shopdb-flask/scripts/check-naming-and-style.sh

#!/usr/bin/env bash
#
# Pre-commit naming + style check for shopdb-flask.
# Enforces CONTRIBUTING.md rules:
#   1. No non-ASCII chars in source (em-dashes, smart quotes, arrows, emojis)
#   2. No banned shorthand identifiers (cfg, ctx, mgr, req, res, env, util, helper)
#      as standalone names (suffix usage like printers_bp, request_obj is allowed)
#   3. No snake_case DB column names in __tablename__ or db.Column attrs
#   4. No snake_case API params in frontend that should match DB column names
#
# Exits non-zero if any violation found.
# Skips: venv/, node_modules/, __pycache__/, frontend/dist/, migrations/versions/,
#        deploy/windows/installer/bundle/ (installer build output)

set -e

REPO_ROOT="$(git rev-parse --show-toplevel 2>/dev/null || echo .)"
cd "$REPO_ROOT"

VIOLATIONS=0

EXCLUDES=(
    --exclude-dir=venv
    --exclude-dir=env
    --exclude-dir=.venv
    --exclude-dir=node_modules
    --exclude-dir=__pycache__
    --exclude-dir=dist
    # The installer's build output: a staged copy of the whole application plus
    # a second SPA build under dist-subpath, which --exclude-dir=dist does not
    # match. Linting it means linting vendored minified JS and failing on
    # characters nobody in this repository wrote.
    --exclude-dir=bundle
    --exclude-dir=dist-subpath
    --exclude-dir=.git
    --exclude-dir=versions
    --exclude-dir=staticdocs
)

INCLUDES_CODE=(
    --include='*.py'
    --include='*.vue'
    --include='*.js'
    --include='*.ts'
)

INCLUDES_ALL=(
    --include='*.py'
    --include='*.vue'
    --include='*.js'
    --include='*.ts'
    --include='*.json'
    --include='*.md'
    --include='*.yaml'
    --include='*.yml'
)

# INCLUDES_ALL, not INCLUDES_CODE: the check only ever covered .py/.vue/.js/.ts,
# so documentation was free to accumulate em-dashes, arrows and smart quotes -
# and did, including in files this repo's own convention forbids them in.
# Markdown, JSON and YAML are now covered too.
echo "==> Checking for non-ASCII characters..."
NON_ASCII=$(grep -rPn '[^\x00-\x7F]' "${EXCLUDES[@]}" "${INCLUDES_ALL[@]}" . 2>/dev/null || true)
if [ -n "$NON_ASCII" ]; then
    echo "FAIL: non-ASCII characters found (em-dashes, smart quotes, arrows, emojis):"
    echo "$NON_ASCII"
    echo
    VIOLATIONS=$((VIOLATIONS + 1))
fi

echo "==> Checking for banned shorthand (standalone)..."
# Match the word as a standalone identifier: not preceded or followed by underscore/word char
# Word boundary in grep is \b but we want to exclude suffix usage like printers_bp
# So: match (^|[^a-zA-Z0-9_])(banned)([^a-zA-Z0-9_]|$)
for word in cfg ctx mgr req res; do
    HITS=$(grep -rPn "(^|[^a-zA-Z0-9_])${word}([^a-zA-Z0-9_]|\$)" "${EXCLUDES[@]}" --include='*.py' --include='*.vue' --include='*.js' --include='*.ts' . 2>/dev/null \
        | grep -vP "(^|[^a-zA-Z0-9_])(request_obj|response_obj)" \
        || true)
    if [ -n "$HITS" ]; then
        echo "FAIL: banned shorthand '$word' (standalone) found:"
        echo "$HITS"
        echo
        VIOLATIONS=$((VIOLATIONS + 1))
    fi
done

echo "==> Checking for snake_case DB tablenames..."
SNAKE_TABLES=$(grep -rPn "__tablename__\s*=\s*['\"][^'\"]*_" "${EXCLUDES[@]}" --include='*.py' . 2>/dev/null || true)
if [ -n "$SNAKE_TABLES" ]; then
    echo "FAIL: snake_case __tablename__ found (must be lowercase concatenated):"
    echo "$SNAKE_TABLES"
    echo
    VIOLATIONS=$((VIOLATIONS + 1))
fi

echo "==> Checking for snake_case DB column attrs..."
SNAKE_COLS=$(grep -rPn "^\s+[a-z]+_[a-z_]+\s*=\s*db\.Column" "${EXCLUDES[@]}" --include='*.py' . 2>/dev/null || true)
if [ -n "$SNAKE_COLS" ]; then
    echo "FAIL: snake_case db.Column attribute found (must match column name, no underscores):"
    echo "$SNAKE_COLS"
    echo
    VIOLATIONS=$((VIOLATIONS + 1))
fi

echo "==> Checking for snake_case ForeignKey targets..."
SNAKE_FK=$(grep -rPn "ForeignKey\(['\"][^'\"]*_[^'\"]*['\"]" "${EXCLUDES[@]}" --include='*.py' . 2>/dev/null || true)
if [ -n "$SNAKE_FK" ]; then
    echo "FAIL: snake_case ForeignKey target found:"
    echo "$SNAKE_FK"
    echo
    VIOLATIONS=$((VIOLATIONS + 1))
fi

echo "==> Checking for snake_case API params in frontend (DB-mirrored fields)..."
SNAKE_FE=$(grep -rPn "params\.(machine_id|location_id|vendor_id|type_id|business_unit_id|model_id|status_id|operating_system_id|asset_id|user_id|is_active|is_shopfloor)" "${EXCLUDES[@]}" --include='*.vue' --include='*.js' --include='*.ts' . 2>/dev/null || true)
if [ -n "$SNAKE_FE" ]; then
    echo "FAIL: snake_case API params in frontend (must match DB column names without underscores):"
    echo "$SNAKE_FE"
    echo
    VIOLATIONS=$((VIOLATIONS + 1))
fi

# ADR-013 Phase 4: a plugin frontend (plugins/<name>/frontend/) may import core
# via the @/ alias or its own tree (./, ../ within the plugin), never another
# plugin's tree and never an escaping ../../ into src. Keeps plugin frontends
# self-contained so a per-site build can drop one cleanly.
echo "==> Checking for cross-plugin / escaping imports in plugin frontends..."
if [ -d plugins ]; then
    PLUGIN_FE_IMPORTS=$(grep -rPn "(import|from)\s+['\"]([^'\"]*\.\./\.\./|[^'\"]*/plugins/)" \
        --include='*.vue' --include='*.js' plugins/*/frontend/ 2>/dev/null || true)
    if [ -n "$PLUGIN_FE_IMPORTS" ]; then
        echo "FAIL: plugin frontend imports must use the @/ alias for core, not"
        echo "      an escaping ../../ or another plugin's path:"
        echo "$PLUGIN_FE_IMPORTS"
        echo
        VIOLATIONS=$((VIOLATIONS + 1))
    fi
fi

# ADR-015: one site's data does not belong in product code. A site host, a site
# FQDN or a site network in plugins/ or shopdb/ ships another site a value it
# cannot see and did not choose. Use a setting with a NEUTRAL default, a
# site-namespaced directory (scripts/site_imports/<site>/), or seed data.
#
# Code generation lives in one module. Seven views were importing qrcode and
# jsbarcode directly, each with its own margin, width and error correction, and
# the answers had already drifted - which on a label means one page scans and
# another does not. frontend/src/utils/codes.js owns that knowledge now; a view
# passes what is specific to its label and nothing else.
echo "==> Checking that only the shared module imports the code libraries..."
CODE_LIB_IMPORTS=$(grep -rn "from ['\"]\(qrcode\|jsbarcode\)['\"]" \
    --include='*.vue' --include='*.js' \
    frontend/src plugins/ 2>/dev/null \
    | grep -v 'frontend/src/utils/codes\.js' \
    | grep -v '\.plugins-staged/' || true)
if [ -n "$CODE_LIB_IMPORTS" ]; then
    echo "FAIL: import qrcode/jsbarcode through @/utils/codes, not directly:"
    echo "$CODE_LIB_IMPORTS"
    echo
    VIOLATIONS=$((VIOLATIONS + 1))
fi

# ADR-017: a marker position is pixels in ONE LEVEL's coordinate space, so a
# payload that emits mapx/mapy without levelid gives the consumer coordinates and
# no drawing to put them on. The consumer then either renders nothing or - worse,
# and what a reasonable implementation does - falls back to the default level,
# drawing one building's ground floor behind a marker positioned for another
# building's mezzanine. That renders perfectly and points at the wrong place.
#
# Eight files emit positions across twenty sites. This is the check that says
# whether all of them travel with their level, because reading them by eye is
# how the twentieth gets missed.
echo "==> Checking that emitted map positions carry their level (ADR-017)..."
# PER OCCURRENCE, not per file. The file-level form passed a module that emitted
# 'mapx' six times and 'levelid' once, and two payloads shipped without a level:
# the toner report and the enforcement report, both feeding a hover preview that
# then said "no level". Every 'mapx' must have a 'levelid' in the same literal -
# the window is wide enough for a comment between them, and no wider.
MISSING_LEVEL=""
while IFS= read -r hit; do
    [ -z "$hit" ] && continue
    file=${hit%%:*}
    line=${hit#*:}; line=${line%%:*}
    if ! sed -n "${line},$((line + 8))p" "$file" | grep -q "'levelid'"; then
        MISSING_LEVEL="$MISSING_LEVEL$file:$line"$'\n'
    fi
done <<EOF
$(grep -rn "'mapx':" --include='*.py' shopdb/ plugins/ scripts/ 2>/dev/null | grep -v '/tests\?/' || true)
EOF
if [ -n "$MISSING_LEVEL" ]; then
    echo "FAIL: these emit 'mapx' with no 'levelid' beside it - a position with no"
    echo "      level cannot be drawn on the right floor plan (ADR-017):"
    echo "$MISSING_LEVEL" | sed 's/^/       /'
    VIOLATIONS=$((VIOLATIONS + 1))
fi

# The same rule for the hover preview: binding coordinates into
# LocationMapTooltip without a level makes it report "no level" for every asset,
# which is exactly what shipped in 0.11.0 - all seven call sites missed it.
TOOLTIP_MISSING=""
for candidate in $(grep -rl "LocationMapTooltip" --include='*.vue' frontend/src plugins/ 2>/dev/null | grep -v plugins-staged || true); do
    grep -q ':left=' "$candidate" || continue
    grep -q ':levelid=' "$candidate" || TOOLTIP_MISSING="$TOOLTIP_MISSING$candidate"$'\n'
done
if [ -n "$TOOLTIP_MISSING" ]; then
    echo "FAIL: these bind LocationMapTooltip coordinates without :levelid, so the"
    echo "      preview cannot know which drawing to use (ADR-017):"
    echo "$TOOLTIP_MISSING" | sed 's/^/       /'
    VIOLATIONS=$((VIOLATIONS + 1))
fi

# ENFORCING. It was report-only while the backlog was worked off, and the hit
# count then did not move for weeks - a rule that only prints is read as no rule.
# Set SITE_LITERALS_ENFORCE=0 to drop back to reporting for a local run.
SITE_LITERALS_ENFORCE=${SITE_LITERALS_ENFORCE:-1}

echo "==> Checking for site-specific literals in product code (ADR-015)..."
# A real site hostname, a site FQDN, a site name or a site network.
#
# SCOPE. It used to look at .py/.vue/.js under plugins/ and shopdb/ only, and
# every literal that actually reached a second site was somewhere else: the
# PowerShell clients, the installer, the seeds, generated JSON. Case-sensitive
# too, so a capitalised spelling of the host passed. Both fixed - the scan is
# only worth having where
# the leaks are.
#
# Fleet-wide vocabulary (gea-shopfloor-*) is NOT matched: it is overridable
# through the pctypemap settings and is not one site's data.
#
# A line may declare itself deliberate with a trailing `ADR-015-OK: <reason>`
# marker. That is for an organisation-wide default that is genuinely right for
# every site and configurable anyway - not for "we will fix it later". The
# marker makes the claim visible in review; silence would not.
# ASSEMBLED FROM FRAGMENTS, deliberately. This script is published, and the
# publication scrub greps for the very strings below - written literally, the
# rule's own definition trips the gate that enforces it. Same reason
# tests/test_docs_publishable.py splits its terms.
# Adjacent quoted strings CONCATENATE: 'ab''cd' is the four-character string
# abcd. The apostrophe-escape form ('ab'\''cd') looks almost identical and is
# not - it embeds a literal ' in the middle, so every pattern below matched
# nothing and this check silently passed everything from the day it was written
# until 2026-08-26. That is how a QA SaaS tenant sat in the shipped defaults for
# months with the gate green. Keep the fragments split (the scrub greps this
# file for the very strings the patterns are made of) and keep them ADJACENT.
SITE_HOST='tsg''wp00525'
SITE_DOMAIN='\.geaero''space\.net'
SITE_NAME='West Jeff''erson'
SITE_NETS='10\.134\.48\.|10\.48\.249\.'
# A named SaaS TENANT is as site-specific as a hostname, and this one shipped as
# the default ServiceNow URL for months: every adopting site would have linked
# its technicians into another organisation's QA instance. The vendor domain
# itself is fine to mention; the tenant is not.
SITE_TENANTS='geaero''spaceqa'
SITE_PATTERNS="${SITE_HOST}|${SITE_DOMAIN}|\bwjs\b|${SITE_NAME}|${SITE_NETS}|${SITE_TENANTS}"
SITE_HITS=$(grep -rPni "$SITE_PATTERNS" "${EXCLUDES[@]}" \
    --include='*.py' --include='*.vue' --include='*.js' \
    --include='*.ps1' --include='*.psm1' --include='*.sh' --include='*.iss' \
    --include='*.json' --include='*.html' \
    plugins/ shopdb/ scripts/ deploy/ frontend/src/ tools/ 2>/dev/null \
    | grep -v '/tests\?/' \
    | grep -v '\.spec\.js:' \
    | grep -v 'site_imports/' \
    | grep -v 'installer/bundle/' \
    | grep -v '\.plugins-staged/' \
    | grep -v 'check-naming-and-style\.sh' \
    | grep -v 'export-github\.sh' \
    | grep -v 'ADR-015-OK' || true)
if [ -n "$SITE_HITS" ]; then
    COUNT=$(echo "$SITE_HITS" | wc -l)
    if [ "$SITE_LITERALS_ENFORCE" = "1" ]; then
        echo "FAIL: $COUNT site-specific literal(s) in product code:"
        echo "$SITE_HITS"
        echo
        VIOLATIONS=$((VIOLATIONS + 1))
    else
        echo "     $COUNT site-specific literal(s) found (report only, not failing):"
        echo "$SITE_HITS" | sed 's/^/       /'
        echo "     See docs/adr/ADR-015-site-specific-configuration.md"
        echo
    fi
fi

if [ "$VIOLATIONS" -gt 0 ]; then
    echo "=================================================="
    echo "$VIOLATIONS naming/style violation(s) found."
    echo "See CONTRIBUTING.md for the full convention."
    echo "=================================================="
    exit 1
fi

echo "==> All naming/style checks passed."
exit 0