issues

Thanks again for such a powerful project.

For deploying to cloudrun from github actions, I'd like to create a service account with minimal permissions.

It would be great to document what those minimum permission that need to be set in the IAM.

I've been thinking about what it might look like to allow command-line arguments to be used to define any of the configuration options in datasette.yml, as alternative and more convenient syntax.

Here's what I've come up with: datasette \ -s settings.sql_time_limit_ms 1000 \ -s plugins.datasette-auth-tokens.manage_tokens true \ -s plugins.datasette-auth-tokens.manage_tokens_database tokens \ mydatabase.db tokens.db Which would be equivalent to datasette.yml containing this: yaml plugins: datasette-auth-tokens: manage_tokens: true manage_tokens_database: tokens settings: sql_time_limit_ms: 1000 More details in https://github.com/simonw/datasette/issues/2143#issuecomment-1690792514

Related: - #262

I've made the change to the table view, now I need the new format to work for arbitrary SQL queries too.

Note that this incorporates both arbitrary SQL queries and canned queries.

Using nginx to redirect public IP to the local uvicorn server as 'normal'. I can't figure out how to redirect such that '/' results in accessing the one db/table I want to serve; redirecting / to /db/table breaks some of the CSS; fooling with base_url doesn't seem to help. Can someone explain this, if it's possible?

From https://docs.datasette.io/en/stable/plugin_hooks.html#render-cell-row-value-column-table-database-datasette:

row - sqlite.Row The SQLite row object that the value being rendered is part of

This appears to actually be a CustomRow, but I think that's unrelated to my issue.

I have a table:

sql CREATE TABLE IF NOT EXISTS "dss_job_stats"( job_id integer not null references dss_job(id) on delete cascade, host text not null, // other columns elided as irrelevant primary key (job_id, host) );

On datasette 0.63.2, the render_cell hook receives a row value that looks like:

CustomRow([('job_id', {'value': 2, 'label': '2'}), ('host', 'cldellow.com')])

I expected the job_id value to be 2, but it's actually {'value': 2, 'label': '2'}.

I can work around this, but was wondering if this was intended behaviour?

I have linked Python3.11 on macOS against recent SQLite that was compiled using -DSQLITE_DQS=0. This option disables interpretation of double-quoted identifiers as string literals, described in the SQLite docs as a "MySQL 3.x misfeature". See https://www.sqlite.org/quirks.html#dblquote for background.

Datasette uses the double-quote syntax in a number of key places, and is thus completely broken in this environment.

My experience was to pip install datasette, then run datasette serve -I my-data.db. When I visit http://127.0.0.1:8001 I get a 500 response.

The error: sqlite3.OperationalError: no such column: geometry_columns

The responsible SQL: 'select 1 from sqlite_master where tbl_name = "geometry_columns"'

I then installed datasette from GitHub master in development mode and changed the offending SQL to use correct quotes: "select 1 from sqlite_master where tbl_name = 'geometry_columns'".

With this change, I get a little further, but have the same problem with the first table name in my database (in my case, "Meta"): OperationalError: no such column: Meta Traceback (most recent call last): File "/Users/gwk/external/datasette/datasette/app.py", line 1522, in route_path response = await view(request, send) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/views/base.py", line 151, in view return await self.dispatch_request(request) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/views/base.py", line 105, in dispatch_request response = await handler(request) ^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/views/index.py", line 70, in get "fts_table": await db.fts_table(table), ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/database.py", line 363, in fts_table return await self.execute_fn(lambda conn: detect_fts(conn, table)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/database.py", line 213, in execute_fn return await asyncio.get_event_loop().run_in_executor( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/py/Python.framework/Versions/3.11/lib/python3.11/concurrent/futures/thread.py", line 58, in run result = self.fn(*self.args, **self.kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/database.py", line 211, in in_thread return fn(conn) ^^^^^^^^ File "/Users/gwk/external/datasette/datasette/database.py", line 363, in <lambda> return await self.execute_fn(lambda conn: detect_fts(conn, table)) ^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/gwk/external/datasette/datasette/utils/__init__.py", line 588, in detect_fts rows = conn.execute(detect_fts_sql(table)).fetchall() ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ sqlite3.OperationalError: no such column: Meta INFO: 127.0.0.1:50258 - "GET / HTTP/1.1" 500 Internal Server Error

I will try to continue playing with this, but I also hope that the datasette developers will enable this mode in a test environment as I am unlikely to be able to exercise all of the SQL in the codebase, or make a pull request very soon.

Note that the DQS setting compile-time option can be overridden at runtime with calls to the C API: sqlite3_db_config(db, SQLITE_DBCONFIG_DQS_DDL, 0, (void*)0); sqlite3_db_config(db, SQLITE_DBCONFIG_DQS_DML, 0, (void*)0);

As far as I can tell, sqlite3_db_config is not exposed in Python, but perhaps we could figure out how to invoke it using ctypes.

I hadn't run into an issue with plugins like datasette-template-sql interfering with the existing context for other features before! Definitely not a good thing.

Originally posted by @simonw in https://github.com/simonw/datasette-write/issues/6#issuecomment-1379490596

There are situations - such as loading SQLite files remotely using HTTP range headers - where counting all of the rows in a table should be avoided entirely.

Also, this chunked inefficiency means that I have to hack the URL to not load tables of a database as it seems to try to load the whole database when I click on a database.

I bet that's because Datasette tries to show a count of all of the rows in each table when it shows the list on that page, which triggers a full table scan.

Would be great to have a setting that turns that feature off, which could then be exposed as a query string option for Datasette Lite.

Originally posted by @simonw in https://github.com/simonw/datasette-lite/issues/49#issuecomment-1256880715

Motivation

• I have a directory of sqlite databases
• I'd like to use immutable mode when opening them for better performance docs

• Currently using this flag throws the following error

  IsADirectoryError: [Errno 21] Is a directory: '/name-of-directory'

Proposal

Immutable flag works for both single files and directories

datasette -i /folder-of-sqlite-files

Datasette maintains an in-memory SQLite database with details of the the databases, tables and columns for all of the attached databases.

The above is from the docs ^. There's two problems here - the number of datasette "instances" in a single server/VM and the size of the database itself. We want the opposite of in-memory, including what happens on SQLlite - documented in https://www.sqlite.org/inmemorydb.html

From the context in https://github.com/simonw/datasette/issues/1150 - does it mean datasette is memory-bound to the size of the dataset - which might be a deal-breaker for many large-scale use cases?

In an extreme case - let's say a single server had 100 SQLlite databases, which would enable 100 "instances" of datasette to run, one per client (e.g. in a SaaS multi-tenant environment). How could we achieve all these goals:

1. Allow any one of these 100 databases to grow to say 2Tb in size
2. Have one datasette instance, which connects to 1 of the 100 instances, based on incoming credentials/tenant ID
3. Minimize memory use entirely - both by datasette and SQLlite, such that almost all operations are executed in real-time on-disk with little to no memory consumption per-tenant, or per-database.

Any ideas appreciated - we're looking to use this in a SaaS type of setting - many instances, single server.

@simonw great work on datasette, in general! Possibly related to https://github.com/simonw/datasette/issues/1480 but we don't want use any kind of serverless infra - this is a long-running VM/server.

Is it possible to deploy a datasette application as a python web app?

In my enterprise, I have option to deploy python apps via Rstudio Connect, and I would like to publish a datasette dashboard for sharing.

I welcome any pointers to converting datasette serve into a python app that can be run as something like python datasette.py --my_data.db

I just watched a demo video for the JupyterLab Chart Editor which wraps the plotly chart editor app in a JupyterLab panel and lets you open a plotly chart JSON file in that editor. Essentially, it pops an HTML app into a panel in JupyterLab, and I think registers the app as a file viewer for a particular file type. (I'm not completely taken by it, tbh, because it means you can do irreproducible things to the chart definition file, but that's another issue).

JupyterLab extensions can also open files from a dialogue as the iframe/html previewer shows: https://github.com/timkpaine/jupyterlab_iframe.

This made me wonder about what datasette integration with JupyterLab might do.

For example, by right-clicking on a CSV file (for which there is already a CSV table view) in the file browser, offer a View / Run as datasette file viewer option that will:

• run the CSV file through csvs-to-sqlite;
• launch the datasette server and display the datasette view in a JupyterLab panel.

(? Create a new SQLite db for each CSV file and launch each datasette view on a new port? Or have a JupyterLab (session?) SQLite db that stores all datasette viewed CSVs and runs on a single port?)

As a freebie, the datasette API would allow you to run efficient SQL queries against the file eg using using pandas.read_sql() queries in a notebook in the same space.

Related:

• JupyterLab extensions docs
• a cookiecutter for wrting JupyterLab extensions using Javascript
• a cookiecutter for writing JupyterLab extensions using Typescript
• tutorial: Let’s Make an xkcd JupyterLab Extension

Many Datasette instances mainly exist to serve a single table - for example:

• https://global-power-plants.datasettes.com/global-power-plants/global-power-plants
• https://laion-aesthetic.datasette.io/laion-aesthetic-6pls/images

It would be neat if the / homepage of those instances could be configured to highlight that specific table.

Or maybe more than one?

Wide tables aren't a huge problem if you know to click and drag right. But it's not at all obvious to do that. (it also tends to blue-select any content as it's dragging.) Depending on column widths, public users might entirely miss all the columns to the right.

There is a scrollbar at the bottom of the table, but I'm displaying ALL my records because it's the only way for datasette-vega to make accurate charts. So that bottom scrollbar is likely to be missed. I wonder if some sort of javascript-y mouseover to an arrow might help, similar to those seen in image carousels. Ah: here's a perfect example:

1. Visit http://google.com
2. Search for: animals endangered
3. Note the 'g-right-button' (in the code) that looks like a right-facing caret in a circle.
4. Click on that and the carousel scrolls right (and 'g-left-button' appears on the left).

Might be tricky to do that on a table, rather than a one-row carousel, but it's worth experimenting with.

Another option is just to put the scrollbars at the top of the table, too.

Meantime, I'm trying to build a button like the "View/hide all columns on https://salaries.news.baltimoresun.com/salaries-be494cf/2019+Maryland+state+salaries Might be nice to have that available by default, with settings in the metadata showing which are on by default.

(I saw some other closed issues related to horizontal scrolling, and admit I don't entirely understand them. For instance, the animated gif at https://github.com/simonw/datasette/issues/998#issuecomment-714117534 confuses me. )

To solve #241 JSON logging.

view-source:https://calands.datasettes.com/calands/CPAD_2020a_SuperUnits.geojson?_sort=id&id__exact=4&_labels=on&_trace=1 is showing me a blank page.

I realized that one of the reasons I'm having trouble committing to nailing down the JSON API for 1.0 is that I don't use it much myself - I use the ?_shape=array one quite often, but I don't have any projects that are using the default, more fully-featured API.

As an experiment I built a Web Component for embedding Datasette tables on pages - https://github.com/simonw/datasette-table - and I think it's actually going to be a really useful tool for helping me dog food the v1.0 API design.

The generated Dockerfile currently looks something like this: ```Dockerfile FROM python:3.8 COPY . /app WORKDIR /app

ENV DATASETTE_SECRET 'edab49cbc5d5f6f33238f54852037e3fee710821960b73edd2ce743454182ae2' RUN pip install -U datasette datasette-auth-passwords datasette-tiddlywiki datasette-graphql RUN datasette inspect fixtures.db other.db --inspect-file inspect-data.json ENV PORT 8080 EXPOSE 8080 CMD datasette serve --host 0.0.0.0 -i fixtures.db -i other.db --cors --inspect-file inspect-data.json --metadata metadata.json --create --port $PORT /data/*.db `` This is still on Python 3.8, and it generates a pretty large image compared to theDockerfile` used for https://hub.docker.com/datasetteproject/datasette - https://github.com/simonw/datasette/blob/0.60.2/Dockerfile

Here's the code that generates it: https://github.com/simonw/datasette/blob/7d24fd405f3c60e4c852c5d746c91aa2ba23cf5b/datasette/utils/init.py#L389-L400

What I'm seeing

With default_cache_ttl = 86400, I see the following:

A table view returns Cache-control: max-age=86400:

A static asset returns no Cache-control header:

What I expected to see

I expected the static asset to return a Cache-control header indicating that this response can be cached.

Why this matters

I'm productionising a Datasette deployment right now and was looking into putting it behind a Varnish instance. I was surprised to see requests for static assets being served from Datasette rather than Varnish, this is what led me to look more closely at the response headers.

While Datasette serves those static assets pretty quickly, I don't see why Datasette should serve them. By their nature, static assets like images and JS files are very cacheable, so it should be easy to serve them from a cache like Varnish.

(Note that Varnish can easily be configured to override this header, enabling caching for static assets. But it would be better if this override was not necessary.)

Discussion

It seems clear to me that serving static assets without a Cache-control header is not ideal.

I see two options here:

A. Static assets use the same logic as table / SQL views to set the Cache-control header based on default_cache_ttl. B. An additional setting for static assets is introduced (default_static_cache_ttl, say).

Inspired by this article: https://cldellow.com/2018/06/22/sqlite-parquet-vtable.html#sqlite-database-indexed--squashed

https://en.wikipedia.org/wiki/SquashFS is "a compressed read-only file system for Linux" - which means it could be a really nice fit for Datasette and its read-only SQLite databases.

It would be interesting to explore a Dockerfile recipe that used SquashFS to compress the SQLite database file that was bundled up by datasette package and friends.

https://docs.datasette.io/en/0.55/spatialite.html was written before I had tools like geojson-to-sqlite and shapefile-to-sqlite.

Found this while working on https://github.com/simonw/datasette-tiddlywiki

Then clicking on /tiddlywiki/tiddlers/%24%3A%2FDefaultTiddlers returns a 404.

https://docs.datasette.io/en/stable/sql_queries.html#named-parameters

I thought that I had seen an example of how to do this example below, but I can't seem to find it

sql select * from bib where bib.bib_record_num in (1008088,1008092)

sql select * from bib where bib.bib_record_num in (:bib_record_numbers)

https://ilsweb.cincinnatilibrary.org/collection-analysis/current_collection-204d100?sql=select%0D%0A++*%0D%0Afrom%0D%0A++bib%0D%0Awhere%0D%0A++bib.bib_record_num+in+%28%3Abib_record_numbers%29&bib_record_numbers=1008088%2C1008092

Or, maybe this isn't a fully supported feature.

It would be great to have a sample showing how to move a single database that has a single table, to the index page. I'm trying it now, and find there is a real depth of Datasette and Python understanding that's required to be successful.

I've got all the basic jinja concepts down... variables, template control structures, template inheritance, template overrides, css, html, the --template-dir and --static arguments, etc.

But copying the table.html file to index.html doesn't work. There are undocumented functions and filters... I can figure some of them out (yay, url_builder.py and utils/init.py!) but it's a slog better handled by a much stronger Python developer.

One sample would make a world of difference. The ideal form of this documentation would be a diff between the default table.html and how that would look if essentially moved to index.html. The use case is for everyone who wants to create a public-facing website to explore a single table at the root directory. (Maybe a second bit of documentation for people who have a single database with multiple tables.)

(Hmm... might be cool to have a setting for that, where it happens automagically! If only one table, then home page is at the table level. if only one database, then home page is at the database level.... as an option.)

I suppose I could ignore this, and somehow do this in the DNS settings once I hook up Vercel to a domain name, maybe.. and remove the breadcrumbs in table.html... but for now, a documentation request in the form of a diff... for viewing a single table (or a single database) at the root.

(Actually, there's probably room for a whole expanded section on templates. Noticed some nice table metadata in one of the datasette examples, for instance... Hmm... maybe a whole library of solutions in one place... maybe a documentation hackathon! If that's of interest, of course it's a separate issue. )

Not currently covered here: https://docs.datasette.io/en/stable/internals.html#await-db-execute-sql

This PR changes the main Dockerfile to use ubuntu:20.10 as base image instead of python:3.9.2-slim-buster (itself based on debian:buster-slim).

The Dockerfile is essentially the one from https://github.com/simonw/datasette/issues/1249#issuecomment-803698983 with some additional cleanups to slim it down.

This fixes a couple of issues: 1. The SQLite version in Debian Buster (2.6.0) doesn't support generated columns 2. Installing SpatiaLite from the Debian sid repositories has the side effect of also installing updates to libc and libstdc++ from sid.

As a bonus, the Docker image becomes smaller:

$ docker image ls REPOSITORY TAG IMAGE ID CREATED SIZE datasette 0.56-ubuntu f7aca255140a 5 hours ago 212MB datasetteproject/datasette 0.56 efb3b282f390 13 days ago 258MB

Reproduction of the first issue

``` $ curl -O https://latest.datasette.io/fixtures.db % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 260k 0 260k 0 0 489k 0 --:--:-- --:--:-- --:--:-- 489k

$ docker run -v pwd:/mnt datasetteproject/datasette:0.56 datasette /mnt/fixtures.db Traceback (most recent call last): File "/usr/local/bin/datasette", line 8, in <module> sys.exit(cli()) File "/usr/local/lib/python3.9/site-packages/click/core.py", line 829, in call return self.main(args, kwargs) File "/usr/local/lib/python3.9/site-packages/click/core.py", line 782, in main rv = self.invoke(ctx) File "/usr/local/lib/python3.9/site-packages/click/core.py", line 1259, in invoke return _process_result(sub_ctx.command.invoke(sub_ctx)) File "/usr/local/lib/python3.9/site-packages/click/core.py", line 1066, in invoke return ctx.invoke(self.callback, ctx.params) File "/usr/local/lib/python3.9/site-packages/click/core.py", line 610, in invoke return callback(args, kwargs) File "/usr/local/lib/python3.9/site-packages/datasette/cli.py", line 544, in serve asyncio.get_event_loop().run_until_complete(check_databases(ds)) File "/usr/local/lib/python3.9/asyncio/base_events.py", line 642, in run_until_complete return future.result() File "/usr/local/lib/python3.9/site-packages/datasette/cli.py", line 584, in check_databases await database.execute_fn(check_connection) File "/usr/local/lib/python3.9/site-packages/datasette/database.py", line 155, in execute_fn return await asyncio.get_event_loop().run_in_executor( File "/usr/local/lib/python3.9/concurrent/futures/thread.py", line 52, in run result = self.fn(*self.args, self.kwargs) File "/usr/local/lib/python3.9/site-packages/datasette/database.py", line 153, in in_thread return fn(conn) File "/usr/local/lib/python3.9/site-packages/datasette/utils/init.py", line 892, in check_connection for r in conn.execute( sqlite3.DatabaseError: malformed database schema (generated_columns) - near "AS": syntax error ```

Here is the SQLite version:

`` $ docker run -vpwd`:/mnt -it datasetteproject/datasette:0.56 /bin/bash root@d9220d3b95dd:/# python3 Python 3.9.2 (default, Mar 27 2021, 02:50:26) [GCC 8.3.0] on linux Type "help", "copyright", "credits" or "license" for more information.

import sqlite3 sqlite3.version '2.6.0' ```

Reproduction of the second issue

$ docker build . -t datasette --build-arg VERSION=0.55 [...snip...] The following packages will be upgraded: libc-bin libc6 libstdc++6 [...snip...] Unpacking libc6:amd64 (2.31-11) over (2.28-10) ... [...snip...] Unpacking libstdc++6:amd64 (10.2.1-6) over (8.3.0-6) ... [...snip...]

Both libc and libstdc++ are backwards compatible, so the image still works, but it will result in a combination of libraries and Python versions that exists only in the Datasette image, so it's likely untested. In addition, since Debian sid is an always-changing rolling-release, the versions of libc, libstdc++, Spatialite, and their dependencies change frequently, so the library versions in the Datasette image will depend on the day when it was built.

Fixes https://github.com/simonw/datasette/issues/1129

It uses view_name of database at the moment which isn't as useful.

Hi,

I've been experimenting with SQLite reading from huge datasets using this excellent Parquet extension from @cldellow. https://cldellow.com/2018/06/22/sqlite-parquet-vtable.html https://github.com/cldellow/sqlite-parquet-vtable

This works really well, but I was keen to see if I could combine datasette with this. Having previously experimented with the spatialite extension I knew that datasette supports loading extensions in the underlying sqlite instance. However I hit a blocker as the current design only allows SELECT statements to be executed and so I am unable to execute the crucial

CREATE VIRTUAL TABLE .........

command that is required to load the data from the parquet file into the table.

It seems like this would be a simple-ish change, but I don't know enough about the architecture of datasette to start implementing this myself? Could this be done as a datasette plugin? or would this require more fundamental changes at initialisation time?

My thoughts are that something at init time could detect that the user was loading a .parquet file and then switch to a mode were it loads that via the "CREATE VIRTUAL TABLE..." rather than loading the .db file in the default case??

I'm happy to contribute code and testing, I just need some pointers on the best approach.

Thanks Darren

Split out from #460 - in order to support large numbers of connected databases the homepage needs to be paginated.

One of the nicest qualities of SQLite as a data format is that you get a single file which you can then backup or share with other people.

Datasette breaks this a little once you start including custom metadata.json or template files and CSS.

It would be cool if there was an optional mechanism for baking that extra configuration into the SQLite file itself. That way entire datasette mini-applications (including canned queries and custom HTML and CSS) could be constructed as single .db files.

Since datasette configuration is all file-based, one way to achieve that would be to support a "datasette_files" table which, if present is used to search for file contents by path.

This is inline with the philosophy described by https://www.sqlite.org/appfileformat.html

When JSON fields are rendered in a jinja2 template, it is handy to be able to manipulate them as data (e.g., iterate over an array of values).

Ansible has a "from_json" function, which just called json.loads. It's a trivial as a datasette plugin, but it seems generally useful. Does it makes sense to add it directly into the app?

https://github.com/simonw/datasette/blob/master/docs/ecosystem.rst

Potential fix for https://github.com/simonw/datasette/issues/859.

Disabling table counts for hidden tables speeds up database page quite a bit. In my setup it reduced load time by 2/3 (~300 -> ~90ms)

We need a much smarter way of handling database connections.

Today, connections are simple: Datasette runs a number of threads (defaults to 3) and each thread gets a threadlocal read-only (or immutable) connection to each attached database - opened on demand.

For Datasette Library (#417) I want to support potentially hundreds of attached databases. Datasette Edit (#567) is going to introduce a need for writable connections too.

I'd also like to be able to run joins across multiple databases (#283) which further complicates things.

Supporting thousands of open SQLite connections at once feels like it won't provide good enough performance (though I should benchmark that to be sure). Some kind of connection pooling is likely to be necessary.