sep)

if len(parts) >= 1 and parts[0] in (".git", ".github"):

return True

if False: # .o files are included in release (downloaded from CI)

return True

return False

def should_prune_dir(path):

"""Directories to not even descend into at all (.git, .github)."""

rel = os.path.relpath(path, CPPMODE_DIR)

parts = rel.split(os.sep)

return len(parts) >= 1 and parts[0] in (".git", ".github")

def bump_version_files(tag):

"""

Updates mode.properties and CppMode.txt to match the given release

tag: sets prettyVersion/prettyversion to the tag exactly as given,

and increments the plain integer "version=" build number by 1 in

both files. Both files get updated together so they never drift out

of sync with each other.

"""

targets = [

(os.path.join(CPPMODE_DIR, "mode.properties"), "prettyVersion"),

(os.path.join(CPPMODE_DIR, "CppMode.txt"), "prettyVersion"),

]

for path, pretty_key in targets:

if not os.path.exists(path):

print(f"WARNING: {path} not found, skipping version bump for it.")

continue

lines = open(path).readlines()

new_lines = []

for line in lines:

if line.startswith("version="):

try:

current = int(line.strip().split("=", 1)[1])

except ValueError:

current = 0

new_lines.append(f"version={current + 1}\n")

elif line.startswith(pretty_key + "="):

new_lines.append(f"{pretty_key}={tag}\n")

else:

new_lines.append(line)

with open(path, "w") as out:

out.writelines(new_lines)

print(f"Updated {path}: {pretty_key}={tag}, version bumped by 1")

def trigger_and_wait_cache_build():

"""Trigger a new build-cache CI run and wait for it to complete."""

import subprocess, json, time

repo = "processing-cpp/processing.cpp"

print("Triggering build-cache CI run...")

r = subprocess.run(

["gh", "workflow", "run", "build-cache.yml", "--repo", repo],

capture_output=True, text=True

)

if r.returncode != 0:

print(f" WARNING: Could not trigger build-cache: {r.stderr.strip()}")

return None

time.sleep(5) # wait for run to register

r2 = subprocess.run(

["gh", "run", "list", "--repo", repo, "--workflow", "build-cache.yml",

"--limit", "1", "--json", "databaseId,status"],

capture_output=True, text=True

)

if r2.returncode != 0: return None

run_id = json.loads(r2.stdout)[0]["databaseId"]

print(f" Waiting for run {run_id} to complete...")

while True:

r3 = subprocess.run(

["gh", "run", "view", str(run_id), "--repo", repo, "--json", "status,conclusion"],

capture_output=True, text=True

)

data = json.loads(r3.stdout)

if data["status"] == "completed":

print(f" Run {run_id} completed with: {data['conclusion']}")

return run_id if data["conclusion"] == "success" else None

print(f" Status: {data['status']}... waiting 15s")

time.sleep(15)

def download_cache_artifacts():

"""Download precompiled .o files from latest successful build-cache CI run."""

import subprocess, json

from pathlib import Path

repo = "processing-cpp/processing.cpp"

root = Path(__file__).parent.parent

platforms = ["linux-x64", "windows-x64", "macos-arm64", "macos-x64"]

print("Downloading precompiled cache from CI artifacts...")

# Find the latest successful build-cache run

r = subprocess.run(

["gh", "run", "list", "--repo", repo, "--workflow", "build-cache.yml",

"--status", "success", "--limit", "1", "--json", "databaseId"],

capture_output=True, text=True

)

if r.returncode != 0 or not r.stdout.strip():

print(" WARNING: Could not find successful build-cache run")

return

run_id = json.loads(r.stdout)[0]["databaseId"]

print(f" Using run {run_id}")

for platform in platforms:

dest = root / "cache" / platform

# Clear existing files so gh run download can overwrite

import shutil

if dest.exists(): shutil.rmtree(dest)

dest.mkdir(parents=True, exist_ok=True)

result = subprocess.run(

["gh", "run", "download", str(run_id), "--repo", repo,

"--name", f"cache-{platform}", "--dir", str(dest)],

capture_output=True, text=True

)

if result.returncode != 0:

print(f" WARNING: {platform}: {result.stderr.strip()[:100]}")

else:

files = [f.name for f in dest.iterdir()]

print(f" {platform}: {files}")

def trigger_and_wait_cache_build():

"""Trigger a new build-cache CI run and wait for it to complete."""

import subprocess, json, time

repo = "processing-cpp/processing.cpp"

print("Triggering build-cache CI run...")

r = subprocess.run(

["gh", "workflow", "run", "build-cache.yml", "--repo", repo],

capture_output=True, text=True

)

if r.returncode != 0:

print(f" WARNING: Could not trigger build-cache: {r.stderr.strip()}")

return

time.sleep(8)

r2 = subprocess.run(

["gh", "run", "list", "--repo", repo, "--workflow", "build-cache.yml",

"--limit", "1", "--json", "databaseId"],

capture_output=True, text=True

)

run_id = json.loads(r2.stdout)[0]["databaseId"]

print(f" Waiting for run {run_id}...")

while True:

r3 = subprocess.run(

["gh", "run", "view", str(run_id), "--repo", repo, "--json", "status,conclusion"],

capture_output=True, text=True

)

data = json.loads(r3.stdout)

if data["status"] == "completed":

print(f" Done: {data['conclusion']}")

return

print(f" {data['status']}... waiting 15s")

time.sleep(15)

def main():

version = sys.argv[1] if len(sys.argv) > 1 else None

if version:

bump_version_files(version)

trigger_and_wait_cache_build()

download_cache_artifacts()

zip_name = "CppMode.zip"

zip_path = os.path.join(OUTPUT_DIR, zip_name)

if os.path.exists(zip_path):

os.remove(zip_path)

included = 0

skipped = 0

with zipfile.ZipFile(zip_path, "w", zipfile.ZIP_DEFLATED) as zf:

for root, dirs, files in os.walk(CPPMODE_DIR):

# Prune .git/.github entirely -- never even descend into

# them. cache/<platform> directories are walked normally

# (their FOLDER STRUCTURE is kept), only individual .o

# files inside them get filtered out below.

dirs[:] = [

d for d in dirs

if not should_prune_dir(os.path.join(root, d))

]

# If this directory ends up with no files going into the

# zip -- either because it was ALREADY empty (e.g. a fresh

# cache/windows-x64/ that's never been built into yet) or

# BECAME empty after filtering out its .o files -- add an

# explicit directory entry so the folder still exists after

# extraction. Zip files don't store empty directories

# unless told to, and this needs to cover both cases, not

# just the "had files, now has none" case.

rel_dir = os.path.relpath(root, CPPMODE_DIR)

kept_any_file_here = any(

not should_skip_file(os.path.join(root, f)) for f in files

)

if not kept_any_file_here:

dir_arcname = os.path.join("CppMode", rel_dir) + "/"

zf.write(root, dir_arcname)

for fname in files:

full_path = os.path.join(root, fname)

if should_skip_file(full_path):

skipped += 1

continue

# Store paths inside the zip relative to CppMode's own

# parent, so extracting the zip recreates a top-level

# "CppMode/" folder rather than dumping files loose.

arcname = os.path.join(

"CppMode", os.path.relpath(full_path, CPPMODE_DIR)

)

zf.write(full_path, arcname)

included += 1

size_mb = os.path.getsize(zip_path) / (1024 * 1024)

print(f"Created {zip_path} ({size_mb:.2f} MB)")

print(f" {included} files included, {skipped} cache files skipped")

if __name__ == "__main__":

main()