diff options
Diffstat (limited to 'webapp.py')
| -rwxr-xr-x | webapp.py | 127 |
1 files changed, 61 insertions, 66 deletions
@@ -1,12 +1,11 @@ #!/usr/bin/python -import contextlib import datetime import optparse -import sqlite3 from wsgiref.simple_server import make_server import jinja2 +import sqlalchemy from werkzeug.exceptions import HTTPException, NotFound from werkzeug.routing import Map, Rule, RequestRedirect from werkzeug.wrappers import Request, Response @@ -99,36 +98,35 @@ class Application(object): return e def get_details(self, package): - with contextlib.closing(self.db.cursor()) as cur: - cur.execute("SELECT id, version, architecture FROM package WHERE name = ?;", - (package,)) - row = cur.fetchone() + with self.db.begin() as conn: + row = conn.execute(sqlalchemy.text("SELECT id, version, architecture FROM package WHERE name = :name;"), + name=package).fetchone() if not row: raise NotFound() pid, version, architecture = row - details = dict(pid=pid, - package=package, - version=version, - architecture=architecture) - cur.execute("SELECT count(filename), sum(size) FROM content WHERE pid = ?;", - (pid,)) - num_files, total_size = cur.fetchone() + row = conn.execute(sqlalchemy.text("SELECT count(filename), sum(size) FROM content WHERE pid = :pid;"), + pid=pid).fetchone() + num_files, total_size = row if total_size is None: total_size = 0 - details.update(dict(num_files=num_files, total_size=total_size)) - return details + return dict(pid=pid, + package=package, + version=version, + architecture=architecture, + num_files=num_files, + total_size=total_size) def get_dependencies(self, pid): - with contextlib.closing(self.db.cursor()) as cur: - cur.execute("SELECT required FROM dependency WHERE pid = ?;", - (pid,)) + with self.db.begin() as conn: + cur = conn.execute(sqlalchemy.text("SELECT required FROM dependency WHERE pid = :pid;"), + pid=pid) return set(row[0] for row in fetchiter(cur)) def cached_sharedstats(self, pid): sharedstats = {} - with contextlib.closing(self.db.cursor()) as cur: - cur.execute("SELECT pid2, package.name, f1.name, f2.name, files, size FROM sharing JOIN package ON sharing.pid2 = package.id JOIN function AS f1 ON sharing.fid1 = f1.id JOIN function AS f2 ON sharing.fid2 = f2.id WHERE pid1 = ? AND f1.eqclass = f2.eqclass;", - (pid,)) + with self.db.begin() as conn: + cur = conn.execute(sqlalchemy.text("SELECT pid2, package.name, f1.name, f2.name, files, size FROM sharing JOIN package ON sharing.pid2 = package.id JOIN function AS f1 ON sharing.fid1 = f1.id JOIN function AS f2 ON sharing.fid2 = f2.id WHERE pid1 = :pid AND f1.eqclass = f2.eqclass;"), + pid=pid) for pid2, package2, func1, func2, files, size in fetchiter(cur): curstats = sharedstats.setdefault( function_combination(func1, func2), list()) @@ -143,11 +141,10 @@ class Application(object): params["dependencies"] = self.get_dependencies(params["pid"]) params["shared"] = self.cached_sharedstats(params["pid"]) params["urlroot"] = ".." - cur = self.db.cursor() - cur.execute("SELECT content.filename, issue.issue FROM content JOIN issue ON content.id = issue.cid WHERE content.pid = ?;", - (params["pid"],)) - params["issues"] = dict(cur.fetchall()) - cur.close() + with self.db.begin() as conn: + cur = conn.execute(sqlalchemy.text("SELECT content.filename, issue.issue FROM content JOIN issue ON content.id = issue.cid WHERE content.pid = :pid;"), + pid=params["pid"]) + params["issues"] = dict(cur.fetchall()) return html_response(package_template.render(params)) def compute_comparison(self, pid1, pid2): @@ -160,35 +157,32 @@ class Application(object): * matches: A mapping from filenames in package 2 (pid2) to a mapping from hash function pairs to hash values. """ - cur = self.db.cursor() - cur.execute("SELECT content.id, content.filename, content.size, hash.hash FROM content JOIN hash ON content.id = hash.cid JOIN duplicate ON content.id = duplicate.cid JOIN function ON hash.fid = function.id WHERE pid = ? AND function.name = 'sha512' ORDER BY size DESC;", - (pid1,)) - cursize = -1 - files = dict() - minmatch = 2 if pid1 == pid2 else 1 - cur2 = self.db.cursor() - for cid, filename, size, hashvalue in fetchiter(cur): - if cursize != size: - for entry in files.values(): - if len(entry["matches"]) >= minmatch: - yield entry - files.clear() - cursize = size - - if hashvalue in files: - files[hashvalue]["filenames"].add(filename) - continue - - entry = dict(filenames=set((filename,)), size=size, matches={}) - files[hashvalue] = entry - - cur2.execute("SELECT fa.name, ha.hash, fb.name, filename FROM hash AS ha JOIN hash AS hb ON ha.hash = hb.hash JOIN content ON hb.cid = content.id JOIN function AS fa ON ha.fid = fa.id JOIN function AS fb ON hb.fid = fb.id WHERE ha.cid = ? AND pid = ? AND fa.eqclass = fb.eqclass;", - (cid, pid2)) - for func1, hashvalue, func2, filename in fetchiter(cur2): - entry["matches"].setdefault(filename, {})[func1, func2] = \ - hashvalue - cur2.close() - cur.close() + with self.db.begin() as conn: + cur = conn.execute(sqlalchemy.text("SELECT content.id, content.filename, content.size, hash.hash FROM content JOIN hash ON content.id = hash.cid JOIN duplicate ON content.id = duplicate.cid JOIN function ON hash.fid = function.id WHERE pid = :pid AND function.name = 'sha512' ORDER BY size DESC;"), + pid=pid1) + cursize = -1 + files = dict() + minmatch = 2 if pid1 == pid2 else 1 + for cid, filename, size, hashvalue in fetchiter(cur): + if cursize != size: + for entry in files.values(): + if len(entry["matches"]) >= minmatch: + yield entry + files.clear() + cursize = size + + if hashvalue in files: + files[hashvalue]["filenames"].add(filename) + continue + + entry = dict(filenames=set((filename,)), size=size, matches={}) + files[hashvalue] = entry + + cur2 = conn.execute(sqlalchemy.text("SELECT fa.name, ha.hash, fb.name, filename FROM hash AS ha JOIN hash AS hb ON ha.hash = hb.hash JOIN content ON hb.cid = content.id JOIN function AS fa ON ha.fid = fa.id JOIN function AS fb ON hb.fid = fb.id WHERE ha.cid = :cid AND pid = :pid AND fa.eqclass = fb.eqclass;"), + cid=cid, pid=pid2) + for func1, hashvalue, func2, filename in fetchiter(cur2): + entry["matches"].setdefault(filename, {})[func1, func2] = \ + hashvalue for entry in files.values(): if len(entry["matches"]) >= minmatch: @@ -208,9 +202,9 @@ class Application(object): return html_response(detail_template.stream(params)) def show_hash(self, function, hashvalue): - with contextlib.closing(self.db.cursor()) as cur: - cur.execute("SELECT package.name, content.filename, content.size, f2.name FROM hash JOIN content ON hash.cid = content.id JOIN package ON content.pid = package.id JOIN function AS f2 ON hash.fid = f2.id JOIN function AS f1 ON f2.eqclass = f1.eqclass WHERE f1.name = ? AND hash = ?;", - (function, hashvalue,)) + with self.db.begin() as conn: + cur = conn.execute(sqlalchemy.text("SELECT package.name, content.filename, content.size, f2.name FROM hash JOIN content ON hash.cid = content.id JOIN package ON content.pid = package.id JOIN function AS f2 ON hash.fid = f2.id JOIN function AS f1 ON f2.eqclass = f1.eqclass WHERE f1.name = :function AND hash = :hashvalue;"), + function=function, hashvalue=hashvalue) entries = [dict(package=package, filename=filename, size=size, function=otherfunc) for package, filename, size, otherfunc in fetchiter(cur)] @@ -221,14 +215,14 @@ class Application(object): return html_response(hash_template.render(params)) def show_source(self, package): - with contextlib.closing(self.db.cursor()) as cur: - cur.execute("SELECT name FROM package WHERE source = ?;", - (package,)) + with self.db.begin() as conn: + cur = conn.execute(sqlalchemy.text("SELECT name FROM package WHERE source = :source;"), + source=package) binpkgs = dict.fromkeys(pkg for pkg, in fetchiter(cur)) if not binpkgs: raise NotFound - cur.execute("SELECT p1.name, p2.name, f1.name, f2.name, sharing.files, sharing.size FROM sharing JOIN package AS p1 ON sharing.pid1 = p1.id JOIN package AS p2 ON sharing.pid2 = p2.id JOIN function AS f1 ON sharing.fid1 = f1.id JOIN function AS f2 ON sharing.fid2 = f2.id WHERE p1.source = ?;", - (package,)) + cur = conn.execute(sqlalchemy.text("SELECT p1.name, p2.name, f1.name, f2.name, sharing.files, sharing.size FROM sharing JOIN package AS p1 ON sharing.pid1 = p1.id JOIN package AS p2 ON sharing.pid2 = p2.id JOIN function AS f1 ON sharing.fid1 = f1.id JOIN function AS f2 ON sharing.fid2 = f2.id WHERE p1.source = :source;"), + source=package) for binary, otherbin, func1, func2, files, size in fetchiter(cur): entry = dict(package=otherbin, funccomb=function_combination(func1, func2), @@ -242,10 +236,11 @@ class Application(object): def main(): parser = optparse.OptionParser() parser.add_option("-d", "--database", action="store", - default="test.sqlite3", - help="path to the sqlite3 database file") + default="sqlite:///test.sqlite3", + help="location of the database") options, args = parser.parse_args() - app = Application(sqlite3.connect(options.database)) + db = sqlalchemy.create_engine(options.database) + app = Application(db) app = SharedDataMiddleware(app, {"/static": ("dedup", "static")}) make_server("0.0.0.0", 8800, app).serve_forever() |