Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

🎉 automatic breadcrumbs from the tag graph #4343

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
+425 −50
Open

🎉 automatic breadcrumbs from the tag graph #4343

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
282 changes: 281 additions & 1 deletion adminSiteServer/app.test.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
import { google } from "googleapis"
import { jest } from "@jest/globals"
import { beforeAll, jest } from "@jest/globals"
// Mock the google docs api to retrieve files from the test-files directory
// AFAICT, we have to do this directly after the import
// and before any other code that might import googleapis
Expand Down Expand Up @@ -43,17 +43,30 @@ import knex, { Knex } from "knex"
import { dbTestConfig } from "../db/tests/dbTestConfig.js"
import {
TransactionCloseMode,
getBestBreadcrumbs,
getParentTagArraysByChildName,
knexReadWriteTransaction,
knexReadonlyTransaction,
setKnexInstance,
} from "../db/db.js"
import { cleanTestDb, TABLES_IN_USE } from "../db/tests/testHelpers.js"
import {
ChartConfigsTableName,
ChartsTableName,
DatasetsTableName,
DbInsertTag,
DbInsertTagGraphNode,
MultiDimDataPagesTableName,
MultiDimXChartConfigsTableName,
TagsTableName,
TagGraphTableName,
VariablesTableName,
TagGraphRootName,
PostsGdocsTableName,
OwidGdocType,
DbInsertPostGdoc,
DbInsertPostGdocXTag,
PostsGdocsXTagsTableName,
} from "@ourworldindata/types"
import path from "path"
import fs from "fs"
Expand Down Expand Up @@ -920,3 +933,270 @@ describe("OwidAdminApp: indicator-level chart configs", () => {
expect(json.success).toBe(false)
})
})

describe.only("OwidAdminApp: tag graph", () => {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Really nice tests!

// prettier-ignore
const dummyTags: DbInsertTag[] = [
{ name: TagGraphRootName, id: 1 },
{ name: "Energy and Environment", id: 2 },
{ name: "Energy", slug: "energy", id: 3 },
{ name: "Nuclear Energy", slug: "nuclear-energy", id: 4 },
{ name: "CO2 & Greenhouse Gas Emissions", slug: "co2-and-greenhouse-gas-emissions", id: 5 },
]

const dummyTagGraph: DbInsertTagGraphNode[] = [
{ parentId: 1, childId: 2 },
{ parentId: 2, childId: 3, weight: 110 },
{ parentId: 2, childId: 5 },
{ parentId: 3, childId: 4 },
{ parentId: 5, childId: 4 },
]

function makeDummyTopicPage(slug: string): DbInsertPostGdoc {
return {
slug,
content: JSON.stringify({
type: OwidGdocType.TopicPage,
authors: [] as string[],
}),
id: slug,
published: 1,
createdAt: new Date(),
publishedAt: new Date(),
markdown: "",
}
}
const dummyTopicPages: DbInsertPostGdoc[] = [
makeDummyTopicPage("energy"),
makeDummyTopicPage("nuclear-energy"),
makeDummyTopicPage("co2-and-greenhouse-gas-emissions"),
]

const dummyPostTags: DbInsertPostGdocXTag[] = [
{ gdocId: "energy", tagId: 3 },
{ gdocId: "nuclear-energy", tagId: 4 },
{ gdocId: "co2-and-greenhouse-gas-emissions", tagId: 5 },
]

beforeEach(async () => {
await testKnexInstance!(TagsTableName).insert(dummyTags)
await testKnexInstance!(TagGraphTableName).insert(dummyTagGraph)
await testKnexInstance!(PostsGdocsTableName).insert(dummyTopicPages)
await testKnexInstance!(PostsGdocsXTagsTableName).insert(dummyPostTags)
})
it("should be able to see all the tags", async () => {
const tags = await fetchJsonFromAdminApi("/tags.json")
expect(tags).toEqual({
tags: [
{
id: 5,
isTopic: 1,
name: "CO2 & Greenhouse Gas Emissions",
slug: "co2-and-greenhouse-gas-emissions",
},
{
id: 3,
isTopic: 1,
name: "Energy",
slug: "energy",
},
{
id: 2,
isTopic: 0,
name: "Energy and Environment",
slug: null,
},
{
id: 4,
isTopic: 1,
name: "Nuclear Energy",
slug: "nuclear-energy",
},
{
id: 1,
isTopic: 0,
name: "tag-graph-root",
slug: null,
},
],
})
})

it("should be able to generate a tag graph", async () => {
const json = await fetchJsonFromAdminApi("/flatTagGraph.json")
expect(json).toEqual({
"1": [
{
childId: 2,
isTopic: 0,
name: "Energy and Environment",
parentId: 1,
weight: 100,
},
],
"2": [
{
childId: 3,
isTopic: 1,
name: "Energy",
parentId: 2,
weight: 110,
},
{
childId: 5,
isTopic: 1,
name: "CO2 & Greenhouse Gas Emissions",
parentId: 2,
weight: 100,
},
],
"3": [
{
childId: 4,
isTopic: 1,
name: "Nuclear Energy",
parentId: 3,
weight: 100,
},
],
"5": [
{
childId: 4,
isTopic: 1,
name: "Nuclear Energy",
parentId: 5,
weight: 100,
},
],
__rootId: 1,
})
})

it("should be able to generate a set of breadcrumbs for a tag", async () => {
await knexReadonlyTransaction(
async (trx) => {
const parentTagArraysByChildName =
await getParentTagArraysByChildName(trx)
const breadcrumbs = getBestBreadcrumbs(
[
{
id: 4,
name: "Nuclear Energy",
slug: "nuclear-energy",
},
],
parentTagArraysByChildName
)
// breadcrumb hrefs are env-dependent, so we just assert on the labels
const labelsOnly = breadcrumbs.map((b) => b.label)
expect(labelsOnly).toEqual(["Energy", "Nuclear Energy"])
},
TransactionCloseMode.KeepOpen,
testKnexInstance
)
})

it("should generate an optimal set of breadcrumbs when given multiple tags", async () => {
await knexReadonlyTransaction(
async (trx) => {
const parentTagArraysByChildName =
await getParentTagArraysByChildName(trx)
const breadcrumbs = getBestBreadcrumbs(
[
{
id: 4,
name: "Nuclear Energy",
slug: "nuclear-energy",
},
{
id: 5,
name: "CO2 & Greenhouse Gas Emissions",
slug: "co2-and-greenhouse-gas-emissions",
},
],
parentTagArraysByChildName
)
// breadcrumb hrefs are env-dependent, so we just assert on the labels
const labelsOnly = breadcrumbs.map((b) => b.label)
expect(labelsOnly).toEqual(["Energy", "Nuclear Energy"])
},
TransactionCloseMode.KeepOpen,
testKnexInstance
)
})
it("should return an empty array when there are no topic tags in any of the tags' ancestors", async () => {
await knexReadonlyTransaction(
async (trx) => {
const parentTagArraysByChildName =
await getParentTagArraysByChildName(trx)
const breadcrumbs = getBestBreadcrumbs(
[
{
id: 2,
name: "Energy and Environment",
slug: "",
},
],
parentTagArraysByChildName
)
// breadcrumb hrefs are env-dependent, so we just assert on the labels
const labelsOnly = breadcrumbs.map((b) => b.label)
expect(labelsOnly).toEqual([])
},
TransactionCloseMode.KeepOpen,
testKnexInstance
)
})
it("when there are two valid paths to a given tag, it selects the longest one", async () => {
await knexReadonlyTransaction(
async (trx) => {
// Here, Women's Employment has 2 paths:
// 1. Poverty and Economic Development > Women's Employment
// 2. Human Rights > Women's Rights > Women's Employment
// prettier-ignore
await testKnexInstance!(TagsTableName).insert([
{ name: "Human Rights", id: 6 },
{ name: "Women's Rights", slug: "womens-rights", id: 7 },
{ name: "Women's Employment", slug: "womens-employment", id: 8 },
{ name: "Poverty and Economic Development", id: 9 },
])
await testKnexInstance!(TagGraphTableName).insert([
{ parentId: 1, childId: 6 },
{ parentId: 6, childId: 7 },
{ parentId: 7, childId: 8 },
{ parentId: 1, childId: 9 },
{ parentId: 9, childId: 8 },
])
await testKnexInstance!(PostsGdocsTableName).insert([
makeDummyTopicPage("womens-rights"),
makeDummyTopicPage("womens-employment"),
])
await testKnexInstance!(PostsGdocsXTagsTableName).insert([
{ gdocId: "womens-rights", tagId: 7 },
{ gdocId: "womens-employment", tagId: 8 },
])

const parentTagArraysByChildName =
await getParentTagArraysByChildName(trx)
const breadcrumbs = getBestBreadcrumbs(
[
{
id: 8,
name: "Women's Employment",
slug: "womens-employment",
},
],
parentTagArraysByChildName
)
// breadcrumb hrefs are env-dependent, so we just assert on the labels
const labelsOnly = breadcrumbs.map((b) => b.label)
expect(labelsOnly).toEqual([
"Women's Rights",
"Women's Employment",
])
},
TransactionCloseMode.KeepOpen,
testKnexInstance
)
})
})
13 changes: 13 additions & 0 deletions baker/SiteBaker.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -597,6 +597,9 @@ export class SiteBaker {
.getPublishedGdocPostsWithTags(knex)
.then((gdocs) => gdocs.map(gdocFromJSON))

const allParentTagArraysByChildName =
await db.getParentTagArraysByChildName(knex)

const gdocsToBake =
slugs !== undefined
? publishedGdocs.filter((gdoc) => slugs.includes(gdoc.slug))
Expand Down Expand Up @@ -630,6 +633,16 @@ export class SiteBaker {
}
publishedGdoc.linkedIndicators = attachments.linkedIndicators

if (
!publishedGdoc.breadcrumbs?.length &&
publishedGdoc.tags?.length
) {
publishedGdoc.breadcrumbs = db.getBestBreadcrumbs(
publishedGdoc.tags,
allParentTagArraysByChildName
)
}

// this is a no-op if the gdoc doesn't have an all-chart block
if ("loadRelatedCharts" in publishedGdoc) {
await publishedGdoc.loadRelatedCharts(knex)
Expand Down
12 changes: 8 additions & 4 deletions baker/algolia/utils/charts.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@ import { isPathRedirectedToExplorer } from "../../../explorerAdminServer/Explore
import { ParsedChartRecordRow, RawChartRecordRow } from "./types.js"
import { excludeNullish } from "@ourworldindata/utils"
import { processAvailableEntities } from "./shared.js"
import { getUniqueNamesFromParentTagArrays } from "@ourworldindata/utils/dist/Util.js"

const computeChartScore = (record: Omit<ChartRecord, "score">): number => {
const { numRelatedArticles, views_7d } = record
Expand Down Expand Up @@ -99,7 +100,8 @@ export const getChartsRecords = async (

const pageviews = await getAnalyticsPageviewsByUrlObj(knex)

const parentTagsByChildName = await db.getParentTagsByChildName(knex)
const parentTagArraysByChildName =
await db.getParentTagArraysByChildName(knex)

const records: ChartRecord[] = []
for (const c of parsedRows) {
Expand All @@ -121,10 +123,12 @@ export const getChartsRecords = async (
fontSize: 10, // doesn't matter, but is a mandatory field
}).plaintext

const parentTags = c.tags.flatMap(
const parentTags = c.tags.flatMap((tagName) => {
const parentTagArrays = parentTagArraysByChildName[tagName]
// a chart can be tagged with a tag that isn't in the tag graph
(tag) => parentTagsByChildName[tag] || []
)
if (!parentTagArrays) return []
return getUniqueNamesFromParentTagArrays(parentTagArrays)
})

const record = {
objectID: c.id.toString(),
Expand Down
Loading
Loading