transloadit 6.1.1 → 6.2.0

package/src/alphalib/types/robots/video-subtitle.ts

@@ -59,7 +59,7 @@ export const meta: RobotMetaInput = {
     },
   },
   example_code_description:
-    'Transcribe an uploaded video into timestamped words, then burn in uppercase social captions with the spoken word highlighted. Captions normally contain up to three words, breaking at sentence endings and pauses longer than 0.4 seconds. A trailing function word can move to the next caption, allowing one extra word there:',
+    'Transcribe an uploaded video into timestamped words, then burn in uppercase social captions with the spoken word highlighted. Captions normally contain up to three words, breaking at sentence endings and pauses longer than 0.4 seconds. Social phrasing can use one extra word to keep a short phrase together or avoid a sentence-final orphan:',
   purpose_sentence: 'adds subtitles and closed captions to videos',
   purpose_verb: 'subtitle',
   purpose_word: 'subtitle',
@@ -89,7 +89,7 @@ Determines if subtitles are added as a separate stream to the video (value \`"ex
       .enum(['social'])
       .optional()
       .describe(`
-Applies a ready-made appearance for burned subtitles: heavy Poppins text and a large font scaled with the video height. JSON captions render in uppercase in the lower third with a thick outline and a soft black shadow. Each caption grows from 85% to full size over 120 milliseconds, then its words stay in place while the highlight changes color. The transcript file is preserved. Existing styling parameters override the preset. ASS files retain their own styles. With timestamped JSON from \`/speech/transcribe\`, the active word is highlighted until the next word or caption starts; other words use \`font_color\`. Trailing commas and periods are omitted from social JSON captions; question marks and exclamation marks are preserved. Word highlighting and grouping require JSON input.
+Applies a ready-made appearance for burned subtitles: heavy Poppins text, a thick outline, and a soft black shadow. JSON captions render in uppercase in the lower third. By default, JSON font sizing uses the shorter side of the video and fits each caption within the safe width, preferring one line and using at most two. Words are never split. Each caption grows from 85% to full size over 120 milliseconds, then its words stay in place while the highlight changes color. The transcript file is preserved. Existing styling parameters override the preset; an explicit \`font_size\` disables automatic fitting. ASS files retain their own styles. With timestamped JSON from \`/speech/transcribe\`, the active word is highlighted until the next word or caption starts; other words use \`font_color\`. Trailing commas and periods are omitted from social JSON captions; question marks and exclamation marks are preserved. Word highlighting and grouping require JSON input.
 `),
     highlight_color: subtitleColorSchema.default('FFD60A').describe(`
 The accent color of the active word in JSON subtitles. Only applies to burned JSON subtitles. With \`"style": "social"\`, the previous word stays highlighted through short gaps until the next word or caption starts. Without the social style, gaps use \`font_color\` for the entire caption.
@@ -102,7 +102,7 @@ ${subtitleColorDescription}
       .min(1)
       .default(3)
       .describe(`
-The target maximum number of timestamped JSON words shown in a caption. Captions break earlier at sentence-ending punctuation or pauses longer than 0.4 seconds. With \`"style": "social"\`, a trailing English function word such as “by,” “to,” or “the” moves to the next caption when possible, allowing at most one extra word in that caption. Single-word captions are preserved when the limit is 1. A caption stays visible until the next caption starts when the pause is at most 0.4 seconds, and disappears at its last word’s end during longer pauses. Does not change SRT, VTT, or ASS cues.
+The target maximum number of timestamped JSON words shown in a caption. Captions break earlier at sentence-ending punctuation or pauses longer than 0.4 seconds. With \`"style": "social"\`, a trailing English function word such as “by,” “to,” or “the” moves to the next caption when possible. Short phrases such as “line of code” can stay together, and a sentence-final single word can join the previous caption. These adjustments allow at most one extra word per caption, without crossing sentence endings or long pauses. Single-word captions are preserved when the limit is 1. A caption stays visible until the next caption starts when the pause is at most 0.4 seconds, and disappears at its last word’s end during longer pauses. Does not change SRT, VTT, or ASS cues.
 `),
     border_style: z
       .enum(['box', 'outline', 'shadow'])
@@ -115,12 +115,13 @@ The color for the subtitle border. The default is opaque black. Only applies to
 
 ${subtitleColorDescription}
 `),
+    // Preserve omitted appearance fields so parsing cannot override the selected style's defaults.
     // TODO: Make font an enum
     font: z
       .string()
-      .default('Arial')
+      .optional()
       .describe(`
-The font family to use. Also includes boldness and style of the font. With \`"style": "social"\`, the default is Poppins, using its ExtraBold weight for JSON captions. An explicit font overrides the style.
+The font family to use. Also includes boldness and style of the font. The default is Arial. With \`"style": "social"\`, the default is ${videoSubtitleSocialStyle.font}, using its ExtraBold weight for JSON captions. An explicit font overrides the style.
 
 [Here](/docs/supported-formats/fonts/) is a list of all supported fonts.
 `),
@@ -133,15 +134,15 @@ ${subtitleColorDescription}
       .number()
       .int()
       .min(1)
-      .default(16)
+      .optional()
       .describe(`
-Specifies the size of the text in the 288-pixel subtitle coordinate space, scaled with the video height. With \`"style": "social"\`, the default is 54. Poppins capital letters are about 7.5% of the video height at this size. An explicit value overrides the style.
+Specifies the size of the text in the 288-pixel subtitle coordinate space, scaled with the video height. The default is 16. With \`"style": "social"\`, the default is ${videoSubtitleSocialStyle.font_size}. For social JSON captions, an omitted value scales this default with the shorter side of the video, increases it by 18% for portrait video, and shrinks captions as needed to fit at most two lines. Poppins capital letters are about 7.5% of the shorter side before fitting, or about 9% for portrait video. An explicit value keeps height-relative sizing and disables automatic fitting.
 `),
     bold: z
       .boolean()
-      .default(false)
+      .optional()
       .describe(`
-Specifies whether the subtitle text should be bold. Only applies to burned subtitles. With \`"style": "social"\`, the default is \`true\`; an explicit \`false\` overrides the style.
+Specifies whether the subtitle text should be bold. Only applies to burned subtitles. The default is \`false\`. With \`"style": "social"\`, the default is \`${videoSubtitleSocialStyle.bold}\`; an explicit \`false\` overrides the style.
 `),
     italic: z
       .boolean()
@@ -157,10 +158,10 @@ Specifies whether the subtitle text should be italic. Only applies to burned sub
       .optional()
       .nullable()
       .describe(`
-Specifies the width of the text outline in the 288-pixel subtitle coordinate space, scaled with the video height. Numeric strings such as \`"4"\` are also accepted. Only applies to burned subtitles with \`"outline"\` or \`"box"\` border styles. If omitted or \`null\`, the default width is used (2 with \`"style": "social"\`, about 3.7 pixels on a video 534 pixels high).
+Specifies the width of the text outline in the 288-pixel subtitle coordinate space, scaled with the video height. Numeric strings such as \`"4"\` are also accepted. Only applies to burned subtitles with \`"outline"\` or \`"box"\` border styles. If omitted or \`null\`, the default width is used (2 with \`"style": "social"\`, proportional to the final fitted font size for social JSON captions).
 `),
     position: positionSchema.default('bottom').describe(`
-Specifies the position of the subtitles. With \`"style": "social"\`, the default bottom position places JSON captions in the lower third. An explicit value overrides the style.
+Specifies the position of the subtitles. With \`"style": "social"\`, the default bottom position places JSON captions in the lower third, with extra bottom clearance on portrait video. An explicit value overrides the style.
 `),
     language: z
       .string()

package/src/alphalib/types/robots/video-thumbs.ts

@@ -94,7 +94,8 @@ The thumbnails are taken at regular intervals, determined by dividing the video
 To extract thumbnails for specific timestamps, use the \`offsets\` parameter.
 `),
     offsets: z
-      .union([z.array(z.number()), z.array(percentageSchema)])
+      .union([z.array(z.number().finite()), z.array(percentageSchema)])
+      .nullable()
       .default([])
       .describe(`
 An array of offsets representing seconds of the file duration, such as \`[ 2, 45, 120 ]\`. Millisecond durations of a file can also be used by using decimal place values. For example, an offset from 1250 milliseconds would be represented with \`1.25\`. Offsets can also be percentage values such as \`[ "2%", "50%", "75%" ]\`.
@@ -102,6 +103,8 @@ An array of offsets representing seconds of the file duration, such as \`[ 2, 45
 This option cannot be used with the \`count\` parameter, and takes precedence if both are specified. Out-of-range offsets are silently ignored.
 
 When \`smart\` is \`true\`, smart selection ignores \`offsets\` and uses \`count\` to select from its own candidate timestamps. If no smart candidates can be extracted, the Robot falls back to standard extraction, where \`offsets\` takes precedence. Use \`smart: false\` to consistently extract the specified timestamps.
+
+Use \`null\` to leave timestamp selection to \`count\` and \`smart\`.
 `),
     format: z
       .enum(['jpeg', 'jpg', 'png'])
@@ -114,18 +117,20 @@ The format of the extracted thumbnail. Supported values are \`"jpg"\`, \`"jpeg"\
       .int()
       .min(1)
       .max(1920)
+      .nullable()
       .optional()
       .describe(`
-The width of the thumbnail, in pixels. Defaults to the original width of the video.
+The width of the thumbnail, in pixels. Omit or use \`null\` for the original width of the video.
 `),
     height: z
       .number()
       .int()
       .min(1)
       .max(1080)
+      .nullable()
       .optional()
       .describe(`
-The height of the thumbnail, in pixels. Defaults to the original height of the video.
+The height of the thumbnail, in pixels. Omit or use \`null\` for the original height of the video.
 `),
     resize_strategy: resize_strategy.describe(`
 One of the [available resize strategies](/docs/topics/resize-strategies/).

package/src/alphalib/types/template.ts

@@ -9,6 +9,7 @@ import type {
   RobotsWithHiddenBotsAndFields,
 } from './robots/_index.ts'
 import type { RobotUse } from './robots/_instructions-primitives.ts'
+import { refineDocumentExtractOcr } from './robots/document-extract.ts'
 import { refineTransloaditImportSelector } from './robots/transloadit-import.ts'
 
 export type Step = RobotsSchema
@@ -26,6 +27,7 @@ export const stepSchema: z.ZodType<Step> = z
   })
   .and(robotsSchema)
   .superRefine(refineTransloaditImportSelector)
+  .superRefine(refineDocumentExtractOcr)
 export const stepsSchema: z.ZodType<Steps> = z
   .record(stepSchema)
   .describe('Contains Assembly Instructions.')
@@ -45,6 +47,7 @@ export const stepSchemaWithHiddenFields: z.ZodType<StepWithHiddenFields> = z
   })
   .and(robotsWithHiddenBotsAndFieldsSchema)
   .superRefine(refineTransloaditImportSelector)
+  .superRefine(refineDocumentExtractOcr)
 export const stepsSchemaWithHiddenFields: z.ZodType<StepsWithHiddenFields> = z
   .record(stepSchemaWithHiddenFields)
   .describe('Contains Assembly Instructions.')

package/src/cli/commands/assemblies.ts

@@ -227,7 +227,7 @@ export async function replay(
 ): Promise<void> {
   if (steps) {
     try {
-      await apiCall(await readStepsInputFile(steps))
+      await apiCall(await readStepsInputFile(steps, output))
     } catch (err) {
       const error = ensureError(err)
       output.error(error.message)
@@ -735,7 +735,9 @@ async function resolveResultDownloadTargets({
         throw new Error('stdout can only receive a single result file')
       }
 
-      throw new Error('file outputs can only receive a single result file')
+      throw new Error(
+        'file outputs can only receive a single result file; use -o <dir>/ to write multiple results to a directory',
+      )
     }
 
     return getSingleResultDownloadTarget(allFiles, outputPath)
@@ -1351,18 +1353,20 @@ export async function create(
   // to avoid zod adding default values that the API may reject
   let effectiveStepsData = stepsData
   if (steps) {
-    effectiveStepsData = await readStepsInputFile(steps)
+    effectiveStepsData = await readStepsInputFile(steps, outputctl)
   }
 
   // Determine output stat async before entering the Promise constructor
   let outstat: StatLike | undefined
   if (resolvedOutput != null) {
+    const isDirectoryOutput =
+      outputMode === 'directory' || resolvedOutput.endsWith('/') || resolvedOutput.endsWith('\\')
     const [err, stat] = await tryCatch(myStat(process.stdout, resolvedOutput))
     if (err && (!isErrnoException(err) || err.code !== 'ENOENT')) throw err
     outstat =
       stat ??
       ({
-        isDirectory: () => outputMode === 'directory',
+        isDirectory: () => isDirectoryOutput,
       } satisfies StatLike)
 
     if (outputMode === 'directory' && stat != null && !stat.isDirectory()) {
@@ -1826,7 +1830,8 @@ export class AssembliesCreateCommand extends AuthenticatedCommand {
   inputs = inputPathsOption()
 
   outputPath = Option.String('--output,-o', {
-    description: 'Specify an output file or directory',
+    description:
+      'Specify an output file or directory; a trailing / or \\ creates a directory recursively',
   })
 
   fields = Option.Array('--field,-f', {
@@ -1945,7 +1950,8 @@ export class RunCommand extends AuthenticatedCommand {
   inputs = inputPathsOption()
 
   outputPath = Option.String('--output,-o', {
-    description: 'Specify an output file or directory',
+    description:
+      'Specify an output file or directory; a trailing / or \\ creates a directory recursively',
   })
 
   fields = Option.Array('--field,-f', {
@@ -2012,7 +2018,7 @@ export class RunCommand extends AuthenticatedCommand {
     }
 
     const { hasFailures, resultUrls } = await create(this.output, this.client, {
-      stepsData: parseStepsInputJson(JSON.stringify(compiled.instructions.steps)),
+      stepsData: parseStepsInputJson(JSON.stringify(compiled.instructions.steps), this.output),
       fields: fieldsMap ?? {},
       watch: this.watch,
       recursive: this.recursive,

package/src/cli/commands/auth.ts

@@ -22,6 +22,7 @@ import {
   requireCliCredentials,
   resolveCliConfig,
 } from '../helpers.ts'
+import { createCliStepsSchema } from '../stepsInput.ts'
 import { UnauthenticatedCommand } from './BaseCommand.ts'
 
 type UrlParamPrimitive = string | number | boolean
@@ -113,6 +114,7 @@ function generateSignature(
   input: string,
   credentials: CliKeySecretCredentials,
   algorithm?: string,
+  output: Pick<IOutputCtl, 'warn'> = console,
 ): OutputResult {
   const { authKey } = credentials
   let params: CliSignatureParams
@@ -120,7 +122,10 @@ function generateSignature(
   if (input === '') {
     params = { auth: { key: authKey } }
   } else {
-    const parsedResult = parseJsonObject(input, cliSignatureParamsSchema)
+    const parsedResult = parseJsonObject(
+      input,
+      cliSignatureParamsSchema.extend({ steps: createCliStepsSchema(output).optional() }),
+    )
     if (!parsedResult.ok) {
       return { ok: false, error: parsedResult.error }
     }
@@ -347,7 +352,7 @@ export class SignatureCommand extends UnauthenticatedCommand {
 
     const { content } = await readCliInput({ allowStdinWhenNoPath: true })
     const rawInput = (content ?? '').trim()
-    const result = generateSignature(rawInput, credentials, this.algorithm)
+    const result = generateSignature(rawInput, credentials, this.algorithm, this.output)
 
     if (result.ok) {
       process.stdout.write(`${result.output}\n`)

package/src/cli/commands/templates.ts

@@ -1,7 +1,7 @@
 import type { TemplateContent } from '../../apiTypes.ts'
 import type { Transloadit } from '../../Transloadit.ts'
 import type { IOutputCtl } from '../OutputCtl.ts'
-import type { TemplateFile } from '../types.ts'
+import type { TemplateFile, TemplateFileData } from '../types.ts'
 
 import fsp from 'node:fs/promises'
 import path from 'node:path'
@@ -13,9 +13,9 @@ import { z } from 'zod'
 
 import { tryCatch } from '../../alphalib/tryCatch.ts'
 import { createReadStream, formatAPIError, streamToBuffer } from '../helpers.ts'
-import { parseStepsInputJson } from '../stepsInput.ts'
+import { createCliStepsSchema, parseStepsInputJson } from '../stepsInput.ts'
 import ModifiedLookup from '../template-last-modified.ts'
-import { ensureError, isTransloaditAPIError, TemplateFileDataSchema } from '../types.ts'
+import { ensureError, isTransloaditAPIError } from '../types.ts'
 import { AuthenticatedCommand } from './BaseCommand.ts'
 
 const rreaddirAsync = promisify(rreaddir)
@@ -62,7 +62,7 @@ export async function create(
   try {
     const buf = await streamToBuffer(createReadStream(file))
 
-    const steps = parseStepsInputJson(buf.toString())
+    const steps = parseStepsInputJson(buf.toString(), output)
 
     const result = await client.createTemplate({
       name,
@@ -107,7 +107,7 @@ export async function modify(
     let newName = name
 
     if (buf.length > 0) {
-      steps = parseStepsInputJson(buf.toString()) as TemplateContent['steps']
+      steps = parseStepsInputJson(buf.toString(), output) as TemplateContent['steps']
     }
 
     if (!name || buf.length === 0) {
@@ -225,6 +225,13 @@ export async function sync(
   client: Transloadit,
   { files, recursive }: TemplateSyncOptions,
 ): Promise<void> {
+  const templateFileDataSchema: z.ZodType<TemplateFileData> = z
+    .object({
+      transloadit_template_id: z.string().optional(),
+      steps: createCliStepsSchema(output).optional(),
+    })
+    .passthrough()
+
   // Promise [String] -- all files in the directory tree
   const relevantFilesNested = await Promise.all(
     files.map(async (file) => {
@@ -263,7 +270,7 @@ export async function sync(
     try {
       const data = await fsp.readFile(file, 'utf8')
       const parsed: unknown = JSON.parse(data)
-      const validated = TemplateFileDataSchema.safeParse(parsed)
+      const validated = templateFileDataSchema.safeParse(parsed)
       if (!validated.success) return null
       return 'transloadit_template_id' in validated.data ? { file, data: validated.data } : null
     } catch (e) {

package/src/cli/stepsInput.ts

@@ -1,14 +1,40 @@
 import type { StepsInput } from '../alphalib/types/template.ts'
+import type { IOutputCtl } from './OutputCtl.ts'
 
 import fsp from 'node:fs/promises'
 
 import { z } from 'zod'
 
-import { stepsSchema } from '../alphalib/types/template.ts'
+import { robotsSchema } from '../alphalib/types/robots/_index.ts'
+import { refineDocumentExtractOcr } from '../alphalib/types/robots/document-extract.ts'
+import { refineTransloaditImportSelector } from '../alphalib/types/robots/transloadit-import.ts'
 
 const stepsWrapperSchema = z.object({ steps: z.record(z.unknown()) })
 
-export function parseStepsInputJson(content: string): StepsInput {
+/** CLI inputs can be partial overrides; API2 validates OCR after merging their runtime context. */
+export function createCliStepsSchema(
+  output: Pick<IOutputCtl, 'warn'> = console,
+): z.ZodType<StepsInput> {
+  const cliStepSchema = robotsSchema
+    .superRefine(refineTransloaditImportSelector)
+    .superRefine((step, context) => {
+      // API2 merges Template defaults before Robot validation. New OCR checks must not reject
+      // partial --steps overrides in a minor release; explicit Zod schemas remain strict.
+      refineDocumentExtractOcr(step, {
+        ...context,
+        addIssue(issue) {
+          const field = [...context.path, ...(issue.path ?? [])].join('.')
+          output.warn(`${field}: ${issue.message ?? 'Invalid OCR dependency.'}`)
+        },
+      })
+    })
+  return z.record(cliStepSchema)
+}
+
+export function parseStepsInputJson(
+  content: string,
+  output: Pick<IOutputCtl, 'warn'> = console,
+): StepsInput {
   const parsed: unknown = JSON.parse(content)
   const wrapped = stepsWrapperSchema.safeParse(parsed)
   let stepsInput = parsed
@@ -21,7 +47,7 @@ export function parseStepsInputJson(content: string): StepsInput {
     }
     stepsInput = validatedWrapper.data.steps
   }
-  const validated = stepsSchema.safeParse(stepsInput)
+  const validated = createCliStepsSchema(output).safeParse(stepsInput)
   if (!validated.success) {
     throw new Error(`Invalid steps format: ${validated.error.message}`)
   }
@@ -42,7 +68,10 @@ export function parseStepsInputJson(content: string): StepsInput {
   ) as StepsInput
 }
 
-export async function readStepsInputFile(filePath: string): Promise<StepsInput> {
+export async function readStepsInputFile(
+  filePath: string,
+  output: Pick<IOutputCtl, 'warn'> = console,
+): Promise<StepsInput> {
   const content = await fsp.readFile(filePath, 'utf8')
-  return parseStepsInputJson(content)
+  return parseStepsInputJson(content, output)
 }

package/src/cli/types.ts

@@ -2,8 +2,6 @@ import type { Steps } from '../alphalib/types/template.ts'
 
 import { z } from 'zod'
 
-import { optionalStepsSchema } from '../alphalib/types/template.ts'
-
 // Zod schemas for runtime validation
 const APIErrorSchema = z.object({
   error: z.string(),
@@ -36,13 +34,6 @@ export interface TemplateFileData {
   [key: string]: unknown // passthrough
 }
 
-export const TemplateFileDataSchema: z.ZodType<TemplateFileData> = z
-  .object({
-    transloadit_template_id: z.string().optional(),
-    steps: optionalStepsSchema,
-  })
-  .passthrough() as z.ZodType<TemplateFileData>
-
 export interface TemplateFile {
   file: string
   data: TemplateFileData