2017-11-10 19:25:54 +00:00
|
|
|
from contextlib import contextmanager
|
?_labels= and ?_label=COL to expand foreign keys in JSON/CSV
These new querystring arguments can be used to request expanded foreign keys
in both JSON and CSV formats.
?_labels=on turns on expansions for ALL foreign key columns
?_label=COLUMN1&_label=COLUMN2 can be used to pick specific columns to expand
e.g. `Street_Tree_List.json?_label=qSpecies&_label=qLegalStatus`
{
"rowid": 233,
"TreeID": 121240,
"qLegalStatus": {
"value" 2,
"label": "Private"
}
"qSpecies": {
"value": 16,
"label": "Sycamore"
}
"qAddress": "91 Commonwealth Ave",
...
}
The labels option also works for the HTML and CSV views.
HTML defaults to `?_labels=on`, so if you pass `?_labels=off` you can disable
foreign key expansion entirely - or you can use `?_label=COLUMN` to request
just specific columns.
If you expand labels on CSV you get additional columns in the output:
`/Street_Tree_List.csv?_label=qLegalStatus`
rowid,TreeID,qLegalStatus,qLegalStatus_label...
1,141565,1,Permitted Site...
2,232565,2,Undocumented...
I also refactored the existing foreign key expansion code.
Closes #233. Refs #266.
2018-06-16 22:18:57 +00:00
|
|
|
from collections import OrderedDict
|
2017-11-10 19:25:54 +00:00
|
|
|
import base64
|
publish_subcommand hook + default plugins mechanism, used for publish heroku/now (#349)
This change introduces a new plugin hook, publish_subcommand, which can be
used to implement new subcommands for the "datasette publish" command family.
I've used this new hook to refactor out the "publish now" and "publish heroku"
implementations into separate modules. I've also added unit tests for these
two publishers, mocking the subprocess.call and subprocess.check_output
functions.
As part of this, I introduced a mechanism for loading default plugins. These
are defined in the new "default_plugins" list inside datasette/app.py
Closes #217 (Plugin support for datasette publish)
Closes #348 (Unit tests for "datasette publish")
Refs #14, #59, #102, #103, #146, #236, #347
2018-07-26 05:15:59 +00:00
|
|
|
import click
|
2017-11-30 07:09:54 +00:00
|
|
|
import hashlib
|
2020-05-27 19:25:52 +00:00
|
|
|
import inspect
|
2017-11-10 19:25:54 +00:00
|
|
|
import json
|
2020-04-10 18:34:09 +00:00
|
|
|
import mergedeep
|
2017-11-13 16:13:38 +00:00
|
|
|
import os
|
2017-11-11 05:55:50 +00:00
|
|
|
import re
|
2017-11-21 18:51:58 +00:00
|
|
|
import shlex
|
2017-11-13 16:13:38 +00:00
|
|
|
import tempfile
|
2017-11-10 19:25:54 +00:00
|
|
|
import time
|
2019-07-03 04:32:55 +00:00
|
|
|
import types
|
2017-12-08 16:06:24 +00:00
|
|
|
import shutil
|
2017-11-10 19:25:54 +00:00
|
|
|
import urllib
|
2018-04-14 10:41:27 +00:00
|
|
|
import numbers
|
2020-04-02 19:30:53 +00:00
|
|
|
import yaml
|
2020-05-27 18:17:43 +00:00
|
|
|
from .shutil_backport import copytree
|
2017-11-10 19:25:54 +00:00
|
|
|
|
2018-08-16 00:58:56 +00:00
|
|
|
try:
|
|
|
|
import pysqlite3 as sqlite3
|
|
|
|
except ImportError:
|
|
|
|
import sqlite3
|
|
|
|
|
2018-04-03 13:39:50 +00:00
|
|
|
# From https://www.sqlite.org/lang_keywords.html
|
2019-05-04 02:15:14 +00:00
|
|
|
reserved_words = set(
|
|
|
|
(
|
|
|
|
"abort action add after all alter analyze and as asc attach autoincrement "
|
|
|
|
"before begin between by cascade case cast check collate column commit "
|
|
|
|
"conflict constraint create cross current_date current_time "
|
|
|
|
"current_timestamp database default deferrable deferred delete desc detach "
|
|
|
|
"distinct drop each else end escape except exclusive exists explain fail "
|
|
|
|
"for foreign from full glob group having if ignore immediate in index "
|
|
|
|
"indexed initially inner insert instead intersect into is isnull join key "
|
|
|
|
"left like limit match natural no not notnull null of offset on or order "
|
|
|
|
"outer plan pragma primary query raise recursive references regexp reindex "
|
|
|
|
"release rename replace restrict right rollback row savepoint select set "
|
|
|
|
"table temp temporary then to transaction trigger union unique update using "
|
|
|
|
"vacuum values view virtual when where with without"
|
|
|
|
).split()
|
|
|
|
)
|
|
|
|
|
|
|
|
SPATIALITE_DOCKERFILE_EXTRAS = r"""
|
2018-05-31 14:16:50 +00:00
|
|
|
RUN apt-get update && \
|
|
|
|
apt-get install -y python3-dev gcc libsqlite3-mod-spatialite && \
|
|
|
|
rm -rf /var/lib/apt/lists/*
|
|
|
|
ENV SQLITE_EXTENSIONS /usr/lib/x86_64-linux-gnu/mod_spatialite.so
|
2019-05-04 02:15:14 +00:00
|
|
|
"""
|
2018-05-31 14:16:50 +00:00
|
|
|
|
2018-04-03 13:39:50 +00:00
|
|
|
|
2018-04-09 00:06:10 +00:00
|
|
|
def urlsafe_components(token):
|
|
|
|
"Splits token on commas and URL decodes each component"
|
2019-05-04 02:15:14 +00:00
|
|
|
return [urllib.parse.unquote_plus(b) for b in token.split(",")]
|
2017-11-10 19:25:54 +00:00
|
|
|
|
|
|
|
|
2018-04-15 21:48:30 +00:00
|
|
|
def path_from_row_pks(row, pks, use_rowid, quote=True):
|
|
|
|
""" Generate an optionally URL-quoted unique identifier
|
|
|
|
for a row from its primary keys."""
|
2017-11-10 19:25:54 +00:00
|
|
|
if use_rowid:
|
2019-05-04 02:15:14 +00:00
|
|
|
bits = [row["rowid"]]
|
2018-04-15 21:48:30 +00:00
|
|
|
else:
|
2018-06-21 14:56:28 +00:00
|
|
|
bits = [
|
2019-05-04 02:15:14 +00:00
|
|
|
row[pk]["value"] if isinstance(row[pk], dict) else row[pk] for pk in pks
|
2018-06-21 14:56:28 +00:00
|
|
|
]
|
2018-04-15 21:48:30 +00:00
|
|
|
if quote:
|
|
|
|
bits = [urllib.parse.quote_plus(str(bit)) for bit in bits]
|
|
|
|
else:
|
|
|
|
bits = [str(bit) for bit in bits]
|
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
return ",".join(bits)
|
2017-11-10 19:25:54 +00:00
|
|
|
|
|
|
|
|
2018-03-30 06:26:22 +00:00
|
|
|
def compound_keys_after_sql(pks, start_index=0):
|
2018-03-30 05:10:09 +00:00
|
|
|
# Implementation of keyset pagination
|
|
|
|
# See https://github.com/simonw/datasette/issues/190
|
|
|
|
# For pk1/pk2/pk3 returns:
|
|
|
|
#
|
|
|
|
# ([pk1] > :p0)
|
|
|
|
# or
|
|
|
|
# ([pk1] = :p0 and [pk2] > :p1)
|
|
|
|
# or
|
|
|
|
# ([pk1] = :p0 and [pk2] = :p1 and [pk3] > :p2)
|
|
|
|
or_clauses = []
|
|
|
|
pks_left = pks[:]
|
|
|
|
while pks_left:
|
|
|
|
and_clauses = []
|
|
|
|
last = pks_left[-1]
|
|
|
|
rest = pks_left[:-1]
|
2019-05-04 02:15:14 +00:00
|
|
|
and_clauses = [
|
|
|
|
"{} = :p{}".format(escape_sqlite(pk), (i + start_index))
|
|
|
|
for i, pk in enumerate(rest)
|
|
|
|
]
|
|
|
|
and_clauses.append(
|
|
|
|
"{} > :p{}".format(escape_sqlite(last), (len(rest) + start_index))
|
|
|
|
)
|
|
|
|
or_clauses.append("({})".format(" and ".join(and_clauses)))
|
2018-03-30 05:10:09 +00:00
|
|
|
pks_left.pop()
|
|
|
|
or_clauses.reverse()
|
2019-05-04 02:15:14 +00:00
|
|
|
return "({})".format("\n or\n".join(or_clauses))
|
2018-03-30 05:10:09 +00:00
|
|
|
|
|
|
|
|
2017-11-10 19:25:54 +00:00
|
|
|
class CustomJSONEncoder(json.JSONEncoder):
|
|
|
|
def default(self, obj):
|
|
|
|
if isinstance(obj, sqlite3.Row):
|
|
|
|
return tuple(obj)
|
|
|
|
if isinstance(obj, sqlite3.Cursor):
|
|
|
|
return list(obj)
|
|
|
|
if isinstance(obj, bytes):
|
|
|
|
# Does it encode to utf8?
|
|
|
|
try:
|
2019-05-04 02:15:14 +00:00
|
|
|
return obj.decode("utf8")
|
2017-11-10 19:25:54 +00:00
|
|
|
except UnicodeDecodeError:
|
|
|
|
return {
|
2019-05-04 02:15:14 +00:00
|
|
|
"$base64": True,
|
|
|
|
"encoded": base64.b64encode(obj).decode("latin1"),
|
2017-11-10 19:25:54 +00:00
|
|
|
}
|
|
|
|
return json.JSONEncoder.default(self, obj)
|
|
|
|
|
|
|
|
|
|
|
|
@contextmanager
|
|
|
|
def sqlite_timelimit(conn, ms):
|
|
|
|
deadline = time.time() + (ms / 1000)
|
2017-11-15 02:41:03 +00:00
|
|
|
# n is the number of SQLite virtual machine instructions that will be
|
|
|
|
# executed between each check. It's hard to know what to pick here.
|
|
|
|
# After some experimentation, I've decided to go with 1000 by default and
|
|
|
|
# 1 for time limits that are less than 50ms
|
|
|
|
n = 1000
|
|
|
|
if ms < 50:
|
|
|
|
n = 1
|
2017-11-10 19:25:54 +00:00
|
|
|
|
|
|
|
def handler():
|
|
|
|
if time.time() >= deadline:
|
|
|
|
return 1
|
2017-11-15 02:41:03 +00:00
|
|
|
|
|
|
|
conn.set_progress_handler(handler, n)
|
2019-04-21 19:02:24 +00:00
|
|
|
try:
|
|
|
|
yield
|
|
|
|
finally:
|
|
|
|
conn.set_progress_handler(None, n)
|
2017-11-10 19:25:54 +00:00
|
|
|
|
|
|
|
|
|
|
|
class InvalidSql(Exception):
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
2017-12-04 04:51:31 +00:00
|
|
|
allowed_sql_res = [
|
2019-05-04 02:15:14 +00:00
|
|
|
re.compile(r"^select\b"),
|
|
|
|
re.compile(r"^explain select\b"),
|
|
|
|
re.compile(r"^explain query plan select\b"),
|
|
|
|
re.compile(r"^with\b"),
|
2019-10-06 17:23:58 +00:00
|
|
|
re.compile(r"^explain with\b"),
|
|
|
|
re.compile(r"^explain query plan with\b"),
|
2017-12-04 04:51:31 +00:00
|
|
|
]
|
2020-05-06 17:18:31 +00:00
|
|
|
allowed_pragmas = (
|
|
|
|
"database_list",
|
|
|
|
"foreign_key_list",
|
|
|
|
"function_list",
|
|
|
|
"index_info",
|
|
|
|
"index_list",
|
|
|
|
"index_xinfo",
|
|
|
|
"page_count",
|
|
|
|
"max_page_count",
|
|
|
|
"page_size",
|
|
|
|
"schema_version",
|
|
|
|
"table_info",
|
|
|
|
"table_xinfo",
|
|
|
|
)
|
|
|
|
disallawed_sql_res = [
|
|
|
|
(
|
|
|
|
re.compile("pragma(?!_({}))".format("|".join(allowed_pragmas))),
|
|
|
|
"Statement may not contain PRAGMA",
|
|
|
|
)
|
|
|
|
]
|
2017-12-04 04:51:31 +00:00
|
|
|
|
|
|
|
|
2017-11-10 19:25:54 +00:00
|
|
|
def validate_sql_select(sql):
|
2020-02-13 06:05:46 +00:00
|
|
|
sql = "\n".join(
|
|
|
|
line for line in sql.split("\n") if not line.strip().startswith("--")
|
|
|
|
)
|
2017-11-10 19:25:54 +00:00
|
|
|
sql = sql.strip().lower()
|
2017-12-04 04:51:31 +00:00
|
|
|
if not any(r.match(sql) for r in allowed_sql_res):
|
2019-05-04 02:15:14 +00:00
|
|
|
raise InvalidSql("Statement must be a SELECT")
|
2017-12-04 04:51:31 +00:00
|
|
|
for r, msg in disallawed_sql_res:
|
|
|
|
if r.search(sql):
|
|
|
|
raise InvalidSql(msg)
|
2017-11-10 20:41:14 +00:00
|
|
|
|
|
|
|
|
2018-06-18 06:03:22 +00:00
|
|
|
def append_querystring(url, querystring):
|
|
|
|
op = "&" if ("?" in url) else "?"
|
2019-05-04 02:15:14 +00:00
|
|
|
return "{}{}{}".format(url, op, querystring)
|
2018-06-18 06:03:22 +00:00
|
|
|
|
|
|
|
|
2018-04-03 14:52:54 +00:00
|
|
|
def path_with_added_args(request, args, path=None):
|
|
|
|
path = path or request.path
|
2017-11-23 04:03:46 +00:00
|
|
|
if isinstance(args, dict):
|
|
|
|
args = args.items()
|
2018-05-14 22:09:09 +00:00
|
|
|
args_to_remove = {k for k, v in args if v is None}
|
2018-05-12 21:35:25 +00:00
|
|
|
current = []
|
2018-05-14 03:02:07 +00:00
|
|
|
for key, value in urllib.parse.parse_qsl(request.query_string):
|
2018-05-14 22:09:09 +00:00
|
|
|
if key not in args_to_remove:
|
2018-05-14 03:02:07 +00:00
|
|
|
current.append((key, value))
|
2019-05-04 02:15:14 +00:00
|
|
|
current.extend([(key, value) for key, value in args if value is not None])
|
2018-05-12 21:35:25 +00:00
|
|
|
query_string = urllib.parse.urlencode(current)
|
2017-11-23 20:32:54 +00:00
|
|
|
if query_string:
|
2019-05-04 02:15:14 +00:00
|
|
|
query_string = "?{}".format(query_string)
|
2018-04-03 14:52:54 +00:00
|
|
|
return path + query_string
|
2017-11-11 05:55:50 +00:00
|
|
|
|
|
|
|
|
2018-05-14 20:42:10 +00:00
|
|
|
def path_with_removed_args(request, args, path=None):
|
2019-03-17 22:55:04 +00:00
|
|
|
query_string = request.query_string
|
|
|
|
if path is None:
|
|
|
|
path = request.path
|
|
|
|
else:
|
|
|
|
if "?" in path:
|
|
|
|
bits = path.split("?", 1)
|
|
|
|
path, query_string = bits
|
2018-05-15 10:11:52 +00:00
|
|
|
# args can be a dict or a set
|
2018-05-14 20:42:10 +00:00
|
|
|
current = []
|
2018-05-15 10:11:52 +00:00
|
|
|
if isinstance(args, set):
|
2019-05-04 02:15:14 +00:00
|
|
|
|
2018-05-15 10:11:52 +00:00
|
|
|
def should_remove(key, value):
|
|
|
|
return key in args
|
2019-05-04 02:15:14 +00:00
|
|
|
|
2018-05-15 10:11:52 +00:00
|
|
|
elif isinstance(args, dict):
|
|
|
|
# Must match key AND value
|
|
|
|
def should_remove(key, value):
|
|
|
|
return args.get(key) == value
|
2019-05-04 02:15:14 +00:00
|
|
|
|
2019-03-17 22:55:04 +00:00
|
|
|
for key, value in urllib.parse.parse_qsl(query_string):
|
2018-05-15 10:11:52 +00:00
|
|
|
if not should_remove(key, value):
|
2018-05-14 20:42:10 +00:00
|
|
|
current.append((key, value))
|
|
|
|
query_string = urllib.parse.urlencode(current)
|
|
|
|
if query_string:
|
2019-05-04 02:15:14 +00:00
|
|
|
query_string = "?{}".format(query_string)
|
2018-05-14 20:42:10 +00:00
|
|
|
return path + query_string
|
|
|
|
|
|
|
|
|
2018-05-15 09:34:45 +00:00
|
|
|
def path_with_replaced_args(request, args, path=None):
|
|
|
|
path = path or request.path
|
|
|
|
if isinstance(args, dict):
|
|
|
|
args = args.items()
|
|
|
|
keys_to_replace = {p[0] for p in args}
|
|
|
|
current = []
|
|
|
|
for key, value in urllib.parse.parse_qsl(request.query_string):
|
|
|
|
if key not in keys_to_replace:
|
|
|
|
current.append((key, value))
|
|
|
|
current.extend([p for p in args if p[1] is not None])
|
|
|
|
query_string = urllib.parse.urlencode(current)
|
|
|
|
if query_string:
|
2019-05-04 02:15:14 +00:00
|
|
|
query_string = "?{}".format(query_string)
|
2018-05-15 09:34:45 +00:00
|
|
|
return path + query_string
|
|
|
|
|
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
_css_re = re.compile(r"""['"\n\\]""")
|
|
|
|
_boring_keyword_re = re.compile(r"^[a-zA-Z_][a-zA-Z0-9_]*$")
|
2017-11-11 05:55:50 +00:00
|
|
|
|
|
|
|
|
|
|
|
def escape_css_string(s):
|
2019-05-04 02:15:14 +00:00
|
|
|
return _css_re.sub(lambda m: "\\{:X}".format(ord(m.group())), s)
|
2017-11-11 07:25:22 +00:00
|
|
|
|
|
|
|
|
2018-04-03 13:39:50 +00:00
|
|
|
def escape_sqlite(s):
|
|
|
|
if _boring_keyword_re.match(s) and (s.lower() not in reserved_words):
|
2017-11-12 23:17:00 +00:00
|
|
|
return s
|
|
|
|
else:
|
2019-05-04 02:15:14 +00:00
|
|
|
return "[{}]".format(s)
|
|
|
|
|
2017-11-12 23:17:00 +00:00
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
def make_dockerfile(
|
|
|
|
files,
|
|
|
|
metadata_file,
|
|
|
|
extra_options,
|
|
|
|
branch,
|
|
|
|
template_dir,
|
|
|
|
plugins_dir,
|
|
|
|
static,
|
|
|
|
install,
|
|
|
|
spatialite,
|
|
|
|
version_note,
|
2019-07-08 02:06:31 +00:00
|
|
|
environment_variables=None,
|
2020-01-29 22:46:43 +00:00
|
|
|
port=8001,
|
2019-05-04 02:15:14 +00:00
|
|
|
):
|
|
|
|
cmd = ["datasette", "serve", "--host", "0.0.0.0"]
|
2019-05-19 22:53:34 +00:00
|
|
|
for filename in files:
|
|
|
|
cmd.extend(["-i", filename])
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--cors", "--inspect-file", "inspect-data.json"])
|
2017-11-13 21:58:34 +00:00
|
|
|
if metadata_file:
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--metadata", "{}".format(metadata_file)])
|
2017-12-09 18:19:39 +00:00
|
|
|
if template_dir:
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--template-dir", "templates/"])
|
2018-04-16 05:22:01 +00:00
|
|
|
if plugins_dir:
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--plugins-dir", "plugins/"])
|
2018-06-17 20:14:55 +00:00
|
|
|
if version_note:
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--version-note", "{}".format(version_note)])
|
2017-12-09 18:19:39 +00:00
|
|
|
if static:
|
|
|
|
for mount_point, _ in static:
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--static", "{}:{}".format(mount_point, mount_point)])
|
2017-11-13 21:58:34 +00:00
|
|
|
if extra_options:
|
|
|
|
for opt in extra_options.split():
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.append("{}".format(opt))
|
2019-05-03 13:59:01 +00:00
|
|
|
cmd = [shlex.quote(part) for part in cmd]
|
|
|
|
# port attribute is a (fixed) env variable and should not be quoted
|
2019-05-04 02:15:14 +00:00
|
|
|
cmd.extend(["--port", "$PORT"])
|
|
|
|
cmd = " ".join(cmd)
|
2017-11-19 18:20:17 +00:00
|
|
|
if branch:
|
2019-05-04 02:15:14 +00:00
|
|
|
install = [
|
|
|
|
"https://github.com/simonw/datasette/archive/{}.zip".format(branch)
|
|
|
|
] + list(install)
|
2018-04-18 14:48:34 +00:00
|
|
|
else:
|
2019-05-04 02:15:14 +00:00
|
|
|
install = ["datasette"] + list(install)
|
2018-04-18 14:48:34 +00:00
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
return """
|
2019-11-12 06:03:09 +00:00
|
|
|
FROM python:3.8
|
2017-11-11 07:25:22 +00:00
|
|
|
COPY . /app
|
|
|
|
WORKDIR /app
|
2018-05-31 14:16:50 +00:00
|
|
|
{spatialite_extras}
|
2019-07-08 02:06:31 +00:00
|
|
|
{environment_variables}
|
2018-07-24 01:51:43 +00:00
|
|
|
RUN pip install -U {install_from}
|
2017-12-07 16:57:31 +00:00
|
|
|
RUN datasette inspect {files} --inspect-file inspect-data.json
|
2020-01-29 22:46:43 +00:00
|
|
|
ENV PORT {port}
|
|
|
|
EXPOSE {port}
|
2019-05-04 02:15:14 +00:00
|
|
|
CMD {cmd}""".format(
|
2019-07-08 02:06:31 +00:00
|
|
|
environment_variables="\n".join(
|
|
|
|
[
|
|
|
|
"ENV {} '{}'".format(key, value)
|
|
|
|
for key, value in (environment_variables or {}).items()
|
|
|
|
]
|
|
|
|
),
|
2019-05-04 02:15:14 +00:00
|
|
|
files=" ".join(files),
|
2019-05-03 13:59:01 +00:00
|
|
|
cmd=cmd,
|
2019-05-04 02:15:14 +00:00
|
|
|
install_from=" ".join(install),
|
|
|
|
spatialite_extras=SPATIALITE_DOCKERFILE_EXTRAS if spatialite else "",
|
2020-01-29 22:46:43 +00:00
|
|
|
port=port,
|
2017-11-11 07:25:22 +00:00
|
|
|
).strip()
|
2017-11-13 16:13:38 +00:00
|
|
|
|
|
|
|
|
|
|
|
@contextmanager
|
2018-04-18 14:48:34 +00:00
|
|
|
def temporary_docker_directory(
|
|
|
|
files,
|
|
|
|
name,
|
|
|
|
metadata,
|
|
|
|
extra_options,
|
|
|
|
branch,
|
|
|
|
template_dir,
|
|
|
|
plugins_dir,
|
|
|
|
static,
|
|
|
|
install,
|
2018-05-31 14:16:50 +00:00
|
|
|
spatialite,
|
2018-06-17 20:14:55 +00:00
|
|
|
version_note,
|
2019-05-04 02:15:14 +00:00
|
|
|
extra_metadata=None,
|
2019-07-08 02:06:31 +00:00
|
|
|
environment_variables=None,
|
2020-01-29 22:46:43 +00:00
|
|
|
port=8001,
|
2018-04-18 14:48:34 +00:00
|
|
|
):
|
2017-11-15 05:02:11 +00:00
|
|
|
extra_metadata = extra_metadata or {}
|
2017-11-13 16:13:38 +00:00
|
|
|
tmp = tempfile.TemporaryDirectory()
|
|
|
|
# We create a datasette folder in there to get a nicer now deploy name
|
|
|
|
datasette_dir = os.path.join(tmp.name, name)
|
|
|
|
os.mkdir(datasette_dir)
|
|
|
|
saved_cwd = os.getcwd()
|
2019-05-04 02:15:14 +00:00
|
|
|
file_paths = [os.path.join(saved_cwd, file_path) for file_path in files]
|
2017-11-13 16:13:38 +00:00
|
|
|
file_names = [os.path.split(f)[-1] for f in files]
|
2017-11-15 05:02:11 +00:00
|
|
|
if metadata:
|
2020-04-02 19:30:53 +00:00
|
|
|
metadata_content = parse_metadata(metadata.read())
|
2017-11-15 05:02:11 +00:00
|
|
|
else:
|
|
|
|
metadata_content = {}
|
2020-04-10 18:34:09 +00:00
|
|
|
# Merge in the non-null values in extra_metadata
|
|
|
|
mergedeep.merge(
|
|
|
|
metadata_content,
|
|
|
|
{key: value for key, value in extra_metadata.items() if value is not None},
|
|
|
|
)
|
2017-11-13 16:13:38 +00:00
|
|
|
try:
|
2017-12-09 18:19:39 +00:00
|
|
|
dockerfile = make_dockerfile(
|
|
|
|
file_names,
|
2019-05-04 02:15:14 +00:00
|
|
|
metadata_content and "metadata.json",
|
2017-12-09 18:19:39 +00:00
|
|
|
extra_options,
|
|
|
|
branch,
|
|
|
|
template_dir,
|
2018-04-16 05:22:01 +00:00
|
|
|
plugins_dir,
|
2017-12-09 18:19:39 +00:00
|
|
|
static,
|
2018-04-18 14:48:34 +00:00
|
|
|
install,
|
2018-05-31 14:16:50 +00:00
|
|
|
spatialite,
|
2018-06-17 20:14:55 +00:00
|
|
|
version_note,
|
2019-07-08 02:06:31 +00:00
|
|
|
environment_variables,
|
2020-01-29 22:46:43 +00:00
|
|
|
port=port,
|
2017-12-09 18:19:39 +00:00
|
|
|
)
|
2017-11-13 16:13:38 +00:00
|
|
|
os.chdir(datasette_dir)
|
2017-11-15 05:02:11 +00:00
|
|
|
if metadata_content:
|
2019-05-04 02:15:14 +00:00
|
|
|
open("metadata.json", "w").write(json.dumps(metadata_content, indent=2))
|
|
|
|
open("Dockerfile", "w").write(dockerfile)
|
2017-11-13 16:13:38 +00:00
|
|
|
for path, filename in zip(file_paths, file_names):
|
2017-12-08 16:06:24 +00:00
|
|
|
link_or_copy(path, os.path.join(datasette_dir, filename))
|
2017-12-09 18:19:39 +00:00
|
|
|
if template_dir:
|
|
|
|
link_or_copy_directory(
|
|
|
|
os.path.join(saved_cwd, template_dir),
|
2019-05-04 02:15:14 +00:00
|
|
|
os.path.join(datasette_dir, "templates"),
|
2017-12-09 18:19:39 +00:00
|
|
|
)
|
2018-04-16 05:22:01 +00:00
|
|
|
if plugins_dir:
|
|
|
|
link_or_copy_directory(
|
|
|
|
os.path.join(saved_cwd, plugins_dir),
|
2019-05-04 02:15:14 +00:00
|
|
|
os.path.join(datasette_dir, "plugins"),
|
2018-04-16 05:22:01 +00:00
|
|
|
)
|
2017-12-09 18:19:39 +00:00
|
|
|
for mount_point, path in static:
|
|
|
|
link_or_copy_directory(
|
2019-05-04 02:15:14 +00:00
|
|
|
os.path.join(saved_cwd, path), os.path.join(datasette_dir, mount_point)
|
2017-12-09 18:19:39 +00:00
|
|
|
)
|
2017-12-08 16:06:24 +00:00
|
|
|
yield datasette_dir
|
2017-11-13 16:13:38 +00:00
|
|
|
finally:
|
|
|
|
tmp.cleanup()
|
|
|
|
os.chdir(saved_cwd)
|
2017-11-15 19:53:00 +00:00
|
|
|
|
2017-11-16 01:34:32 +00:00
|
|
|
|
2019-04-07 03:11:08 +00:00
|
|
|
def detect_primary_keys(conn, table):
|
|
|
|
" Figure out primary keys for a table. "
|
|
|
|
table_info_rows = [
|
|
|
|
row
|
2019-05-04 02:15:14 +00:00
|
|
|
for row in conn.execute('PRAGMA table_info("{}")'.format(table)).fetchall()
|
2019-04-07 03:11:08 +00:00
|
|
|
if row[-1]
|
|
|
|
]
|
|
|
|
table_info_rows.sort(key=lambda row: row[-1])
|
|
|
|
return [str(r[1]) for r in table_info_rows]
|
|
|
|
|
|
|
|
|
2019-04-07 02:56:07 +00:00
|
|
|
def get_outbound_foreign_keys(conn, table):
|
2019-05-04 02:15:14 +00:00
|
|
|
infos = conn.execute("PRAGMA foreign_key_list([{}])".format(table)).fetchall()
|
2019-04-07 02:56:07 +00:00
|
|
|
fks = []
|
|
|
|
for info in infos:
|
|
|
|
if info is not None:
|
|
|
|
id, seq, table_name, from_, to_, on_update, on_delete, match = info
|
2019-05-04 02:15:14 +00:00
|
|
|
fks.append(
|
|
|
|
{"other_table": table_name, "column": from_, "other_column": to_}
|
|
|
|
)
|
2019-04-07 02:56:07 +00:00
|
|
|
return fks
|
|
|
|
|
|
|
|
|
2017-11-16 01:34:32 +00:00
|
|
|
def get_all_foreign_keys(conn):
|
2019-05-04 02:15:14 +00:00
|
|
|
tables = [
|
|
|
|
r[0] for r in conn.execute('select name from sqlite_master where type="table"')
|
|
|
|
]
|
2017-11-16 01:34:32 +00:00
|
|
|
table_to_foreign_keys = {}
|
|
|
|
for table in tables:
|
2019-05-04 02:15:14 +00:00
|
|
|
table_to_foreign_keys[table] = {"incoming": [], "outgoing": []}
|
2017-11-16 01:34:32 +00:00
|
|
|
for table in tables:
|
2019-05-04 02:15:14 +00:00
|
|
|
infos = conn.execute("PRAGMA foreign_key_list([{}])".format(table)).fetchall()
|
2017-11-16 01:34:32 +00:00
|
|
|
for info in infos:
|
|
|
|
if info is not None:
|
|
|
|
id, seq, table_name, from_, to_, on_update, on_delete, match = info
|
2017-11-17 16:18:26 +00:00
|
|
|
if table_name not in table_to_foreign_keys:
|
|
|
|
# Weird edge case where something refers to a table that does
|
|
|
|
# not actually exist
|
|
|
|
continue
|
2019-05-04 02:15:14 +00:00
|
|
|
table_to_foreign_keys[table_name]["incoming"].append(
|
|
|
|
{"other_table": table, "column": to_, "other_column": from_}
|
|
|
|
)
|
|
|
|
table_to_foreign_keys[table]["outgoing"].append(
|
|
|
|
{"other_table": table_name, "column": from_, "other_column": to_}
|
|
|
|
)
|
2017-11-16 01:34:32 +00:00
|
|
|
|
|
|
|
return table_to_foreign_keys
|
2017-11-19 16:59:26 +00:00
|
|
|
|
|
|
|
|
2018-04-11 21:20:25 +00:00
|
|
|
def detect_spatialite(conn):
|
2019-05-04 02:15:14 +00:00
|
|
|
rows = conn.execute(
|
|
|
|
'select 1 from sqlite_master where tbl_name = "geometry_columns"'
|
|
|
|
).fetchall()
|
2018-04-11 21:20:25 +00:00
|
|
|
return len(rows) > 0
|
|
|
|
|
|
|
|
|
2018-04-29 00:04:32 +00:00
|
|
|
def detect_fts(conn, table):
|
2017-11-19 16:59:26 +00:00
|
|
|
"Detect if table has a corresponding FTS virtual table and return it"
|
|
|
|
rows = conn.execute(detect_fts_sql(table)).fetchall()
|
|
|
|
if len(rows) == 0:
|
|
|
|
return None
|
|
|
|
else:
|
|
|
|
return rows[0][0]
|
|
|
|
|
|
|
|
|
|
|
|
def detect_fts_sql(table):
|
2019-05-04 02:15:14 +00:00
|
|
|
return r"""
|
2017-11-19 16:59:26 +00:00
|
|
|
select name from sqlite_master
|
|
|
|
where rootpage = 0
|
2017-11-20 06:18:07 +00:00
|
|
|
and (
|
|
|
|
sql like '%VIRTUAL TABLE%USING FTS%content="{table}"%'
|
2019-09-03 00:32:27 +00:00
|
|
|
or sql like '%VIRTUAL TABLE%USING FTS%content=[{table}]%'
|
2017-11-24 22:51:00 +00:00
|
|
|
or (
|
|
|
|
tbl_name = "{table}"
|
2017-12-07 04:54:25 +00:00
|
|
|
and sql like '%VIRTUAL TABLE%USING FTS%'
|
2017-11-24 22:51:00 +00:00
|
|
|
)
|
2017-11-20 06:18:07 +00:00
|
|
|
)
|
2019-05-04 02:15:14 +00:00
|
|
|
""".format(
|
|
|
|
table=table
|
|
|
|
)
|
2017-11-20 06:03:24 +00:00
|
|
|
|
|
|
|
|
2019-04-10 15:17:19 +00:00
|
|
|
def detect_json1(conn=None):
|
|
|
|
if conn is None:
|
|
|
|
conn = sqlite3.connect(":memory:")
|
|
|
|
try:
|
|
|
|
conn.execute("SELECT json('{}')")
|
|
|
|
return True
|
|
|
|
except Exception:
|
|
|
|
return False
|
|
|
|
|
|
|
|
|
2019-04-07 01:58:51 +00:00
|
|
|
def table_columns(conn, table):
|
|
|
|
return [
|
|
|
|
r[1]
|
|
|
|
for r in conn.execute(
|
|
|
|
"PRAGMA table_info({});".format(escape_sqlite(table))
|
|
|
|
).fetchall()
|
|
|
|
]
|
|
|
|
|
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
filter_column_re = re.compile(r"^_filter_column_\d+$")
|
2017-11-23 04:03:46 +00:00
|
|
|
|
|
|
|
|
|
|
|
def filters_should_redirect(special_args):
|
|
|
|
redirect_params = []
|
2017-11-23 20:32:54 +00:00
|
|
|
# Handle _filter_column=foo&_filter_op=exact&_filter_value=...
|
2019-05-04 02:15:14 +00:00
|
|
|
filter_column = special_args.get("_filter_column")
|
|
|
|
filter_op = special_args.get("_filter_op") or ""
|
|
|
|
filter_value = special_args.get("_filter_value") or ""
|
|
|
|
if "__" in filter_op:
|
|
|
|
filter_op, filter_value = filter_op.split("__", 1)
|
2017-11-23 20:32:54 +00:00
|
|
|
if filter_column:
|
|
|
|
redirect_params.append(
|
2019-05-04 02:15:14 +00:00
|
|
|
("{}__{}".format(filter_column, filter_op), filter_value)
|
2017-11-23 20:32:54 +00:00
|
|
|
)
|
2019-05-04 02:15:14 +00:00
|
|
|
for key in ("_filter_column", "_filter_op", "_filter_value"):
|
2017-11-23 20:32:54 +00:00
|
|
|
if key in special_args:
|
|
|
|
redirect_params.append((key, None))
|
2017-11-23 04:03:46 +00:00
|
|
|
# Now handle _filter_column_1=name&_filter_op_1=contains&_filter_value_1=hello
|
|
|
|
column_keys = [k for k in special_args if filter_column_re.match(k)]
|
|
|
|
for column_key in column_keys:
|
2019-05-04 02:15:14 +00:00
|
|
|
number = column_key.split("_")[-1]
|
2017-11-23 04:03:46 +00:00
|
|
|
column = special_args[column_key]
|
2019-05-04 02:15:14 +00:00
|
|
|
op = special_args.get("_filter_op_{}".format(number)) or "exact"
|
|
|
|
value = special_args.get("_filter_value_{}".format(number)) or ""
|
|
|
|
if "__" in op:
|
|
|
|
op, value = op.split("__", 1)
|
2017-11-23 20:32:54 +00:00
|
|
|
if column:
|
2019-05-04 02:15:14 +00:00
|
|
|
redirect_params.append(("{}__{}".format(column, op), value))
|
|
|
|
redirect_params.extend(
|
|
|
|
[
|
|
|
|
("_filter_column_{}".format(number), None),
|
|
|
|
("_filter_op_{}".format(number), None),
|
|
|
|
("_filter_value_{}".format(number), None),
|
|
|
|
]
|
|
|
|
)
|
2017-11-23 04:03:46 +00:00
|
|
|
return redirect_params
|
2017-11-29 17:05:24 +00:00
|
|
|
|
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
whitespace_re = re.compile(r"\s")
|
2017-11-29 17:05:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
def is_url(value):
|
|
|
|
"Must start with http:// or https:// and contain JUST a URL"
|
|
|
|
if not isinstance(value, str):
|
|
|
|
return False
|
2019-05-04 02:15:14 +00:00
|
|
|
if not value.startswith("http://") and not value.startswith("https://"):
|
2017-11-29 17:05:24 +00:00
|
|
|
return False
|
|
|
|
# Any whitespace at all is invalid
|
|
|
|
if whitespace_re.search(value):
|
|
|
|
return False
|
|
|
|
return True
|
2017-11-30 07:09:54 +00:00
|
|
|
|
|
|
|
|
2019-05-04 02:15:14 +00:00
|
|
|
css_class_re = re.compile(r"^[a-zA-Z]+[_a-zA-Z0-9-]*$")
|
|
|
|
css_invalid_chars_re = re.compile(r"[^a-zA-Z0-9_\-]")
|
2017-11-30 07:09:54 +00:00
|
|
|
|
|
|
|
|
|
|
|
def to_css_class(s):
|
|
|
|
"""
|
|
|
|
Given a string (e.g. a table name) returns a valid unique CSS class.
|
|
|
|
For simple cases, just returns the string again. If the string is not a
|
|
|
|
valid CSS class (we disallow - and _ prefixes even though they are valid
|
|
|
|
as they may be confused with browser prefixes) we strip invalid characters
|
|
|
|
and add a 6 char md5 sum suffix, to make sure two tables with identical
|
|
|
|
names after stripping characters don't end up with the same CSS class.
|
|
|
|
"""
|
|
|
|
if css_class_re.match(s):
|
|
|
|
return s
|
2019-05-04 02:15:14 +00:00
|
|
|
md5_suffix = hashlib.md5(s.encode("utf8")).hexdigest()[:6]
|
2017-11-30 07:09:54 +00:00
|
|
|
# Strip leading _, -
|
2019-05-04 02:15:14 +00:00
|
|
|
s = s.lstrip("_").lstrip("-")
|
2017-11-30 07:09:54 +00:00
|
|
|
# Replace any whitespace with hyphens
|
2019-05-04 02:15:14 +00:00
|
|
|
s = "-".join(s.split())
|
2017-11-30 07:09:54 +00:00
|
|
|
# Remove any remaining invalid characters
|
2019-05-04 02:15:14 +00:00
|
|
|
s = css_invalid_chars_re.sub("", s)
|
2017-11-30 07:09:54 +00:00
|
|
|
# Attach the md5 suffix
|
|
|
|
bits = [b for b in (s, md5_suffix) if b]
|
2019-05-04 02:15:14 +00:00
|
|
|
return "-".join(bits)
|
2017-12-08 16:06:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
def link_or_copy(src, dst):
|
|
|
|
# Intended for use in populating a temp directory. We link if possible,
|
|
|
|
# but fall back to copying if the temp directory is on a different device
|
|
|
|
# https://github.com/simonw/datasette/issues/141
|
|
|
|
try:
|
|
|
|
os.link(src, dst)
|
2017-12-09 18:19:39 +00:00
|
|
|
except OSError:
|
2017-12-08 16:06:24 +00:00
|
|
|
shutil.copyfile(src, dst)
|
2017-12-09 18:19:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
def link_or_copy_directory(src, dst):
|
|
|
|
try:
|
2020-05-27 18:17:43 +00:00
|
|
|
copytree(src, dst, copy_function=os.link, dirs_exist_ok=True)
|
2017-12-09 18:19:39 +00:00
|
|
|
except OSError:
|
2020-05-27 18:17:43 +00:00
|
|
|
copytree(src, dst, dirs_exist_ok=True)
|
2018-04-16 05:22:01 +00:00
|
|
|
|
|
|
|
|
|
|
|
def module_from_path(path, name):
|
|
|
|
# Adapted from http://sayspy.blogspot.com/2011/07/how-to-import-module-from-just-file.html
|
2019-07-03 04:32:55 +00:00
|
|
|
mod = types.ModuleType(name)
|
2018-04-16 05:22:01 +00:00
|
|
|
mod.__file__ = path
|
2019-05-04 02:15:14 +00:00
|
|
|
with open(path, "r") as file:
|
|
|
|
code = compile(file.read(), path, "exec", dont_inherit=True)
|
2018-04-16 05:22:01 +00:00
|
|
|
exec(code, mod.__dict__)
|
|
|
|
return mod
|
2018-04-19 05:24:48 +00:00
|
|
|
|
|
|
|
|
2019-05-01 23:01:56 +00:00
|
|
|
async def resolve_table_and_format(table_and_format, table_exists, allowed_formats=[]):
|
2019-05-04 02:15:14 +00:00
|
|
|
if "." in table_and_format:
|
2018-06-15 06:51:23 +00:00
|
|
|
# Check if a table exists with this exact name
|
2019-03-31 18:02:22 +00:00
|
|
|
it_exists = await table_exists(table_and_format)
|
|
|
|
if it_exists:
|
2018-06-15 06:51:23 +00:00
|
|
|
return table_and_format, None
|
2019-05-01 23:01:56 +00:00
|
|
|
|
2018-06-15 06:51:23 +00:00
|
|
|
# Check if table ends with a known format
|
2019-05-04 02:15:14 +00:00
|
|
|
formats = list(allowed_formats) + ["csv", "jsono"]
|
2019-05-01 23:01:56 +00:00
|
|
|
for _format in formats:
|
2018-06-15 06:51:23 +00:00
|
|
|
if table_and_format.endswith(".{}".format(_format)):
|
2019-05-04 02:15:14 +00:00
|
|
|
table = table_and_format[: -(len(_format) + 1)]
|
2018-06-15 06:51:23 +00:00
|
|
|
return table, _format
|
|
|
|
return table_and_format, None
|
|
|
|
|
|
|
|
|
|
|
|
def path_with_format(request, format, extra_qs=None):
|
|
|
|
qs = extra_qs or {}
|
|
|
|
path = request.path
|
|
|
|
if "." in request.path:
|
|
|
|
qs["_format"] = format
|
|
|
|
else:
|
|
|
|
path = "{}.{}".format(path, format)
|
|
|
|
if qs:
|
|
|
|
extra = urllib.parse.urlencode(sorted(qs.items()))
|
|
|
|
if request.query_string:
|
2019-05-04 02:15:14 +00:00
|
|
|
path = "{}?{}&{}".format(path, request.query_string, extra)
|
2018-06-15 06:51:23 +00:00
|
|
|
else:
|
|
|
|
path = "{}?{}".format(path, extra)
|
|
|
|
elif request.query_string:
|
|
|
|
path = "{}?{}".format(path, request.query_string)
|
|
|
|
return path
|
2018-06-16 16:44:31 +00:00
|
|
|
|
|
|
|
|
?_labels= and ?_label=COL to expand foreign keys in JSON/CSV
These new querystring arguments can be used to request expanded foreign keys
in both JSON and CSV formats.
?_labels=on turns on expansions for ALL foreign key columns
?_label=COLUMN1&_label=COLUMN2 can be used to pick specific columns to expand
e.g. `Street_Tree_List.json?_label=qSpecies&_label=qLegalStatus`
{
"rowid": 233,
"TreeID": 121240,
"qLegalStatus": {
"value" 2,
"label": "Private"
}
"qSpecies": {
"value": 16,
"label": "Sycamore"
}
"qAddress": "91 Commonwealth Ave",
...
}
The labels option also works for the HTML and CSV views.
HTML defaults to `?_labels=on`, so if you pass `?_labels=off` you can disable
foreign key expansion entirely - or you can use `?_label=COLUMN` to request
just specific columns.
If you expand labels on CSV you get additional columns in the output:
`/Street_Tree_List.csv?_label=qLegalStatus`
rowid,TreeID,qLegalStatus,qLegalStatus_label...
1,141565,1,Permitted Site...
2,232565,2,Undocumented...
I also refactored the existing foreign key expansion code.
Closes #233. Refs #266.
2018-06-16 22:18:57 +00:00
|
|
|
class CustomRow(OrderedDict):
|
|
|
|
# Loose imitation of sqlite3.Row which offers
|
|
|
|
# both index-based AND key-based lookups
|
2018-06-21 14:56:28 +00:00
|
|
|
def __init__(self, columns, values=None):
|
?_labels= and ?_label=COL to expand foreign keys in JSON/CSV
These new querystring arguments can be used to request expanded foreign keys
in both JSON and CSV formats.
?_labels=on turns on expansions for ALL foreign key columns
?_label=COLUMN1&_label=COLUMN2 can be used to pick specific columns to expand
e.g. `Street_Tree_List.json?_label=qSpecies&_label=qLegalStatus`
{
"rowid": 233,
"TreeID": 121240,
"qLegalStatus": {
"value" 2,
"label": "Private"
}
"qSpecies": {
"value": 16,
"label": "Sycamore"
}
"qAddress": "91 Commonwealth Ave",
...
}
The labels option also works for the HTML and CSV views.
HTML defaults to `?_labels=on`, so if you pass `?_labels=off` you can disable
foreign key expansion entirely - or you can use `?_label=COLUMN` to request
just specific columns.
If you expand labels on CSV you get additional columns in the output:
`/Street_Tree_List.csv?_label=qLegalStatus`
rowid,TreeID,qLegalStatus,qLegalStatus_label...
1,141565,1,Permitted Site...
2,232565,2,Undocumented...
I also refactored the existing foreign key expansion code.
Closes #233. Refs #266.
2018-06-16 22:18:57 +00:00
|
|
|
self.columns = columns
|
2018-06-21 14:56:28 +00:00
|
|
|
if values:
|
|
|
|
self.update(values)
|
?_labels= and ?_label=COL to expand foreign keys in JSON/CSV
These new querystring arguments can be used to request expanded foreign keys
in both JSON and CSV formats.
?_labels=on turns on expansions for ALL foreign key columns
?_label=COLUMN1&_label=COLUMN2 can be used to pick specific columns to expand
e.g. `Street_Tree_List.json?_label=qSpecies&_label=qLegalStatus`
{
"rowid": 233,
"TreeID": 121240,
"qLegalStatus": {
"value" 2,
"label": "Private"
}
"qSpecies": {
"value": 16,
"label": "Sycamore"
}
"qAddress": "91 Commonwealth Ave",
...
}
The labels option also works for the HTML and CSV views.
HTML defaults to `?_labels=on`, so if you pass `?_labels=off` you can disable
foreign key expansion entirely - or you can use `?_label=COLUMN` to request
just specific columns.
If you expand labels on CSV you get additional columns in the output:
`/Street_Tree_List.csv?_label=qLegalStatus`
rowid,TreeID,qLegalStatus,qLegalStatus_label...
1,141565,1,Permitted Site...
2,232565,2,Undocumented...
I also refactored the existing foreign key expansion code.
Closes #233. Refs #266.
2018-06-16 22:18:57 +00:00
|
|
|
|
|
|
|
def __getitem__(self, key):
|
|
|
|
if isinstance(key, int):
|
|
|
|
return super().__getitem__(self.columns[key])
|
|
|
|
else:
|
|
|
|
return super().__getitem__(key)
|
|
|
|
|
|
|
|
def __iter__(self):
|
|
|
|
for column in self.columns:
|
|
|
|
yield self[column]
|
|
|
|
|
|
|
|
|
2018-06-16 16:44:31 +00:00
|
|
|
def value_as_boolean(value):
|
2019-05-04 02:15:14 +00:00
|
|
|
if value.lower() not in ("on", "off", "true", "false", "1", "0"):
|
2018-06-16 16:44:31 +00:00
|
|
|
raise ValueAsBooleanError
|
2019-05-04 02:15:14 +00:00
|
|
|
return value.lower() in ("on", "true", "1")
|
2018-06-16 16:44:31 +00:00
|
|
|
|
|
|
|
|
|
|
|
class ValueAsBooleanError(ValueError):
|
|
|
|
pass
|
2018-06-18 03:21:02 +00:00
|
|
|
|
|
|
|
|
|
|
|
class WriteLimitExceeded(Exception):
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
class LimitedWriter:
|
|
|
|
def __init__(self, writer, limit_mb):
|
|
|
|
self.writer = writer
|
|
|
|
self.limit_bytes = limit_mb * 1024 * 1024
|
|
|
|
self.bytes_count = 0
|
|
|
|
|
2019-06-24 03:13:09 +00:00
|
|
|
async def write(self, bytes):
|
2018-06-18 03:21:02 +00:00
|
|
|
self.bytes_count += len(bytes)
|
|
|
|
if self.limit_bytes and (self.bytes_count > self.limit_bytes):
|
2019-05-04 02:15:14 +00:00
|
|
|
raise WriteLimitExceeded(
|
|
|
|
"CSV contains more than {} bytes".format(self.limit_bytes)
|
|
|
|
)
|
2019-06-24 03:13:09 +00:00
|
|
|
await self.writer.write(bytes)
|
2018-07-24 03:07:57 +00:00
|
|
|
|
|
|
|
|
|
|
|
_infinities = {float("inf"), float("-inf")}
|
|
|
|
|
|
|
|
|
|
|
|
def remove_infinites(row):
|
|
|
|
if any((c in _infinities) if isinstance(c, float) else 0 for c in row):
|
2019-05-04 02:15:14 +00:00
|
|
|
return [None if (isinstance(c, float) and c in _infinities) else c for c in row]
|
2018-07-24 03:07:57 +00:00
|
|
|
return row
|
publish_subcommand hook + default plugins mechanism, used for publish heroku/now (#349)
This change introduces a new plugin hook, publish_subcommand, which can be
used to implement new subcommands for the "datasette publish" command family.
I've used this new hook to refactor out the "publish now" and "publish heroku"
implementations into separate modules. I've also added unit tests for these
two publishers, mocking the subprocess.call and subprocess.check_output
functions.
As part of this, I introduced a mechanism for loading default plugins. These
are defined in the new "default_plugins" list inside datasette/app.py
Closes #217 (Plugin support for datasette publish)
Closes #348 (Unit tests for "datasette publish")
Refs #14, #59, #102, #103, #146, #236, #347
2018-07-26 05:15:59 +00:00
|
|
|
|
|
|
|
|
|
|
|
class StaticMount(click.ParamType):
|
2019-11-26 02:31:42 +00:00
|
|
|
name = "mount:directory"
|
publish_subcommand hook + default plugins mechanism, used for publish heroku/now (#349)
This change introduces a new plugin hook, publish_subcommand, which can be
used to implement new subcommands for the "datasette publish" command family.
I've used this new hook to refactor out the "publish now" and "publish heroku"
implementations into separate modules. I've also added unit tests for these
two publishers, mocking the subprocess.call and subprocess.check_output
functions.
As part of this, I introduced a mechanism for loading default plugins. These
are defined in the new "default_plugins" list inside datasette/app.py
Closes #217 (Plugin support for datasette publish)
Closes #348 (Unit tests for "datasette publish")
Refs #14, #59, #102, #103, #146, #236, #347
2018-07-26 05:15:59 +00:00
|
|
|
|
|
|
|
def convert(self, value, param, ctx):
|
|
|
|
if ":" not in value:
|
|
|
|
self.fail(
|
|
|
|
'"{}" should be of format mountpoint:directory'.format(value),
|
2019-05-04 02:15:14 +00:00
|
|
|
param,
|
|
|
|
ctx,
|
publish_subcommand hook + default plugins mechanism, used for publish heroku/now (#349)
This change introduces a new plugin hook, publish_subcommand, which can be
used to implement new subcommands for the "datasette publish" command family.
I've used this new hook to refactor out the "publish now" and "publish heroku"
implementations into separate modules. I've also added unit tests for these
two publishers, mocking the subprocess.call and subprocess.check_output
functions.
As part of this, I introduced a mechanism for loading default plugins. These
are defined in the new "default_plugins" list inside datasette/app.py
Closes #217 (Plugin support for datasette publish)
Closes #348 (Unit tests for "datasette publish")
Refs #14, #59, #102, #103, #146, #236, #347
2018-07-26 05:15:59 +00:00
|
|
|
)
|
2019-07-11 16:13:19 +00:00
|
|
|
path, dirpath = value.split(":", 1)
|
|
|
|
dirpath = os.path.abspath(dirpath)
|
publish_subcommand hook + default plugins mechanism, used for publish heroku/now (#349)
This change introduces a new plugin hook, publish_subcommand, which can be
used to implement new subcommands for the "datasette publish" command family.
I've used this new hook to refactor out the "publish now" and "publish heroku"
implementations into separate modules. I've also added unit tests for these
two publishers, mocking the subprocess.call and subprocess.check_output
functions.
As part of this, I introduced a mechanism for loading default plugins. These
are defined in the new "default_plugins" list inside datasette/app.py
Closes #217 (Plugin support for datasette publish)
Closes #348 (Unit tests for "datasette publish")
Refs #14, #59, #102, #103, #146, #236, #347
2018-07-26 05:15:59 +00:00
|
|
|
if not os.path.exists(dirpath) or not os.path.isdir(dirpath):
|
|
|
|
self.fail("%s is not a valid directory path" % value, param, ctx)
|
|
|
|
return path, dirpath
|
2019-02-06 04:53:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
def format_bytes(bytes):
|
|
|
|
current = float(bytes)
|
|
|
|
for unit in ("bytes", "KB", "MB", "GB", "TB"):
|
|
|
|
if current < 1024:
|
|
|
|
break
|
|
|
|
current = current / 1024
|
|
|
|
if unit == "bytes":
|
|
|
|
return "{} {}".format(int(current), unit)
|
|
|
|
else:
|
|
|
|
return "{:.1f} {}".format(current, unit)
|
2019-06-24 03:13:09 +00:00
|
|
|
|
|
|
|
|
2019-12-29 18:48:13 +00:00
|
|
|
_escape_fts_re = re.compile(r'\s+|(".*?")')
|
|
|
|
|
|
|
|
|
|
|
|
def escape_fts(query):
|
|
|
|
# If query has unbalanced ", add one at end
|
|
|
|
if query.count('"') % 2:
|
|
|
|
query += '"'
|
|
|
|
bits = _escape_fts_re.split(query)
|
|
|
|
bits = [b for b in bits if b and b != '""']
|
|
|
|
return " ".join(
|
|
|
|
'"{}"'.format(bit) if not bit.startswith('"') else bit for bit in bits
|
|
|
|
)
|
|
|
|
|
|
|
|
|
2019-06-24 03:13:09 +00:00
|
|
|
class RequestParameters(dict):
|
|
|
|
def get(self, name, default=None):
|
|
|
|
"Return first value in the list, if available"
|
|
|
|
try:
|
|
|
|
return super().get(name)[0]
|
|
|
|
except (KeyError, TypeError):
|
|
|
|
return default
|
|
|
|
|
|
|
|
def getlist(self, name, default=None):
|
|
|
|
"Return full list"
|
|
|
|
return super().get(name, default)
|
2020-02-15 17:56:48 +00:00
|
|
|
|
|
|
|
|
|
|
|
class ConnectionProblem(Exception):
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
class SpatialiteConnectionProblem(ConnectionProblem):
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
def check_connection(conn):
|
|
|
|
tables = [
|
|
|
|
r[0]
|
|
|
|
for r in conn.execute(
|
|
|
|
"select name from sqlite_master where type='table'"
|
|
|
|
).fetchall()
|
|
|
|
]
|
|
|
|
for table in tables:
|
|
|
|
try:
|
|
|
|
conn.execute("PRAGMA table_info({});".format(escape_sqlite(table)),)
|
|
|
|
except sqlite3.OperationalError as e:
|
|
|
|
if e.args[0] == "no such module: VirtualSpatialIndex":
|
|
|
|
raise SpatialiteConnectionProblem(e)
|
|
|
|
else:
|
|
|
|
raise ConnectionProblem(e)
|
2020-04-02 19:30:53 +00:00
|
|
|
|
|
|
|
|
|
|
|
class BadMetadataError(Exception):
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
def parse_metadata(content):
|
|
|
|
# content can be JSON or YAML
|
|
|
|
try:
|
|
|
|
return json.loads(content)
|
|
|
|
except json.JSONDecodeError:
|
|
|
|
try:
|
|
|
|
return yaml.safe_load(content)
|
|
|
|
except yaml.YAMLError:
|
|
|
|
raise BadMetadataError("Metadata is not valid JSON or YAML")
|
2020-05-27 19:25:52 +00:00
|
|
|
|
|
|
|
|
|
|
|
def call_with_supported_arguments(fn, **kwargs):
|
|
|
|
parameters = inspect.signature(fn).parameters.keys()
|
|
|
|
call_with = []
|
|
|
|
for parameter in parameters:
|
|
|
|
if parameter not in kwargs:
|
2020-05-28 05:57:05 +00:00
|
|
|
raise TypeError(
|
|
|
|
"{} requires parameters {}, missing: {}".format(
|
|
|
|
fn, tuple(parameters), set(parameters) - set(kwargs.keys())
|
|
|
|
)
|
|
|
|
)
|
2020-05-27 19:25:52 +00:00
|
|
|
call_with.append(kwargs[parameter])
|
|
|
|
return fn(*call_with)
|