This isn't actually working - the outer code attempts to send an .interrupt() call to the connection object via the connections thread-local, which doesn't work because it's a thread-local so the connection isn't visible to that code.

Need to figure out how to communicate with that thread properly.

Also a test that fails in this particular case would be a good idea!

(Without the apt-get update ... SpatiaLite line it's 125MB)

Final version of Dockerfile which installs the specified version from GitHub:

docker build . -t datasette-spatialite --build-arg VERSION=0.55

```dockerfile FROM python:3.9.2-slim-buster as build

Version of Datasette to install, e.g. 0.55

docker build . -t datasette --build-arg VERSION=0.55

ARG VERSION

software-properties-common provides add-apt-repository

which we need in order to install a more recent release

of libsqlite3-mod-spatialite from the sid distribution

RUN apt-get update && \ apt-get -y --no-install-recommends install software-properties-common && \ add-apt-repository "deb http://httpredir.debian.org/debian sid main" && \ apt-get update && \ apt-get -t sid install -y --no-install-recommends libsqlite3-mod-spatialite && \ apt-get remove -y software-properties-common && \ apt clean && \ rm -rf /var/lib/apt && \ rm -rf /var/lib/dpkg

RUN pip install https://github.com/simonw/datasette/archive/refs/tags/${VERSION}.zip && \ find /usr/local/lib -name 'pycache' | xargs rm -r && \ rm -rf /root/.cache/pip

EXPOSE 8001 CMD ["datasette"] ``` Run against 0.55 this produces an image of 262MB

I think part of the reason it's smaller is that I ran pip install datasette instead of using COPY . /datasette followed by pip install /datasette.

Considering the image on Docker Hub is 383MB, I'm happy with getting that down to 262MB. I'm going to stop looking for new optimizations here.

I tried adding apt-get remove -y software-properties-common && to remove software-properties-common but it made no difference to the image size.

I tried copying just the mod_spatialite.so file into a second stage build but it failed. So I ran bash in a working image and used ldd to figure out what it was linked to: root@39683f91e588:/usr/lib/x86_64-linux-gnu# ldd mod_spatialite.so linux-vdso.so.1 (0x00007ffd021f4000) libxml2.so.2 => /usr/lib/x86_64-linux-gnu/libxml2.so.2 (0x00007f5c75412000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f5c753f0000) libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f5c752ac000) libminizip.so.1 => /usr/lib/x86_64-linux-gnu/libminizip.so.1 (0x00007f5c750a0000) librttopo.so.1 => /usr/lib/x86_64-linux-gnu/librttopo.so.1 (0x00007f5c75028000) libfreexl.so.1 => /usr/lib/x86_64-linux-gnu/libfreexl.so.1 (0x00007f5c7501c000) libproj.so.19 => /usr/lib/x86_64-linux-gnu/libproj.so.19 (0x00007f5c74ca7000) libz.so.1 => /lib/x86_64-linux-gnu/libz.so.1 (0x00007f5c74a89000) libsqlite3.so.0 => /usr/lib/x86_64-linux-gnu/libsqlite3.so.0 (0x00007f5c74967000) libgeos_c.so.1 => /usr/lib/x86_64-linux-gnu/libgeos_c.so.1 (0x00007f5c7492b000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f5c74766000) libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f5c74760000) libicuuc.so.67 => /usr/lib/x86_64-linux-gnu/libicuuc.so.67 (0x00007f5c74575000) liblzma.so.5 => /lib/x86_64-linux-gnu/liblzma.so.5 (0x00007f5c7454d000) /lib64/ld-linux-x86-64.so.2 (0x00007f5c75d49000) libtiff.so.5 => /usr/lib/x86_64-linux-gnu/libtiff.so.5 (0x00007f5c744c7000) libcurl-gnutls.so.4 => /usr/lib/x86_64-linux-gnu/libcurl-gnutls.so.4 (0x00007f5c74439000) libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007f5c7426c000) libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007f5c74250000) libgeos-3.9.0.so => /usr/lib/x86_64-linux-gnu/libgeos-3.9.0.so (0x00007f5c74040000) libicudata.so.67 => /usr/lib/x86_64-linux-gnu/libicudata.so.67 (0x00007f5c72527000) libwebp.so.6 => /usr/lib/x86_64-linux-gnu/libwebp.so.6 (0x00007f5c724bc000) libzstd.so.1 => /usr/lib/x86_64-linux-gnu/libzstd.so.1 (0x00007f5c7241c000) libjbig.so.0 => /usr/lib/x86_64-linux-gnu/libjbig.so.0 (0x00007f5c7220e000) libjpeg.so.62 => /usr/lib/x86_64-linux-gnu/libjpeg.so.62 (0x00007f5c72188000) libdeflate.so.0 => /usr/lib/x86_64-linux-gnu/libdeflate.so.0 (0x00007f5c7216c000) libnghttp2.so.14 => /usr/lib/x86_64-linux-gnu/libnghttp2.so.14 (0x00007f5c72144000) libidn2.so.0 => /usr/lib/x86_64-linux-gnu/libidn2.so.0 (0x00007f5c72125000) librtmp.so.1 => /usr/lib/x86_64-linux-gnu/librtmp.so.1 (0x00007f5c71f08000) libssh2.so.1 => /usr/lib/x86_64-linux-gnu/libssh2.so.1 (0x00007f5c71eda000) libpsl.so.5 => /usr/lib/x86_64-linux-gnu/libpsl.so.5 (0x00007f5c71ec5000) libnettle.so.6 => /usr/lib/x86_64-linux-gnu/libnettle.so.6 (0x00007f5c71e8d000) libgnutls.so.30 => /usr/lib/x86_64-linux-gnu/libgnutls.so.30 (0x00007f5c71ce0000) libgssapi_krb5.so.2 => /usr/lib/x86_64-linux-gnu/libgssapi_krb5.so.2 (0x00007f5c71c93000) libkrb5.so.3 => /usr/lib/x86_64-linux-gnu/libkrb5.so.3 (0x00007f5c71bb3000) libk5crypto.so.3 => /usr/lib/x86_64-linux-gnu/libk5crypto.so.3 (0x00007f5c71b7f000) libcom_err.so.2 => /lib/x86_64-linux-gnu/libcom_err.so.2 (0x00007f5c71b77000) libldap_r-2.4.so.2 => /usr/lib/x86_64-linux-gnu/libldap_r-2.4.so.2 (0x00007f5c71b23000) liblber-2.4.so.2 => /usr/lib/x86_64-linux-gnu/liblber-2.4.so.2 (0x00007f5c71b12000) libunistring.so.2 => /usr/lib/x86_64-linux-gnu/libunistring.so.2 (0x00007f5c7198e000) libhogweed.so.4 => /usr/lib/x86_64-linux-gnu/libhogweed.so.4 (0x00007f5c71955000) libgmp.so.10 => /usr/lib/x86_64-linux-gnu/libgmp.so.10 (0x00007f5c718d0000) libgcrypt.so.20 => /lib/x86_64-linux-gnu/libgcrypt.so.20 (0x00007f5c717b2000) libp11-kit.so.0 => /usr/lib/x86_64-linux-gnu/libp11-kit.so.0 (0x00007f5c71683000) libtasn1.so.6 => /usr/lib/x86_64-linux-gnu/libtasn1.so.6 (0x00007f5c71470000) libkrb5support.so.0 => /usr/lib/x86_64-linux-gnu/libkrb5support.so.0 (0x00007f5c71461000) libkeyutils.so.1 => /lib/x86_64-linux-gnu/libkeyutils.so.1 (0x00007f5c71458000) libresolv.so.2 => /lib/x86_64-linux-gnu/libresolv.so.2 (0x00007f5c7143e000) libsasl2.so.2 => /usr/lib/x86_64-linux-gnu/libsasl2.so.2 (0x00007f5c71421000) libgpg-error.so.0 => /lib/x86_64-linux-gnu/libgpg-error.so.0 (0x00007f5c713fe000) libffi.so.6 => /usr/lib/x86_64-linux-gnu/libffi.so.6 (0x00007f5c713f4000)

```dockerfile FROM python:3.9.2-slim-buster as build

software-properties-common provides add-apt-repository

RUN apt-get update && \ apt-get -y --no-install-recommends install software-properties-common && \ add-apt-repository "deb http://httpredir.debian.org/debian sid main" && \ apt-get update && \ apt-get -t sid install -y --no-install-recommends libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/apt && \ rm -rf /var/lib/dpkg

RUN pip install datasette && \ find /usr/local/lib -name 'pycache' | xargs rm -r && \ rm -rf /root/.cache/pip

EXPOSE 8001 CMD ["datasette"] ``` 262 MB

RUN pip install datasette && \ find /usr/local/lib -name '__pycache__' | xargs rm -r That dropped it to 265MB.

Adding --no-install-recommends dropped it to 275MB

I wrote a bunch of tips on creating smaller Docker images here: https://simonwillison.net/2018/Nov/19/smaller-python-docker-images/

Does --no-install-recommends make a difference?

Replacing rm -rf /var/lib/{apt,dpkg,cache,log}/ with rm -rf /var/lib/apt && \ rm -rf /var/lib/dpkg Got the size down to 305MB.

Considering the image on Docker Hub right now is 383MB this is actually an improvement.

This Dockerfile: ```dockerfile FROM python:3.9.2-slim-buster as build

software-properties-common provides add-apt-repository

RUN apt-get update && \ apt-get -y install software-properties-common && \ add-apt-repository "deb http://httpredir.debian.org/debian sid main" && \ apt-get update && \ apt-get -t sid install -y libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

RUN pip install datasette

EXPOSE 8001 CMD ["datasette"] ``` Produces a 344MB image that includes a working SpatiaLite 5.0 module. And weirdly... it doesn't exhibit the hanging bug!

This Dockerfile: ```dockerfile FROM python:3.9.2-slim-buster as build

Setup build dependencies

RUN apt update \ && apt install -y python3-dev build-essential wget libxml2-dev libproj-dev \ libminizip-dev libgeos-dev libsqlite3-dev zlib1g-dev pkg-config git \ && apt clean

RUN wget "https://www.sqlite.org/2021/sqlite-autoconf-3340100.tar.gz" && tar xzf sqlite-autoconf-3340100.tar.gz \ && cd sqlite-autoconf-3340100 && ./configure --disable-static --enable-fts5 --enable-json1 \ CFLAGS="-g -O2 -DSQLITE_ENABLE_FTS3=1 -DSQLITE_ENABLE_FTS3_PARENTHESIS -DSQLITE_ENABLE_FTS4=1 -DSQLITE_ENABLE_RTREE=1 -DSQLITE_ENABLE_JSON1" \ && make && make install

RUN wget "http://www.gaia-gis.it/gaia-sins/freexl-1.0.6.tar.gz" && tar zxf freexl-1.0.6.tar.gz \ && cd freexl-1.0.6 && ./configure && make && make install

RUN wget "http://www.gaia-gis.it/gaia-sins/libspatialite-5.0.1.tar.gz" && tar zxf libspatialite-5.0.1.tar.gz \ && cd libspatialite-5.0.1 && ./configure --disable-rttopo && make && make install

RUN wget "http://www.gaia-gis.it/gaia-sins/readosm-sources/readosm-1.1.0.tar.gz" && tar zxf readosm-1.1.0.tar.gz && cd readosm-1.1.0 && ./configure && make && make install

RUN wget "http://www.gaia-gis.it/gaia-sins/spatialite-tools-5.0.0.tar.gz" && tar zxf spatialite-tools-5.0.0.tar.gz \ && cd spatialite-tools-5.0.0 && ./configure --disable-rttopo && make && make install

Add local code to the image instead of fetching from pypi.

COPY . /datasette

RUN pip install /datasette

RUN pip install datasette

FROM python:3.9.2-slim-buster

Copy python dependencies and spatialite libraries

COPY --from=build /usr/local/lib/ /usr/local/lib/

Copy executables

COPY --from=build /usr/local/bin /usr/local/bin

Copy spatial extensions

COPY --from=build /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu

ENV LD_LIBRARY_PATH=/usr/local/lib

EXPOSE 8001 CMD ["datasette"] ``` Produced a 448MB image.

Ideally I'd like to use the Debian stable python:3.9.2-slim-buster base image but install SpatiaLite from Debian unstable here: https://packages.debian.org/sid/libspatialite7

This pattern might let me do that: https://github.com/helmesjo/cpp_bash_utils/blob/f031e926249f8e2d7f260f22dc8974c6d5be11fe/docker/images/linux-gcc.dockerfile#L20-L24

I can remove this code too: https://github.com/simonw/datasette/blob/6f41c8a2bef309a66588b2875c3e24d26adb4850/datasette/database.py#L190-L192

I figured out the cause of the hang in #1268 - it was caused by select count(*) from SpatialIndex interacting badly with the set_progress_handler() mechanism I was using to implement query time limits. #1271 has a replacement for that using asyncio.wait_for() and conn.interrupt() which should resolve the SpatiaLite issue too.

Closing this because I've figured out the root of the problem now, and I have a potential solution.

1270 looks promising, and I don't want to leave open a security hole where someone could potentially hang Datasette with a nasty count(*) query.

Confirmed that the interrupt() based cancellation mechanism fixes the SpatiaLite issue in #1268!

I think the implementation for this goes here: https://github.com/simonw/datasette/blob/6f41c8a2bef309a66588b2875c3e24d26adb4850/datasette/database.py#L146-L157

I figured out a similar pattern in datasette-ripgrep here: https://github.com/simonw/datasette-ripgrep/blob/0.7/datasette_ripgrep/init.py#L63-L71

Also worth trying is to change this code: python n = 1000 if ms < 50: n = 1 What happens with n = 10 instead?

This may not be necessary if using `.interrupt() for SQLite timeouts in #1270 works.

Even if I implement that workaround in #1269 I'm concerned that this could still allow users to deliberately crash Datasette (if it's running SpatiaLite 5.0) by executing select count(*) from SpatialIndex.

That interrupt timeout mechanism is worth digging into further.

Got two new TILs out of this:

• Tracing every executed Python statement
• Running gdb against a Python process in a running Docker container

If I want to avoid counting virtual tables, I need to detect which tables are virtual tables.

The safest way to do this is probably to pull the sql for every table and then, in Python, check for values that start with create virtual table after converting to lower case, using any number of spaces.

This would catch things like CREATE virtual TABLE which might be missed by a SQL like query.

So one fix could be to avoid running counts for anything that turns out to be a virtual table.

That's in this code here: https://github.com/simonw/datasette/blob/c4f1ec7f33fd7d5b93f0f895dafb5351cc3bfc5b/datasette/database.py#L221-L241

... and sure enough, adding this code fixed the problem: diff diff --git a/datasette/database.py b/datasette/database.py index 3579cce..b466b12 100644 --- a/datasette/database.py +++ b/datasette/database.py @@ -224,6 +226,9 @@ class Database: # Try to get counts for each table, $limit timeout for each count counts = {} for table in await self.table_names(): + if table == "SpatialIndex": + counts[table] = 0 + continue try: table_count = ( await self.execute(

To figure out which SQL query triggers the problem I added this code to write to a log file: python with sqlite_timelimit(conn, time_limit_ms): try: cursor = conn.cursor() with open("/tmp/sql.log", "ab", buffering=0) as fp: fp.write(("{}: {}\n".format(sql, params)).encode("utf-8")) cursor.execute(sql, params if params is not None else {}) I had to use ab binary mode because Python doesn't allow buffering=0 for non-binary file operations.

With the log enabled, I used docker exec -it 589ae68de943 bash to attach to the running container and tail -f /tmp/sql.log to see the logs. Here's where it broke:

select count(*) from [idx_civici_geom_parent]: None select count(*) from [sqlite_stat1]: None select count(*) from [sqlite_stat3]: None select count(*) from [SpatialIndex]: None So attempting to run a count(*) against the SpatialIndex virtual table is the thing that triggers the bug.

Maybe I could implement SQLite query timeouts using the interrupt() method instead of the progress handler hack I'm currently using?

https://stackoverflow.com/questions/43240496/python-sqlite3-how-to-quickly-and-cleanly-interrupt-long-running-query-with-e has some tips.

I tried building a container where the conn.set_progress_handler(handler, n) line was commented out... and it fixed the bug.

I had to run docker kill 16197781a7b5 to kill the broken container - Ctrl+C in the Datasette console window didn't do anything.

Using https://til.simonwillison.net/docker/attach-bash-to-running-container - I figured out how to run gdb. I had to use --privileged here because otherwise gdb showed a "Could not attach to process" error. ``` docker exec --privileged -it 16197781a7b5 bash

apt-get install gdb python3-dbg

gdb /usr/bin/python3 -p 20

This paused the process. I tried running this: (gdb) py-bt Traceback (most recent call first): File "/usr/lib/python3.8/asyncio/base_events.py", line 1845, in _run_once if handle._cancelled: File "/usr/lib/python3.8/asyncio/base_events.py", line 570, in run_forever self._run_once() File "/usr/lib/python3.8/asyncio/base_events.py", line 603, in run_until_complete self.run_forever() File "/usr/local/lib/python3.8/dist-packages/uvicorn/server.py", line 49, in run loop.run_until_complete(self.serve(sockets=sockets)) File "/usr/local/lib/python3.8/dist-packages/uvicorn/main.py", line 386, in run server.run() File "/usr/local/lib/python3.8/dist-packages/datasette/cli.py", line 575, in serve uvicorn.run(ds.app(), uvicorn_kwargs) File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 610, in invoke return callback(*args, kwargs) File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 1066, in invoke return ctx.invoke(self.callback, ctx.params) File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 1259, in invoke return _process_result(sub_ctx.command.invoke(sub_ctx)) File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 782, in main rv = self.invoke(ctx) File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 829, in call return self.main(*args, kwargs) File "/usr/local/bin/datasette", line 8, in <module> sys.exit(cli()) <built-in method exec of module object at remote 0x7f0981a280e0> File "/usr/lib/python3.8/trace.py", line 450, in runctx exec(cmd, globals, locals) File "/usr/lib/python3.8/trace.py", line 6632, in main File "/usr/lib/python3.8/trace.py", line 756, in <module> main() <built-in method exec of module object at remote 0x7f0981a280e0> File "/usr/lib/python3.8/runpy.py", line 343, in _run_code File "/usr/lib/python3.8/runpy.py", line 450, in _run_module_as_main ``` Not sure if that's useful or not.

So I think there's a bug in the way the set_progress_handler() mechanism works when used in conjunction with SpatiaLite 5.0 on Linux.

Hitting http://localhost:8001/tuscany_housenumbers triggers the bug. It gets stuck in a loop that looks like this:

Which looks to me like this code: https://github.com/simonw/datasette/blob/8e18c7943181f228ce5ebcea48deb59ce50bee1f/datasette/utils/init.py#L139-L158

Hitting http://localhost:8001/ successfully shows the homepage (after a lot more scrolling).

Here's a fun debugging trick:

docker run -it -p 8001:8001 -v `pwd`:/mnt datasette-spatialite:latest bash
root@16197781a7b5:/# python3 -m trace --trace $(which datasette) \
  -p 8001 -h 0.0.0.0 /mnt/tuscany_housenumbers.sqlite \
  --load-extension=spatialite

A huge amount of stuff scrolls past as Datasette starts up, since we are tracing every executed line of Python.

After about a minute it's finished starting and gets to this point:

selectors.py(452): if timeout is None: selectors.py(454): elif timeout <= 0: selectors.py(459): timeout = math.ceil(timeout * 1e3) * 1e-3 selectors.py(464): max_ev = max(len(self._fd_to_key), 1) selectors.py(466): ready = [] selectors.py(467): try: selectors.py(468): fd_event_list = self._selector.poll(timeout, max_ev) Now I can make some HTTP requests against it.

gdb may be able to help debug this: https://www.podoliaka.org/2016/04/10/debugging-cpython-gdb/

I'll spin off a separate ticket to investigate the hang.

Thanks for doing this - I've used alternativeto.net a bunch in the past, it's great to see Datasette listed there.

This does raise some interesting philosophical questions: three years into the project I'm still not entirely sure what Datasette competes with! Could be SQLite desktop packages, could be visualization software like Tableau, could even be something like Airtable (given a few more plugins).

It will be interesting to see how the alternativeto listing evolves, maybe it will help me answer that question!

Well this is frustrating. I finally found a Dockerfile that worked and installed an Ubuntu pre-compiled SpatiaLite module that would load... ```dockerfile FROM ubuntu:20.10 as install_spatialite

RUN apt update && \ apt install -y libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

FROM ubuntu:20.10

RUN apt update && \ apt install -y python3-pip && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

RUN pip install datasette

Copy spatial extensions

COPY --from=install_spatialite /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/

ENV LD_LIBRARY_PATH=/usr/local/lib

EXPOSE 8001 CMD ["datasette"] ``` (Which produced a 550MB image)

And when I ran Datasette I got that same error where the database listing page hangs! docker run -p 8001:8001 -v `pwd`:/mnt datasette-spatialite:latest datasette -p 8001 -h 0.0.0.0 /mnt/tuscany_housenumbers.sqlite --load-extension=spatialite

Here's why: datasette % docker run -it -p 8001:8001 -v `pwd`:/mnt datasette-spatialite:latest bash root@3430352ff378:/# datasette bash: /usr/local/bin/datasette: /usr/bin/python3: bad interpreter: No such file or directory

Got this error attempting to run Datasette (with or without SpatiaLite): standard_init_linux.go:219: exec user process caused: no such file or directory

I'll try using ubuntu:20.10 for everything: ```dockerfile FROM ubuntu:20.10 as install_spatialite

RUN apt update && \ apt install -y libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

FROM ubuntu:20.10 as build

RUN apt update && \ apt install -y python3-pip && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

RUN pip install datasette

COPY . /datasette

RUN pip install /datasette

FROM ubuntu:20.10

Copy python dependencies and spatialite libraries

COPY --from=build /usr/local/lib/ /usr/local/lib/

Copy executables

COPY --from=build /usr/local/bin /usr/local/bin

Copy spatial extensions

COPY --from=install_spatialite /usr/lib/x86_64-linux-gnu/mod_spatialite.so /usr/lib/x86_64-linux-gnu/mod_spatialite.so

ENV LD_LIBRARY_PATH=/usr/local/lib

EXPOSE 8001 CMD ["datasette"] ```

I tried copying just the mod_spatialite.so file: ```dockerfile FROM ubuntu:20.10 as install_spatialite

RUN apt update && \ apt install -y libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

FROM python:3.9.2-slim as build

RUN pip install datasette

COPY . /datasette

RUN pip install /datasette

FROM python:3.9.2-slim

Copy python dependencies and spatialite libraries

COPY --from=build /usr/local/lib/ /usr/local/lib/

Copy executables

COPY --from=build /usr/local/bin /usr/local/bin

Copy spatial extensions

COPY --from=install_spatialite /usr/lib/x86_64-linux-gnu/mod_spatialite.so /usr/lib/x86_64-linux-gnu/mod_spatialite.so

ENV LD_LIBRARY_PATH=/usr/local/lib

EXPOSE 8001 CMD ["datasette"] But when I ran Datasette with `--load-extension=spatialite` I got this: File "/usr/local/lib/python3.9/site-packages/datasette/database.py", line 151, in in_thread self.ds._prepare_connection(conn, self.name) File "/usr/local/lib/python3.9/site-packages/datasette/app.py", line 502, in _prepare_connection conn.execute(f"SELECT load_extension('{extension}')") sqlite3.OperationalError: /usr/lib/x86_64-linux-gnu/mod_spatialite.so.so: cannot open shared object file: No such file or directory ```

... except my clever image using SpatiaLite installed for Ubuntu doesn't actually work:

datasette % docker run -p 8001:8001 -v `pwd`:/mnt datasette-spatialite:latest datasette -p 8001 -h 0.0.0.0 /mnt/fixtures.db File "/usr/local/lib/python3.9/sqlite3/dbapi2.py", line 27, in <module> from _sqlite3 import * ImportError: /lib/x86_64-linux-gnu/libm.so.6: version `GLIBC_2.29' not found (required by /usr/lib/x86_64-linux-gnu/libsqlite3.so.0)

Which is great, because the image on Docker Hub right now is 383MB.

Here's a trick: install SpatiaLite in ubuntu:20.10 and then copy it into the final python:3.9.2-slim image.

```dockerfile FROM ubuntu:20.10 as install_spatialite

RUN apt update && \ apt install -y libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/

FROM python:3.9.2-slim as build

RUN pip install datasette

COPY . /datasette

RUN pip install /datasette

FROM python:3.9.2-slim

Copy python dependencies and spatialite libraries

COPY --from=build /usr/local/lib/ /usr/local/lib/

Copy executables

COPY --from=build /usr/local/bin /usr/local/bin

Copy spatial extensions

COPY --from=install_spatialite /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu

ENV LD_LIBRARY_PATH=/usr/local/lib

EXPOSE 8001 CMD ["datasette"] ``` That produced a 265MB image.

I tried that with just python3-pip (removing libsqlite3-mod-spatialite) and got 435MB.

Building a Dockerfile containing just FROM ubuntu:20.10 gave me 79.5MB.

Building this one: ```dockerfile FROM ubuntu:20.10

Setup build dependencies

RUN apt update && \ apt install -y python3-pip libsqlite3-mod-spatialite && \ apt clean && \ rm -rf /var/lib/{apt,dpkg,cache,log}/ ``` Resulted in a 515MB image.

It's pretty big though. I tried this version which avoids copying junk from my laptop in:

```dockerfile FROM ubuntu:20.10

Setup build dependencies

RUN apt update && apt install -y python3-pip libsqlite3-mod-spatialite && apt clean

RUN pip install datasette

EXPOSE 8001 CMD ["datasette"] And got this: datasette % docker images datasette-spatialite
REPOSITORY TAG IMAGE ID CREATED SIZE datasette-spatialite latest 0796950653c2 2 seconds ago 528MB ```

This is the shortest Dockerfile that appeared to give me a working SpatiaLite module: ```dockerfile FROM ubuntu:20.10

Setup build dependencies

RUN apt update && apt install -y python3-pip libsqlite3-mod-spatialite && apt clean

Add local code to the image instead of fetching from pypi.

COPY . /datasette

RUN pip install /datasette

RUN rm -rf /datasette

EXPOSE 8001 CMD ["datasette"] ```

To debug I'm running:

docker run -it -p 8001:8001 -v `pwd`:/mnt datasette-spatialite:latest bash

This gets me a shell I can use.

I'm messing around with the Dockerfile and after each change I'm running:

docker build . -t datasette-spatialite

And then:

docker run -p 8001:8001 -v `pwd`:/mnt datasette-spatialite:latest datasette -p 8001 -h 0.0.0.0 /mnt/fixtures.db

Actually for the loadable module I think I need https://packages.ubuntu.com/groovy/libsqlite3-mod-spatialite

So I'm going to try 20.10 and see where that gets me.

https://pythonspeed.com/articles/base-image-python-docker-images/ suggests using python:3.9-slim-buster or ubuntu:20.04 - but 20.04 is focal which still has SpatiaLite 4.3.0a-6build1 - It's 20.10 that has 5.0: https://packages.ubuntu.com/groovy/libspatialite-dev

Debian sid has it too: https://packages.debian.org/sid/libspatialite-dev

Ubuntu Groovy has a package for SpatiaLite 5 - I could try using that instead: https://packages.ubuntu.com/groovy/libspatialite-dev

Here's something odd: when I run datasette tuscany_housenumbers.sqlite --load-extension=spatialite on macOS against SpatiaLite installed using Homebrew (which reports "spatialite": "5.0.0" on the /-/versions page) I don't get any weird errors at all, everything works fine.

But when I tried compiling SpatiaLite inside the Docker container I had hanging errors on some pages.

This is using https://www.gaia-gis.it/gaia-sins/knn/tuscany_housenumbers.7z from the SpatiaLite KNN tutorial at https://www.gaia-gis.it/fossil/libspatialite/wiki?name=KNN