@cj-tech-master/excelts 7.6.0 → 8.0.0

package/dist/browser/modules/markdown/parse/index.js

@@ -0,0 +1,428 @@
+/**
+ * Markdown Table Parser
+ *
+ * Parses Markdown tables into structured data.
+ *
+ * Supports:
+ * - Standard GFM (GitHub Flavored Markdown) table syntax
+ * - Column alignment detection via separator row
+ * - Escaped pipes (`\|`) in cell content
+ * - Tables with or without leading/trailing pipes
+ * - Tolerant parsing (mismatched column counts, extra whitespace)
+ * - Multiline cell content via `<br>` / `<br/>` / `<br />` tags
+ *
+ * @example
+ * ```ts
+ * const result = parseMarkdown("| Name | Age |\n| --- | --- |\n| Alice | 30 |");
+ * // result.headers = ["Name", "Age"]
+ * // result.rows = [["Alice", "30"]]
+ * // result.alignments = ["none", "none"]
+ * ```
+ */
+import { BR_TAG_REGEX, LINEBREAK_REGEX, UNESCAPE_REGEX } from "../constants.js";
+import { MarkdownParseError } from "../errors.js";
+// =============================================================================
+// Character Codes (avoid repeated charCodeAt comparisons with magic numbers)
+// =============================================================================
+const CH_PIPE = 0x7c; // |
+const CH_BACKSLASH = 0x5c; // \
+const CH_COLON = 0x3a; // :
+const CH_DASH = 0x2d; // -
+const CH_SPACE = 0x20; // space
+const CH_TAB = 0x09; // tab
+function resolveParseOpts(options) {
+    return {
+        trim: options.trim !== false,
+        unescape: options.unescape !== false,
+        skipEmpty: options.skipEmptyRows !== false,
+        maxRows: options.maxRows,
+        convertBr: options.convertBr === true
+    };
+}
+// =============================================================================
+// Internal Helpers
+// =============================================================================
+/**
+ * Split a Markdown table row into cell values.
+ * Handles escaped pipes (`\|`) correctly by scanning character by character.
+ *
+ * Optimized: uses start/end index tracking with `slice()` instead of
+ * character-by-character string concatenation to avoid O(n²) worst case.
+ */
+function splitRow(line) {
+    const cells = [];
+    const len = line.length;
+    // Skip leading pipe
+    let i = len > 0 && line.charCodeAt(0) === CH_PIPE ? 1 : 0;
+    // Check for trailing pipe (to exclude it from the last cell)
+    // Must count consecutive backslashes before the pipe:
+    // even count (0, 2, 4...) → backslashes are all escaped, pipe is real
+    // odd count (1, 3, 5...) → last backslash escapes the pipe
+    let end = len;
+    if (len > 1 && line.charCodeAt(len - 1) === CH_PIPE) {
+        let backslashCount = 0;
+        let k = len - 2;
+        while (k >= 0 && line.charCodeAt(k) === CH_BACKSLASH) {
+            backslashCount++;
+            k--;
+        }
+        if (backslashCount % 2 === 0) {
+            end = len - 1;
+        }
+    }
+    // Track the start of the current cell segment
+    // We collect segments (between escape sequences) to minimize allocations
+    let segStart = i;
+    let current = "";
+    let hasEscape = false;
+    while (i < end) {
+        const ch = line.charCodeAt(i);
+        if (ch === CH_BACKSLASH && i + 1 < end) {
+            // Escape sequence: flush preceding segment, add escape pair
+            hasEscape = true;
+            if (i > segStart) {
+                current += line.slice(segStart, i);
+            }
+            current += line.slice(i, i + 2);
+            i += 2;
+            segStart = i;
+        }
+        else if (ch === CH_PIPE) {
+            // Cell boundary: flush and push
+            if (hasEscape) {
+                if (i > segStart) {
+                    current += line.slice(segStart, i);
+                }
+                cells.push(current);
+                current = "";
+                hasEscape = false;
+            }
+            else {
+                cells.push(line.slice(segStart, i));
+            }
+            i++;
+            segStart = i;
+        }
+        else {
+            i++;
+        }
+    }
+    // Push the last cell
+    if (hasEscape) {
+        if (end > segStart) {
+            current += line.slice(segStart, end);
+        }
+        cells.push(current);
+    }
+    else {
+        cells.push(line.slice(segStart, end));
+    }
+    return cells;
+}
+/**
+ * Determine column alignment from a separator cell.
+ *
+ * - `:---:` → center
+ * - `:---`  → left
+ * - `---:`  → right
+ * - `---`   → none
+ */
+function parseAlignment(cell) {
+    const trimmed = cell.trim();
+    const tLen = trimmed.length;
+    if (tLen === 0) {
+        return "none";
+    }
+    const leftColon = trimmed.charCodeAt(0) === CH_COLON;
+    const rightColon = trimmed.charCodeAt(tLen - 1) === CH_COLON;
+    if (leftColon && rightColon) {
+        return "center";
+    }
+    if (leftColon) {
+        return "left";
+    }
+    if (rightColon) {
+        return "right";
+    }
+    return "none";
+}
+/**
+ * Check if a cell string is a valid separator cell.
+ * Hand-rolled check replacing regex for better performance.
+ * Pattern: optional whitespace, optional colon, one or more dashes, optional colon, optional whitespace.
+ */
+function isSeparatorCell(cell) {
+    const len = cell.length;
+    let i = 0;
+    // Skip leading whitespace
+    while (i < len) {
+        const ch = cell.charCodeAt(i);
+        if (ch !== CH_SPACE && ch !== CH_TAB) {
+            break;
+        }
+        i++;
+    }
+    // Optional leading colon
+    if (i < len && cell.charCodeAt(i) === CH_COLON) {
+        i++;
+    }
+    // At least one dash required
+    const dashStart = i;
+    while (i < len && cell.charCodeAt(i) === CH_DASH) {
+        i++;
+    }
+    if (i === dashStart) {
+        return false;
+    }
+    // Optional trailing colon
+    if (i < len && cell.charCodeAt(i) === CH_COLON) {
+        i++;
+    }
+    // Skip trailing whitespace
+    while (i < len) {
+        const ch = cell.charCodeAt(i);
+        if (ch !== CH_SPACE && ch !== CH_TAB) {
+            return false;
+        }
+        i++;
+    }
+    return true;
+}
+/**
+ * Check if a line is a valid separator row.
+ * A separator row consists entirely of cells matching the pattern `:?-+:?`.
+ */
+function isSeparatorRow(cells) {
+    if (cells.length === 0) {
+        return false;
+    }
+    for (let i = 0; i < cells.length; i++) {
+        if (!isSeparatorCell(cells[i])) {
+            return false;
+        }
+    }
+    return true;
+}
+/**
+ * Process cell content: trim, optionally unescape, and optionally convert `<br>` to newlines.
+ */
+function processCell(value, opts) {
+    let result = opts.trim ? value.trim() : value;
+    if (opts.unescape) {
+        result = result.replace(UNESCAPE_REGEX, "$1");
+    }
+    if (opts.convertBr) {
+        result = result.replace(BR_TAG_REGEX, "\n");
+    }
+    return result;
+}
+/**
+ * Normalize a row to the expected column count.
+ * - If row has fewer cells, pad with empty strings
+ * - If row has more cells, truncate
+ */
+function normalizeRow(cells, columnCount, opts) {
+    const row = new Array(columnCount);
+    for (let i = 0; i < columnCount; i++) {
+        row[i] = i < cells.length ? processCell(cells[i], opts) : "";
+    }
+    return row;
+}
+/**
+ * Check if a row is empty (all cells are empty strings).
+ */
+function isEmptyRow(row) {
+    for (let i = 0; i < row.length; i++) {
+        if (row[i] !== "") {
+            return false;
+        }
+    }
+    return true;
+}
+/**
+ * Check if a line could be part of a table (contains a pipe character).
+ */
+function isTableLine(line) {
+    return line.indexOf("|") !== -1;
+}
+/**
+ * Check if a line starts with a pipe (after optional leading whitespace).
+ * Used to determine the table's "piped" style for data row validation.
+ */
+function startsWithPipe(line) {
+    const len = line.length;
+    let i = 0;
+    while (i < len) {
+        const ch = line.charCodeAt(i);
+        if (ch !== CH_SPACE && ch !== CH_TAB) {
+            return ch === CH_PIPE;
+        }
+        i++;
+    }
+    return false;
+}
+/**
+ * Check if a line could be a separator candidate (contains a dash).
+ */
+function hasDash(line) {
+    return line.indexOf("-") !== -1;
+}
+// =============================================================================
+// Core Table Parser (shared between parseMarkdown and parseMarkdownAll)
+// =============================================================================
+/**
+ * Attempt to parse a table starting at line index `startLine`.
+ *
+ * Returns `{ result, endLine }` if a valid table starts here, or `null` otherwise.
+ */
+function parseTableAt(lines, startLine, lineCount, opts) {
+    if (startLine >= lineCount - 1) {
+        return null;
+    }
+    const line = lines[startLine].trim();
+    // Skip empty lines and non-table content
+    if (line === "" || !isTableLine(line)) {
+        return null;
+    }
+    // Candidate header row
+    const headerCells = splitRow(line);
+    if (headerCells.length < 1) {
+        return null;
+    }
+    // Check if the next line is a valid separator row
+    const separatorLine = lines[startLine + 1].trim();
+    if (separatorLine === "" || !hasDash(separatorLine)) {
+        return null;
+    }
+    const separatorCells = splitRow(separatorLine);
+    if (!isSeparatorRow(separatorCells)) {
+        return null;
+    }
+    // Valid table found — extract headers and alignments
+    const columnCount = headerCells.length;
+    const headers = new Array(columnCount);
+    const alignments = new Array(columnCount);
+    for (let c = 0; c < columnCount; c++) {
+        headers[c] = processCell(headerCells[c], opts);
+        alignments[c] = c < separatorCells.length ? parseAlignment(separatorCells[c]) : "none";
+    }
+    // Determine if this is a "piped" table (header starts with `|`).
+    // When the header has a leading pipe, data rows must also start with `|`.
+    // This prevents prose like "This has a | pipe" from being swallowed as data.
+    const piped = startsWithPipe(line);
+    // Parse data rows
+    const rows = [];
+    let j = startLine + 2;
+    for (; j < lineCount; j++) {
+        const dataLine = lines[j].trim();
+        // Stop at empty line or non-table content (end of table)
+        if (dataLine === "" || !isTableLine(dataLine)) {
+            break;
+        }
+        // For piped tables, data rows must also start with `|`
+        if (piped && !startsWithPipe(dataLine)) {
+            break;
+        }
+        // Check maxRows limit
+        if (opts.maxRows !== undefined && rows.length >= opts.maxRows) {
+            // Skip remaining table rows for parseMarkdownAll to correctly advance
+            while (j < lineCount) {
+                const remaining = lines[j].trim();
+                if (remaining === "" || !isTableLine(remaining)) {
+                    break;
+                }
+                if (piped && !startsWithPipe(remaining)) {
+                    break;
+                }
+                j++;
+            }
+            break;
+        }
+        const dataCells = splitRow(dataLine);
+        const row = normalizeRow(dataCells, columnCount, opts);
+        if (opts.skipEmpty && isEmptyRow(row)) {
+            continue;
+        }
+        rows.push(row);
+    }
+    return { result: { headers, rows, alignments }, endLine: j };
+}
+// =============================================================================
+// Main Parser
+// =============================================================================
+/**
+ * Parse a Markdown table string into structured data.
+ *
+ * The parser looks for the GFM table pattern:
+ * 1. A header row (pipe-delimited cells)
+ * 2. A separator row (dashes with optional colons for alignment)
+ * 3. Zero or more data rows
+ *
+ * Non-table content before and after the table is ignored.
+ *
+ * @param input - Markdown string containing a table
+ * @param options - Parse options
+ * @returns Parsed table data with headers, rows, and alignments
+ *
+ * @throws {MarkdownParseError} When no valid table is found in the input
+ *
+ * @example
+ * ```ts
+ * // Basic table
+ * const result = parseMarkdown("| Name | Age |\n| --- | --- |\n| Alice | 30 |");
+ *
+ * // With alignment
+ * const result = parseMarkdown("| Left | Center | Right |\n|:---|:---:|---:|\n|a|b|c|");
+ * // result.alignments = ["left", "center", "right"]
+ *
+ * // From a larger Markdown document
+ * const result = parseMarkdown(markdownDoc); // Finds the first table
+ *
+ * // With options
+ * const result = parseMarkdown(input, { trim: false, maxRows: 100 });
+ * ```
+ */
+export function parseMarkdown(input, options = {}) {
+    const opts = resolveParseOpts(options);
+    const lines = input.split(LINEBREAK_REGEX);
+    const lineCount = lines.length;
+    for (let i = 0; i < lineCount - 1; i++) {
+        const parsed = parseTableAt(lines, i, lineCount, opts);
+        if (parsed) {
+            return parsed.result;
+        }
+    }
+    throw new MarkdownParseError("No valid Markdown table found in input", lineCount > 0 ? lineCount : 1);
+}
+/**
+ * Parse all Markdown tables from a document.
+ *
+ * @param input - Markdown string containing one or more tables
+ * @param options - Parse options (maxRows applies per table)
+ * @returns Array of parsed tables
+ *
+ * @example
+ * ```ts
+ * const tables = parseMarkdownAll(markdownDoc);
+ * console.log(`Found ${tables.length} tables`);
+ * tables.forEach((t, i) => console.log(`Table ${i}: ${t.headers.join(", ")}`));
+ * ```
+ */
+export function parseMarkdownAll(input, options = {}) {
+    const opts = resolveParseOpts(options);
+    const lines = input.split(LINEBREAK_REGEX);
+    const lineCount = lines.length;
+    const tables = [];
+    let i = 0;
+    while (i < lineCount - 1) {
+        const parsed = parseTableAt(lines, i, lineCount, opts);
+        if (parsed) {
+            tables.push(parsed.result);
+            i = parsed.endLine;
+        }
+        else {
+            i++;
+        }
+    }
+    return tables;
+}

package/dist/browser/modules/markdown/types.d.ts

@@ -0,0 +1,130 @@
+/**
+ * Markdown Types
+ *
+ * Centralized type definitions for the Markdown module.
+ */
+/**
+ * Column alignment in a Markdown table.
+ * Determined by colon placement in the separator row:
+ * - `:---`  = left
+ * - `:---:` = center
+ * - `---:`  = right
+ * - `---`   = none (defaults to left in most renderers)
+ */
+export type MarkdownAlignment = "left" | "center" | "right" | "none";
+/**
+ * Result of parsing a Markdown table.
+ */
+export interface MarkdownParseResult {
+    /** Parsed header row (column names) */
+    headers: string[];
+    /** Parsed data rows (each row is an array of cell values) */
+    rows: string[][];
+    /** Column alignments extracted from the separator row */
+    alignments: MarkdownAlignment[];
+}
+/**
+ * Markdown parsing options.
+ */
+export interface MarkdownParseOptions {
+    /**
+     * Trim whitespace from cell values.
+     * @default true
+     */
+    trim?: boolean;
+    /**
+     * Unescape Markdown escape sequences in cell values (e.g. `\|` -> `|`).
+     * @default true
+     */
+    unescape?: boolean;
+    /**
+     * Skip empty rows in the output.
+     * @default true
+     */
+    skipEmptyRows?: boolean;
+    /**
+     * Maximum number of data rows to parse (excludes header).
+     * Useful for previewing large tables.
+     */
+    maxRows?: number;
+    /**
+     * Convert `<br>`, `<br/>`, `<br />` tags in cell content to newline characters.
+     * Useful for round-tripping multiline cell content through Markdown.
+     * @default false
+     */
+    convertBr?: boolean;
+}
+/**
+ * Column configuration for Markdown formatting.
+ */
+export interface MarkdownColumnConfig {
+    /** Column header text */
+    header: string;
+    /**
+     * Column alignment.
+     * @default "left"
+     */
+    alignment?: MarkdownAlignment;
+    /**
+     * Minimum column width (in characters, excluding padding).
+     * The actual width will be the maximum of this value and the widest cell content.
+     * @default 3 (minimum for separator `---`)
+     */
+    minWidth?: number;
+}
+/**
+ * Markdown formatting options.
+ */
+export interface MarkdownFormatOptions {
+    /**
+     * Column configuration. When provided, overrides auto-detected headers and alignment.
+     * Can be an array of strings (header names) or MarkdownColumnConfig objects.
+     */
+    columns?: (string | MarkdownColumnConfig)[];
+    /**
+     * Default alignment for columns without explicit alignment.
+     * @default "left"
+     */
+    alignment?: MarkdownAlignment;
+    /**
+     * Align columns to equal width by padding cell content with spaces.
+     * When false, disables width-alignment padding but retains the single space
+     * around cell content required by most Markdown renderers (`| value |`).
+     * @default true
+     */
+    padding?: boolean;
+    /**
+     * Include a trailing newline at the end of the output.
+     * @default true
+     */
+    trailingNewline?: boolean;
+    /**
+     * Escape pipe characters (`|`) and backslashes (`\`) in cell content.
+     * @default true
+     */
+    escapeContent?: boolean;
+    /**
+     * Custom value-to-string converter.
+     * Called for each cell value before formatting.
+     * Return value is used as the cell text.
+     */
+    stringify?: (value: unknown) => string;
+}
+/**
+ * Unified Markdown options for Workbook integration.
+ * Combines parse and format options with worksheet-specific settings.
+ */
+export interface MarkdownOptions extends MarkdownParseOptions, MarkdownFormatOptions {
+    /** Name of the worksheet to read from or write to */
+    sheetName?: string;
+    /** ID of the worksheet to read from or write to */
+    sheetId?: number;
+    /** Custom value mapper for parsing (MD string -> cell value) */
+    map?: (value: string, column: number) => unknown;
+    /** Date format string for formatting Date values */
+    dateFormat?: string;
+    /** Use UTC for date formatting */
+    dateUTC?: boolean;
+    /** Include empty rows in output */
+    includeEmptyRows?: boolean;
+}

package/dist/browser/modules/markdown/types.js

@@ -0,0 +1,6 @@
+/**
+ * Markdown Types
+ *
+ * Centralized type definitions for the Markdown module.
+ */
+export {};

package/dist/cjs/index.js

@@ -18,7 +18,7 @@ var __exportStar = (this && this.__exportStar) || function(m, exports) {
 };
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.PdfRenderError = exports.PdfError = exports.PageSizes = exports.excelToPdf = exports.pdf = exports.uint8ArrayToString = exports.stringToUint8Array = exports.toUint8Array = exports.concatUint8Arrays = exports.getRootCause = exports.getErrorChain = exports.errorToJSON = exports.toError = exports.BaseError = exports.getSupportedFormats = exports.DateFormatter = exports.DateParser = exports.xmlDecode = exports.xmlEncode = exports.uint8ArrayToBase64 = exports.base64ToUint8Array = exports.excelToDate = exports.dateToExcel = exports.encodeRange = exports.decodeRange = exports.encodeCell = exports.decodeCell = exports.encodeRow = exports.decodeRow = exports.encodeCol = exports.decodeCol = exports.DefinedNames = exports.createCsvFormatterStream = exports.createCsvParserStream = exports.CsvFormatterStream = exports.CsvParserStream = exports.WorksheetReader = exports.WorksheetWriter = exports.WorkbookReader = exports.WorkbookWriter = exports.FormCheckbox = exports.DataValidations = exports.Table = exports.Image = exports.Range = exports.Cell = exports.Column = exports.Row = exports.Worksheet = exports.Workbook = void 0;
-exports.MaxItemsExceededError = exports.ImageError = exports.TableError = exports.PivotTableError = exports.WorksheetNameError = exports.XmlParseError = exports.InvalidValueTypeError = exports.MergeConflictError = exports.RowOutOfBoundsError = exports.ColumnOutOfBoundsError = exports.InvalidAddressError = exports.ExcelStreamStateError = exports.ExcelNotSupportedError = exports.ExcelDownloadError = exports.ExcelFileError = exports.isExcelError = exports.ExcelError = exports.isPdfError = exports.PdfStructureError = exports.PdfFontError = void 0;
+exports.MarkdownParseError = exports.MarkdownError = exports.MaxItemsExceededError = exports.ImageError = exports.TableError = exports.PivotTableError = exports.WorksheetNameError = exports.XmlParseError = exports.InvalidValueTypeError = exports.MergeConflictError = exports.RowOutOfBoundsError = exports.ColumnOutOfBoundsError = exports.InvalidAddressError = exports.ExcelStreamStateError = exports.ExcelNotSupportedError = exports.ExcelDownloadError = exports.ExcelFileError = exports.isExcelError = exports.ExcelError = exports.isPdfError = exports.PdfStructureError = exports.PdfFontError = void 0;
 var workbook_1 = require("./modules/excel/workbook.js");
 Object.defineProperty(exports, "Workbook", { enumerable: true, get: function () { return workbook_1.Workbook; } });
 var worksheet_1 = require("./modules/excel/worksheet.js");
@@ -147,3 +147,7 @@ Object.defineProperty(exports, "PivotTableError", { enumerable: true, get: funct
 Object.defineProperty(exports, "TableError", { enumerable: true, get: function () { return errors_2.TableError; } });
 Object.defineProperty(exports, "ImageError", { enumerable: true, get: function () { return errors_2.ImageError; } });
 Object.defineProperty(exports, "MaxItemsExceededError", { enumerable: true, get: function () { return errors_2.MaxItemsExceededError; } });
+// Markdown errors
+var errors_3 = require("./modules/markdown/errors.js");
+Object.defineProperty(exports, "MarkdownError", { enumerable: true, get: function () { return errors_3.MarkdownError; } });
+Object.defineProperty(exports, "MarkdownParseError", { enumerable: true, get: function () { return errors_3.MarkdownParseError; } });

package/dist/cjs/modules/archive/compression/compress.browser.js

@@ -134,7 +134,7 @@ async function processWithStrategy(strategy, data, options) {
         });
     }
     // Fallback to pure JS implementation.
-    return strategy.jsFallback(data);
+    return strategy.jsFallback(data, options.level);
 }
 // =============================================================================
 // Public API
@@ -159,7 +159,7 @@ function compressSync(data, options = {}) {
     if (level === 0) {
         return data;
     }
-    return (0, deflate_fallback_1.deflateRawCompressed)(data);
+    return (0, deflate_fallback_1.deflateRawCompressed)(data, level);
 }
 /**
  * Decompress data using browser's native DecompressionStream or JS fallback
@@ -272,7 +272,7 @@ async function gzip(data, options = {}) {
  */
 function gzipSync(data, options = {}) {
     const level = options.level ?? defaults_1.DEFAULT_COMPRESS_LEVEL;
-    const deflated = level === 0 ? (0, deflate_fallback_1.deflateRawStore)(data) : (0, deflate_fallback_1.deflateRawCompressed)(data);
+    const deflated = level === 0 ? (0, deflate_fallback_1.deflateRawStore)(data) : (0, deflate_fallback_1.deflateRawCompressed)(data, level);
     return wrapGzip(deflated, data);
 }
 /**
@@ -348,7 +348,7 @@ async function zlib(data, options = {}) {
  */
 function zlibSync(data, options = {}) {
     const level = options.level ?? defaults_1.DEFAULT_COMPRESS_LEVEL;
-    const deflated = level === 0 ? (0, deflate_fallback_1.deflateRawStore)(data) : (0, deflate_fallback_1.deflateRawCompressed)(data);
+    const deflated = level === 0 ? (0, deflate_fallback_1.deflateRawStore)(data) : (0, deflate_fallback_1.deflateRawCompressed)(data, level);
     return wrapZlib(deflated, data, level);
 }
 /**