kallithea Files · docs/dev/dbmigrations.rst

Files @ caef0be39948

Branch filter:

Location: kallithea/docs/dev/dbmigrations.rst

caef0be39948 2.9 KiB text/prs.fallenstein.rst Show Annotation Show as Raw Download as Raw

FUJIWARA Katsunori

search: make "repository:" condition work as expected

Before this revision, "repository:foo" condition at searching for
"File contents" or "File names" shows files in repositories below.

- foo
- foo/bar
- foo-bar
- and so on ...

Whoosh library, which is used to parse text for indexing and seaching,
does:

- treat almost all non-alphanumeric characters as delimiter both at
indexing search items and at parsing search condition
- make each fields for a search item be indexed by multiple values

For example, files in "foo/bar" repository are indexed by "foo" and
"bar" in "repository" field. This tokenization make "repository:foo"
search condition match against files in "foo/bar" repository, too.

In addition to it, using plain TEXT also causes unintentional
ignorance of "stop words" in search conditions. For example, "this",
"a", "you", and so on are ignored at indexing and parsing, because
these are too generic words (from point of view of generic "text
search").

This issue can't be resolved by using ID instead of TEXT for
"repository" of SCHEMA, like as previous revisions for JOURNAL_SCHEMA,
because:

- highlight-ing file content requires SCHEMA to support "positions"
feature, but using ID instead of TEXT disables it
- using ID violates current case-insensitive search policy, because
it preserves case of text

To make "repository:" condition work as expected, this revision
explicitly specifies "analyzer", which does:

- avoid tokenization
- match case-insensitively
- avoid removing "stop words" from text

This revision requires full re-building index tables, because indexing
schema is changed.

BTW, "repository:" condition at searching for "Commit messages" uses
CHGSETS_SCHEMA instead of SCHEMA. The former uses ID for "repository",
and it does:

- avoid issues by tokenization and removing "stop words"

- disable "positions" feature of CHGSETS_SCHEMA

But highlight-ing file content isn't needed at searching for
"Commit messages". Therefore, this can be ignored.

- preserve case of text

This violates current case-insensitive search policy, This issue
will be fixed by subsequent revision, because fixing it isn't so
simple.

=======================
Database schema changes
=======================

Kallithea uses Alembic for :ref:`database migrations <upgrade_db>`
(upgrades and downgrades).

If you are developing a Kallithea feature that requires database schema
changes, you should make a matching Alembic database migration script:

1. :ref:`Create a Kallithea configuration and database <setup>` for testing
   the migration script, or use existing ``development.ini`` setup.

   Ensure that this database is up to date with the latest database
   schema *before* the changes you're currently developing. (Do not
   create the database while your new schema changes are applied.)

2. Create a separate throwaway configuration for iterating on the actual
   database changes::

    paster make-config Kallithea temp.ini

   Edit the file to change database settings. SQLite is typically fine,
   but make sure to change the path to e.g. ``temp.db``, to avoid
   clobbering any existing database file.

3. Make your code changes (including database schema changes in ``db.py``).

4. After every database schema change, recreate the throwaway database
   to test the changes::

    rm temp.db
    paster setup-db temp.ini --repos=/var/repos --user=doe --email doe@example.com --password=123456 --no-public-access --force-yes
    paster repo-scan temp.ini

5. Once satisfied with the schema changes, auto-generate a draft Alembic
   script using the development database that has *not* been upgraded.
   (The generated script will upgrade the database to match the code.)

   ::

    alembic -c development.ini revision -m "area: add cool feature" --autogenerate

6. Edit the script to clean it up and fix any problems.

   Note that for changes that simply add columns, it may be appropriate
   to not remove them in the downgrade script (and instead do nothing),
   to avoid the loss of data. Unknown columns will simply be ignored by
   Kallithea versions predating your changes.

7. Run ``alembic -c development.ini upgrade head`` to apply changes to
   the (non-throwaway) database, and test the upgrade script. Also test
   downgrades.

   The included ``development.ini`` has full SQL logging enabled. If
   you're using another configuration file, you may want to enable it
   by setting ``level = DEBUG`` in section ``[handler_console_sql]``.

The Alembic migration script should be committed in the same revision as
the database schema (``db.py``) changes.

See the `Alembic documentation`__ for more information, in particular
the tutorial and the section about auto-generating migration scripts.

.. __: http://alembic.zzzcomputing.com/en/latest/


Troubleshooting
---------------

* If ``alembic --autogenerate`` responds "Target database is not up to
  date", you need to either first use Alembic to upgrade the database
  to the most recent version (before your changes), or recreate the
  database from scratch (without your schema changes applied).