issues

Hi Simon.

It seems that datasette currently has wildcard-searches disabled by default (along with the boolean search-options, NEAR-queries and more, and despite the docs). If I try out the search-url provided in the docs (https://fara.datasettes.com/fara/FARA_All_ShortForms?_search=manafort), it does not handle wildcard-searches, and I'm unable to make it work on my datasette-instance.

I would argue that wildcard-searches is such a standard query, that it should be enabled by default. Requiring "_searchmode=raw" when using prefix-searches seems unnecessary. Plus: What happens to non-ascii searches when using "_searchmode=raw"? Is the "escape_fts"-function from datasette.utils ignored?

Thanks!

/Claus

WIP

This PR adds the ability to import emails from a Gmail mbox export from Google Takeout.

This is my first PR to a datasette/dogsheep repo. I've tested this on my personal Google Takeout mbox with ~520,000 emails going back to 2004. This took around ~20 minutes to process.

To provide some feedback on the progress of the import I added the "rich" python module. I'm happy to remove that if adding a dependency is discouraged. However, I think it makes a nice addition to give feedback on the progress of a long import.

Do we want to log emails that have errors when trying to import them?

Dealing with encodings with emails is a bit tricky. I'm very open to feedback on how to deal with those better. As well as any other feedback for improvements.

My sqlite is getting too big to be processed by datasette (more than 10 minutes waiting to load) so I am working with duckdb and is waaaaay faster. I think the fastest embeddable database actually.

https://duckdb.org/

Taking into account DuckDb is SQLite based it would be GREAT to use it with datasette.

is that possible?

Regards and thanks for a superb job

Following the readme, on big sur, and having entered my auth creds via dogsheep-photos s3-auth:

(venv) (base) Robins-MacBook:datasette robin$ dogsheep-photos upload photos.db     ~/Pictures/Photos\ /Users/robin/Pictures/Library.photoslibrary --dry-run

Fetching existing keys from S3...
Traceback (most recent call last):
  File "/Users/robin/datasette/venv/bin/dogsheep-photos", line 8, in <module>
    sys.exit(cli())
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/click/core.py", line 829, in __call__
    return self.main(*args, **kwargs)
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/click/core.py", line 782, in main
    rv = self.invoke(ctx)
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/click/core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/click/core.py", line 1066, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/click/core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/dogsheep_photos/cli.py", line 96, in upload
    key.split(".")[0] for key in get_all_keys(client, creds["photos_s3_bucket"])
  File "/Users/robin/datasette/venv/lib/python3.8/site-packages/dogsheep_photos/utils.py", line 46, in get_all_keys
    for row in page["Contents"]:
KeyError: 'Contents'

Possibly since the bucket is in EU (London) eu-west-2 and this into is not requested?

From takeout, I only exported my Gmail account. Ideally I could parse this into sqlite via this tool.

It seems that custom pages aren't routing properly when the base_url setting is used.

To reproduce, with Datasette 0.55.

Create a templates/pages/custom.html with some text.

mkdir -p templates/pages/
echo "Hello, world!" > templates/pages/custom.html

Start Datasette.

datasette --template-dir templates/

Visit http://localhost:8001/custom and see "Hello, world!".

Start Datasette with a base_url.

datasette --template-dir templates/ --setting base_url /prefix/

Visit http://localhost:8001/prefix/custom and see a "Database not found: custom" 404.

Note that like all routes, http://localhost:8001/custom still works when run with base_url.

This is pinned to v1 at the moment.

This isn't particularly straight-forward - all the more reason for Datasette to implement it for you. This article is helpful: http://charlesleifer.com/blog/using-sqlite-full-text-search-with-python/

SQLite has the ability to attach multiple databases to a single connection and then run joins across multiple databases.

Since Datasette supports more than one database, this would make a pretty neat feature.

Remove extra [

Thank you to release this bravo project.
When i use this project on multi table db, I want to implement
convenient search on column name from different tables.
I want to develop a meta table to save the meta data of different columns
of different tables and search on this meta table to get rows from the
data table (which the meta table describes)
does this project provide some simple function on it ?

You can think a have a knowledge graph about the table in the db,
and i save this knowledge graph into the db with fts enabled.

I just wrote this code:

https://github.com/simonw/datasette/blob/47eb885cc2c3aafa03645c330c6f597bee9b3b25/tests/test_facets.py#L334-L335

It would be nice if you didn't have to separately instantiate a database object here.

I sometimes don't get the suggestion for facet-by-array for columns that contain arrays. I think it may be because they have empty spaces in them - or perhaps it's because the null detection doesn't actually work.

My httpx pull request adding raw_path support was just merged: https://github.com/encode/httpx/pull/1357 - but it's not in a release yet.

I'm going to mark these tests as xfail so I can land this change - I'll remove that once an httpx release comes out that I can use to get the tests passing.

_Originally posted by @simonw in https://github.com/simonw/datasette/pull/1000#issuecomment-706263157_

Following our conversation last week, want to note this here before I forget.

I've had a couple situations where I'd like to do a bunch of updates in an async event loop, but I run into SQLite's issues with concurrent writes. This feels like something sqlite-utils could help with.

PeeWee ORM has a SQLite write queue that might be a good model. It's using threads or gevent, but I think that approach would translate well enough to asyncio.

Happy to help with this, too.

Imagine a table (imported from a nested JSON file) where one of the columns contains values that look like this:

{"email": "anonymous@noreply.airtable.com", "id": "usrROSHARE0000000", "name": "Anonymous"}

The sqlite-utils extract command already uses single text values in a column to populate a new table. It would not be much of a stretch for it to be able to use JSON instead, including specifying which of those values should be used as the primary key in the new table.

Facets are a tremendously useful feature, especially for people peeking at the database for the first time and still having little knowledge about the details of the data. It is of great assistance to discover interesting features to explore futher in advanced queries.

Yet, it seems it's impossible to use facets when running a custom SQL query, be it from the little gear icons in column names, the facet suggestions at the top (hidden when performing a custom query), or by appending a facet code to the URL.

Is there a technical limitation, or is this something that could be unlocked easily?

Suggestion from office hours.

Someone asked "what are good example plugins I can look at?" and I realized that the answer is to look through the example links on https://docs.datasette.io/en/stable/plugin_hooks.html - but that tip should be written down somewhere on the https://docs.datasette.io/en/stable/writing_plugins.html page.

When I'm generating a derived table in python, I often drop the table and create it from scratch. However, the first time I generate the table, it doesn't exist, so the drop raises an exception. That means more boilerplate.

I was going to submit a pull request that adds an "if_exists" option to the drop method of tables and views.

However, for a utility like sqlite_utils, perhaps the "IF EXISTS" SQL semantics is what you want most of the time, and thus should be the default.

What do you think?

I ran this:

db["Reports"].add_foreign_key("Reported by ID", "Reporters", "id")

And got this:

~/jupyter-venv/lib/python3.9/site-packages/sqlite_utils/db.py in add_foreign_keys(self, foreign_keys)
    616         # Have to VACUUM outside the transaction to ensure .foreign_keys property
    617         # can see the newly created foreign key.
--> 618         self.vacuum()
    619 
    620     def index_foreign_keys(self):

~/jupyter-venv/lib/python3.9/site-packages/sqlite_utils/db.py in vacuum(self)
    629 
    630     def vacuum(self):
--> 631         self.execute("VACUUM;")
    632 
    633 

~/jupyter-venv/lib/python3.9/site-packages/sqlite_utils/db.py in execute(self, sql, parameters)
    234             return self.conn.execute(sql, parameters)
    235         else:
--> 236             return self.conn.execute(sql)
    237 
    238     def executescript(self, sql):

DatabaseError: database disk image is malformed

Original title: Easier way to update a row returned from .rows

Here's a surprisingly hard problem I ran into while trying to implement #239 - given a row returned by db[table].rows how can you update that row?

The problem is that the db[table].update(...) method requires a primary key. But if you have a row from the db[table].rows iterator it might not even contain the primary key - provided the table is a rowid table.

Instead, currently, you need to introspect the table and, if rowid is a primary key, explicitly include that in the select= argument to table.rows_where(...) - otherwise it will not be returned.

A utility mechanism to make this easier would be very welcome.

Refs #239. Still needs documentation and CLI implementation.

I use datasette in an iframe inside another HTML file that contains other ways to represent my data (mostly leaflets maps built with R on summarized data), and the datasette iframe is a tab in that page.

This particular use prevents users to access the full URLs of their datasette views and queries, which is a shame because the way datasette handles URLs to make every view or query easy to share is awesome. I know how to get the URL from the context menu of my browser, but I don't think many visitors would do it or even notice that datasette uses permalinks for pretty much every action they do. Would it be possible to add a "Share link" button to the interface, either in datasette itself or in a plugin?

Error as above, however I do have python3.8 and the readme indicates this is supported.

(venv) (base) Robins-MacBook:datasette robin$ ls /usr/local/opt/python3/bin/

.. pip3 python3 python3.8

Got this exception:

ERROR: conn=<sqlite3.Connection object at 0x10ea68e40>, sql = 'select Address, Affiliation, County, [Has Report], [Latest report notes], [Latest report yes], Latitude, [Location Type], Longitude, Name, id, [Appointment scheduling instructions], [Availability Info], [Latest report] from locations where rowid in (\n select locations.rowid from locations, json_each(locations.Availability Info) j\n where j.value = :p0\n ) and "Latest report yes" = :p1 order by id limit 101', params = {'p0': 'Yes: appointment required', 'p1': '1'}: near "Info": syntax error

Suggested by @frankieroberto in https://github.com/simonw/datasette/issues/782#issuecomment-782746755

The implementation in #1232 is ready to land. It's the simplest-thing-that-could-possibly-work: you can run datasette one.db two.db three.db --crossdb and then use the /_memory page to run joins across tables from multiple databases.

It only works on the first 10 databases that were passed to the command-line. This means that if you have a Datasette instance with hundreds of attached databases (see Datasette Library) this won't be particularly useful for you.

So... a better, future version of this feature would be one that lets you join across databases on command - maybe by hitting /_memory?attach=db1&attach=db2 to get a special connection.

_Originally posted by @simonw in https://github.com/simonw/datasette/issues/283#issuecomment-781665560_

enable_fts and its related methods only work on tables, not views.

Could those methods and possibly others move up to the Queryable superclass?

https://prettier.io/ - I'm going to go with 2 spaces.

The default JSON just isn't right. I find myself using ?_shape=array for almost everything I build against the API.

Working on my PR i noticed that tests are very slow.

The plain pytest run took about 37 minutes for me.
However i could shave of about 10 minutes from that if i used pytest-xdist to parallelize execution.
pytest -n 8 is run only in 28 minutes on my machine.

I can create a PR to mention that in your documentation.
This will be a simple change to add pytest-xdist to requirements and change a command to run pytest in documentation.

Does that make sense to you?

After a bit more investigation it looks like python-xdist is not an answer. It creates a race condition for tests that try to clead temp dir before run.

Profiling shows that most time is spent on conn.executescript(TABLES) in make_app_client function. Which makes sense.

Perhaps the better approach would be look at the app_client fixture which is already session scoped, but not used by all test cases.
And/or use conn = sqlite3.connect(":memory:") which is much faster.
And/or truncate tables after each TC instead of deleting the file and re-creating them.

I can take a look which is the best approach if you give the go-ahead.

https://latest.datasette.io/fixtures?sql=select%0D%0A++*%0D%0Afrom%0D%0A++%5Bfoo%5D

Would be useful if this page showed you the invalid SQL you entered so you can edit it and try again.

fixes #1224

This PR ensures all databases found in a configuration directory that match the files in inspect-data.json will be set to immutable as outlined in https://docs.datasette.io/en/latest/settings.html#configuration-directory-mode

specifically on building the datasette instance it checks:
- if immutables is an empty tuple - as passed by the cli code
- if immutables is the default function value None - when it's not explicitly set

And correctly builds the immutable database list from the inspect-data[file] keys.

Note for this to work the inspect-data.json file must contain file paths which are relative to the configuration directory otherwise the file paths won't match and the dbs won't be set to immutable.

I couldn't find an easy way to test this due to the way make_app_client works, happy to take directions on adding a test for this.

I've updated the relevant docs as well, i.e. use the inspect cli cmd from the config directory path to create the relevant file

cd $config_dir
datasette inspect *.db --inspect-file=inspect-data.json

https://docs.datasette.io/en/latest/performance.html#using-datasette-inspect

This will enable cross-database joins, as seen in https://github.com/simonw/datasette/issues/283

Also refs #113

https://www.sqlite.org/lang_attach.html

Maybe something like this:

db.attach("other_db", "other_db.db")

Currently uses 3.7.2:

https://github.com/simonw/datasette/blob/73bed175631a79e13a521eee82f8451dd0477eb3/Dockerfile#L1

There's a security fix for Python which it would be good to ship in this image (even though I'm reasonably confident it doesn't affect Datasette): https://bugs.python.org/issue42938

Refs #283. Still needs:

• Unit test for --crossdb queries
• Show warning on console if it truncates at ten databases (or on web interface)
• Show connected databases on the /_memory database page
• Documentation
• https://latest.datasette.io/ demo should demonstrate this feature

Got this error:

Step 6/9 : RUN datasette inspect fixtures.db extra database.db --inspect-file inspect-data.json
 ---> Running in db9da0068592
Usage: datasette inspect [OPTIONS] [FILES]...
Try 'datasette inspect --help' for help.

Error: Invalid value for '[FILES]...': Path 'extra' does not exist.
The command '/bin/sh -c datasette inspect fixtures.db extra database.db --inspect-file inspect-data.json' returned a non-zero code: 2
ERROR
ERROR: build step 0 "gcr.io/cloud-builders/docker" failed: step exited with non-zero status: 2

While working on the demo for #1232, using this deploy command:

GITHUB_SHA=crossdb datasette publish cloudrun fixtures.db 'extra database.db' \
            -m fixtures.json \
            --plugins-dir=plugins \
            --branch=$GITHUB_SHA \
            --version-note=$GITHUB_SHA \
            --extra-options="--setting template_debug 1 --crossdb" \
            --install=pysqlite3-binary \
            --service=datasette-latest-crossdb

I tried running a Locust load test against Datasette and hit an error message about a failure to create tables because they already existed. I think this means there are race conditions in the new refresh_schemas() mechanism added in #1150.

Currently throws an ugly error message:

(datasette-graphql) datasette-graphql % datasette fivethirtyeight.db -p 80094
INFO:     Started server process [45497]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
Traceback (most recent call last):
  File "/Users/simon/.local/share/virtualenvs/datasette-graphql-n1OSJCS8/bin/datasette", line 8, in <module>
    sys.exit(cli())
...
    server = await loop.create_server(
  File "/Users/simon/.pyenv/versions/3.8.2/lib/python3.8/asyncio/base_events.py", line 1461, in create_server
    sock.bind(sa)
OverflowError: bind(): port must be 0-65535.

I filtered a data set on some criteria and obtain 265 results, split over three pages (100, 100, 65), and reazlized that Vega plots are only applied to the results displayed on the current page, instead of the whole filtered data, e.g., 100 on page 1, 100 on page 2, 65 on page 3. Is there a way to force the graphs to consider all results instead of just the page, considering that pages rarely represent sensible information?

Likewise, while the cluster map does show all results on the first page, if you go to next pages, it will show all remaining results except the previous page(s), e.g., 265 on page 1, 165 on page 2, 65 on page 3.

In both cases, I don't see many situations where one would like to represent the data this way, and it might even lead to interpretation errors when viewing the data. Am I missing some cases where this would be best? Perhaps a clickable option to subset visual representations according visible pages vs. display all search results would do?

[Edit] Oh, I just saw the "Load all" button under the cluster map as well as the setting to alter the max number or results. So I guess this issue only is about the Vega charts.

Spotted this in the aspw documentation: https://github.com/rogerbinns/apsw/blob/3.34.0-r1/doc/conf.py#L29-L36

extlinks={
    'cvstrac': ('https://sqlite.org/cvstrac/tktview?tn=%s',
                'SQLite ticket #'),
    'sqliteapi': ('https://sqlite.org/c3ref/%s.html', 'XXYouShouldNotSeeThisXX'),
    'issue': ('https://github.com/rogerbinns/apsw/issues/%s',
              'APSW issue '),
    'source': ('https://github.com/rogerbinns/apsw/blob/master/%s',
               ''),
    }

Which lets you link to issues like this:

:issue:`268`

Thanks for what seems like a truly great suite of libraries. I wanted to try out Datasette, but never got more than half way through your YouTube video with the SF tree dataset. Whenever I try to extract a column, I get a sqlite3.OperationalError: table sqlite_master may not be modified error from Python. This snippet reproduces the error on my system, Python 3.9.1 and sqlite-utils 3.5 on an M1 Macbook Pro running in rosetta mode:

curl "https://data.nasa.gov/resource/y77d-th95.json" | \
    sqlite-utils insert meteorites.db meteorites - --pk=id
sqlite-utils extract meteorites.db meteorites  recclass

I have tried googling the problem, but all I've found is that this might be a problem with the sqlite3 database running in defensive mode, but I definitely can't know for sure. Does the problem seem familiar to you?

I recently discovered datasette thanks to your great talk at FOSDEM and would like to use it for some projects. However, when trying to use it on databases created from some csv ot tsv files, I am sometimes getting this issue when going to http://127.0.0.1:8001/databasetest/databasetest and I don't exactly understand what it refers to.

So far, I couldn't find anything relevant when reviewing the raw text files that could explain this issue, nor could I find something obvious between the files that generate this issue and those that don't. Does the error ring a bell and, if so, could you please point me to the right direction?

$ datasette databasetest.db 
INFO:     Started server process [1408482]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://127.0.0.1:8001 (Press CTRL+C to quit)
INFO:     127.0.0.1:56394 - "GET / HTTP/1.1" 200 OK
INFO:     127.0.0.1:56394 - "GET /-/static/app.css?4e362c HTTP/1.1" 200 OK
INFO:     127.0.0.1:56396 - "GET /-/static-plugins/datasette_vega/main.2acbb312.css HTTP/1.1" 200 OK
INFO:     127.0.0.1:56398 - "GET /-/static-plugins/datasette_vega/main.08f5d3d8.js HTTP/1.1" 200 OK
Traceback (most recent call last):
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/app.py", line 1099, in route_path
    response = await view(request, send)
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/views/base.py", line 147, in view
    request, **request.scope["url_route"]["kwargs"]
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/views/base.py", line 121, in dispatch_request
    return await handler(request, *args, **kwargs)
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/views/base.py", line 260, in get
    request, database, hash, correct_hash_provided, **kwargs
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/views/base.py", line 434, in view_get
    request, database, hash, **kwargs
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/views/table.py", line 782, in data
    suggested_facets.extend(await facet.suggest())
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/facets.py", line 168, in suggest
    and any(r["n"] > 1 for r in distinct_values)
  File "/home/kabouik/.local/lib/python3.7/site-packages/datasette/facets.py", line 168, in <genexpr>
    and any(r["n"] > 1 for r in distinct_values)
TypeError: '>' not supported between instances of 'str' and 'int'
INFO:     127.0.0.1:56402 - "GET /databasetest/databasetest HTTP/1.1" 500 Internal Server Error
INFO:     127.0.0.1:56402 - "GET /-/static/app.css?4e362c HTTP/1.1" 200 OK
INFO:     127.0.0.1:56404 - "GET / HTTP/1.1" 200 OK
INFO:     127.0.0.1:56404 - "GET /-/static/app.css?4e362c HTTP/1.1" 200 OK
INFO:     127.0.0.1:56406 - "GET /-/static-plugins/datasette_vega/main.2acbb312.css HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /-/static-plugins/datasette_vega/main.08f5d3d8.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /databasetest HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /-/static/app.css?4e362c HTTP/1.1" 200 OK
INFO:     127.0.0.1:56404 - "GET /-/static-plugins/datasette_vega/main.2acbb312.css HTTP/1.1" 200 OK
INFO:     127.0.0.1:56406 - "GET /-/static/codemirror-5.57.0.min.css HTTP/1.1" 200 OK
INFO:     127.0.0.1:56410 - "GET /-/static-plugins/datasette_vega/main.08f5d3d8.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56414 - "GET /-/static/codemirror-5.57.0-sql.min.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56412 - "GET /-/static/codemirror-5.57.0.min.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /-/static/sql-formatter-2.3.3.min.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /databasetest?sql=select+*+from+databasetest HTTP/1.1" 200 OK
INFO:     127.0.0.1:56410 - "GET /-/static/app.css?4e362c HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /-/static-plugins/datasette_vega/main.2acbb312.css HTTP/1.1" 200 OK
INFO:     127.0.0.1:56412 - "GET /-/static/codemirror-5.57.0.min.css HTTP/1.1" 200 OK
INFO:     127.0.0.1:56404 - "GET /-/static/sql-formatter-2.3.3.min.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56406 - "GET /-/static/codemirror-5.57.0.min.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56414 - "GET /-/static-plugins/datasette_vega/main.08f5d3d8.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56408 - "GET /-/static/codemirror-5.57.0-sql.min.js HTTP/1.1" 200 OK
INFO:     127.0.0.1:56410 - "GET /databasetest.json?sql=select+*+from+databasetest&_shape=array&_shape=array HTTP/1.1" 200 OK
^CINFO:     Shutting down
INFO:     Waiting for application shutdown.
INFO:     Application shutdown complete.
INFO:     Finished server process [1408482]

Note that there is no error if I go to http://127.0.0.1:8001/databasetest and then click on Run SQL.

Feel free to close this issue - I mostly added it for reference for future folks that run into this :)

I have a CSV file with one column that has very long strings. When i try to import this file via the insert command I get the following error:

sqlite-utils insert database.db table_name file_with_large_column.csv

Traceback (most recent call last):
  File "/usr/local/bin/sqlite-utils", line 10, in <module>
    sys.exit(cli())
  File "/usr/local/lib/python3.7/site-packages/click/core.py", line 829, in __call__
    return self.main(*args, **kwargs)
  File "/usr/local/lib/python3.7/site-packages/click/core.py", line 782, in main
    rv = self.invoke(ctx)
  File "/usr/local/lib/python3.7/site-packages/click/core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/usr/local/lib/python3.7/site-packages/click/core.py", line 1066, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/usr/local/lib/python3.7/site-packages/click/core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "/usr/local/lib/python3.7/site-packages/sqlite_utils/cli.py", line 774, in insert
    default=default,
  File "/usr/local/lib/python3.7/site-packages/sqlite_utils/cli.py", line 705, in insert_upsert_implementation
    docs, pk=pk, batch_size=batch_size, alter=alter, **extra_kwargs
  File "/usr/local/lib/python3.7/site-packages/sqlite_utils/db.py", line 1852, in insert_all
    first_record = next(records)
  File "/usr/local/lib/python3.7/site-packages/sqlite_utils/cli.py", line 703, in <genexpr>
    docs = (decode_base64_values(doc) for doc in docs)
  File "/usr/local/lib/python3.7/site-packages/sqlite_utils/cli.py", line 681, in <genexpr>
    docs = (dict(zip(headers, row)) for row in reader)
_csv.Error: field larger than field limit (131072)

Built with the docker image datasetteproject/datasette:0.54 with the following versions:

# sqlite-utils --version
sqlite-utils, version 3.4.1

# datasette --version
datasette, version 0.54

It appears this is a known issue reading in csv files in python and doesn't look to be modifiable through system / env vars (i may be very wrong on this).

Noting that using sqlite3 import command work without error (not using the python csv reader)

sqlite3 database.db
sqlite> .mode csv
sqlite> .import file_with_large_column.csv table_name

Sadly I couldn't see an easy way around this while using the cli as it appears this value needs to be changed in python code. FWIW I've switched to using https://datasette.io/tools/csvs-to-sqlite for importing csv data and it's working well.

Finally, I'm loving https://datasette.io/ thank you very much for an amazing tool and data ecosytem 🙇‍♀️

I came across this while about to open an issue and PR against the documentation for batch_size, which is a bit incomplete.

As mentioned in #145, while:

SQLITE_MAX_VARIABLE_NUMBER ... defaults to 999 for SQLite versions prior to 3.32.0 (2020-05-22) or 32766 for SQLite versions after 3.32.0.

it is common that it is increased at compile time. Debian-based systems, for example, seem to ship with a version of sqlite compiled with SQLITE_MAX_VARIABLE_NUMBER set to 250,000, and I believe this is the case for homebrew installations too.

In working to understand what batch_size was actually doing and why, I realized that by setting SQLITE_MAX_VARS in db.py to match the value my sqlite was compiled with (I'm on Debian), I was able to decrease the time to insert_all() my test data set (~128k records across 7 tables) from ~26.5s to ~3.5s. Given that this about .05% of my total dataset, this is time I am keen to save...

Unfortunately, it seems that sqlite3 in the python standard library doesn't expose the get_limit() C API (even though pysqlite used to), so it's hard to know what value sqlite has been compiled with (note that this could mean, I suppose, that it's less than 999, and even hardcoding SQLITE_MAX_VARS to the conservative default might not be adequate. It can also be lowered -- but not raised -- at runtime). The best I could come up with is echo "" | sqlite3 -cmd ".limits variable_number" (only available in sqlite >= 2015-05-07 (3.8.10)).

Obviously this couldn't be relied upon in sqlite_utils, but I wonder what your opinion would be about exposing SQLITE_MAX_VARS as a user-configurable parameter (with suitable "here be dragons" warnings)? I'm going to go ahead and monkey-patch it for my purposes in any event, but it seems like it might be worth considering.

In several applications I've been experimenting with alternate formatting of datasette query results. Lately I've found that CSS grids work very well and seem quite general for formatting rows. In CSS I use grid templates to define the layout of each record and the regions for each field, hiding the fields I don't want. It's pretty flexible and looks good. It's also a great basis for highly responsive layout.

I initially thought I'd only use this feature for record detail views, but now I use it for index views as well.

However, there are some limitations: * With the existing table templates, it seems that you can change the display property on the enclosing table, tbody, and tr to make them be grid-like, but that seems hacky (convert table and tbody to be display: block and tr to be display: grid). * More significantly, it's very nice to have the column name available when rendering each record to display headers/field labels. The existing templates don't do that, so a custom _table template is necessary. * I don't know if any plugins are sensitive to whether data is rendered as a table or not since I'm not completely clear how plugins get their data. * Regardless, you need custom CSS to take full advantage of grids. I don't have a proposal on how to integrate them more deeply.

It would be helpful to at least have an official example or test that used a grid layout for records to make sure nothing in datasette breaks with it.

https://bl.iro.bl.uk/work/ns/3037474a-761c-456d-a00c-9ef3c6773f4c has a fascinating CSV file that doesn't have a header row - it starts like this:

Computation and measurement of turbulent flow through idealized turbine blade passages,,"Loizou, Panos A.",https://isni.org/isni/0000000136122593,,University of Manchester,https://isni.org/isni/0000000121662407,1989,Thesis (Ph.D.),,Physical Sciences,,,https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.232781,
"Prolactin and growth hormone secretion in normal, hyperprolactinaemic and acromegalic man",,"Prescott, R. W. G.",https://isni.org/isni/0000000134992122,,University of Newcastle upon Tyne,https://isni.org/isni/0000000104627212,1983,Thesis (Ph.D.),,Biological Sciences,,,https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.232784,

It would be useful if sqlite-utils insert ... --csv had a mechanism for importing files like this one.

I have a csv file where one of the fields is so large it is throwing an exception with this error and stops loading:
_csv.Error: field larger than field limit (131072)

The stack trace occurs here: https://github.com/simonw/sqlite-utils/blob/3.1/sqlite_utils/cli.py#L633

There is a way to handle this that helps:
https://stackoverflow.com/questions/15063936/csv-error-field-larger-than-field-limit-131072

One issue I had with this problem was sqlite-utils only provides limited context as to where the problem line is.
There is the progress bar, but that is by percent rather than by line number. It would have been helpful if it could have provided a line number.

Also, it would have been useful if it had allowed the loading to continue with later lines.

I just spotted that csv.Sniffer in the Python standard library has a .has_header(sample) method which detects if the first row appears to be a header or not, which is interesting. https://docs.python.org/3/library/csv.html#csv.Sniffer

_Originally posted by @simonw in https://github.com/simonw/sqlite-utils/issues/228#issuecomment-778812050_

Hi,

I ran into a problem when trying to create a database from my Apple Healthkit data using healthkit-to-sqlite. The program crashed because of an invalid insert statement that was generated for table rDistanceCycling.

The actual problem turned out to be in sqlite-utils. Table.insert_all processes the data to be inserted in chunks of rows and checks for every chunk which columns are used, and it will collect all column names in the variable all_columns. The collection of columns is done using a nested list comprehension that is not completely correct.

I'm using a Windows machine and had to make a few adjustments to the tests in order to be able to run them because they had a posix dependency.

Thanks, kind regards,

Frans

# this is a (condensed) chunk of data from my Apple healthkit export that caused the problem.
# the 3 last items in the chunk have additional keys: metadata_HKMetadataKeySyncVersion and metadata_HKMetadataKeySyncIdentifier

chunk = [{'sourceName': 'AppleÂ\xa0Watch van Frans', 'sourceVersion': '7.0.1',
          'device': '<<HKDevice: 0x281cf6c70>, name:Apple Watch, manufacturer:Apple Inc., model:Watch, hardware:Watch3,4, software:7.0.1>',
          'unit': 'km', 'creationDate': '2020-10-10 12:29:09 +0100', 'startDate': '2020-10-10 12:29:06 +0100',
          'endDate': '2020-10-10 12:29:07 +0100', 'value': '0.00518016'},
         {'sourceName': 'AppleÂ\xa0Watch van Frans', 'sourceVersion': '7.0.1',
          'device': '<<HKDevice: 0x281cf6c70>, name:Apple Watch, manufacturer:Apple Inc., model:Watch, hardware:Watch3,4, software:7.0.1>',
          'unit': 'km', 'creationDate': '2020-10-10 12:29:10 +0100', 'startDate': '2020-10-10 12:29:07 +0100',
          'endDate': '2020-10-10 12:29:08 +0100', 'value': '0.00544049'},
         {'sourceName': 'AppleÂ\xa0Watch van Frans', 'sourceVersion': '6.2.6',
          'device': '<<HKDevice: 0x281cf83e0>, name:Apple Watch, manufacturer:Apple Inc., model:Watch, hardware:Watch3,4, software:6.2.6>',
          'unit': 'km', 'creationDate': '2020-10-14 05:54:12 +0100', 'startDate': '2020-07-15 16:40:50 +0100',
          'endDate': '2020-07-15 16:42:49 +0100', 'value': '0.952092', 'metadata_HKMetadataKeySyncVersion': '1',
          'metadata_HKMetadataKeySyncIdentifier': '3:674DBCDB-3FE8-40D1-9FC1-E54A2B413805:616520450.99823:616520569.99360:119'},
         {'sourceName': 'AppleÂ\xa0Watch van Frans', 'sourceVersion': '6.2.6',
          'device': '<<HKDevice: 0x281cf83e0>, name:Apple Watch, manufacturer:Apple Inc., model:Watch, hardware:Watch3,4, software:6.2.6>',
          'unit': 'km', 'creationDate': '2020-10-14 05:54:12 +0100', 'startDate': '2020-07-15 16:42:49 +0100',
          'endDate': '2020-07-15 16:44:51 +0100', 'value': '0.848983', 'metadata_HKMetadataKeySyncVersion': '1',
          'metadata_HKMetadataKeySyncIdentifier': '3:674DBCDB-3FE8-40D1-9FC1-E54A2B413805:616520569.99360:616520691.98826:119'},
         {'sourceName': 'AppleÂ\xa0Watch van Frans', 'sourceVersion': '6.2.6',
          'device': '<<HKDevice: 0x281cf83e0>, name:Apple Watch, manufacturer:Apple Inc., model:Watch, hardware:Watch3,4, software:6.2.6>',
          'unit': 'km', 'creationDate': '2020-10-14 05:54:12 +0100', 'startDate': '2020-07-15 16:44:51 +0100',
          'endDate': '2020-07-15 16:46:50 +0100', 'value': '0.834403', 'metadata_HKMetadataKeySyncVersion': '1',
          'metadata_HKMetadataKeySyncIdentifier': '3:674DBCDB-3FE8-40D1-9FC1-E54A2B413805:616520691.98826:616520810.98305:119'}]



def all_columns_old():
    all_columns = [col for col in chunk[0]]
    all_columns += [column for record in chunk
                           for column in record if column not in all_columns]
    return all_columns


def all_columns_new():
    all_columns = [col for col in chunk[0]]
    for record in chunk:
        all_columns += [column for column in record if column not in all_columns]
    return all_columns



if __name__ == '__main__':
    from pprint import pprint

    print('problem: ')
    pprint(all_columns_old())
    print('\nfix: ')
    pprint(all_columns_new())

Reported by @nieuwenhoven in #225 along with a proposed fix.

I'm going to try and get the test suite to run in Windows on GitHub Actions.

_Originally posted by @simonw in https://github.com/simonw/sqlite-utils/issues/225#issuecomment-778834504_

Refs #232

I'm going to add a offset= parameter to support this case. Thanks for the suggestion!

_Originally posted by @simonw in https://github.com/simonw/sqlite-utils/issues/224#issuecomment-778828495_

The limit can be passed as a string to the query builder to have an offset. I have tested it using the shorthand limit=f"15, 30", the standard syntax should work too.

The insert command currently results in string types for every column - at least when used against CSV or TSV inputs.

It would be useful if you could do the following:

• automatically detects the column types based on eg the first 1000 records
• explicitly state the rule for specific columns

--detect-types could work for the former - or it could do that by default and allow opt-out using --no-detect-types

For specific columns maybe this:

sqlite-utils insert db.db images images.tsv \
  --tsv \
  -c id int \
  -c score float

Hi,
If I run

docker run -p 8001:8001 -v `pwd`:/mnt \                                                                         1 ↵
    datasetteproject/datasette \
    datasette -p 8001 -h 0.0.0.0 fixtures.db

I have

Error: Invalid value for '[FILES]...': Path 'fixtures.db' does not exist.

If I run test -f fixtures.db && echo "it exists." I have it exists..

What's my error?

Thank you

As a total beginner here how could you access data from the sqlite table to run other commands.

What I am thinking is I want to get all the repos in an organization then using the repo list pull all the commit messages for each repo.

I love this project by the way!

Say I have a /databases/ directory with multiple sqlite db files in that dir (1.db & 2.db) and an inspect-data.json file.

If I start datasette via datasette -h 0.0.0.0 /databases/ then the resulting databases are set to is_mutable: true as inspected via http://127.0.0.1:8001/-/databases.json

I don't want to have to list out the databases by name, e.g. datasette -i /databases/1.db -i /databases/2.db as i want the system to autodetect the sqlite dbs i have in the configuration directory

According to the docs outlined in https://docs.datasette.io/en/latest/settings.html?highlight=immutable#configuration-directory-mode this should be possible

inspect-data.json the result of running datasette inspect - any database files listed here will be treated as immutable, so they should not be changed while Datasette is running

I believe that if the inspect-json.json file present, then in theory the databases will be automatically set to immutable via this code https://github.com/simonw/datasette/blob/9603d893b9b72653895318c9104d754229fdb146/datasette/app.py#L211-L216

However it appears the Click Multiple Options will return a tuple via https://github.com/simonw/datasette/blob/9603d893b9b72653895318c9104d754229fdb146/datasette/cli.py#L311-L317

The resulting tuple is passed to the Datasette app via kwargs and overrides the behaviour to set the databases to immutable via this arg https://github.com/simonw/datasette/blob/9603d893b9b72653895318c9104d754229fdb146/datasette/app.py#L182

If you think this is a bug and needs fixing, I am willing to make a PR to check for the empty immutable tuple before calling the Datasette class initializer as I think leaving that class interface alone is the best path here.

Thoughts?

Also - i'm loving Datasette, it truly is a wonderful tool, thank you :)

Having installed in a venv I get:

(venv) (base) Robins-MacBook:datasette robin$ photo-to-sqlite apple-photos photos.db

-bash: photo-to-sqlite: command not found

This test was failing when run inside the Docker container: test_searchable[/fixtures/searchable.json?_search=te*+AND+do*&_searchmode=raw-expected_rows3],

with this error:

    def test_searchable(app_client, path, expected_rows):
        response = app_client.get(path)
>       assert expected_rows == response.json["rows"]
E       AssertionError: assert [[1, 'barry c...sel', 'puma']] == []
E         Left contains 2 more items, first extra item: [1, 'barry cat', 'terry dog', 'panther']
E         Full diff:
E         + []
E         - [[1, 'barry cat', 'terry dog', 'panther'],
E         -  [2, 'terry dog', 'sara weasel', 'puma']]

The issue was that the version of sqlite3 built inside the Docker container was built with FTS3 and FTS4 enabled, but without the
SQLITE_ENABLE_FTS3_PARENTHESIS compile option passed, which adds support for using AND and NOT within match expressions (see https://sqlite.org/fts3.html#compiling_and_enabling_fts3_and_fts4 and https://www.sqlite.org/compile.html).

Without this, the AND used in the search in this test was being interpreted as a literal string, and so no matches were found. Adding this compile option fixes this.

I actually ran into this issue because the same test was failing when I ran the test suite on my own machine, outside of Docker, and so I eventually tracked this down to my system sqlite3 also being compiled without this option.

I wonder if this is a sign of a slightly deeper issue, that Datasette can silently behave differently based on the version and compilation of sqlite3 it is being used with. On my own system I fixed the test suite by running pip install pysqlite3-binary, so that this would be picked up instead of the sqlite package, as this seems to be compiled using this option, . Maybe using pysqlite3-binary could be installed/recommended by default so a more deterministic version of sqlite is used? Or there could be some feature detection done on the available sqlite version, to know what features are available and can be used/tested?

This should be pretty easy because Uvicorn supports them already. Need a good mechanism for testing it - https://pypi.org/project/trustme/ looks ideal.

Bugfix 1:

Traceback (most recent call last):
File "d:\anaconda3\lib\runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "d:\anaconda3\lib\runpy.py", line 87, in _run_code
exec(code, run_globals)
File "D:\Anaconda3\Scripts\evernote-to-sqlite.exe__main__.py", line 7, in <module>
File "d:\anaconda3\lib\site-packages\click\core.py", line 829, in call
File "d:\anaconda3\lib\site-packages\click\core.py", line 782, in main
rv = self.invoke(ctx)
File "d:\anaconda3\lib\site-packages\click\core.py", line 1259, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
return ctx.invoke(self.callback, ctx.params)
File "d:\anaconda3\lib\site-packages\click\core.py", line 610, in invoke
return callback(*args, kwargs)
File "d:\anaconda3\lib\site-packages\evernote_to_sqlite\cli.py", line 30, in enex
for tag, note in find_all_tags(fp, ["note"], progress_callback=bar.update):
File "d:\anaconda3\lib\site-packages\evernote_to_sqlite\utils.py", line 11, in find_all_tags
chunk = fp.read(1024 * 1024)
UnicodeDecodeError: 'gbk' codec can't decode byte 0xa4 in position 383: illegal multibyte sequence

Bugfix 2:

Traceback (most recent call last):
File "D:\Anaconda3\Scripts\evernote-to-sqlite-script.py", line 33, in <module>
sys.exit(load_entry_point('evernote-to-sqlite==0.3', 'console_scripts', 'evernote-to-sqlite')())
File "D:\Anaconda3\lib\site-packages\click\core.py", line 829, in call
return self.main(args, kwargs)
File "D:\Anaconda3\lib\site-packages\click\core.py", line 782, in main
rv = self.invoke(ctx)
File "D:\Anaconda3\lib\site-packages\click\core.py", line 1259, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "D:\Anaconda3\lib\site-packages\click\core.py", line 1066, in invoke
return ctx.invoke(self.callback, ctx.params)
File "D:\Anaconda3\lib\site-packages\click\core.py", line 610, in invoke
return callback(args, **kwargs)
File "D:\Anaconda3\lib\site-packages\evernote_to_sqlite-0.3-py3.8.egg\evernote_to_sqlite\cli.py", line 31, in enex
File "D:\Anaconda3\lib\site-packages\evernote_to_sqlite-0.3-py3.8.egg\evernote_to_sqlite\utils.py", line 28, in save_note
AttributeError: 'NoneType' object has no attribute 'text'

I got the following error when executing evernote-to-sqlite enex evernote.db evernote.enex

... 
  File "evernote_to_sqlite/cli.py", line 31, in enex
    save_note(db, note)
  File "evernote_to_sqlite/utils.py", line 28, in save_note
    updated = note.find("updated").text
AttributeError: 'NoneType' object has no attribute 'text'

Seems that in some cases the updated element is not added to the note, this is a part of the problematic note:

<created>20201019T074518Z</created>
<note-attributes>
   <source>web.clip7</source>
   <source-application>webclipper.evernote</source-application>
</note-attributes>

running evernote-to-sqlite 0.2 on windows 10. Command:

evernote-to-sqlite enex evernote.db MyNotes.enex

I get the followinng error:

File "C:\Users\marti\AppData\Roaming\Python\Python38\site-packages\evernote_to_sqlite\utils.py", line 46, in save_note
note_id = db["notes"].insert(row, hash_id="id", replace=True, alter=True).last_pk
TypeError: insert() got an unexpected keyword argument 'replace'

Removing replace=True,

Leads to below error:

note_id = db["notes"].insert(row, hash_id="id", alter=True).last_pk
File "C:\Users\marti\AppData\Roaming\Python\Python38\site-packages\sqlite_utils\db.py", line 924, in insert
return self.insert_all(
File "C:\Users\marti\AppData\Roaming\Python\Python38\site-packages\sqlite_utils\db.py", line 1046, in insert_all
result = self.db.conn.execute(sql, values)
sqlite3.IntegrityError: UNIQUE constraint failed: notes.id

I encountered a parse error if the enex file contained š or  

Run command:
evernote-to-sqlite enex evernote.db evernote.enex

Traceback (most recent call last):
...
  File "evernote_to_sqlite/cli.py", line 31, in enex
    save_note(db, note)
  File "evernote_to_sqlite/utils.py", line 35, in save_note
    content = ET.tostring(ET.fromstring(content_xml)).decode("utf-8")
  File "/usr/lib/python3.8/xml/etree/ElementTree.py", line 1320, in XML
    parser.feed(text)
xml.etree.ElementTree.ParseError: undefined entity š: line 3, column 35

Workaround:

sed -i 's/š//g' evernote.enex
sed -i 's/ //g' evernote.enex

I am on Windows 10 using Windows Subsystem for Linux, Python 3.8. I installed evernote-to-sqlite via pipx (in a venv). I tried using enex files from the latest version of Evernote for Windows (10.6.9 which only lets you export 50 notes at a time) and from Legacy Evernote (6.25.2.9198 which lets you export all your notes at once). The enex file from latest evernote gives this error:

File "/usr/lib/python3.8/xml/etree/ElementTree.py", line 1320, in XML parser.feed(text)
xml.etree.ElementTree.ParseError: XML or text declaration not at start of entity: line 2, column 6

The enex file from Legacy Evernote gives this error:

File "/home/david/.local/pipx/venvs/evernote-to-sqlite/lib/python3.8/site-packages/evernote_to_sqlite/utils.py", line 28, in save_note
updated = note.find("updated").text
AttributeError: 'NoneType' object has no attribute 'text'

https://latest.datasette.io/fixtures?sql=select+*+from+compound_three_primary_keys&_size=10 - _size=10 does not do anything at the moment. It would be useful if it did.

Would also be good if it persisted in a hidden form field.

https://github.com/emeryberger/scalene looks like an interesting profiling tool.

Here's the error I get when running pip install pocket-to-sqlite:

  Could not fetch URL https://pypi.python.org/simple/pocket-to-sqlite/: There was a problem confirming the ssl certificate: [SSL: TLSV1_ALERT_PROTOCOL_VERSION] tlsv1 alert protocol version (_ssl.c:661) - skipping
  Could not find a version that satisfies the requirement pocket-to-sqlite (from versions: )
No matching distribution found for pocket-to-sqlite

Does this require python 3?

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 started seeing weird errors, caused by this line: https://github.com/simonw/sqlite-utils/blob/f8010ca78fed8c5fca6cde19658ec09fdd468420/sqlite_utils/cli.py#L1-L3

That was added by accident in 1b666f9315d4ea6bb332b2e75e48480c26100199

I'm surprised the tests didn't catch this!

https://bruxellesdata.opendatasoft.com/explore/dataset/dog-toilets/export/?location=12,50.85802,4.38054 says:

CSV uses semicolon (;) as a separator.

Would be useful to be able to do this:

sqlite-utils insert places.db places places.csv --delimiter ';'

--delimiter could imply --csv

Your code is full of open files that are never closed, especially when you deal with reading/writing json/yaml files.

If you run python with warnings enabled this problem becomes evident.
This probably contributes to some memory leaks in long running datasettes if the GC will not 'collect' those resources properly.

This is easily fixed by using a context manager instead of just using open:

with open('some_file', 'w') as opened_file:
    opened_file.write('string')

In some newer parts of the code you use Path objects 'read_text' and 'write_text' functions which close the file properly and are prefered in some cases.

If you want I can create a PR for all places i found this pattern in.

Bellow is a fraction of places where i found a ResourceWarning:

update-docs-help.py:
  20          actual = actual.replace("Usage: cli ", "Usage: datasette ")
  21:         open(docs_path / filename, "w").write(actual)
  22  

datasette\app.py:
  210          ):
  211:             inspect_data = json.load((config_dir / "inspect-data.json").open())
  212              if immutables is None:

  266          if config_dir and (config_dir / "settings.json").exists() and not config:
  267:             config = json.load((config_dir / "settings.json").open())
  268          self._settings = dict(DEFAULT_SETTINGS, **(config or {}))

  445              self._app_css_hash = hashlib.sha1(
  446:                 open(os.path.join(str(app_root), "datasette/static/app.css"))
  447                  .read()

datasette\cli.py:
  130      else:
  131:         out = open(inspect_file, "w")
  132      loop = asyncio.get_event_loop()

  459      if inspect_file:
  460:         inspect_data = json.load(open(inspect_file))
  461  

Add support for compound foreign keys, as per issue #117

Not sure if this is the right approach. In particular I'm unsure about:

• the new ForeignKey class, which replaces the namedtuple in order to ensure that column and other_column are forced into tuples. The class does the job, but doesn't feel very elegant.
• I haven't rewritten guess_foreign_table to take account of multiple columns, so it just checks for the first column in the foreign key definition. This isn't ideal.
• I haven't added any ability to the CLI to add compound foreign keys, it's only in the python API at the moment.

The PR also contains a minor related change that columns and tables are always quoted in foreign key definitions.

I have a database that I only want to read from; when instructing datasette to treat the database as immutable my defined canned queries disappear. Are these two features incompatible or have I hit an unintended bug? Thanks for datasette in any way, it's a joy to use!

The order in which databases are attached to Datasette matters - it affects the homepage, and it's beginning to influence how certain plugins work (see https://github.com/simonw/datasette-tiles/issues/8).

Two years ago in cccea85be6aaaeadb31f3b588ec7f732628815f5 I made /-/databases return things in alphabetical order, to fix a test failure in Python 3.5.

Python 3.5 is no longer supported, so this is no longer necessary - and this behaviour should now be treated as a bug.

In my database I have 13300 quote tweets, but eta 3600 have quoted_status empty.

I fetched some of them using https://api.twitter.com/1.1/statuses/show.json?id=xx and they did have ids of quoted tweets.

Really nasty bug, caused by #1194 fix in 07e163561592c743e4117f72102fcd350a600909

Navigate to this page: https://github-to-sqlite.dogsheep.net/github/labels?_search=help&_sort=id

Click "Apply" to submit the form and the resulting URL is https://github-to-sqlite.dogsheep.net/github/labels?_search=help&_sort=id&_search=help&_sort=id

That's because the (truncated) HTML for the form looks like this:

    ... <input id="_search" type="search" name="_search" value="help">
    ...
            <div class="select-wrapper small-screen-only">
                <select name="_sort" id="sort_by">
                    <option value="">Sort...</option>
                            <option value="id" selected>Sort by id</option>
                            <option value="node_id">Sort by node_id</option>
                            ...
                </select>
            </div>
            ...
            <input type="hidden" name="_search" value="help">
            <input type="hidden" name="_sort" value="id">
        <input type="submit" value="Apply">

This page https://datasette-tiles-demo.datasette.io/San_Francisco/tiles is 2MB because of all of the base64 images. Gzipped it's 1.5MB.

Since Datasette is usually deployed without a frontend gzipping proxy, Datasette itself needs to solve for this.

Gzipping everything won't work because some endpoints - the all-rows CSV endpoint and the download-database endpoint - are streaming and hence can't be buffered-and-gzipped.

Context manager with open closes the files after usage. Fixes: https://github.com/simonw/datasette/issues/1208

When the object is already a pathlib.Path i used read_text
write_text functions

In some cases pathlib.Path.open were used in context manager,
it is basically the same as builtin open.

Tests are passing: 850 passed, 5 xfailed, 10 xpassed

This PR changes the display of facets to hopefully make them more readable.

Requested fix in https://github.com/dogsheep/github-to-sqlite/pull/59

The stargazers and workflows commands always require an auth file, even when using a GITHUB_TOKEN. Other commands don't require the auth file.

This may be solved or a feature already, but I couldn't figure it out, is it possible to extract and store also full text from the saved pages? The same way that Pocket parses the text, it'd be amazing to be able to store (and thus make searchable later) the text.

Thank you very much for the project, it's such an amazing idea!

From #50 - right now it will crash with an error of it hits the rate limit. Since the rate limit information (including reset time) is available in the headers it could automatically sleep and try again instead.

If you're writing tests for a Datasette plugin and you get a 500 error from inside Datasette, you can cause Datasette to open a PDB session within the application server code by doing this:

ds = Datasette([db_path], pdb=True)
response = await ds.client.get("/")

You'll need to run pytest -s to interact with the debugger, otherwise you'll get an error.

v0.54 500 error in sql query template; code worked in v0.53; found a workaround

schema:
CREATE TABLE "talks" ("talk" TEXT,"series" INTEGER, "talkdate" TEXT)
CREATE TABLE "series" ("id" INTEGER PRIMARY KEY, "series" TEXT, talks_list TEXT default '', website TEXT default '');

Live example of correctly rendered template in v.053: https://cosmotalks-cy6xkkbezq-uw.a.run.app/cosmotalks/talks/1

Description of problem: I needed 'sql select' code in a custom row-mydatabase-mytable.html template to lookup the series name for a foreign key integer value in the talks table. So metadata.json specifies the datasette-template-sql plugin.

The code below worked perfectly in v0.53 (just the relevant sql statement part is shown; full code is here):

{# custom addition #}  
{% for row in display_rows %}  
    ...  
    {% set sname = sql("select series from series where id = ?", [row.series]) %}  
    <strong>Series name: {{ sname[0].series }}  
    ...  
{% endfor %}  
{# End of custom addition #}  

In v0.54, that code resulted in a 500 error with a 'no such table series' message. A second query in that template also did not work but the above is fully illustrative of the problem.

All templates were up-to-date along with datasette v0.54.

Workaround: After fiddling around with trying different things, what worked was the syntax from Querying a different database from the datasette-template-sql github repo to add the database name to the sql statement:

{% set sname = sql("select series from series where id = ?", [row.series], database="cosmotalks") %}

Though this was found to work, it should not be necessary to add database="cosmotalks" since per the datasette-template-sql README, it's only needed when querying a different database, but here it's a table within the same database.

For consistency with /_internal - and because then we don't need to escape the : characters.

This change would need to be in before Datasette 1.0. I could land it earlier and set up redirects from the old URLs though.

I requested a Twitter export yesterday, and unfortunately they seem to have changed it such that twitter-to-sqlite import can't handle it anymore 😢

So far I've ran into two issues. The first was easy to work around, but the second will take more investigation. If I can find the time I'll keep working on it and update this issue accordingly.

The issues (so far):

1. Data seems to have moved to a data/ subdirectory

Running twitter-to-sqlite import on the raw zip file reports a bunch of "not yet implemented" errors, and then exits without actually importing anything:

❯ twitter-to-sqlite import tarchive.db twitter.zip
...
data/manifest: not yet implemented
data/account-creation-ip: not yet implemented
data/account-suspension: not yet implemented
... (dozens of more lines like this, including critical stuff like data/tweets) ...

(tarchive.db now exists, but is empty)

Workaround: unpack the zip file, and run twitter-to-sqlite import tarchive.db path/to/archive/data

That gets further, but:

2. Some schema(s?) have changed

At least, the blocks schema seems different now:

❯ twitter-to-sqlite import tarchive.db archive/data
direct-messages-group: not yet implemented
branch-links: not yet implemented
periscope-expired-broadcasts: not yet implemented
direct-messages: not yet implemented
mute: not yet implemented
Traceback (most recent call last):
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/bin/twitter-to-sqlite", line 8, in <module>
    sys.exit(cli())
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/click/core.py", line 829, in __call__
    return self.main(*args, **kwargs)
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/click/core.py", line 782, in main
    rv = self.invoke(ctx)
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/click/core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/click/core.py", line 1066, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/click/core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/twitter_to_sqlite/cli.py", line 772, in import_
    archive.import_from_file(db, filepath.name, open(filepath, "rb").read())
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/twitter_to_sqlite/archive.py", line 215, in import_from_file
    to_insert = transformer(data)
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/twitter_to_sqlite/archive.py", line 115, in lists_member
    return {"lists-member": _list_from_common(data)}
  File "/Users/jacob/Library/Caches/pypoetry/virtualenvs/jacobian-dogsheep-4AXaN4tu-py3.8/lib/python3.8/site-packages/twitter_to_sqlite/archive.py", line 200, in _list_from_common
    for url in block["userListInfo"]["urls"]:
KeyError: 'urls'

That's as far as I got before I needed to work on something else. I'll report back if I get further!

Next challenge: figure out how to use the Database class from https://github.com/simonw/datasette/blob/0.53/datasette/database.py for an in-memory database which persists data for the duration of the lifetime of the server, and allows access to that in-memory database from multiple threads in a way that lets them see each other's changes.

_Originally posted by @simonw in https://github.com/simonw/datasette/issues/1150#issuecomment-747768112_

Most Datasette instances only host a single database, but the current homepage design assumes that it should leave plenty of space for multiple databases:

https://user-images.githubusercontent.com/9599/95024344-5b51fd80-0637-11eb-8a11-40bad16f6907.png">

Reconsider this design - should the default show more information?

The Covid-19 Datasette homepage looks particularly sparse I think: https://covid-19.datasettes.com/

https://user-images.githubusercontent.com/9599/95024391-876d7e80-0637-11eb-8f19-ef38e4c87d2a.png">

to keep the data write from failing if it came across an error during processing. In particular when trying to convert my HealthKit zip file (and that of my wife's) it would consistently error out with the following:

db.py 1709 insert_chunk
result = self.db.execute(query, params)

db.py 226 execute
return self.conn.execute(sql, parameters)

sqlite3.OperationalError:
too many SQL variables

---------------------------------------------------------------------------------------------------------------------------------------------------------------------
db.py 1709 insert_chunk
result = self.db.execute(query, params)

db.py 226 execute
return self.conn.execute(sql, parameters)

sqlite3.OperationalError:
too many SQL variables

---------------------------------------------------------------------------------------------------------------------------------------------------------------------
db.py 1709 insert_chunk
result = self.db.execute(query, params)

db.py 226 execute
return self.conn.execute(sql, parameters)

sqlite3.OperationalError:
table rBodyMass has no column named metadata_HKWasUserEntered

---------------------------------------------------------------------------------------------------------------------------------------------------------------------
healthkit-to-sqlite 8 <module>
sys.exit(cli())

core.py 829 __call__
return self.main(*args, **kwargs)

core.py 782 main
rv = self.invoke(ctx)

core.py 1066 invoke
return ctx.invoke(self.callback, **ctx.params)

core.py 610 invoke
return callback(*args, **kwargs)

cli.py 57 cli
convert_xml_to_sqlite(fp, db, progress_callback=bar.update, zipfile=zf)

utils.py 42 convert_xml_to_sqlite
write_records(records, db)

utils.py 143 write_records
db[table].insert_all(

db.py 1899 insert_all
self.insert_chunk(

db.py 1720 insert_chunk
self.insert_chunk(

db.py 1720 insert_chunk
self.insert_chunk(

db.py 1714 insert_chunk
result = self.db.execute(query, params)

db.py 226 execute
return self.conn.execute(sql, parameters)

sqlite3.OperationalError:
table rBodyMass has no column named metadata_HKWasUserEntered

Adding the try / except in the write_records seems to fix that issue.

These will incorporate the release notes from the alpha, much expanded: https://github.com/simonw/datasette/releases/tag/0.54a0

Refs #1201