--dirs option for scanning directories for SQLite databases

[simonw]

Refs #417.

[simonw]

One design issue: how to pick neat unique names for database files in a file hierarchy?

Here's what I have so far:

datasette/datasette/app.py

Lines 231 to 237 in fe6f9e6

	self.add_database(
	str(filepath)
	.replace("../", "")
	.replace("/", "_")
	.replace(".db", ""),
	Database(self, filepath, is_mutable=True),
	)

For these files:

../travel-old.db
../sf-tree-history/trees.db
../library-of-congress/records-from-df.db

It made these names:

travel-old
sf-tree-history_trees
library-of-congress_records-from-df

Maybe this is good enough? Needs some tests.

[simonw]

This can take a LONG time to run, and at the moment it's blocking and prevents Datasette from starting up.

It would be much better if this ran in a thread, or an asyncio task. Probably have to be a thread because there's no easy async version of pathlib.Path.glob() that I've seen.

[simonw]

Another problem: if any of the found databases use SpatiaLite then Datasette will fail to start at all.

It should skip them instead.

The select * from sqlite_master check apparently isn't quite enough to catch this case.

[simonw]

I tried running the scan_dirs() method in a thread and got an interesting error while trying to load the homepage: RuntimeError: OrderedDict mutated during iteration

Makes sense - I had a thread that added an item to that dictionary right while the homepage was attempting to run this code:

datasette/datasette/views/index.py

Lines 24 to 27 in efa54b4

	async def get(self, request, as_format):
	databases = []
	for name, db in self.ds.databases.items():
	table_names = await db.table_names()

[simonw]

So I need to ensure the ds.databases data structure is manipulated in a thread-safe manner.

Mainly I need to ensure that it is locked during iterations over it, then unlocked at the end.

Trickiest part is probably ensuring there is a test that proves this is working - I feel like I got lucky encountering that RuntimeError as early as I did.

[simonw]

... or maybe I can cheat and wrap the access to self.ds.databases.items() in list(), so I'm iterating over an atomically-created list of those things instead? I'll try that first.

[simonw]

... cheating like this seems to work:

for name, db in list(self.ds.databases.items()):

Python built-in operations are supposedly threadsafe, so in this case I can grab a copy of the list atomically (I think) and then safely iterate over it.

Seems to work in my testing. Wish I could prove it with a unit test though.

[simonw]

Interesting new problem: hitting Ctrl+C no longer terminates the problem provided that scan_dirs() thread is still running.

https://stackoverflow.com/questions/49992329/the-workers-in-threadpoolexecutor-is-not-really-daemon has clues. The workers are only meant to exit when their worker queues are empty.

But... I want to run the worker every 10 seconds. How do I do that without having it loop forever and hence never quit?

[simonw]

https://gist.github.com/clchiou/f2608cbe54403edb0b13 might work.

[simonw]

It think the fix is to use an old-fashioned threading module daemon thread directly. That should exit cleanly when the program exits.

[simonw]

I've figured out how to tell if a database is safe to open or not:

select sql from sqlite_master where sql like 'CREATE VIRTUAL TABLE%';

This returns the SQL definitions for virtual tables. The bit after using tells you what they need.

Run this against a SpatiaLite database and you get the following:

CREATE VIRTUAL TABLE SpatialIndex USING VirtualSpatialIndex()
CREATE VIRTUAL TABLE ElementaryGeometries USING VirtualElementary()

Run it against an Apple Photos photos.db file (found with find ~/Library | grep photos.db) and you get this (partial list):

CREATE VIRTUAL TABLE RidList_VirtualReader using RidList_VirtualReaderModule
CREATE VIRTUAL TABLE Array_VirtualReader using Array_VirtualReaderModule
CREATE VIRTUAL TABLE LiGlobals_VirtualBufferReader using VirtualBufferReaderModule
CREATE VIRTUAL TABLE RKPlace_RTree using rtree (modelId,minLongitude,maxLongitude,minLatitude,maxLatitude)

For a database with FTS4 you get:

CREATE VIRTUAL TABLE "docs_fts" USING FTS4 (
    [title], [content], content="docs"
)

FTS5:

CREATE VIRTUAL TABLE [FARA_All_Registrants_fts] USING FTS5 (
                [Name], [Address_1], [Address_2],
                content=[FARA_All_Registrants]
            )

So I can use this to figure out all of the using pieces and then compare them to a list of known support ones.

[simonw]

Need to support .sqlite files here too

[simonw]

Also consider .sqlite - using pathlib .stem here would be better.

[simonw]

This should be configurable - 10 is an OK default, but cases like simonw/jupyterserverproxy-datasette-demo#2 where it's running in a Binder session would benefit from faster updates.

[simonw]

Here's the new utility function I should be using to verify database files that I find:

datasette/datasette/utils/__init__.py

Lines 773 to 787 in 6aa516d

	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)

[simonw]

While running it against a nested directory with a TON of databases I kept seeing errors like this:

Traceback (most recent call last):
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 121, in route_path
    return await view(new_scope, receive, send)
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 193, in view
    request, **scope["url_route"]["kwargs"]
  File "/Users/simonw/Dropbox/Development/datasette/datasette/views/index.py", line 58, in get
    tables[table]["num_relationships_for_sorting"] = count
KeyError: 'primary-candidates-2018/rep_candidates'

And

Traceback (most recent call last):
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 121, in route_path
    return await view(new_scope, receive, send)
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 193, in view
    request, **scope["url_route"]["kwargs"]
  File "/Users/simonw/Dropbox/Development/datasette/datasette/views/index.py", line 58, in get
    tables[table]["num_relationships_for_sorting"] = count
KeyError: 'space_used'

[simonw]

I also eventually get this error:

Traceback (most recent call last):
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 121, in route_path
    return await view(new_scope, receive, send)
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 336, in inner_static
    await asgi_send_file(send, full_path, chunk_size=chunk_size)
  File "/Users/simonw/Dropbox/Development/datasette/datasette/utils/asgi.py", line 303, in asgi_send_file
    async with aiofiles.open(str(filepath), mode="rb") as fp:
  File "/Users/simonw/.local/share/virtualenvs/datasette-oJRYYJuA/lib/python3.7/site-packages/aiofiles/base.py", line 78, in __aenter__
  File "/Users/simonw/.local/share/virtualenvs/datasette-oJRYYJuA/lib/python3.7/site-packages/aiofiles/threadpool/__init__.py", line 35, in _open
  File "/usr/local/Cellar/python/3.7.5/Frameworks/Python.framework/Versions/3.7/lib/python3.7/concurrent/futures/thread.py", line 57, in run
OSError: [Errno 24] Too many open files: '/Users/simonw/Dropbox/Development/datasette/datasette/static/app.css'

[simonw]

I think I can tell what the current file limit is like so:

In [1]: import resource                                                                                                                                                                   

In [2]: resource.getrlimit(resource.RLIMIT_NOFILE)                                                                                                                                        
Out[2]: (256, 9223372036854775807)

So maybe I should have Datasette refuse to open more database files than that number minus 5 (to give me some spare room for opening CSS files etc).