Bookshelf/umbra-bookshelf/tests/verify_structure.py

#!/usr/bin/env python3
"""Static verification for the merged umbra-bookshelf plugin.

Checks:
  1. PHP syntax of every file
  2. No class declared twice (fatal error in PHP)
  3. Every class referenced by new X / X:: actually exists
  4. Every require_once path resolves on disk
  5. Every wp_enqueue_* asset path resolves on disk (no 404s)
  6. No leftover pre-rename identifiers
  7. Preserved storage/API contracts still present
"""
import io
import os
import re
import subprocess
import sys

ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
failures = []
notes = []


def rel(p):
    return os.path.relpath(p, ROOT)


def walk(exts):
    for dirpath, _, files in os.walk(ROOT):
        for f in files:
            if f.endswith(exts):
                yield os.path.join(dirpath, f)


# ---------------------------------------------------------------- 1. php -l
php = [f for f in walk('.php')]
for f in php:
    r = subprocess.run(['php', '-l', f], capture_output=True, text=True)
    if 'No syntax errors' not in r.stdout:
        failures.append('PHP SYNTAX: %s\n%s' % (rel(f), r.stdout + r.stderr))
print('1. php -l: %d files checked' % len(php))

# ------------------------------------------------- 2. duplicate class decls
decls = {}
for f in php:
    src = io.open(f, encoding='utf-8', errors='replace').read()
    for m in re.finditer(r'^\s*(?:final\s+|abstract\s+)?class\s+([A-Za-z_][A-Za-z0-9_]*)',
                         src, re.M):
        decls.setdefault(m.group(1), []).append(rel(f))
dupes = {c: fs for c, fs in decls.items() if len(fs) > 1}
for c, fs in sorted(dupes.items()):
    failures.append('DUPLICATE CLASS: %s declared in %s' % (c, ', '.join(fs)))
print('2. class declarations: %d unique, %d duplicated'
      % (len(decls), len(dupes)))

# --------------------------------------------- 3. referenced classes exist
refs = set()
for f in php:
    src = io.open(f, encoding='utf-8', errors='replace').read()
    for m in re.finditer(r'\bnew\s+([A-Z][A-Za-z0-9_]*)\s*\(', src):
        refs.add(m.group(1))
    for m in re.finditer(r"\b([A-Z][A-Za-z0-9_]*)::", src):
        refs.add(m.group(1))

# names defined by this plugin, or provided by WordPress
wp_provided = {'BD_BookShelfHelper',  # created via class_alias()
               'WP_Error', 'WP_Query', 'WP_Post', 'WP_REST_Response',
               'WP_REST_Request', 'WP_REST_Server', 'WP_Term', 'wpdb',
               'ReflectionClass', 'DateTime', 'Exception', 'ArrayObject'}
known_external = {'GeminiClient',
                 # browser JS globals, matched by the `new X(` heuristic
                 # when scanning inline <script> blocks inside PHP files
                 'FormData', 'URL', 'Image', 'Audio'}
missing = sorted(r for r in refs
                 if r not in decls
                 and r not in wp_provided
                 and r not in known_external)
for r in missing:
    failures.append('UNDEFINED CLASS REFERENCED: %s' % r)
print('3. class references: %d referenced, %d undefined' % (len(refs), len(missing)))

# --------------------------------------------------- 4. require_once paths
for f in php:
    src = io.open(f, encoding='utf-8', errors='replace').read()
    for m in re.finditer(r"require(?:_once)?\s+([^;]+);", src):
        expr = m.group(1)
        pm = re.search(r"UMBRA_BOOKSHELF_DIR\s*\.\s*'([^']+)'", expr)
        if pm:
            target = os.path.join(ROOT, pm.group(1).lstrip('/'))
            if not os.path.isfile(target):
                failures.append('MISSING REQUIRE: %s -> %s' % (rel(f), pm.group(1)))
        elif "__DIR__" in expr:
            pm = re.search(r"__DIR__\s*\.\s*'([^']+)'", expr)
            if pm:
                target = os.path.join(os.path.dirname(f), pm.group(1).lstrip('/'))
                if not os.path.isfile(target):
                    failures.append('MISSING REQUIRE: %s -> %s'
                                    % (rel(f), pm.group(1)))
print('4. require_once targets verified')

# ------------------------------------------------------- 5. asset 404 check
assets = 0
for f in list(walk('.php')):
    src = io.open(f, encoding='utf-8', errors='replace').read()
    for m in re.finditer(
            r"UMBRA_BOOKSHELF_URL\s*\.\s*'([^']+\.(?:css|js|png|jpg|svg))'",
            src):
        target = os.path.join(ROOT, m.group(1))
        assets += 1
        if not os.path.isfile(target):
            failures.append('ASSET 404: %s -> %s' % (rel(f), m.group(1)))
print('5. enqueued assets: %d references, all resolved' % assets)

# -------------------------------------------- 6. leftover pre-rename tokens
stale_pat = re.compile(r'\b(?:BD_[A-Z][A-Za-z0-9_]*|BDM_[A-Za-z0-9_]*)\b')
stale = []
# Scan shipped plugin source only. Excluded on purpose:
#   spec/archive/ -- historical pre-merge specs, must keep the old names
#   tests/        -- asserts that the old names are GONE, so it mentions them
SCAN_DIRS = ('includes', 'admin', 'public', 'library', 'email')
scan_files = []
for d in SCAN_DIRS:
    for dirpath, _, files in os.walk(os.path.join(ROOT, d)):
        for f in files:
            if f.endswith(('.php', '.js', '.css')):
                scan_files.append(os.path.join(dirpath, f))
for f in ['umbra-bookshelf.php', 'uninstall.php']:
    scan_files.append(os.path.join(ROOT, f))
scan_files = [f for f in scan_files if os.path.isfile(f)]
for f in scan_files:
    src = io.open(f, encoding='utf-8', errors='replace').read()
    for i, line in enumerate(src.splitlines(), 1):
        for m in stale_pat.finditer(line):
            # BD_BookShelfHelper is intentionally kept as the legacy theme alias
            if m.group(0) == 'BD_BookShelfHelper':
                notes.append('%s:%d keeps legacy alias %s'
                             % (rel(f), i, m.group(0)))
                continue
            stale.append('%s:%d %s' % (rel(f), i, m.group(0)))
for s in sorted(set(stale)):
    failures.append('STALE IDENTIFIER: %s' % s)
print('6. stale identifiers: %d (excluding documented legacy alias)'
      % len(set(stale)))

# ------------------------------------- 7. preserved contracts still present
contracts = [
    ('bd_books', 'table name'),
    ('bd_series', 'table name'),
    ('bd_bookshelf_db_version', 'option key'),
    ('bd_bookshelf_activation_time', 'option key'),
    ('_bd_bookshelf_book_id', 'post meta key'),
    ('bd_bookshelf_import_posts', 'nonce action'),
    ('bd_bookshelf_import_csv', 'nonce/ajax action'),
    ('bd_bookshelf_sync_ratings', 'cron hook'),
    ('bd_bookshelf_goodreads_refresh', 'cron hook'),
    ('bd_bookshelf_amazon_scrape', 'cron hook'),
    ('bdm_update_links', 'ajax action'),
    ('bdm_autosave', 'ajax action'),
    ('bdm_ajax_signup', 'ajax action'),
    ('bd_bookshelf_consent', 'cookie name'),
]
blob = ''
for f in walk(('.php', '.js')):
    blob += io.open(f, encoding='utf-8', errors='replace').read()
for tok, kind in contracts:
    if tok not in blob:
        failures.append('LOST CONTRACT (%s): %s' % (kind, tok))
print('7. preserved contracts: %d checked' % len(contracts))

# ------------------------------------------------------------------- report
print()
for n in sorted(set(notes)):
    print('note: %s' % n)
print()
if failures:
    print('FAIL (%d)' % len(failures))
    for f in failures:
        print('  - %s' % f)
    sys.exit(1)
print('PASS - all static checks clean')