Merge pull request #61 from ajcwebdev/rss

Add `--date` and `--lastDays` RSS options

Author: ajcwebdev

docs/examples.md

@@ -181,10 +181,43 @@ Process multiple RSS feeds:
 
 ```bash
 npm run as -- \
-  --rss "https://ajcwebdev.substack.com/feed" \
-  --rss "https://feeds.transistor.fm/fsjam-podcast/" \
   --last 1 \
-  --whisper tiny
+  --whisper tiny \
+  --rss "https://ajcwebdev.substack.com/feed" \
+  "https://feeds.transistor.fm/fsjam-podcast/"
+```
+
+Download episodes from a specific date:
+
+```bash
+npm run as -- \
+  --rss "https://ajcwebdev.substack.com/feed" \
+  --date 2021-05-10
+```
+
+Download episodes from multiple dates:
+
+```bash
+npm run as -- \
+  --rss "https://ajcwebdev.substack.com/feed" \
+  --date 2021-05-10 2022-05-10
+```
+
+Download episodes from multiple dates on multiple RSS feeds:
+
+```bash
+npm run as -- \
+  --rss "https://ajcwebdev.substack.com/feed" \
+  "https://feeds.transistor.fm/fsjam-podcast/" \
+  --date 2021-05-10 2022-05-10
+```
+
+Download episodes from a specific number of previous days, for example to download episodes from the last 7 days:
+
+```bash
+npm run as -- \
+  --rss "https://ajcwebdev.substack.com/feed" \
+  --lastDays 7
 ```
 
 ## Language Model (LLM) Options

src/cli/commander.ts

@@ -68,6 +68,8 @@ program
   .option('--order <order>', 'Specify the order for RSS feed processing (newest or oldest)')
   .option('--skip <number>', 'Number of items to skip when processing RSS feed', parseInt)
   .option('--last <number>', 'Number of most recent items to process (overrides --order and --skip)', parseInt)
+  .option('--date <dates...>', 'Process items from these dates (YYYY-MM-DD)')
+  .option('--lastDays <number>', 'Number of days to look back for items', parseInt)
   .option('--info', 'Skip processing and write metadata to JSON objects (supports --urls, --rss, --playlist, --channel)')
   // Transcription service options
   .option('--whisper [model]', 'Use Whisper.cpp for transcription with optional model specification')

src/commands/process-rss.ts

@@ -136,6 +136,23 @@ function selectItemsToProcess(items: RSSItem[], options: ProcessingOptions): RSS
     return matchedItems
   }
 
+  if (options.lastDays !== undefined) {
+    const now = new Date()
+    const cutoff = new Date(now.getTime() - (options.lastDays * 24 * 60 * 60 * 1000))
+  
+    const matchedItems = items.filter((item) => {
+      const itemDate = new Date(item.publishDate)
+      return itemDate >= cutoff
+    })
+    return matchedItems
+  }
+  
+  if (options.date && options.date.length > 0) {
+    const selectedDates = new Set(options.date)
+    const matchedItems = items.filter((item) => selectedDates.has(item.publishDate))
+    return matchedItems
+  }  
+
   if (options.last) {
     return items.slice(0, options.last)
   }
@@ -242,6 +259,10 @@ export async function processRSS(
     }
 
     const itemsToProcess = selectItemsToProcess(items, options)
+    if (itemsToProcess.length === 0) {
+      l(wait(`\nNo items found matching the provided criteria for this feed. Skipping...`))
+      return
+    }
     logProcessingStatus(items.length, itemsToProcess.length, options)
     await processItems(itemsToProcess, options, llmServices, transcriptServices)
   } catch (error) {

src/types/main.ts

@@ -99,6 +99,12 @@ export type ProcessingOptions = {
   /** Number of most recent items to process (overrides --order and --skip). */
   last?: number
 
+  /** Date of RSS items to process. */
+  date?: string[]
+
+  /** Number of previous days to check for RSS items to process. */
+  lastDays?: number
+
   /** Whether to run in interactive mode. */
   interactive?: boolean
 }

src/utils/validate-option.ts

@@ -64,4 +64,30 @@ export function validateRSSOptions(options: ProcessingOptions): void {
     err("Error: The --order option must be either 'newest' or 'oldest'.")
     process.exit(1)
   }
+
+  if (options.lastDays !== undefined) {
+    if (!Number.isInteger(options.lastDays) || options.lastDays < 1) {
+      err('Error: The --lastDays option must be a positive integer.')
+      process.exit(1)
+    }
+    if (options.last !== undefined || options.skip !== undefined || options.order !== undefined || (options.date && options.date.length > 0)) {
+      err('Error: The --lastDays option cannot be used with --last, --skip, --order, or --date.')
+      process.exit(1)
+    }
+  }
+
+  if (options.date && options.date.length > 0) {
+    const dateRegex = /^\d{4}-\d{2}-\d{2}$/
+    for (const d of options.date) {
+      if (!dateRegex.test(d)) {
+        err(`Error: Invalid date format "${d}". Please use YYYY-MM-DD format.`)
+        process.exit(1)
+      }
+    }
+
+    if (options.last !== undefined || options.skip !== undefined || options.order !== undefined) {
+      err('Error: The --date option cannot be used with --last, --skip, or --order.')
+      process.exit(1)
+    }
+  }
 }