web-csv-toolbox 0.4.0 → 0.5.1

package/lib/index.js

@@ -1,430 +0,0 @@
-const FieldDelimiter = Symbol.for("web-csv-toolbox.FieldDelimiter");
-const RecordDelimiter = Symbol.for("web-csv-toolbox.RecordDelimiter");
-const Field = Symbol.for("web-csv-toolbox.Field");
-const CR = "\r";
-const CRLF = "\r\n";
-const LF = "\n";
-const COMMA = ",";
-const DOUBLE_QUATE = '"';
-function assertCommonOptions(options) {
-  if (typeof options.quotation === "string" && options.quotation.length === 0) {
-    throw new Error("quotation must not be empty");
-  }
-  if (typeof options.demiliter === "string" && options.demiliter.length === 0) {
-    throw new Error("demiliter must not be empty");
-  }
-  if (options.quotation.includes(LF) || options.quotation.includes(CR)) {
-    throw new Error("quotation must not include CR or LF");
-  }
-  if (options.demiliter.includes(LF) || options.demiliter.includes(CR)) {
-    throw new Error("demiliter must not include CR or LF");
-  }
-  if (
-    options.demiliter.includes(options.quotation) ||
-    options.quotation.includes(options.demiliter)
-  ) {
-    throw new Error(
-      "demiliter and quotation must not include each other as a substring",
-    );
-  }
-}
-function escapeRegExp(v) {
-  return v.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
-}
-class LexerTransformer extends TransformStream {
-  #demiliter;
-  #demiliterLength;
-  #quotation;
-  #quotationLength;
-  #matcher;
-  #buffer = "";
-  get demiliter() {
-    return this.#demiliter;
-  }
-  get quotation() {
-    return this.#quotation;
-  }
-  constructor({ demiliter = COMMA, quotation = DOUBLE_QUATE } = {}) {
-    assertCommonOptions({ demiliter, quotation });
-    super({
-      transform: (chunk, controller) => {
-        if (chunk.length !== 0) {
-          this.#buffer += chunk;
-          for (const token of this.#tokens({ flush: false })) {
-            controller.enqueue(token);
-          }
-        }
-      },
-      flush: (controller) => {
-        for (const token of this.#tokens({ flush: true })) {
-          controller.enqueue(token);
-        }
-      },
-    });
-    this.#demiliter = demiliter;
-    this.#demiliterLength = demiliter.length;
-    this.#quotation = quotation;
-    this.#quotationLength = quotation.length;
-    const d = escapeRegExp(demiliter);
-    const q = escapeRegExp(quotation);
-    this.#matcher = new RegExp(
-      `^(?:(?!${q})(?!${d})(?![\\r\\n]))([\\S\\s\\uFEFF\\xA0]+?)(?=${q}|${d}|\\r|\\n|$)`,
-    );
-  }
-  *#tokens({ flush }) {
-    let currentField = null;
-    for (let token; (token = this.#nextToken({ flush })); ) {
-      switch (token.type) {
-        case Field:
-          if (currentField) {
-            currentField.value += token.value;
-          } else {
-            currentField = token;
-          }
-          break;
-        case FieldDelimiter:
-          if (currentField) {
-            yield currentField;
-            currentField = null;
-          }
-          yield token;
-          break;
-        case RecordDelimiter:
-          if (currentField) {
-            yield currentField;
-            currentField = null;
-          }
-          yield token;
-          break;
-      }
-    }
-    if (currentField) {
-      yield currentField;
-    }
-  }
-  #nextToken({ flush = false } = {}) {
-    if (this.#buffer.length === 0) {
-      return null;
-    }
-    if (this.#buffer.startsWith(CRLF)) {
-      this.#buffer = this.#buffer.slice(2);
-      return { type: RecordDelimiter, value: CRLF };
-    }
-    if (this.#buffer.startsWith(LF)) {
-      this.#buffer = this.#buffer.slice(1);
-      return { type: RecordDelimiter, value: LF };
-    }
-    if (this.#buffer.startsWith(this.#demiliter)) {
-      this.#buffer = this.#buffer.slice(this.#demiliterLength);
-      return { type: FieldDelimiter, value: this.#demiliter };
-    }
-    if (this.#buffer.startsWith(this.#quotation)) {
-      if (flush === false && this.#buffer.endsWith(this.#quotation)) {
-        return null;
-      }
-      return this.extractQuotedString(flush);
-    }
-    const match = this.#matcher.exec(this.#buffer);
-    if (match) {
-      if (flush === false && match[0].length === this.#buffer.length) {
-        return null;
-      }
-      this.#buffer = this.#buffer.slice(match[0].length);
-      return { type: Field, value: match[0] };
-    }
-    return null;
-  }
-  extractQuotedString(flush) {
-    let end = this.#quotationLength;
-    let value = "";
-    while (end < this.#buffer.length) {
-      if (
-        this.#buffer.slice(end, end + this.#quotationLength) ===
-          this.quotation &&
-        this.#buffer.slice(
-          end + this.#quotationLength,
-          end + this.#quotationLength * 2,
-        ) === this.quotation
-      ) {
-        value += this.quotation;
-        end += this.#quotationLength * 2;
-        continue;
-      }
-      if (
-        this.#buffer.slice(end, end + this.#quotationLength) === this.quotation
-      ) {
-        if (
-          flush === false &&
-          end + this.#quotationLength < this.#buffer.length &&
-          this.#buffer.slice(
-            end + this.#quotationLength,
-            this.#demiliterLength,
-          ) !== this.demiliter &&
-          this.#buffer.slice(
-            end + this.#quotationLength,
-            end + this.#quotationLength + 2,
-          ) !== CRLF &&
-          this.#buffer.slice(
-            end + this.#quotationLength,
-            end + this.#quotationLength + 1,
-          ) !== LF
-        ) {
-          return null;
-        }
-        this.#buffer = this.#buffer.slice(end + this.#quotationLength);
-        return { type: Field, value };
-      }
-      value += this.#buffer[end];
-      end++;
-    }
-    return null;
-  }
-}
-class RecordAssemblerTransformar extends TransformStream {
-  #fieldIndex = 0;
-  #row = [];
-  #header;
-  #darty = false;
-  constructor(options = {}) {
-    super({
-      transform: (token, controller) => {
-        switch (token.type) {
-          case Field:
-            this.#darty = true;
-            this.#row[this.#fieldIndex] = token.value;
-            break;
-          case FieldDelimiter:
-            this.#fieldIndex++;
-            break;
-          case RecordDelimiter:
-            if (this.#header === undefined) {
-              this.#setHeader(this.#row);
-            } else {
-              if (this.#darty) {
-                const record = Object.fromEntries(
-                  this.#header
-                    .filter((v) => v)
-                    .map((header, index) => [header, this.#row.at(index)]),
-                );
-                controller.enqueue(record);
-              }
-            }
-            this.#fieldIndex = 0;
-            this.#row = new Array(this.#header?.length);
-            this.#darty = false;
-            break;
-        }
-      },
-      flush: (controller) => {
-        if (this.#fieldIndex !== 0 && this.#header !== undefined) {
-          if (this.#darty) {
-            const record = Object.fromEntries(
-              this.#header
-                .filter((v) => v)
-                .map((header, index) => [header, this.#row.at(index)]),
-            );
-            controller.enqueue(record);
-          }
-        }
-      },
-    });
-    if (options.header !== undefined && Array.isArray(options.header)) {
-      this.#setHeader(options.header);
-    }
-  }
-  #setHeader(header) {
-    this.#header = header;
-    if (this.#header.length === 0) {
-      throw new Error("The header must not be empty.");
-    }
-    if (new Set(this.#header).size !== this.#header.length) {
-      throw new Error("The header must not contain duplicate fields.");
-    }
-  }
-}
-class SingleValueReadableStream extends ReadableStream {
-  constructor(value) {
-    super({
-      start(controller) {
-        controller.enqueue(value);
-        controller.close();
-      },
-    });
-  }
-}
-async function toArray(...args) {
-  const rows = [];
-  for await (const row of this(...args)) {
-    rows.push(row);
-  }
-  return rows;
-}
-async function* parseStringStream(stream, options) {
-  let controller;
-  const readable = new ReadableStream({
-    start: (controller_) => (controller = controller_),
-  });
-  await stream
-    .pipeThrough(new LexerTransformer(options))
-    .pipeThrough(new RecordAssemblerTransformar(options))
-    .pipeTo(
-      new WritableStream({
-        write: (row) => controller.enqueue(row),
-        close: () => controller.close(),
-      }),
-    );
-  const reader = readable.getReader();
-  try {
-    while (true) {
-      const { value, done } = await reader.read();
-      if (done) break;
-      yield value;
-    }
-  } finally {
-    reader.releaseLock();
-  }
-}
-((parseStringStream) => {
-  Object.defineProperty(parseStringStream, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseStringStream || (parseStringStream = {}));
-async function* parseString(csv, options) {
-  yield* parseStringStream(new SingleValueReadableStream(csv), options);
-}
-((parseString) => {
-  Object.defineProperty(parseString, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseString || (parseString = {}));
-async function* parseUint8ArrayStream(stream, options) {
-  const { charset, fatal, ignoreBOM, decomposition } = options ?? {};
-  yield* parseStringStream(
-    [
-      ...(decomposition ? [new DecompressionStream(decomposition)] : []),
-      new TextDecoderStream(charset, { fatal, ignoreBOM }),
-    ].reduce((stream, transformer) => stream.pipeThrough(transformer), stream),
-    options,
-  );
-}
-((parseUint8ArrayStream) => {
-  Object.defineProperty(parseUint8ArrayStream, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseUint8ArrayStream || (parseUint8ArrayStream = {}));
-function parseUint8Array(bytes, options) {
-  return parseUint8ArrayStream(new SingleValueReadableStream(bytes), options);
-}
-((parseUint8Array) => {
-  Object.defineProperty(parseUint8Array, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseUint8Array || (parseUint8Array = {}));
-function parseArrayBuffer(buffer, options) {
-  return parseUint8Array(new Uint8Array(buffer), options);
-}
-((parseArrayBuffer) => {
-  Object.defineProperty(parseArrayBuffer, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseArrayBuffer || (parseArrayBuffer = {}));
-function parseMime(contentType) {
-  const [type, ...parameters] = contentType.split(";");
-  const result = {
-    type: type.trim(),
-    parameters: {},
-  };
-  for (const paramator of parameters) {
-    const [key, value] = paramator.split("=");
-    result.parameters[key.trim()] = value.trim();
-  }
-  return result;
-}
-function parseResponse(response, options) {
-  const { headers } = response;
-  const contentType = headers.get("content-type") ?? "text/csv";
-  const mime = parseMime(contentType);
-  if (mime.type !== "text/csv") {
-    throw new Error(`Invalid mime type: ${contentType}`);
-  }
-  const decomposition = headers.get("content-encoding") ?? undefined;
-  const charset = mime.parameters.charset ?? "utf-8";
-  if (response.body === null) {
-    throw new Error("Response body is null");
-  }
-  return parseUint8ArrayStream(response.body, {
-    decomposition,
-    charset,
-    ...options,
-  });
-}
-((parseResponse) => {
-  Object.defineProperty(parseResponse, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseResponse || (parseResponse = {}));
-async function* parseStream(stream, options) {
-  const [branch1, branch2] = stream.tee();
-  const reader1 = branch1.getReader();
-  const { value: firstChunk } = await reader1.read();
-  reader1.releaseLock();
-  if (typeof firstChunk === "string") {
-    yield* parseStringStream(branch2, options);
-  } else if (firstChunk instanceof Uint8Array) {
-    yield* parseUint8ArrayStream(branch2, options);
-  }
-}
-((parseStream) => {
-  Object.defineProperty(parseStream, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parseStream || (parseStream = {}));
-async function* parse(csv, options) {
-  if (typeof csv === "string") {
-    yield* parseString(csv, options);
-  } else if (csv instanceof Uint8Array) {
-    yield* parseUint8Array(csv, options);
-  } else if (csv instanceof ArrayBuffer) {
-    yield* parseArrayBuffer(csv, options);
-  } else if (csv instanceof ReadableStream) {
-    yield* parseStream(csv, options);
-  } else if (csv instanceof Response) {
-    yield* parseResponse(csv, options);
-  }
-}
-((parse) => {
-  Object.defineProperty(parse, "toArray", {
-    enumerable: true,
-    writable: false,
-    value: toArray,
-  });
-})(parse || (parse = {}));
-export {
-  Field,
-  FieldDelimiter,
-  LexerTransformer,
-  RecordAssemblerTransformar,
-  RecordDelimiter,
-  parse,
-  parseArrayBuffer,
-  parseResponse,
-  parseStream,
-  parseString,
-  parseStringStream,
-  parseUint8Array,
-  parseUint8ArrayStream,
-};

package/lib/index.umd.js

@@ -1 +0,0 @@
-!function(e,t){"object"==typeof exports&&"undefined"!=typeof module?t(exports):"function"==typeof define&&define.amd?define(["exports"],t):t((e="undefined"!=typeof globalThis?globalThis:e||self).CSV={})}(this,(function(e){"use strict";const t=Symbol.for("web-csv-toolbox.FieldDelimiter"),i=Symbol.for("web-csv-toolbox.RecordDelimiter"),r=Symbol.for("web-csv-toolbox.Field"),n="\r\n",o="\n";function s(e){return e.replace(/[.*+?^${}()|[\]\\]/g,"\\$&")}class a extends TransformStream{#e;#t;#i;#r;#n;#o="";get demiliter(){return this.#e}get quotation(){return this.#i}constructor({demiliter:e=",",quotation:t='"'}={}){!function(e){if("string"==typeof e.quotation&&0===e.quotation.length)throw new Error("quotation must not be empty");if("string"==typeof e.demiliter&&0===e.demiliter.length)throw new Error("demiliter must not be empty");if(e.quotation.includes(o)||e.quotation.includes("\r"))throw new Error("quotation must not include CR or LF");if(e.demiliter.includes(o)||e.demiliter.includes("\r"))throw new Error("demiliter must not include CR or LF");if(e.demiliter.includes(e.quotation)||e.quotation.includes(e.demiliter))throw new Error("demiliter and quotation must not include each other as a substring")}({demiliter:e,quotation:t}),super({transform:(e,t)=>{if(0!==e.length){this.#o+=e;for(const e of this.#s({flush:!1}))t.enqueue(e)}},flush:e=>{for(const t of this.#s({flush:!0}))e.enqueue(t)}}),this.#e=e,this.#t=e.length,this.#i=t,this.#r=t.length;const i=s(e),r=s(t);this.#n=new RegExp(`^(?:(?!${r})(?!${i})(?![\\r\\n]))([\\S\\s\\uFEFF\\xA0]+?)(?=${r}|${i}|\\r|\\n|$)`)}*#s({flush:e}){let n=null;for(let o;o=this.#a({flush:e});)switch(o.type){case r:n?n.value+=o.value:n=o;break;case t:case i:n&&(yield n,n=null),yield o}n&&(yield n)}#a({flush:e=!1}={}){if(0===this.#o.length)return null;if(this.#o.startsWith(n))return this.#o=this.#o.slice(2),{type:i,value:n};if(this.#o.startsWith(o))return this.#o=this.#o.slice(1),{type:i,value:o};if(this.#o.startsWith(this.#e))return this.#o=this.#o.slice(this.#t),{type:t,value:this.#e};if(this.#o.startsWith(this.#i))return!1===e&&this.#o.endsWith(this.#i)?null:this.extractQuotedString(e);const s=this.#n.exec(this.#o);return s?!1===e&&s[0].length===this.#o.length?null:(this.#o=this.#o.slice(s[0].length),{type:r,value:s[0]}):null}extractQuotedString(e){let t=this.#r,i="";for(;t<this.#o.length;)if(this.#o.slice(t,t+this.#r)!==this.quotation||this.#o.slice(t+this.#r,t+2*this.#r)!==this.quotation){if(this.#o.slice(t,t+this.#r)===this.quotation)return!1===e&&t+this.#r<this.#o.length&&this.#o.slice(t+this.#r,this.#t)!==this.demiliter&&this.#o.slice(t+this.#r,t+this.#r+2)!==n&&this.#o.slice(t+this.#r,t+this.#r+1)!==o?null:(this.#o=this.#o.slice(t+this.#r),{type:r,value:i});i+=this.#o[t],t++}else i+=this.quotation,t+=2*this.#r;return null}}class u extends TransformStream{#u=0;#l=[];#h;#f=!1;constructor(e={}){super({transform:(e,n)=>{switch(e.type){case r:this.#f=!0,this.#l[this.#u]=e.value;break;case t:this.#u++;break;case i:if(void 0===this.#h)this.#c(this.#l);else if(this.#f){const e=Object.fromEntries(this.#h.filter((e=>e)).map(((e,t)=>[e,this.#l.at(t)])));n.enqueue(e)}this.#u=0,this.#l=new Array(this.#h?.length),this.#f=!1}},flush:e=>{if(0!==this.#u&&void 0!==this.#h&&this.#f){const t=Object.fromEntries(this.#h.filter((e=>e)).map(((e,t)=>[e,this.#l.at(t)])));e.enqueue(t)}}}),void 0!==e.header&&Array.isArray(e.header)&&this.#c(e.header)}#c(e){if(this.#h=e,0===this.#h.length)throw new Error("The header must not be empty.");if(new Set(this.#h).size!==this.#h.length)throw new Error("The header must not contain duplicate fields.")}}class l extends ReadableStream{constructor(e){super({start(t){t.enqueue(e),t.close()}})}}async function h(...e){const t=[];for await(const i of this(...e))t.push(i);return t}async function*f(e,t){let i;const r=new ReadableStream({start:e=>i=e});await e.pipeThrough(new a(t)).pipeThrough(new u(t)).pipeTo(new WritableStream({write:e=>i.enqueue(e),close:()=>i.close()}));const n=r.getReader();try{for(;;){const{value:e,done:t}=await n.read();if(t)break;yield e}}finally{n.releaseLock()}}async function*c(e,t){yield*f(new l(e),t)}async function*d(e,t){const{charset:i,fatal:r,ignoreBOM:n,decomposition:o}=t??{};yield*f([...o?[new DecompressionStream(o)]:[],new TextDecoderStream(i,{fatal:r,ignoreBOM:n})].reduce(((e,t)=>e.pipeThrough(t)),e),t)}function m(e,t){return d(new l(e),t)}function b(e,t){return m(new Uint8Array(e),t)}function y(e,t){const{headers:i}=e,r=i.get("content-type")??"text/csv",n=function(e){const[t,...i]=e.split(";"),r={type:t.trim(),parameters:{}};for(const e of i){const[t,i]=e.split("=");r.parameters[t.trim()]=i.trim()}return r}(r);if("text/csv"!==n.type)throw new Error(`Invalid mime type: ${r}`);const o=i.get("content-encoding")??void 0,s=n.parameters.charset??"utf-8";if(null===e.body)throw new Error("Response body is null");return d(e.body,{decomposition:o,charset:s,...t})}async function*p(e,t){const[i,r]=e.tee(),n=i.getReader(),{value:o}=await n.read();n.releaseLock(),"string"==typeof o?yield*f(r,t):o instanceof Uint8Array&&(yield*d(r,t))}async function*g(e,t){"string"==typeof e?yield*c(e,t):e instanceof Uint8Array?yield*m(e,t):e instanceof ArrayBuffer?yield*b(e,t):e instanceof ReadableStream?yield*p(e,t):e instanceof Response&&(yield*y(e,t))}!function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(f||(f={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(c||(c={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(d||(d={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(m||(m={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(b||(b={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(y||(y={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(p||(p={})),function(e){Object.defineProperty(e,"toArray",{enumerable:!0,writable:!1,value:h})}(g||(g={})),e.Field=r,e.FieldDelimiter=t,e.LexerTransformer=a,e.RecordAssemblerTransformar=u,e.RecordDelimiter=i,e.parse=g,e.parseArrayBuffer=b,e.parseResponse=y,e.parseStream=p,e.parseString=c,e.parseStringStream=f,e.parseUint8Array=m,e.parseUint8ArrayStream=d}));