add configurable download mirrors

admin/.env.example

@@ -15,4 +15,8 @@ REDIS_PORT=6379
 # Storage path for NOMAD content (ZIM files, maps, etc.)
 # On Windows dev, use an absolute path like: C:/nomad-storage
 # On Linux production, use: /opt/project-nomad/storage
-NOMAD_STORAGE_PATH=/opt/project-nomad/storage
+NOMAD_STORAGE_PATH=/opt/project-nomad/storage
+# Optional: rewrite download URLs to one or more mirrors using JSON source-prefix => target-prefix mappings.
+# Example for Kiwix:
+# DOWNLOAD_MIRROR_RULES={"https://download.kiwix.org/zim/":"https://cdimage.debian.org/mirror/kiwix.org/zim/"}
+DOWNLOAD_MIRROR_RULES=

admin/app/jobs/run_download_job.ts

@@ -7,6 +7,7 @@ import { DockerService } from '#services/docker_service'
 import { ZimService } from '#services/zim_service'
 import { MapService } from '#services/map_service'
 import { EmbedFileJob } from './embed_file_job.js'
+import { rewriteDownloadUrl } from '../utils/download_mirrors.js'
 
 export class RunDownloadJob {
   static get queue() {
@@ -18,7 +19,7 @@ export class RunDownloadJob {
   }
 
   static getJobId(url: string): string {
-    return createHash('sha256').update(url).digest('hex').slice(0, 16)
+    return createHash('sha256').update(rewriteDownloadUrl(url)).digest('hex').slice(0, 16)
   }
 
   async handle(job: Job) {
@@ -124,10 +125,11 @@ export class RunDownloadJob {
   static async dispatch(params: RunDownloadJobParams) {
     const queueService = new QueueService()
     const queue = queueService.getQueue(this.queue)
-    const jobId = this.getJobId(params.url)
+    const normalizedParams = { ...params, url: rewriteDownloadUrl(params.url) }
+    const jobId = this.getJobId(normalizedParams.url)
 
     try {
-      const job = await queue.add(this.key, params, {
+      const job = await queue.add(this.key, normalizedParams, {
         jobId,
         attempts: 3,
         backoff: { type: 'exponential', delay: 2000 },
@@ -137,7 +139,7 @@ export class RunDownloadJob {
       return {
         job,
         created: true,
-        message: `Dispatched download job for URL ${params.url}`,
+        message: `Dispatched download job for URL ${normalizedParams.url}`,
       }
     } catch (error) {
       if (error.message.includes('job already exists')) {
@@ -145,7 +147,7 @@ export class RunDownloadJob {
         return {
           job: existing,
           created: false,
-          message: `Job already exists for URL ${params.url}`,
+          message: `Job already exists for URL ${normalizedParams.url}`,
         }
       }
       throw error

admin/app/services/collection_manifest_service.ts

@@ -12,10 +12,12 @@ import {
   getFileStatsIfExists,
   ZIM_STORAGE_PATH,
 } from '../utils/fs.js'
+import { rewriteDownloadUrl } from '../utils/download_mirrors.js'
 import type {
   ManifestType,
   ZimCategoriesSpec,
   MapsSpec,
+  WikipediaSpec,
   CategoryWithStatus,
   CollectionWithStatus,
   SpecResource,
@@ -77,7 +79,7 @@ export class CollectionManifestService {
   async getCachedSpec<T>(type: ManifestType): Promise<T | null> {
     const manifest = await CollectionManifest.find(type)
     if (!manifest) return null
-    return manifest.spec_data as T
+    return this.applyDownloadMirrors(type, manifest.spec_data) as T
   }
 
   async getSpecWithFallback<T>(type: ManifestType): Promise<T | null> {
@@ -170,6 +172,48 @@ export class CollectionManifestService {
     return undefined
   }
 
+  private applyDownloadMirrors(
+    type: ManifestType,
+    spec: ZimCategoriesSpec | MapsSpec | WikipediaSpec
+  ): ZimCategoriesSpec | MapsSpec | WikipediaSpec {
+    if (type === 'zim_categories') {
+      return {
+        ...spec,
+        categories: (spec as ZimCategoriesSpec).categories.map((category) => ({
+          ...category,
+          tiers: category.tiers.map((tier) => ({
+            ...tier,
+            resources: tier.resources.map((resource) => ({
+              ...resource,
+              url: rewriteDownloadUrl(resource.url),
+            })),
+          })),
+        })),
+      }
+    }
+
+    if (type === 'maps') {
+      return {
+        ...spec,
+        collections: (spec as MapsSpec).collections.map((collection) => ({
+          ...collection,
+          resources: collection.resources.map((resource) => ({
+            ...resource,
+            url: rewriteDownloadUrl(resource.url),
+          })),
+        })),
+      }
+    }
+
+    return {
+      ...spec,
+      options: (spec as WikipediaSpec).options.map((option) => ({
+        ...option,
+        url: option.url ? rewriteDownloadUrl(option.url) : option.url,
+      })),
+    }
+  }
+
   // ---- Filename parsing ----
 
   static parseZimFilename(filename: string): { resource_id: string; version: string } | null {

admin/app/services/collection_update_service.ts

@@ -4,6 +4,7 @@ import axios from 'axios'
 import InstalledResource from '#models/installed_resource'
 import { RunDownloadJob } from '../jobs/run_download_job.js'
 import { ZIM_STORAGE_PATH } from '../utils/fs.js'
+import { rewriteDownloadUrl } from '../utils/download_mirrors.js'
 import { join } from 'path'
 import type {
   ResourceUpdateCheckRequest,
@@ -49,12 +50,15 @@ export class CollectionUpdateService {
         timeout: 15000,
       })
 
-      logger.info(
-        `[CollectionUpdateService] Update check complete: ${response.data.length} update(s) available`
-      )
+      const updates = response.data.map((update) => ({
+        ...update,
+        download_url: rewriteDownloadUrl(update.download_url),
+      }))
+
+      logger.info(`[CollectionUpdateService] Update check complete: ${updates.length} update(s) available`)
 
       return {
-        updates: response.data,
+        updates,
         checked_at: new Date().toISOString(),
       }
     } catch (error) {
@@ -82,8 +86,10 @@ export class CollectionUpdateService {
   async applyUpdate(
     update: ResourceUpdateInfo
   ): Promise<{ success: boolean; jobId?: string; error?: string }> {
+    const downloadUrl = rewriteDownloadUrl(update.download_url)
+
     // Check if a download is already in progress for this URL
-    const existingJob = await RunDownloadJob.getByUrl(update.download_url)
+    const existingJob = await RunDownloadJob.getByUrl(downloadUrl)
     if (existingJob) {
       const state = await existingJob.getState()
       if (state === 'active' || state === 'waiting' || state === 'delayed') {
@@ -98,7 +104,7 @@ export class CollectionUpdateService {
     const filepath = this.buildFilepath(update, filename)
 
     const result = await RunDownloadJob.dispatch({
-      url: update.download_url,
+      url: downloadUrl,
       filepath,
       timeout: 30000,
       allowedMimeTypes:

admin/app/services/docker_service.ts

@@ -4,6 +4,7 @@ import logger from '@adonisjs/core/services/logger'
 import { inject } from '@adonisjs/core'
 import transmit from '@adonisjs/transmit/services/main'
 import { doResumableDownloadWithRetry } from '../utils/downloads.js'
+import { rewriteDownloadUrl } from '../utils/download_mirrors.js'
 import { join } from 'path'
 import { ZIM_STORAGE_PATH } from '../utils/fs.js'
 import { SERVICE_NAMES } from '../../constants/service_names.js'
@@ -614,8 +615,9 @@ export class DockerService {
      * At least one .zim file must be available before we can start the kiwix container.
      * We'll download the lightweight mini Wikipedia Top 100 zim file for this purpose.
      **/
-    const WIKIPEDIA_ZIM_URL =
+    const WIKIPEDIA_ZIM_URL = rewriteDownloadUrl(
       'https://github.com/Crosstalk-Solutions/project-nomad/raw/refs/heads/main/install/wikipedia_en_100_mini_2025-06.zim'
+    )
     const filename = 'wikipedia_en_100_mini_2025-06.zim'
     const filepath = join(process.cwd(), ZIM_STORAGE_PATH, filename)
     logger.info(`[DockerService] Kiwix Serve pre-install: Downloading ZIM file to ${filepath}`)

admin/app/services/map_service.ts

@@ -13,6 +13,7 @@ import {
   getFile,
   ensureDirectoryExists,
 } from '../utils/fs.js'
+import { rewriteDownloadUrl } from '../utils/download_mirrors.js'
 import { join, resolve, sep } from 'path'
 import urlJoin from 'url-join'
 import { RunDownloadJob } from '#jobs/run_download_job'
@@ -61,9 +62,9 @@ export class MapService implements IMapService {
       'https://github.com/Crosstalk-Solutions/project-nomad-maps/raw/refs/heads/master/'
     )
 
-    const resolvedURL = url ? new URL(url) : defaultTarFileURL
+    const resolvedURL = rewriteDownloadUrl((url ? new URL(url) : defaultTarFileURL).toString())
     await doResumableDownloadWithRetry({
-      url: resolvedURL.toString(),
+      url: resolvedURL,
       filepath: tempTarPath,
       timeout: 30000,
       max_retries: 2,
@@ -245,7 +246,7 @@ export class MapService implements IMapService {
 
       // Perform a HEAD request to get the content length
       const { default: axios } = await import('axios')
-      const response = await axios.head(url)
+      const response = await axios.head(rewriteDownloadUrl(url))
 
       if (response.status !== 200) {
         throw new Error(`Failed to fetch file info: ${response.status} ${response.statusText}`)

admin/app/services/zim_service.ts

@@ -16,19 +16,17 @@ import {
   listDirectoryContents,
   ZIM_STORAGE_PATH,
 } from '../utils/fs.js'
+import { rewriteDownloadUrl } from '../utils/download_mirrors.js'
 import { join, resolve, sep } from 'path'
 import { WikipediaOption, WikipediaState } from '../../types/downloads.js'
-import vine from '@vinejs/vine'
-import { wikipediaOptionsFileSchema } from '#validators/curated_collections'
 import WikipediaSelection from '#models/wikipedia_selection'
 import InstalledResource from '#models/installed_resource'
 import { RunDownloadJob } from '#jobs/run_download_job'
 import { SERVICE_NAMES } from '../../constants/service_names.js'
 import { CollectionManifestService } from './collection_manifest_service.js'
-import type { CategoryWithStatus } from '../../types/collections.js'
+import type { CategoryWithStatus, WikipediaSpec } from '../../types/collections.js'
 
 const ZIM_MIME_TYPES = ['application/x-zim', 'application/x-openzim', 'application/octet-stream']
-const WIKIPEDIA_OPTIONS_URL = 'https://raw.githubusercontent.com/Crosstalk-Solutions/project-nomad/refs/heads/main/collections/wikipedia.json'
 
 @inject()
 export class ZimService {
@@ -106,7 +104,9 @@ export class ZimService {
       }
 
       // downloadLink['href'] will end with .meta4, we need to remove that to get the actual download URL
-      const download_url = downloadLink['href'].substring(0, downloadLink['href'].length - 6)
+      const download_url = rewriteDownloadUrl(
+        downloadLink['href'].substring(0, downloadLink['href'].length - 6)
+      )
       const file_name = download_url.split('/').pop() || `${entry.title}.zim`
       const sizeBytes = parseInt(downloadLink['length'], 10)
 
@@ -361,20 +361,14 @@ export class ZimService {
   // Wikipedia selector methods
 
   async getWikipediaOptions(): Promise<WikipediaOption[]> {
-    try {
-      const response = await axios.get(WIKIPEDIA_OPTIONS_URL)
-      const data = response.data
-
-      const validated = await vine.validate({
-        schema: wikipediaOptionsFileSchema,
-        data,
-      })
-
-      return validated.options
-    } catch (error) {
-      logger.error(`[ZimService] Failed to fetch Wikipedia options:`, error)
+    const manifestService = new CollectionManifestService()
+    const spec = await manifestService.getSpecWithFallback<WikipediaSpec>('wikipedia')
+    if (!spec) {
+      logger.error('[ZimService] Failed to fetch Wikipedia options: no spec available')
       throw new Error('Failed to fetch Wikipedia options')
     }
+
+    return spec.options
   }
 
   async getWikipediaSelection(): Promise<WikipediaSelection | null> {

admin/app/utils/download_mirrors.ts

@@ -0,0 +1,140 @@
+type DownloadMirrorRule = {
+  source: string
+  target: string
+}
+
+let cachedMirrorRulesRaw: string | undefined
+let cachedMirrorRules: DownloadMirrorRule[] = []
+
+function normalizeMirrorPrefix(urlString: string): string {
+  const parsed = new URL(urlString)
+  parsed.search = ''
+  parsed.hash = ''
+  return parsed.toString()
+}
+
+function joinMirrorUrl(targetPrefix: string, suffix: string): string {
+  if (!suffix) return targetPrefix
+
+  if (targetPrefix.endsWith('/') && suffix.startsWith('/')) {
+    return `${targetPrefix}${suffix.slice(1)}`
+  }
+
+  if (!targetPrefix.endsWith('/') && !suffix.startsWith('/')) {
+    return `${targetPrefix}/${suffix}`
+  }
+
+  return `${targetPrefix}${suffix}`
+}
+
+function normalizeDownloadMirrorRule(source: string, target: string): DownloadMirrorRule {
+  const normalizedSource = normalizeMirrorPrefix(source)
+  const normalizedTarget = normalizeMirrorPrefix(target)
+
+  return {
+    source: normalizedSource,
+    target: normalizedTarget,
+  }
+}
+
+export function parseDownloadMirrorRules(raw?: string | null): DownloadMirrorRule[] {
+  if (!raw?.trim()) {
+    return []
+  }
+
+  let parsed: unknown
+  try {
+    parsed = JSON.parse(raw)
+  } catch (error) {
+    console.warn(
+      `[download_mirrors] Ignoring invalid DOWNLOAD_MIRROR_RULES JSON: ${
+        error instanceof Error ? error.message : String(error)
+      }`
+    )
+    return []
+  }
+
+  const rules: DownloadMirrorRule[] = []
+
+  if (Array.isArray(parsed)) {
+    for (const entry of parsed) {
+      if (
+        typeof entry !== 'object' ||
+        entry === null ||
+        !('source' in entry) ||
+        !('target' in entry) ||
+        typeof entry.source !== 'string' ||
+        typeof entry.target !== 'string'
+      ) {
+        console.warn('[download_mirrors] Ignoring malformed mirror rule in DOWNLOAD_MIRROR_RULES array')
+        continue
+      }
+
+      try {
+        rules.push(normalizeDownloadMirrorRule(entry.source, entry.target))
+      } catch (error) {
+        console.warn(
+          `[download_mirrors] Ignoring invalid mirror rule ${JSON.stringify(entry)}: ${
+            error instanceof Error ? error.message : String(error)
+          }`
+        )
+      }
+    }
+  } else if (parsed && typeof parsed === 'object') {
+    for (const [source, target] of Object.entries(parsed as Record<string, unknown>)) {
+      if (typeof target !== 'string') {
+        console.warn(
+          `[download_mirrors] Ignoring mirror rule for ${source}: target must be a string`
+        )
+        continue
+      }
+
+      try {
+        rules.push(normalizeDownloadMirrorRule(source, target))
+      } catch (error) {
+        console.warn(
+          `[download_mirrors] Ignoring invalid mirror rule ${source}: ${
+            error instanceof Error ? error.message : String(error)
+          }`
+        )
+      }
+    }
+  } else {
+    console.warn(
+      '[download_mirrors] Ignoring DOWNLOAD_MIRROR_RULES because it must be a JSON object or array'
+    )
+    return []
+  }
+
+  return rules.sort((a, b) => b.source.length - a.source.length)
+}
+
+export function rewriteDownloadUrlWithRules(url: string, rules: DownloadMirrorRule[]): string {
+  for (const rule of rules) {
+    if (!url.startsWith(rule.source)) {
+      continue
+    }
+
+    return joinMirrorUrl(rule.target, url.slice(rule.source.length))
+  }
+
+  return url
+}
+
+export function getConfiguredDownloadMirrorRules(): DownloadMirrorRule[] {
+  const raw = process.env.DOWNLOAD_MIRROR_RULES
+
+  if (raw === cachedMirrorRulesRaw) {
+    return cachedMirrorRules
+  }
+
+  cachedMirrorRulesRaw = raw
+  cachedMirrorRules = parseDownloadMirrorRules(raw)
+  return cachedMirrorRules
+}
+
+export function rewriteDownloadUrl(url: string): string {
+  return rewriteDownloadUrlWithRules(url, getConfiguredDownloadMirrorRules())
+}
+
+export type { DownloadMirrorRule }

admin/start/env.ts

@@ -19,6 +19,7 @@ export default await Env.create(new URL('../', import.meta.url), {
   URL: Env.schema.string(),
   LOG_LEVEL: Env.schema.string(),
   INTERNET_STATUS_TEST_URL: Env.schema.string.optional(),
+  DOWNLOAD_MIRROR_RULES: Env.schema.string.optional(),
 
   /*
   |----------------------------------------------------------

install/management_compose.yaml

@@ -44,6 +44,10 @@ services:
       - REDIS_HOST=redis
       # If you change the Redis port, make sure to update this accordingly
       - REDIS_PORT=6379
+      # Optional: rewrite one or more download URL prefixes to mirrors.
+      # Example:
+      # - DOWNLOAD_MIRROR_RULES={"https://download.kiwix.org/zim/":"https://cdimage.debian.org/mirror/kiwix.org/zim/"}
+      - DOWNLOAD_MIRROR_RULES=
     depends_on:
       mysql:
         condition: service_healthy
@@ -117,4 +121,4 @@ services:
 
 volumes:
   nomad-update-shared:
-    driver: local
+    driver: local