It is increasingly being useful configuration options, when it started out as purely metadata.

Could cause confusion with the --config mechanism though - maybe that should be called "settings" instead?

Idea is to have my existing scraper at https://github.com/simonw/scrape-hacker-news-by-domain also write to my private Datasette Cloud account, then create an atom feed from it.

Related: - #1768

I think this base image looks right: 3.11.0-slim-bullseye

The getting started docs currently suggest you try this to browse your Chrome history:

datasette ~/Library/Application\ Support/Google/Chrome/Default/History

But if Chrome is running you will likely get this error:

sqlite3.OperationalError: database is locked

Turns out there's a workaround for this which I just spotted on the SQLite forum:

You can do this using a URI filename: sqlite3 'file:places.sqlite?mode=ro&nolock=1' That opens the file places.sqlite in read-only mode with locking disabled. This isn't safe, in that changes to the database made by other corrections are likely to cause this connection to return incorrect results or crash. Read-only mode should at least mean that you don't corrupt the database in the process.

I'm going to refactor this stuff out and document it so it can be easily used by plugins:

https://github.com/simonw/datasette/blob/4a4164b81191dec35e423486a208b05a9edc65e4/datasette/views/base.py#L69-L103

Originally posted by @simonw in https://github.com/simonw/datasette/issues/1660#issuecomment-1074136176

Refs #1661.

Refs #1439.

• [x] Build dash_encode / dash_decode functions
• [x] Use dash encoding for row primary keys
• [x] Use dash encoding for ?_next= pagination tokens
• [x] Use dash encoding for table names in URLs
• [x] Use dash encoding for database name
• ~~Implement redirects from previous % URLs that replace those with -~~ - separate issue: #1650

There are only a few classes that plugins need to import. It would be nice if these imports were as short and memorable as possible.

For example: python from datasette.app import Datasette from datasette.utils.asgi import Response Could both become: python from datasette import Datasette from datasette import Response

Similar bug to #1525 (and #1506 before it). Start on https://latest.datasette.io/fixtures/facetable?_facet=_neighborhood - then select a neighborhood - then try to remove that filter using the little "x" and submitting the form again.

Every single piece of code I've written against the write APIs has used the block=True option to wait for the result.

Without that, it instead fires the write into the queue but then continues even before it has finished executing.

block=True should clearly be the default behaviour here!

Refs #878 - starting out with the new AsyncBase class implementing a pytest-inspired asyncio parallel execution mechanism.

Clicking "Hide this column" on fips on https://covid-19.datasettes.com/covid/ny_times_us_counties shows this error:

https://covid-19.datasettes.com/covid/ny_times_us_counties?_nocol=fips

Invalid SQL

no such column: fips

The reason is that https://covid-19.datasettes.com/-/metadata sets up the following:

json "ny_times_us_counties": { "sort_desc": "date", "facets": [ "state", "county", "fips" ], It's setting fips as a default facet, which breaks if you attempt to remove the column using ?_nocol.

Original title: "More" link for facets that shows the full set of results

The simplest way to do this will be to have it link to a generated SQL query.

Originally posted by @simonw in https://github.com/simonw/datasette/issues/1332#issuecomment-846479062

Don't think this has been covered here yet. I'm a little stumped with this one and can't tell if it's a bug or I have something misconfigured.

Oddly, when running locally the usual list of tables populates (i.e. at /charts a list of tables in charts.db). But when on the web server it throws an Invalid SQL error and "no such table: pragma_database_list" below.

All the url endpoints seem to work fine aside from this - individual tables (/charts/chart_one), as well as stored queries (/charts/query_one).

Not sure if this has anything to do with upgrading to Datasette 0.55, or something to do with our setup, which uses a metadata build script similar to the one for the 538 server, or something else.

Spotted this while testing https://github.com/simonw/datasette/issues/1249#issuecomment-808998719_

https://hub.docker.com/r/datasetteproject/datasette/tags?page=1&ordering=last_updated isn't showing the tags for any version more recent than 0.54.1 - we are up to 0.56 now.

But the :latest tag is for the new 0.56 release.

Feature request: Would it be possible to add an "edit this query" button on canned queries? Clicking it would open the canned query as an editable sql query. I think the intent is to have named parameters to allow this, but sometimes you just gotta rewrite it?

Refs #1023 datasette % git grep '"/"' -- '*.html' datasette/templates/error.html: <a href="/">home</a> datasette/templates/patterns.html: <a href="/">home</a> / datasette/templates/query.html: <a href="/">home</a> /

Several plugin hooks now expect plugins to return data in a specific shape - notably the new output format hook and the custom facet hook.

These use Python dictionaries right now but that's quite error prone: it would be good to have a mechanism that supported a more structured format.

Full list of current hooks is here: https://datasette.readthedocs.io/en/latest/plugins.html#plugin-hooks

dbstat is a table that is usually available on SQLite giving statistics about the database. For example:

https://fivethirtyeight.datasettes.com/fivethirtyeight?sql=SELECT+*+FROM+%22dbstat%22+WHERE+name%3D%27bachelorette%2Fbachelorette%27%3B

| name | path | pageno | pagetype | ncell | payload | unused | mx_payload | pgoffset | pgsize | |---------------------------|--------|----------|------------|---------|-----------|----------|--------------|------------|----------| | bachelorette/bachelorette | / | 89 | internal | 13 | 0 | 3981 | 0 | 360448 | 4096 | | bachelorette/bachelorette | /000/ | 91 | leaf | 66 | 3792 | 32 | 74 | 368640 | 4096 | | bachelorette/bachelorette | /001/ | 92 | leaf | 67 | 3800 | 14 | 74 | 372736 | 4096 | | bachelorette/bachelorette | /002/ | 93 | leaf | 65 | 3717 | 46 | 70 | 376832 | 4096 | | bachelorette/bachelorette | /003/ | 94 | leaf | 68 | 3742 | 6 | 71 | 380928 | 4096 | | bachelorette/bachelorette | /004/ | 95 | leaf | 70 | 3696 | 42 | 66 | 385024 | 4096 | | bachelorette/bachelorette | /005/ | 96 | leaf | 69 | 3721 | 22 | 71 | 389120 | 4096 | | bachelorette/bachelorette | /006/ | 97 | leaf | 70 | 3737 | 1 | 72 | 393216 | 4096 | | bachelorette/bachelorette | /007/ | 98 | leaf | 69 | 3728 | 15 | 69 | 397312 | 4096 | | bachelorette/bachelorette | /008/ | 99 | leaf | 73 | 3715 | 8 | 64 | 401408 | 4096 | | bachelorette/bachelorette | /009/ | 100 | leaf | 73 | 3705 | 18 | 62 | 405504 | 4096 | | bachelorette/bachelorette | /00a/ | 101 | leaf | 75 | 3681 | 32 | 62 | 409600 | 4096 | | bachelorette/bachelorette | /00b/ | 102 | leaf | 77 | 3694 | 9 | 62 | 413696 | 4096 | | bachelorette/bachelorette | /00c/ | 103 | leaf | 74 | 3673 | 45 | 62 | 417792 | 4096 | | bachelorette/bachelorette | /00d/ | 104 | leaf | 5 | 228 | 3835 | 48 | 421888 | 4096 |

Other than direct select * from dbsat queries it is completely invisible.

It would be cool if https://fivethirtyeight.datasettes.com/fivethirtyeight/dbstat didn't 404 (on databases for which that table was available).

https://github.com/simonw/datasette/runs/1043709494?check_suite_focus=true

• [x] This step should not run if a release is an alpha or beta
• [x] When it DOES run it should work
• [x] See it work for both an alpha and a non-alpha release, then close this ticket

e.g. https://travis-ci.org/github/simonw/datasette/jobs/717123725

Here's how it broke: --2020-08-12 03:08:17-- https://www.gaia-gis.it/gaia-sins/freexl-1.0.5.tar.gz Resolving www.gaia-gis.it (www.gaia-gis.it)... 212.83.162.51 Connecting to www.gaia-gis.it (www.gaia-gis.it)|212.83.162.51|:443... connected. HTTP request sent, awaiting response... 404 Not Found 2020-08-12 03:08:18 ERROR 404: Not Found. The command '/bin/sh -c wget "https://www.gaia-gis.it/gaia-sins/freexl-1.0.5.tar.gz" && tar zxf freexl-1.0.5.tar.gz && cd freexl-1.0.5 && ./configure && make && make install' returned a non-zero code: 8 The command "docker build -f Dockerfile -t $REPO:$TRAVIS_TAG ." exited with 8. 0.07s$ docker tag $REPO:$TRAVIS_TAG $REPO:latest Error response from daemon: No such image: [secure]/datasette:0.47.1 The command "docker tag $REPO:$TRAVIS_TAG $REPO:latest" exited with 1. 0.08s$ docker push $REPO The push refers to repository [docker.io/[secure]/datasette] An image does not exist locally with the tag: [secure]/datasette The command "docker push $REPO" exited with 1. cache.2 store build cache

Lots of detail in https://github.com/simonw/datasette-search-all/issues/2 - short version is that I have a plugin with its own index.html template and Datasette intermittently fails to load it and uses the default index.html that ships with Datasette instead.

Related:

• 689: "Templates considered" comment broken in >=0.35

• 693: Variables from extra_template_vars() not exposed in _context=1 (may as well fix this while I'm in there)

It's 3.26.0 at the moment: https://github.com/simonw/datasette/blob/af9cd4ca64652fae262e6f7b5d201f6e0adc989b/Dockerfile#L9-L11 Most recent release is 3.31.1: https://www.sqlite.org/releaselog/3_31_1.html

Sometimes a plugin will need to render a template for some custom UI. We need a documented API for doing this, which ensures that everything will work correctly if you extend base.html etc.

See also #576. This could be a .render() method on the Datasette class, but that feels a bit weird - should that class also take responsibility for rendering?

Cool project!

This fixes #136 using the suggested sql formatter library. I included the minified version in the bundle and added the relevant scripts to the codemirror includes instead of adding new files, though I could also add new files. I wanted to keep it all together, since the result of the format needs access to the editor in order to properly update the codemirror instance.

On attempting install on a Win 7 PC with py 3.6.2 (Anaconda dist) I get the error:

Collecting uvloop>=0.5.3 (from Sanic==0.7.0->datasette) Downloading uvloop-0.9.1.tar.gz (1.8MB) 100% |¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦| 1.8MB 12.8MB/s Complete output from command python setup.py egg_info: Traceback (most recent call last): File "<string>", line 1, in <module> File "C:\Users\RCole\AppData\Local\Temp\pip-build-juakfqt8\uvloop\setup.py ", line 10, in <module> raise RuntimeError('uvloop does not support Windows at the moment') RuntimeError: uvloop does not support Windows at the moment

Only available if arbitrary SQL is enabled (the default).

?_where=id in (1,2,3)&_where=id in (select tag_id from tags)

Allows any table (or view) page to have arbitrary additional extra_where clauses defined using the URL!

This would be extremely useful for building JavaScript applications against the Datasette API that only need on extra tiny bit of SQL but still want to benefit from other table view features like faceting.

Would be nice if this could take :named parameters and have them filled in via querystring as well.

https://sebest.github.io/post/using-travis-ci-to-build-docker-images/

The generated Docker images can then just inherit from that. This will speed up deploys as no need to pip install anything.

• [x] Ship that image to Docker Hub
• [ ] Update the generated Dockerfile to use it

Imagine if this:

https://stateless-datasets-jykibytogk.now.sh/flights-07d1283/airports.jsono?country__contains=gu&_group_count=country

Turned into this:

https://stateless-datasets-jykibytogk.now.sh/flights-07d1283?sql=select%20country,%20count(*)%20as%20group_count_country%20from%20airports%20where%20country%20like%20%27%gu%%27%20group%20by%20country%20order%20by%20group_count_country%20desc

This would involve introducing a new precedent of query string arguments that start with an _ having special meanings. While we're at it, could try adding _fields=x,y,z

Tasks:

• [x] Get initial version working
• [ ] Refactor code to not just "pretend to be a view"
• [ ] Get foreign key relationships expanded

Consider https://datasette-issue-190-compound-pks.now.sh/compound-pks-9aafe8f/compound_primary_key

The next= link is to d,v:

https://datasette-issue-190-compound-pks.now.sh/compound-pks-9aafe8f/compound_primary_key?_next=d%2Cv

But that page starts with:

The next key in the sequence should be d,w. Also we should return the full a-z of the ones that start with the letter e - in this example we only return e-w, e-x, e-y and e-z