chatboxai/chatbox · warning · Error

No readable text content found in EPUB file

Error message

No readable text content found in EPUB file

What it means

Thrown at the top of CustomGemini.paint() when `isGeminiImageModel(this.options.model.modelId)` is false. isGeminiImageModel (in definitions/image-models.ts) returns true only when the model id contains both 'gemini' and 'image'. paint() is the image-generation entry point, so calling it on a text-only Gemini model is a programmer/UX error, not a network condition — the guard fires before any API call.

Source

Thrown at src/main/file-parser.ts:309

            return plainText || null
          } catch (chapterError) {
            log.warn(`Failed to read chapter ${chapter.id}, skipping:`, chapterError)
            return null // Return null for failed chapters to continue processing
          }
        }

        // Extract text from all chapters using concurrent processing
        log.info(`Starting concurrent processing of ${epub.flow.length} chapters with concurrency: 8`)

        const chapterResults = await concurrentMap(epub.flow as { id: string }[], processChapter, 8)
        const chapterTexts = chapterResults.filter((text: string | null) => text !== null) as string[]
        log.info(`Successfully processed ${chapterTexts.length}/${epub.flow.length} chapters`)

        const fullText = chapterTexts.join('\n\n')

        if (!fullText) {
          throw new Error('No readable text content found in EPUB file')
        }

        log.info(`Successfully extracted ${fullText.length} characters from ${chapterTexts.length} chapters`)
        resolve(fullText)
      } catch (error) {
        log.error('Error extracting EPUB content:', error)
        reject(error)
      }
    })

    epub.parse()
  })
}

View on GitHub (pinned to 81571269ad)

Solutions

1. Before calling paint(), gate the UI on `isGeminiImageModel(modelId)` (model id must include both 'gemini' and 'image') and only enable the paint action for matching models.
2. Switch the selected model to a known Gemini image model such as `gemini-2.0-flash-exp-image-generation` or `gemini-2.5-flash-image-preview`.
3. If you maintain a custom provider, ensure the model id for image-capable models contains the substring 'image' so the heuristic matches.
4. Update image-models.ts if Google ships a new image model whose id does not match the `gemini`+`image` rule.

Example fix

// before
await model.paint({ prompt, num: 1 })  // throws on text models

// after
guard before calling:
if (!isGeminiImageModel(session.modelId)) {
  showWarning('Select a Gemini image model to generate images')
} else {
  await model.paint({ prompt, num: 1 })
}

function canPaint(modelId: string): boolean {
  return modelId.includes('gemini') && modelId.includes('image')
}
// before invoking paint:
if (!canPaint(session.modelId)) disablePaintUI()

import { isGeminiImageModel } from '../image-models'
function assertImageCapable(modelId: string): void {
  if (!isGeminiImageModel(modelId)) throw new Error(`Model ${modelId} cannot generate images`)
}

try { await gemini.paint({ prompt, num: 1 }) }
catch (e) {
  if (e instanceof ApiError && e.message.includes('does not support image generation')) {
    showWarning('Switch to a Gemini image model to paint')
    return []
  }
  throw e
}

When it happens

Trigger: UI invokes paint() with the currently selected model id even though it is a text model (e.g. `gemini-2.0-flash`, `gemini-1.5-pro`); a saved session has a non-image Gemini model selected when the user clicks the paint button; model list refresh replaced an image model id with its text counterpart; the model id does not contain the literal substring 'image' (e.g. a renamed alias).

Common situations: User selects 'gemini-2.5-flash' (text) and opens the image-generation panel expecting it to work; a custom Gemini provider exposes a model under a nickname whose underlying id lacks 'image'; version upgrade renamed the image model (e.g. from `gemini-2.0-flash-exp-image-generation` to something without 'image' in the id) so isGeminiImageModel no longer recognises it.

Related errors

• Embedding batch failed: expected ${batchTexts.length}, got $
• Image generation requires an active chat session.
• Missing --prompt.
• --style must be vivid or natural.
• Image model is not available: ${provider}/${modelId}. Use "c

AI-assisted analysis of chatboxai/chatbox@81571269ad (2026-08-12). Data as JSON: /api/errors/bf3189fec35fdbbc. Report an issue: GitHub.