feat(web): adjust web to new pathogen.json format

nextstrain/nextclade_data#97
nextstrain · Oct 24, 2023 · a529628 · a529628
1 parent b2b5e70
commit a529628
Show file tree

Hide file tree

Showing 7 changed files with 83 additions and 44 deletions.
diff --git a/packages_rs/nextclade-web/src/components/Main/DatasetInfo.tsx b/packages_rs/nextclade-web/src/components/Main/DatasetInfo.tsx
@@ -1,4 +1,4 @@
-import { isNil } from 'lodash'
+import { isNil, last } from 'lodash'
 import { darken } from 'polished'
 import React, { useMemo } from 'react'
 import { Badge } from 'reactstrap'
@@ -12,6 +12,7 @@ import {
   DATASET_ID_UNDETECTED,
   numberAutodetectResultsAtom,
 } from 'src/state/autodetect.state'
+import { AnyType, attrBoolMaybe, attrStrMaybe } from 'src/types'
 import type { Dataset } from 'src/types'
 import styled from 'styled-components'
 
@@ -77,8 +78,7 @@ export interface DatasetInfoProps {
 
 export function DatasetInfo({ dataset }: DatasetInfoProps) {
   const { t } = useTranslationSafe()
-  const { attributes, official, deprecated, enabled, experimental, path, version } = dataset
-  const { name, reference } = attributes
+  const { attributes, path, version } = dataset
 
   const updatedAt = useMemo(() => {
     let updatedAt = version?.updatedAt ? formatDateIsoUtcSimple(version?.updatedAt) : 'unknown'
@@ -88,10 +88,6 @@ export function DatasetInfo({ dataset }: DatasetInfoProps) {
     return updatedAt
   }, [version?.tag, version?.updatedAt])
 
-  if (!enabled) {
-    return null
-  }
-
   if (path === DATASET_ID_UNDETECTED) {
     return <DatasetUndetectedInfo />
   }
@@ -104,10 +100,10 @@ export function DatasetInfo({ dataset }: DatasetInfoProps) {
 
       <FlexRight>
         <DatasetName>
-          <span>{name.valueFriendly ?? name.value ?? path}</span>
+          <span>{attrStrMaybe(attributes, 'name') ?? path}</span>
 
           <span className="d-flex ml-auto">
-            {official ? (
+            {path.startsWith('nextstrain') ? (
               <DatasetInfoBadge
                 className="ml-2 my-auto"
                 color="success"
@@ -125,7 +121,7 @@ export function DatasetInfo({ dataset }: DatasetInfoProps) {
               </DatasetInfoBadge>
             )}
 
-            {experimental && (
+            {attrBoolMaybe(attributes, 'experimental') && (
               <DatasetInfoBadge
                 className="ml-2 my-auto"
                 color="warning"
@@ -135,7 +131,7 @@ export function DatasetInfo({ dataset }: DatasetInfoProps) {
               </DatasetInfoBadge>
             )}
 
-            {deprecated && (
+            {attrBoolMaybe(attributes, 'deprecated') && (
               <DatasetInfoBadge
                 className="ml-2 my-auto"
                 color="secondary"
@@ -147,19 +143,23 @@ export function DatasetInfo({ dataset }: DatasetInfoProps) {
           </span>
         </DatasetName>
 
-        <DatasetInfoLine>
-          {t('Reference: {{ name }} ({{ accession }})', {
-            name: reference.valueFriendly ?? 'Untitled',
-            accession: reference.value,
-          })}
-        </DatasetInfoLine>
+        <DatasetInfoLine>{t('Reference: {{ ref }})', { ref: formatReference(attributes) })}</DatasetInfoLine>
         <DatasetInfoLine>{t('Updated at: {{updated}}', { updated: updatedAt })}</DatasetInfoLine>
         <DatasetInfoLine>{t('Dataset name: {{name}}', { name: path })}</DatasetInfoLine>
       </FlexRight>
     </Container>
   )
 }
 
+function formatReference(attributes: Record<string, AnyType> | undefined) {
+  const name = attrStrMaybe(attributes, 'reference name') ?? 'unknown'
+  const accession = attrStrMaybe(attributes, 'reference accession')
+  if (accession) {
+    return `${name} (${accession})`
+  }
+  return name
+}
+
 export function DatasetUndetectedInfo() {
   const { t } = useTranslationSafe()
 
@@ -181,7 +181,7 @@ export interface DatasetInfoCircleProps {
 
 function DatasetInfoAutodetectProgressCircle({ dataset }: DatasetInfoCircleProps) {
   const { attributes, path } = dataset
-  const { name } = attributes
+  const name = attrStrMaybe(attributes, 'name') ?? last(path.split('/')) ?? '?'
 
   const circleBg = useMemo(() => darken(0.1)(colorHash(path, { saturation: 0.5, reverse: true })), [path])
   const records = useRecoilValue(autodetectResultsByDatasetAtom(path))
@@ -190,7 +190,7 @@ function DatasetInfoAutodetectProgressCircle({ dataset }: DatasetInfoCircleProps
   const { circleText, countText, percentage } = useMemo(() => {
     if (isNil(records)) {
       return {
-        circleText: (firstLetter(name.valueFriendly ?? name.value) ?? ' ').toUpperCase(),
+        circleText: (firstLetter(name) ?? ' ').toUpperCase(),
         percentage: 0,
         countText: '\u00A0',
       }
@@ -203,7 +203,7 @@ function DatasetInfoAutodetectProgressCircle({ dataset }: DatasetInfoCircleProps
       return { circleText, percentage, countText }
     }
     return { circleText: `0%`, percentage: 0, countText: `0 / ${numberAutodetectResults}` }
-  }, [records, name.value, name.valueFriendly, numberAutodetectResults])
+  }, [records, numberAutodetectResults, name])
 
   return (
     <>

diff --git a/packages_rs/nextclade-web/src/components/Main/DatasetSelectorList.tsx b/packages_rs/nextclade-web/src/components/Main/DatasetSelectorList.tsx
@@ -2,6 +2,8 @@ import { get, isNil, sortBy } from 'lodash'
 import { lighten } from 'polished'
 import React, { forwardRef, useCallback, useEffect, useMemo, useRef } from 'react'
 import { useRecoilState, useRecoilValue } from 'recoil'
+import styled from 'styled-components'
+import { areDatasetsEqual, attrStrMaybe, Dataset } from 'src/types'
 import { ListGenericCss } from 'src/components/Common/List'
 import { DatasetInfo } from 'src/components/Main/DatasetInfo'
 import { search } from 'src/helpers/search'
@@ -11,9 +13,6 @@ import {
   autodetectRunStateAtom,
   groupByDatasets,
 } from 'src/state/autodetect.state'
-import type { Dataset } from 'src/types'
-import { areDatasetsEqual } from 'src/types'
-import styled from 'styled-components'
 
 export interface DatasetSelectorListProps {
   datasets: Dataset[]
@@ -61,10 +60,10 @@ export function DatasetSelectorList({
       [...autodetectResult.itemsStartWith, ...autodetectResult.itemsInclude, ...autodetectResult.itemsNotInclude],
       searchTerm,
       (dataset) => [
-        dataset.attributes.name.value,
-        dataset.attributes.name.valueFriendly ?? '',
-        dataset.attributes.reference.value,
-        dataset.attributes.reference.valueFriendly ?? '',
+        dataset.path,
+        attrStrMaybe(dataset.attributes, 'name') ?? '',
+        attrStrMaybe(dataset.attributes, 'reference name') ?? '',
+        attrStrMaybe(dataset.attributes, 'reference accession') ?? '',
       ],
     )
   }, [autodetectResult, searchTerm])

diff --git a/packages_rs/nextclade-web/src/io/fetchDatasetsIndex.ts b/packages_rs/nextclade-web/src/io/fetchDatasetsIndex.ts
@@ -9,10 +9,6 @@ import { axiosFetch } from 'src/io/axiosFetch'
 const MINIMIZER_INDEX_ALGO_VERSION = 'v1'
 const PACKAGE_VERSION = process.env.PACKAGE_VERSION ?? ''
 
-export function isEnabled(dataset: Dataset) {
-  return dataset.enabled
-}
-
 export function isCompatible(dataset: Dataset): boolean {
   const minVersion = dataset.version?.compatibility?.web ?? PACKAGE_VERSION
   return semver.gte(PACKAGE_VERSION, minVersion)
@@ -35,7 +31,7 @@ export function fileUrlsToAbsolute(datasetServerUrl: string, dataset: Dataset):
 
 export function getLatestCompatibleEnabledDatasets(datasetServerUrl: string, datasetsIndexJson: DatasetsIndexV2Json) {
   const datasets = datasetsIndexJson.collections
-    .flatMap((collection) => collection.datasets.filter(isEnabled).filter(isCompatible).filter(isLatest))
+    .flatMap((collection) => collection.datasets.filter(isCompatible).filter(isLatest))
     .map((dataset) => fileUrlsToAbsolute(datasetServerUrl, dataset))
   return { datasets }
 }

diff --git a/packages_rs/nextclade-web/src/io/fetchSingleDatasetFromUrl.ts b/packages_rs/nextclade-web/src/io/fetchSingleDatasetFromUrl.ts
@@ -1,7 +1,6 @@
 import urljoin from 'url-join'
 import { concurrent } from 'fasy'
-
-import { Dataset, VirusProperties } from 'src/types'
+import { attrStrMaybe, Dataset, VirusProperties } from 'src/types'
 import { removeTrailingSlash } from 'src/io/url'
 import { axiosFetch, axiosHead } from 'src/io/axiosFetch'
 import { sanitizeError } from 'src/helpers/sanitizeError'
@@ -24,9 +23,9 @@ export async function fetchSingleDatasetFromUrl(
 
   const datasets = [currentDataset]
   const defaultDataset = currentDataset
-  const currentDatasetName = currentDataset.attributes.name.value
+  const currentDatasetName = currentDataset.path
   const defaultDatasetName = currentDatasetName
-  const defaultDatasetNameFriendly = currentDataset.attributes.name.valueFriendly ?? currentDatasetName
+  const defaultDatasetNameFriendly = attrStrMaybe(currentDataset.attributes, 'name') ?? currentDatasetName
 
   await concurrent.forEach(
     async ([filename, fileUrl]) => {

diff --git a/packages_rs/nextclade-web/src/state/results.state.ts b/packages_rs/nextclade-web/src/state/results.state.ts
@@ -333,7 +333,7 @@ export const analysisStatusGlobalAtom = atom({
         switch (status) {
           case AlgorithmGlobalStatus.started:
             void getPromise(datasetCurrentAtom).then((dataset) => {
-              plausible('Run started', { props: { dataset: dataset?.attributes.name.value ?? 'unknown' } })
+              plausible('Run started', { props: { 'dataset v3': dataset?.path ?? 'unknown' } })
             })
             break
 
@@ -342,8 +342,8 @@ export const analysisStatusGlobalAtom = atom({
               ([results, dataset]) => {
                 plausible('Run completed', {
                   props: {
-                    sequences: results.length,
-                    dataset: dataset?.attributes.name.value ?? 'unknown',
+                    'sequences': results.length,
+                    'dataset v3': dataset?.path ?? 'unknown',
                   },
                 })
               },

diff --git a/packages_rs/nextclade-web/src/types.ts b/packages_rs/nextclade-web/src/types.ts
@@ -1,6 +1,7 @@
-import { isNil, range, sumBy } from 'lodash'
+import { isNil, isNumber, isFinite, isString, range, sumBy, isBoolean, get } from 'lodash'
 import type {
   Aa,
+  AnyType,
   Cds,
   CdsSegment,
   Dataset,
@@ -10,13 +11,13 @@ import type {
   LetterRangeFor_AaAnd_Position, // eslint-disable-line camelcase
   LetterRangeFor_NucAnd_Position, // eslint-disable-line camelcase
   NextcladeErrorOutputs,
+  NextcladeOutputs,
   NextcladeResult,
   Nuc,
   NucSub,
   NucSubLabeled,
   PrivateNucMutations,
   RangeFor_Position, // eslint-disable-line camelcase
-  NextcladeOutputs,
 } from 'src/gen/_SchemaRoot'
 import { StrictOmit } from 'ts-essentials'
 
@@ -139,3 +140,27 @@ export interface AlgorithmInput {
 export function areDatasetsEqual(left?: Dataset, right?: Dataset): boolean {
   return !isNil(left?.path) && !isNil(right?.path) && left?.path === right?.path
 }
+
+export function anyAsStrMaybe(x: AnyType | undefined): string | undefined {
+  return isString(x) ? x : undefined
+}
+
+export function anyAsNumberMaybe(x: AnyType | undefined): number | undefined {
+  return isNumber(x) && isFinite(x) ? x : undefined
+}
+
+export function anyAsBoolMaybe(x: AnyType | undefined): boolean | undefined {
+  return isBoolean(x) ? x : undefined
+}
+
+export function attrStrMaybe(attributes: Record<string, AnyType> | undefined, name: string): string | undefined {
+  return anyAsStrMaybe(get(attributes, name))
+}
+
+export function attrNumberMaybe(attributes: Record<string, AnyType> | undefined, name: string): number | undefined {
+  return anyAsNumberMaybe(get(attributes, name))
+}
+
+export function attrBoolMaybe(attributes: Record<string, AnyType> | undefined, name: string): boolean | undefined {
+  return anyAsBoolMaybe(get(attributes, name))
+}
diff --git a/packages_rs/nextclade/src/utils/any.rs b/packages_rs/nextclade/src/utils/any.rs
@@ -1,17 +1,37 @@
-use derive_more::Display;
+use crate::io::json::{json_stringify, JsonPretty};
 use eyre::{eyre, Report};
 use schemars::JsonSchema;
 use serde::{Deserialize, Serialize};
+use std::collections::BTreeMap;
+use std::fmt::{Display, Formatter};
 
-#[derive(Debug, Display, Clone, Serialize, Deserialize, JsonSchema)]
+#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
 #[serde(rename_all = "camelCase")]
 #[serde(untagged)]
 pub enum AnyType {
   String(String),
   Int(isize),
   Float(f64),
   Bool(bool),
-  Other(serde_json::Value),
+  Array(Vec<AnyType>),
+  Object(BTreeMap<String, AnyType>),
+  Null,
+}
+
+#[allow(clippy::map_err_ignore)]
+impl Display for AnyType {
+  fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+    let s = match self {
+      AnyType::String(x) => x.clone(),
+      AnyType::Int(x) => x.to_string(),
+      AnyType::Float(x) => x.to_string(),
+      AnyType::Bool(x) => x.to_string(),
+      AnyType::Array(x) => json_stringify(&x, JsonPretty(false)).map_err(|_| std::fmt::Error)?,
+      AnyType::Object(x) => json_stringify(&x, JsonPretty(false)).map_err(|_| std::fmt::Error)?,
+      AnyType::Null => "null".to_owned(),
+    };
+    write!(f, "{s}")
+  }
 }
 
 impl AnyType {