@transloadit/node 4.1.3

package/src/alphalib/types/robots/document-convert.ts

@@ -0,0 +1,302 @@
+import { z } from 'zod'
+
+import type { RobotMetaInput } from './_instructions-primitives.ts'
+import { interpolateRobot, robotBase, robotUse } from './_instructions-primitives.ts'
+
+export const meta: RobotMetaInput = {
+  allowed_for_url_transform: true,
+  bytescount: 1,
+  discount_factor: 1,
+  discount_pct: 0,
+  example_code: {
+    steps: {
+      converted: {
+        robot: '/document/convert',
+        use: ':original',
+        format: 'pdf',
+      },
+    },
+  },
+  example_code_description: 'Convert uploaded files to PDF documents:',
+  extended_description: `
+> [!Note]
+> This <dfn>Robot</dfn> can convert files to PDF, but cannot convert PDFs to different formats. If you want to convert PDFs to say, JPEG or TIFF, use [🤖/image/resize](/docs/robots/image-resize/). If you want to turn them into text files or recognize (OCR) them to make them searchable, reach out, as we have a new <dfn>Robot</dfn> in the works for this.
+
+Sometimes, a certain file type might not support what you are trying to accomplish. Perhaps your company is trying to automate document formatting, but it only works with docx, so all your docs need to be converted. Or maybe your stored jpg files are taking up too much space and you want a lighter format. Whatever the case, we have you covered.
+
+Using this <dfn>Robot</dfn>, you can bypass the issues that certain file types may bring, by converting your file into the most suitable format. This also works in conjunction with our other <dfn>Robots</dfn>, allowing for even greater versatility when using our services.
+
+> ![Warning]
+> A general rule of this <dfn>Robot</dfn> is that converting files into an alien format category will result in an error. For example, SRT files can be converted into the VTT format, but not to an image.
+
+The following file formats can be converted from:
+
+- \`ai\`
+- \`csv\`
+- \`doc\`
+- \`docx\`
+- \`eps\`
+- \`gif\`
+- \`html\`
+- \`jpg\`
+- \`latex\`
+- \`md\`
+- \`oda\`
+- \`odd\`
+- \`odt\`
+- \`ott\`
+- \`png\`
+- \`pot\`
+- \`pps\`
+- \`ppt\`
+- \`pptx\`
+- \`ppz\`
+- \`ps\`
+- \`rtf\`
+- \`rtx\`
+- \`svg\`
+- \`text\`
+- \`txt\`
+- \`xhtml\`
+- \`xla\`
+- \`xls\`
+- \`xlsx\`
+- \`xml\`
+`,
+  minimum_charge: 1048576,
+  output_factor: 1,
+  override_lvl1: 'Document Processing',
+  purpose_sentence: 'converts documents into different formats',
+  purpose_verb: 'convert',
+  purpose_word: 'convert',
+  purpose_words: 'Convert documents into different formats',
+  service_slug: 'document-processing',
+  slot_count: 12,
+  title: 'Convert documents into different formats',
+  typical_file_size_mb: 0.8,
+  typical_file_type: 'document',
+  name: 'DocumentConvertRobot',
+  priceFactor: 1,
+  // This slot count needs to be unique, because unoconv can only process one document at a time,
+  // and is also only included in WorkerSlotCalculator::slotsThatFit() when
+  // we have enough idle unoconv daemons.
+  // We do not want a queue of this Robot to block any other Robot's jobs.
+  queueSlotCount: 32,
+  minimumCharge: 1048576,
+  lazyLoad: true,
+  installVersionFile: process.env.API2_UNOCONV_INSTALL_VERSION_FILE || '',
+  isAllowedForUrlTransform: true,
+  trackOutputFileSize: true,
+  // we cannot use coreConfig.numUnoconvDaemons, because it does not live in alphalib
+  numDaemons: 8,
+  isInternal: false,
+  removeJobResultFilesFromDiskRightAfterStoringOnS3: false,
+  stage: 'ga',
+}
+
+export const robotDocumentConvertInstructionsSchema = robotBase
+  .merge(robotUse)
+  .extend({
+    robot: z.literal('/document/convert').describe(`
+> [!Note]
+> This Robot can convert files to PDF, but cannot convert PDFs to different formats. If you want to convert PDFs to say, JPEG or TIFF, use [🤖/image/resize](/docs/robots/image-resize/). If you want to turn them into text files or recognize (OCR) them to make them searchable, reach out, as we have a new Robot in the works for this.
+
+Sometimes, a certain file type might not support what you are trying to accomplish. Perhaps your company is trying to automate document formatting, but it only works with docx, so all your docs need to be converted. Or maybe your stored jpg files are taking up too much space and you want a lighter format. Whatever the case, we have you covered.
+
+Using this Robot, you can bypass the issues that certain file types may bring, by converting your file into the most suitable format. This also works in conjunction with our other Robots, allowing for even greater versatility when using our services.
+
+> [!Warning]
+> A general rule of this Robot is that converting files into an alien format category will result in an error. For example, SRT files can be converted into the VTT format, but not to an image.
+
+The following file formats can be converted from:
+
+- \`ai\`
+- \`csv\`
+- \`doc\`
+- \`docx\`
+- \`eps\`
+- \`gif\`
+- \`html\`
+- \`jpg\`
+- \`latex\`
+- \`md\`
+- \`oda\`
+- \`odd\`
+- \`odt\`
+- \`ott\`
+- \`png\`
+- \`pot\`
+- \`pps\`
+- \`ppt\`
+- \`pptx\`
+- \`ppz\`
+- \`ps\`
+- \`rtf\`
+- \`rtx\`
+- \`svg\`
+- \`text\`
+- \`txt\`
+- \`xhtml\`
+- \`xla\`
+- \`xls\`
+- \`xlsx\`
+- \`xml\`
+`),
+    format: z
+      .enum([
+        'ai',
+        'csv',
+        'doc',
+        'docx',
+        'eps',
+        'gif',
+        'html',
+        'jpeg',
+        'jpg',
+        'latex',
+        'oda',
+        'odd',
+        'odt',
+        'ott',
+        'pdf',
+        'png',
+        'pot',
+        'pps',
+        'ppt',
+        'pptx',
+        'ppz',
+        'ps',
+        'rtf',
+        'rtx',
+        'srt',
+        'svg',
+        'text',
+        'txt',
+        'vtt',
+        'xhtml',
+        'xla',
+        'xls',
+        'xlsx',
+        'xml',
+      ])
+      .describe(`
+The desired format for document conversion.
+`),
+    markdown_format: z
+      .enum(['commonmark', 'gfm'])
+      .default('gfm')
+      .describe(`
+Markdown can be represented in several [variants](https://www.iana.org/assignments/markdown-variants/markdown-variants.xhtml), so when using this Robot to transform Markdown into HTML please specify which revision is being used.
+`),
+    markdown_theme: z
+      .enum(['bare', 'github'])
+      .default('github')
+      .describe(`
+This parameter overhauls your Markdown files styling based on several canned presets.
+`),
+    pdf_margin: z
+      .string()
+      .default('6.25mm,6.25mm,14.11mm,6.25mm')
+      .describe(`
+PDF Paper margins, separated by \`,\` and with units.
+
+We support the following unit values: \`px\`, \`in\`, \`cm\`, \`mm\`.
+
+Currently this parameter is only supported when converting from \`html\`.
+`),
+    pdf_print_background: z
+      .boolean()
+      .default(true)
+      .describe(`
+Print PDF background graphics.
+
+Currently this parameter is only supported when converting from \`html\`.
+`),
+    pdf_format: z
+      .enum(['A0', 'A1', 'A2', 'A3', 'A4', 'A5', 'A6', 'Ledger', 'Legal', 'Letter', 'Tabloid'])
+      .default('Letter')
+      .describe(`
+PDF paper format.
+
+Currently this parameter is only supported when converting from \`html\`.
+`),
+    pdf_display_header_footer: z
+      .boolean()
+      .default(false)
+      .describe(`
+Display PDF header and footer.
+
+Currently this parameter is only supported when converting from \`html\`.
+`),
+    pdf_header_template: z
+      .string()
+      .optional()
+      .describe(`
+HTML template for the PDF print header.
+
+Should be valid HTML markup with following classes used to inject printing values into them:
+- \`date\` formatted print date
+- \`title\` document title
+- \`url\` document location
+- \`pageNumber\` current page number
+- \`totalPages\` total pages in the document
+
+Currently this parameter is only supported when converting from \`html\`, and requires \`pdf_display_header_footer\` to be enabled.
+
+To change the formatting of the HTML element, the \`font-size\` must be specified in a wrapper. For example, to center the page number at the top of a page you'd use the following HTML for the header template:
+
+\`\`\`html
+<div style="font-size: 15px; width: 100%; text-align: center;"><span class="pageNumber"></span></div>
+\`\`\`
+`),
+    pdf_footer_template: z
+      .string()
+      .optional()
+      .describe(`
+HTML template for the PDF print footer.
+
+Should use the same format as the \`pdf_header_template\`.
+
+Currently this parameter is only supported when converting from \`html\`, and requires \`pdf_display_header_footer\` to be enabled.
+
+To change the formatting of the HTML element, the \`font-size\` must be specified in a wrapper. For example, to center the page number in the footer you'd use the following HTML for the footer template:
+
+\`\`\`html
+<div style="font-size: 15px; width: 100%; text-align: center;"><span class="pageNumber"></span></div>
+\`\`\`
+`),
+  })
+  .strict()
+
+export const robotDocumentConvertInstructionsWithHiddenFieldsSchema =
+  robotDocumentConvertInstructionsSchema.extend({
+    result: z
+      .union([z.literal('debug'), robotDocumentConvertInstructionsSchema.shape.result])
+      .optional(),
+  })
+
+export type RobotDocumentConvertInstructions = z.infer<
+  typeof robotDocumentConvertInstructionsSchema
+>
+export type RobotDocumentConvertInstructionsWithHiddenFields = z.infer<
+  typeof robotDocumentConvertInstructionsWithHiddenFieldsSchema
+>
+
+export const interpolatableRobotDocumentConvertInstructionsSchema = interpolateRobot(
+  robotDocumentConvertInstructionsSchema,
+)
+export type InterpolatableRobotDocumentConvertInstructions =
+  InterpolatableRobotDocumentConvertInstructionsInput
+
+export type InterpolatableRobotDocumentConvertInstructionsInput = z.input<
+  typeof interpolatableRobotDocumentConvertInstructionsSchema
+>
+
+export const interpolatableRobotDocumentConvertInstructionsWithHiddenFieldsSchema =
+  interpolateRobot(robotDocumentConvertInstructionsWithHiddenFieldsSchema)
+export type InterpolatableRobotDocumentConvertInstructionsWithHiddenFields = z.infer<
+  typeof interpolatableRobotDocumentConvertInstructionsWithHiddenFieldsSchema
+>
+export type InterpolatableRobotDocumentConvertInstructionsWithHiddenFieldsInput = z.input<
+  typeof interpolatableRobotDocumentConvertInstructionsWithHiddenFieldsSchema
+>

package/src/alphalib/types/robots/document-merge.ts

@@ -0,0 +1,114 @@
+import { z } from 'zod'
+
+import type { RobotMetaInput } from './_instructions-primitives.ts'
+import { interpolateRobot, robotBase, robotUse } from './_instructions-primitives.ts'
+
+export const meta: RobotMetaInput = {
+  allowed_for_url_transform: true,
+  bytescount: 1,
+  discount_factor: 1,
+  discount_pct: 0,
+  example_code: {
+    steps: {
+      merged: {
+        robot: '/document/merge',
+        use: {
+          steps: [':original'],
+          bundle_steps: true,
+        },
+      },
+    },
+  },
+  example_code_description: 'Merge all uploaded PDF documents into one:',
+  extended_description: `
+> ![Note]
+> This <dfn>Robot</dfn> can merge PDF files only at the moment.
+
+Input files are sorted alphanumerically unless you provide the as-syntax in the "use" parameter. For example:
+
+\`\`\`json
+{
+  "use": [
+    { "name": "my_step_name", "as": "document_2" },
+    { "name": "my_other_step_name", "as": "document_1" }
+  ]
+}
+\`\`\`
+`,
+  minimum_charge: 1048576,
+  output_factor: 1,
+  override_lvl1: 'Document Processing',
+  purpose_sentence: 'concatenates several PDF documents into a single file',
+  purpose_verb: 'convert',
+  purpose_word: 'convert',
+  purpose_words: 'Merge documents into one',
+  service_slug: 'document-processing',
+  slot_count: 10,
+  title: 'Merge documents into one',
+  typical_file_size_mb: 0.8,
+  typical_file_type: 'document',
+  name: 'DocumentMergeRobot',
+  priceFactor: 1,
+  queueSlotCount: 10,
+  minimumCharge: 1048576,
+  isAllowedForUrlTransform: true,
+  trackOutputFileSize: true,
+  isInternal: false,
+  removeJobResultFilesFromDiskRightAfterStoringOnS3: false,
+  stage: 'ga',
+}
+
+export const robotDocumentMergeInstructionsSchema = robotBase
+  .merge(robotUse)
+  .extend({
+    robot: z.literal('/document/merge'),
+    input_passwords: z
+      .array(z.string())
+      .default([])
+      .describe(`
+An array of passwords for the input documents, in case they are encrypted. The order of passwords must match the order of the documents as they are passed to the /document/merge step.
+
+This can be achieved via our as-syntax using "document_1", "document_2", etc if provided. See the demos below.
+
+If the as-syntax is not used in the "use" parameter, the documents are sorted alphanumerically based on their filename, and in that order input passwords should be provided.
+`),
+    output_password: z
+      .string()
+      .optional()
+      .describe(`
+If not empty, encrypts the output file and makes it accessible only by typing in this password.
+`),
+  })
+  .strict()
+
+export const robotDocumentMergeInstructionsWithHiddenFieldsSchema =
+  robotDocumentMergeInstructionsSchema.extend({
+    result: z
+      .union([z.literal('debug'), robotDocumentMergeInstructionsSchema.shape.result])
+      .optional(),
+  })
+
+export type RobotDocumentMergeInstructions = z.infer<typeof robotDocumentMergeInstructionsSchema>
+export type RobotDocumentMergeInstructionsWithHiddenFields = z.infer<
+  typeof robotDocumentMergeInstructionsWithHiddenFieldsSchema
+>
+
+export const interpolatableRobotDocumentMergeInstructionsSchema = interpolateRobot(
+  robotDocumentMergeInstructionsSchema,
+)
+export type InterpolatableRobotDocumentMergeInstructions =
+  InterpolatableRobotDocumentMergeInstructionsInput
+
+export type InterpolatableRobotDocumentMergeInstructionsInput = z.input<
+  typeof interpolatableRobotDocumentMergeInstructionsSchema
+>
+
+export const interpolatableRobotDocumentMergeInstructionsWithHiddenFieldsSchema = interpolateRobot(
+  robotDocumentMergeInstructionsWithHiddenFieldsSchema,
+)
+export type InterpolatableRobotDocumentMergeInstructionsWithHiddenFields = z.infer<
+  typeof interpolatableRobotDocumentMergeInstructionsWithHiddenFieldsSchema
+>
+export type InterpolatableRobotDocumentMergeInstructionsWithHiddenFieldsInput = z.input<
+  typeof interpolatableRobotDocumentMergeInstructionsWithHiddenFieldsSchema
+>

package/src/alphalib/types/robots/document-ocr.ts

@@ -0,0 +1,120 @@
+import { z } from 'zod'
+
+import type { RobotMetaInput } from './_instructions-primitives.ts'
+import {
+  aiProviderSchema,
+  granularitySchema,
+  interpolateRobot,
+  robotBase,
+  robotUse,
+} from './_instructions-primitives.ts'
+
+export const meta: RobotMetaInput = {
+  allowed_for_url_transform: true,
+  bytescount: 1,
+  discount_factor: 1,
+  discount_pct: 0,
+  example_code: {
+    steps: {
+      recognized: {
+        robot: '/document/ocr',
+        use: ':original',
+        provider: 'gcp',
+      },
+    },
+  },
+  example_code_description: 'Recognize text in an uploaded document and save it to a JSON file:',
+  extended_description: `
+> [!Warning]
+> Transloadit aims to be deterministic, but this <dfn>Robot</dfn> uses third-party AI services. The providers (AWS, GCP) will evolve their models over time, giving different responses for the same input PDFs. Avoid relying on exact responses in your tests and application.
+
+> [!Note]
+> Currently, this <dfn>Robot</dfn> only supports character recognition for PDFs. To use this <dfn>Robot</dfn> with other document formats, use [/document/convert](/docs/robots/document-convert/) first to convert the document into a PDF.
+`,
+  minimum_charge: 1048576,
+  output_factor: 1,
+  override_lvl1: 'Artificial Intelligence',
+  purpose_sentence: 'recognizes text in documents and returns it in a machine-readable format',
+  purpose_verb: 'recognize',
+  purpose_word: 'recognize text',
+  purpose_words: 'Recognize text in documents (OCR)',
+  service_slug: 'artificial-intelligence',
+  slot_count: 10,
+  title: 'Recognize text in documents',
+  typical_file_size_mb: 0.8,
+  typical_file_type: 'document',
+  name: 'DocumentOcrRobot',
+  priceFactor: 1,
+  queueSlotCount: 10,
+  minimumChargeUsdPerDocumentOcrPage: {
+    aws: 0.02,
+    gcp: 0.015,
+  },
+  isAllowedForUrlTransform: true,
+  trackOutputFileSize: true,
+  isInternal: false,
+  removeJobResultFilesFromDiskRightAfterStoringOnS3: false,
+  stage: 'ga',
+}
+
+export const robotDocumentOcrInstructionsSchema = robotBase
+  .merge(robotUse)
+  .extend({
+    robot: z.literal('/document/ocr').describe(`
+With this <dfn>Robot</dfn>, you can detect and extract text from PDFs using optical character recognition (OCR).
+
+For example, you can use the results to obtain the content of invoices, legal documents or restaurant menus. You can also pass the text down to other <dfn>Robots</dfn> to filter documents that contain (or do not contain) certain phrases.
+`),
+    provider: aiProviderSchema.describe(`
+Which AI provider to leverage. Valid values are \`"aws"\` and \`"gcp"\`.
+
+Transloadit outsources this task and abstracts the interface so you can expect the same data structures, but different latencies and information being returned. Different cloud vendors have different areas they shine in, and we recommend to try out and see what yields the best results for your use case.
+
+AWS supports detection for the following languages: English, Arabic, Russian, German, French, Italian, Portuguese and Spanish. GCP allows for a wider range of languages, with varying levels of support which can be found on the [official documentation](https://cloud.google.com/vision/docs/languages/).
+`),
+    granularity: granularitySchema.describe(`
+Whether to return a full response including coordinates for the text (\`"full"\`), or a flat list of the extracted phrases (\`"list"\`). This parameter has no effect if the \`format\` parameter is set to \`"text"\`.
+`),
+    format: z
+      .enum(['json', 'meta', 'text'])
+      .default('json')
+      .describe(`
+In what format to return the extracted text.
+- \`"json"\` returns a JSON file.
+- \`"meta"\` does not return a file, but stores the data inside Transloadit's file object (under \`\${file.meta.recognized_text}\`, which is an array of strings) that's passed around between encoding <dfn>Steps</dfn>, so that you can use the values to burn the data into videos, filter on them, etc.
+- \`"text"\` returns the recognized text as a plain UTF-8 encoded text file.
+`),
+  })
+  .strict()
+
+export const robotDocumentOcrInstructionsWithHiddenFieldsSchema =
+  robotDocumentOcrInstructionsSchema.extend({
+    result: z
+      .union([z.literal('debug'), robotDocumentOcrInstructionsSchema.shape.result])
+      .optional(),
+  })
+
+export type RobotDocumentOcrInstructions = z.infer<typeof robotDocumentOcrInstructionsSchema>
+export type RobotDocumentOcrInstructionsWithHiddenFields = z.infer<
+  typeof robotDocumentOcrInstructionsWithHiddenFieldsSchema
+>
+
+export const interpolatableRobotDocumentOcrInstructionsSchema = interpolateRobot(
+  robotDocumentOcrInstructionsSchema,
+)
+export type InterpolatableRobotDocumentOcrInstructions =
+  InterpolatableRobotDocumentOcrInstructionsInput
+
+export type InterpolatableRobotDocumentOcrInstructionsInput = z.input<
+  typeof interpolatableRobotDocumentOcrInstructionsSchema
+>
+
+export const interpolatableRobotDocumentOcrInstructionsWithHiddenFieldsSchema = interpolateRobot(
+  robotDocumentOcrInstructionsWithHiddenFieldsSchema,
+)
+export type InterpolatableRobotDocumentOcrInstructionsWithHiddenFields = z.infer<
+  typeof interpolatableRobotDocumentOcrInstructionsWithHiddenFieldsSchema
+>
+export type InterpolatableRobotDocumentOcrInstructionsWithHiddenFieldsInput = z.input<
+  typeof interpolatableRobotDocumentOcrInstructionsWithHiddenFieldsSchema
+>

package/src/alphalib/types/robots/document-split.ts

@@ -0,0 +1,78 @@
+import { z } from 'zod'
+
+import type { RobotMetaInput } from './_instructions-primitives.ts'
+import { interpolateRobot, robotBase, robotUse } from './_instructions-primitives.ts'
+
+export const meta: RobotMetaInput = {
+  allowed_for_url_transform: true,
+  bytescount: 1,
+  discount_factor: 1,
+  discount_pct: 0,
+  example_code_description: 'Extract single or multiple pages from a PDF document:',
+  minimum_charge: 2097152,
+  output_factor: 1,
+  override_lvl1: 'Document Processing',
+  purpose_sentence: 'extracts pages from documents',
+  purpose_verb: 'extract',
+  purpose_word: 'extracts pages',
+  purpose_words: 'Extracts pages',
+  service_slug: 'document-processing',
+  slot_count: 10,
+  title: 'Extract pages from a document',
+  typical_file_size_mb: 0.8,
+  typical_file_type: 'document',
+  name: 'DocumentSplitRobot',
+  priceFactor: 1,
+  queueSlotCount: 10,
+  minimumCharge: 1048576,
+  isAllowedForUrlTransform: true,
+  trackOutputFileSize: true,
+  isInternal: false,
+  removeJobResultFilesFromDiskRightAfterStoringOnS3: false,
+  stage: 'ga',
+}
+
+export const robotDocumentSplitInstructionsSchema = robotBase
+  .merge(robotUse)
+  .extend({
+    robot: z.literal('/document/split'),
+    pages: z
+      .union([z.string(), z.array(z.string())])
+      .describe(
+        'The pages to select from the input PDF and to be included in the output PDF. Each entry can be a single page number (e.g. 5), or a range (e.g. `5-10`). Page numbers start at 1. By default all pages are extracted.',
+      )
+      .optional(),
+  })
+  .strict()
+
+export const robotDocumentSplitInstructionsWithHiddenFieldsSchema =
+  robotDocumentSplitInstructionsSchema.extend({
+    result: z
+      .union([z.literal('debug'), robotDocumentSplitInstructionsSchema.shape.result])
+      .optional(),
+  })
+
+export type RobotDocumentSplitInstructions = z.infer<typeof robotDocumentSplitInstructionsSchema>
+export type RobotDocumentSplitInstructionsWithHiddenFields = z.infer<
+  typeof robotDocumentSplitInstructionsWithHiddenFieldsSchema
+>
+
+export const interpolatableRobotDocumentSplitInstructionsSchema = interpolateRobot(
+  robotDocumentSplitInstructionsSchema,
+)
+export type InterpolatableRobotDocumentSplitInstructions =
+  InterpolatableRobotDocumentSplitInstructionsInput
+
+export type InterpolatableRobotDocumentSplitInstructionsInput = z.input<
+  typeof interpolatableRobotDocumentSplitInstructionsSchema
+>
+
+export const interpolatableRobotDocumentSplitInstructionsWithHiddenFieldsSchema = interpolateRobot(
+  robotDocumentSplitInstructionsWithHiddenFieldsSchema,
+)
+export type InterpolatableRobotDocumentSplitInstructionsWithHiddenFields = z.infer<
+  typeof interpolatableRobotDocumentSplitInstructionsWithHiddenFieldsSchema
+>
+export type InterpolatableRobotDocumentSplitInstructionsWithHiddenFieldsInput = z.input<
+  typeof interpolatableRobotDocumentSplitInstructionsWithHiddenFieldsSchema
+>