Jobs: add a nullable jobset_id foreign key to Jobsets.

Dependencies

• [2] FZR7ET5D Jobsets: add a SERIAL, unique, non-null id column
• [3] CCIORPJ5 When manually scheduling an eval, force re-instantiation of store derivations
• [4] MWXDVTS3 Copy the flake migration from the flake branch
• [5] KQAQ4FIF Update Schema classes
• [6] X27GNHDV * Basic job info in the database.
• [7] SB2V735V Keep track of the database schema version
• [8] PCKLFRT5 Support push notification of repository changes
• [9] IGR322YP sql: Generate models from postgresql
• [10] FYO6NECE hydra
• [11] JM3DPYOM generated schema with new dbix class schema loader, grrrrrr
• [12] QLOLZHRX Allow a per-jobset check interval
• [13] DH3KNBAV Merge remote-tracking branch 'upstream/who-broke-builds' into upstream-master
• [14] K3HODXGH Check all inputs for blame but only email selected inputs
• [15] T5BIOVJE Add support for tracking custom metrics
• [16] W4G5MZZS hydra-evaluator improvements
• [17] PCD3ZH6Z Partially revert 1c20cfdf2403feb78cef515faf15c04d5c9f17bd
• [18] S5PV6IIM * Represent jobs explicitly in the DB.
• [19] Y6AHH4TH Remove the logfile and logSize columns from the database
• [20] GEADFVZ5 hydra-queue-runner: Improved scheduling
• [21] KSBB33RE Add a dashboard
• [22] SS4TZXNU Distinguish between permanent evaluation errors and transient input errors
• [23] LZVO64YG Merge in the first bits of the API work
• [24] W5OAZWPD Drop the errorMsg column in the Jobs table
• [*] KN3VYE5P * Cleaned up the foreign key constraints.
• [*] N22GPKYT * Put info about logs / build products in the DB.
• [*] D5QIOJGP * Move everything up one directory.

Change contents

• edit in src/lib/Hydra/Schema/Jobs.pm at line 49
  [3.13837]

  [3.13851]
  +

  + =head2 jobset_id

  +

  + data_type: 'integer'

  + is_foreign_key: 1

  + is_nullable: 1
• edit in src/lib/Hydra/Schema/Jobs.pm at line 68
  [3.15510]

  [3.1774]
  + "jobset_id",

  + { data_type => "integer", is_foreign_key => 1, is_nullable => 1 },
• edit in src/lib/Hydra/Schema/Jobs.pm at line 140
  [26.1352]

  [26.1352]
  + "Hydra::Schema::Jobsets",

  + { id => "jobset_id" },

  + {

  + is_deferrable => 0,

  + join_type => "LEFT",

  + on_delete => "CASCADE",

  + on_update => "NO ACTION",

  + },

  + );

  +

  + =head2 jobset_project_jobset

  +

  + Type: belongs_to

  +

  + Related object: L<Hydra::Schema::Jobsets>

  +

  + =cut

  +

  + __PACKAGE__->belongs_to(

  + "jobset_project_jobset",
• replacement in src/lib/Hydra/Schema/Jobs.pm at line 200
  [3.1556]→[3.3858:4000](∅→∅)
  − # Created by DBIx::Class::Schema::Loader v0.07049 @ 2020-02-06 12:22:36

  − # DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:sYa6dZNK+stMAnTH0Tmn8A
  [3.1556]

  [3.2756]
  + # Created by DBIx::Class::Schema::Loader v0.07049 @ 2020-02-06 12:30:58

  + # DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:dFusVjxb423gIEoadAw9sw
• edit in src/lib/Hydra/Schema/Jobsets.pm at line 277
  [3.2717]

  [3.19391]
  +

  + =head2 jobs_jobset_ids

  +

  + Type: has_many
• replacement in src/lib/Hydra/Schema/Jobsets.pm at line 282
  [3.19392]→[3.19392:19404](∅→∅)
  − =head2 jobs
  [3.19392]

  [3.19404]
  + Related object: L<Hydra::Schema::Jobs>

  +

  + =cut

  +

  + __PACKAGE__->has_many(

  + "jobs_jobset_ids",

  + "Hydra::Schema::Jobs",

  + { "foreign.jobset_id" => "self.id" },

  + undef,

  + );

  +

  + =head2 jobs_project_jobsets
• replacement in src/lib/Hydra/Schema/Jobsets.pm at line 302
  [3.2740]→[3.2740:2750](∅→∅)
  − "jobs",
  [3.2740]

  [3.2750]
  + "jobs_project_jobsets",
• replacement in src/lib/Hydra/Schema/Jobsets.pm at line 396
  [3.22628]→[2.420:562](∅→∅)
  − # Created by DBIx::Class::Schema::Loader v0.07049 @ 2020-02-09 15:25:17

  − # DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:1CMOaLf9fYRdJBlYiivmxA
  [3.22628]

  [2.562]
  + # Created by DBIx::Class::Schema::Loader v0.07049 @ 2020-02-09 15:26:15

  + # DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:DH1jX0smo2rFvyr4V+qJcw

  +

  + =head2 jobs

  +

  + Type: has_many

  +

  + Related object: L<Hydra::Schema::Jobs>

  +

  + =cut

  +

  + __PACKAGE__->has_many(

  + "jobs",

  + "Hydra::Schema::Jobs",

  + {

  + "foreign.jobset" => "self.name",

  + "foreign.project" => "self.project",

  + },

  + undef,

  + );
• edit in src/sql/hydra.sql at line 149
  [26.6902]

  [26.6902]
  + jobset_id integer null,
• edit in src/sql/hydra.sql at line 153
  [26.7316]

  [26.7316]
  + foreign key (jobset_id) references Jobsets(id) on delete cascade,
• file addition: upgrade-60.sql (----------)
  [28.3004]
  + -- Add the jobset_id columns to the Jobs table. This will go

  + -- quickly, since the field is nullable. Note this is just part one of

  + -- this migration. Future steps involve a piecemeal backfilling, and

  + -- then making the column non-null.

  +

  + ALTER TABLE Jobs

  + ADD COLUMN jobset_id integer NULL,

  + ADD FOREIGN KEY (jobset_id)

  + REFERENCES Jobsets(id)

  + ON DELETE CASCADE;