allucanget/jobs
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

simplify caching

Browse Source
This commit is contained in:
georg.sinn-schirwitz
2025-08-29 16:31:10 +02:00
parent d5f686bf1e
commit 2dac771d47
4 changed files with 102 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

66
tests/test_cache_paths.py Normal file
View File

@@ -0,0 +1,66 @@
import os
import tempfile
import pytest
import web.db as db
from web.utils import get_cache_dir
# Skip unless explicitly enabled (MySQL integration expected)
if not os.getenv("RUN_DB_TESTS"):
pytest.skip("Set RUN_DB_TESTS=1 to run cache path integration tests",
allow_module_level=True)
def test_db_sync_inserts_relative_paths(tmp_path, monkeypatch):
# arrange: create a temporary cache dir and a fake html file
cache_dir = tmp_path / "cache"
cache_dir.mkdir()
f = cache_dir / "example.org_path_to_page_123.html"
f.write_text("<html>ok</html>")
# point app at this cache dir
monkeypatch.setenv("PYTEST_CACHE_DIR", str(cache_dir))
# monkeypatch get_cache_dir used by db functions
monkeypatch.setattr('web.utils.get_cache_dir', lambda: str(cache_dir))
# ensure DB initialized
db.db_init()
# act
db.db_sync_cached_pages(str(cache_dir))
# assert: DB contains relative path, not absolute
rows = db.db_get_all_cached_pages()
assert any(r['file_path'] == os.path.relpath(
str(f), start=str(cache_dir)) for r in rows)
def test_normalize_cached_page_paths_converts_absolute(tmp_path, monkeypatch):
cache_dir = tmp_path / "cache"
cache_dir.mkdir()
# create an actual file
f = cache_dir / "site_example_page_1.html"
f.write_text("<html>ok</html>")
monkeypatch.setattr('web.utils.get_cache_dir', lambda: str(cache_dir))
db.db_init()
abs_fp = str(f)
rel_fp = os.path.relpath(abs_fp, start=str(cache_dir))
# Insert an absolute path row directly (simulate legacy data)
with db._ensure_session() as session:
session.execute(
db.text("INSERT INTO cached_pages(file_path, url_guess, last_modified, size_bytes, job_id) VALUES(:fp, :ug, :lm, :sz, :jid)"),
{"fp": abs_fp, "ug": "https://example.org/page1.html",
"lm": None, "sz": 10, "jid": None}
)
session.commit()
# normalize should convert absolute to relative
changed = db.normalize_cached_page_paths()
assert changed >= 1
rows = db.db_get_all_cached_pages()
assert any(r['file_path'] == rel_fp for r in rows)

1
web/app.py
View File

@@ -255,7 +255,6 @@ csrf.exempt(set_favorite)
@app.route('/scrape', methods=['GET']) @app.route('/scrape', methods=['GET'])
def scrape(): def scrape():
"""Trigger the web scraping process.""" """Trigger the web scraping process."""
# Run the full scraper orchestration (fetch listings, sync cache, process jobs)
scraper() scraper()
return jsonify({"status": "Scraping completed"}) return jsonify({"status": "Scraping completed"})

51
web/db.py
View File

@@ -24,6 +24,7 @@ from web.utils import (
normalize_job_id, normalize_job_id,
now_iso, now_iso,
get_cache_path, get_cache_path,
get_cache_dir,
get_mysql_config, get_mysql_config,
) )
@@ -280,12 +281,14 @@ def upsert_job_details(job_data: Dict[str, Any]):
def upsert_cached_page(*, file_path: str, url_guess: Optional[str], last_modified: Optional[str], size_bytes: Optional[int], job_id: Optional[int]): def upsert_cached_page(*, file_path: str, url_guess: Optional[str], last_modified: Optional[str], size_bytes: Optional[int], job_id: Optional[int]):
# Always store absolute paths # Store file paths relative to the cache directory (keeps DB portable)
cache_dir = os.path.abspath(get_cache_dir())
abs_fp = os.path.abspath(file_path) abs_fp = os.path.abspath(file_path)
rel_fp = os.path.relpath(abs_fp, start=cache_dir)
with _ensure_session() as session: with _ensure_session() as session:
obj = session.get(CachedPage, abs_fp) obj = session.get(CachedPage, rel_fp)
if obj is None: if obj is None:
obj = CachedPage(file_path=abs_fp) obj = CachedPage(file_path=rel_fp)
session.add(obj) session.add(obj)
setattr(obj, "url_guess", url_guess) setattr(obj, "url_guess", url_guess)
setattr(obj, "last_modified", last_modified) setattr(obj, "last_modified", last_modified)
@@ -295,10 +298,12 @@ def upsert_cached_page(*, file_path: str, url_guess: Optional[str], last_modifie
def remove_cached_page(file_path: str): def remove_cached_page(file_path: str):
# Accept either relative or absolute; remove both variants just in case # Accept absolute or relative input but DB keys are stored relative to cache dir
cache_dir = os.path.abspath(get_cache_dir())
abs_fp = os.path.abspath(file_path) abs_fp = os.path.abspath(file_path)
rel_fp = os.path.relpath(abs_fp, start=cache_dir)
with _ensure_session() as session: with _ensure_session() as session:
obj = session.get(CachedPage, abs_fp) obj = session.get(CachedPage, rel_fp)
if obj: if obj:
session.delete(obj) session.delete(obj)
session.commit() session.commit()
@@ -306,7 +311,8 @@ def remove_cached_page(file_path: str):
def db_remove_cached_url(url: str): def db_remove_cached_url(url: str):
"""Remove a cached page by URL.""" """Remove a cached page by URL."""
abs_fp = get_cache_path(url) # Compute absolute path for the URL and delegate to remove_cached_page
abs_fp = os.path.abspath(get_cache_path(url))
try: try:
remove_cached_page(abs_fp) remove_cached_page(abs_fp)
except Exception: except Exception:
@@ -353,15 +359,17 @@ def db_sync_cached_pages(cache_dir: str):
"""Scan cache_dir and upsert page metadata into cached_pages table.""" """Scan cache_dir and upsert page metadata into cached_pages table."""
if not os.path.isdir(cache_dir): if not os.path.isdir(cache_dir):
return return
db_cache = db_get_all_cached_pages() abs_cache = os.path.abspath(cache_dir)
for root, _, files in os.walk(cache_dir): # read existing DB keys once for quick membership tests
db_cache_paths = {c["file_path"] for c in db_get_all_cached_pages()}
for root, _, files in os.walk(abs_cache):
for name in files: for name in files:
if not name.lower().endswith(".html"): if not name.lower().endswith(".html"):
continue continue
fp = os.path.abspath(os.path.join(root, name)) fp = os.path.abspath(os.path.join(root, name))
if fp in [c["file_path"] for c in db_cache]: rel_fp = os.path.relpath(fp, start=abs_cache)
if rel_fp in db_cache_paths:
continue continue
try: try:
stat = os.stat(fp) stat = os.stat(fp)
mtime = datetime.fromtimestamp(stat.st_mtime).isoformat() mtime = datetime.fromtimestamp(stat.st_mtime).isoformat()
@@ -371,22 +379,23 @@ def db_sync_cached_pages(cache_dir: str):
size = None size = None
url_guess = get_url_from_filename(name) url_guess = get_url_from_filename(name)
job_id = url_to_job_id(url_guess) job_id = url_to_job_id(url_guess)
upsert_cached_page(file_path=fp, url_guess=url_guess, upsert_cached_page(file_path=rel_fp, url_guess=url_guess,
last_modified=mtime, size_bytes=size, job_id=job_id) last_modified=mtime, size_bytes=size, job_id=job_id)
def normalize_cached_page_paths() -> int: def normalize_cached_page_paths() -> int:
"""Ensure all cached_pages.file_path values are absolute. Returns number of rows updated/normalized.""" """Ensure all cached_pages.file_path values are absolute. Returns number of rows updated/normalized."""
# Convert any absolute paths in DB to relative paths (relative to cache dir)
changed = 0 changed = 0
abs_cache = os.path.abspath(get_cache_dir())
with _ensure_session() as session: with _ensure_session() as session:
rows = session.execute(text( rows = session.execute(text(
"SELECT file_path, url_guess, last_modified, size_bytes, job_id FROM cached_pages")).fetchall() "SELECT file_path, url_guess, last_modified, size_bytes, job_id FROM cached_pages")).fetchall()
for (fp, url_guess, last_modified, size_bytes, job_id) in rows: for (fp, url_guess, last_modified, size_bytes, job_id) in rows:
if not os.path.isabs(fp): if os.path.isabs(fp):
abs_fp = os.path.abspath(fp) rel_fp = os.path.relpath(fp, start=abs_cache)
# Upsert under absolute path, then remove the relative entry
upsert_cached_page( upsert_cached_page(
file_path=abs_fp, file_path=rel_fp,
url_guess=url_guess, url_guess=url_guess,
last_modified=last_modified, last_modified=last_modified,
size_bytes=size_bytes, size_bytes=size_bytes,
@@ -454,7 +463,9 @@ ORDER BY d.posted_time DESC
"timestamp": row[7], "timestamp": row[7],
"posted_time": row[8], "posted_time": row[8],
"url": row[9], "url": row[9],
# file_path is stored relative to cache dir; provide both forms
"file_path": row[10], "file_path": row[10],
"file_path_abs": os.path.join(os.path.abspath(get_cache_dir()), row[10]) if row[10] else None,
"last_modified": row[11], "last_modified": row[11],
"url_guess": row[12], "url_guess": row[12],
"url_guess_stale": row[13], "url_guess_stale": row[13],
@@ -463,6 +474,16 @@ ORDER BY d.posted_time DESC
return jobs return jobs
def db_get_cached_abs_path(db_file_path: Optional[str]) -> Optional[str]:
"""Return absolute cache file path given a DB-stored (relative) file_path.
Returns None if input is falsy.
"""
if not db_file_path:
return None
return os.path.join(os.path.abspath(get_cache_dir()), db_file_path)
def db_get_all_job_urls() -> List[str]: def db_get_all_job_urls() -> List[str]:
"""Return list of job URLs from job_listings.""" """Return list of job URLs from job_listings."""
with _ensure_session() as session: with _ensure_session() as session:

1
web/utils.py
View File

@@ -203,7 +203,6 @@ def get_cache_path(url: str) -> str:
def cache_page(url: str, content: str): def cache_page(url: str, content: str):
"""Cache the page content with a timestamp.""" """Cache the page content with a timestamp."""
cache_path = get_cache_path(url) cache_path = get_cache_path(url)
os.makedirs(os.path.dirname(cache_path), exist_ok=True)
with open(cache_path, "w", encoding="utf-8") as f: with open(cache_path, "w", encoding="utf-8") as f:
f.write(content) f.write(content)
# Update the file's modification time to the current time # Update the file's modification time to the current time