| 1 | from __future__ import annotations |
| 2 | |
| 3 | import csv |
| 4 | import io |
| 5 | import json |
| 6 | import re |
| 7 | import zipfile |
| 8 | from pathlib import Path |
| 9 | from typing import Any |
| 10 | from xml.sax.saxutils import escape |
| 11 | import xml.etree.ElementTree as ET |
| 12 | |
| 13 | from plugins._office.helpers import document_store, pptx_writer |
| 14 | |
| 15 | |
| 16 | W_NS = "http://schemas.openxmlformats.org/wordprocessingml/2006/main" |
| 17 | A_NS = "http://schemas.openxmlformats.org/drawingml/2006/main" |
| 18 | P_NS = "http://schemas.openxmlformats.org/presentationml/2006/main" |
| 19 | R_NS = "http://schemas.openxmlformats.org/officeDocument/2006/relationships" |
| 20 | CT_NS = "http://schemas.openxmlformats.org/package/2006/content-types" |
| 21 | REL_NS = "http://schemas.openxmlformats.org/package/2006/relationships" |
| 22 | XML_NS = "http://www.w3.org/XML/1998/namespace" |
| 23 | ODF_OFFICE_NS = "urn:oasis:names:tc:opendocument:xmlns:office:1.0" |
| 24 | ODF_TEXT_NS = "urn:oasis:names:tc:opendocument:xmlns:text:1.0" |
| 25 | ODF_TABLE_NS = "urn:oasis:names:tc:opendocument:xmlns:table:1.0" |
| 26 | ODF_DRAW_NS = "urn:oasis:names:tc:opendocument:xmlns:drawing:1.0" |
| 27 | ODF_PRESENTATION_NS = "urn:oasis:names:tc:opendocument:xmlns:presentation:1.0" |
| 28 | ODS_DIRECT_EDIT_ROW_LIMIT = 10000 |
| 29 | ODS_DIRECT_EDIT_COLUMN_LIMIT = 1024 |
| 30 | |
| 31 | for prefix, namespace in { |
| 32 | "w": W_NS, |
| 33 | "a": A_NS, |
| 34 | "p": P_NS, |
| 35 | "r": R_NS, |
| 36 | "office": ODF_OFFICE_NS, |
| 37 | "text": ODF_TEXT_NS, |
| 38 | "table": ODF_TABLE_NS, |
| 39 | "draw": ODF_DRAW_NS, |
| 40 | "presentation": ODF_PRESENTATION_NS, |
| 41 | }.items(): |
| 42 | ET.register_namespace(prefix, namespace) |
| 43 | |
| 44 | |
| 45 | def qn(namespace: str, tag: str) -> str: |
| 46 | return f"{{{namespace}}}{tag}" |
| 47 | |
| 48 | |
| 49 | def read_artifact(doc: dict[str, Any], max_chars: int = 12000) -> dict[str, Any]: |
| 50 | """Extract compact editable content from an Office artifact.""" |
| 51 | path = Path(doc["path"]) |
| 52 | ext = str(doc["extension"]).lower() |
| 53 | if ext == "md": |
| 54 | raise ValueError("Office artifact reads are not available for Markdown; use text_editor.") |
| 55 | if ext == "odt": |
| 56 | content = _read_odt(path) |
| 57 | elif ext == "ods": |
| 58 | content = _read_ods(path) |
| 59 | elif ext == "odp": |
| 60 | content = _read_odp(path) |
| 61 | elif ext == "docx": |
| 62 | content = _read_docx(path) |
| 63 | elif ext == "xlsx": |
| 64 | content = _read_xlsx(path) |
| 65 | elif ext == "pptx": |
| 66 | content = _read_pptx(path) |
| 67 | else: |
| 68 | raise ValueError(f"Unsupported document format: {ext}") |
| 69 | |
| 70 | return _trim_payload(content, max_chars=max_chars) |
| 71 | |
| 72 | |
| 73 | def edit_artifact( |
| 74 | doc: dict[str, Any], |
| 75 | operation: str = "", |
| 76 | content: str = "", |
| 77 | find: str = "", |
| 78 | replace: str = "", |
| 79 | sheet: str = "", |
| 80 | cells: Any = None, |
| 81 | rows: Any = None, |
| 82 | chart: Any = None, |
| 83 | slides: Any = None, |
| 84 | **kwargs: Any, |
| 85 | ) -> tuple[dict[str, Any], dict[str, Any]]: |
| 86 | """Apply a direct saved edit to an Office artifact and return updated metadata.""" |
| 87 | path = Path(doc["path"]) |
| 88 | ext = str(doc["extension"]).lower() |
| 89 | operation, content, find, replace, cells, rows, chart, slides, kwargs = _normalize_edit_inputs( |
| 90 | operation=operation, |
| 91 | content=content, |
| 92 | find=find, |
| 93 | replace=replace, |
| 94 | cells=cells, |
| 95 | rows=rows, |
| 96 | chart=chart, |
| 97 | slides=slides, |
| 98 | kwargs=kwargs, |
| 99 | ) |
| 100 | op = normalize_operation(operation, content=content, find=find, cells=cells, rows=rows, chart=chart, slides=slides) |
| 101 | if op in {"append_text", "prepend_text"} and content == "": |
| 102 | raise ValueError(f"content is required for {op}") |
| 103 | before = path.read_bytes() |
| 104 | |
| 105 | invalidate_sessions = bool(kwargs.pop("invalidate_sessions", False)) |
| 106 | if ext == "md": |
| 107 | raise ValueError("Office artifact edits are not available for Markdown; use text_editor.") |
| 108 | elif ext == "odt": |
| 109 | updated, details = _edit_odt(before, op, content=content, find=find, replace=replace, **kwargs) |
| 110 | elif ext == "ods": |
| 111 | updated, details = _edit_ods(before, op, content=content, find=find, replace=replace, sheet=sheet, cells=cells, rows=rows, **kwargs) |
| 112 | elif ext == "odp": |
| 113 | updated, details = _edit_odp(before, op, content=content, find=find, replace=replace, slides=slides, **kwargs) |
| 114 | elif ext == "docx": |
| 115 | updated, details = _edit_docx(before, op, content=content, find=find, replace=replace, **kwargs) |
| 116 | elif ext == "xlsx": |
| 117 | updated, details = _edit_xlsx(path, op, content=content, find=find, replace=replace, sheet=sheet, cells=cells, rows=rows, chart=chart, **kwargs) |
| 118 | elif ext == "pptx": |
| 119 | updated, details = _edit_pptx(before, op, content=content, find=find, replace=replace, slides=slides, **kwargs) |
| 120 | else: |
| 121 | raise ValueError(f"Direct edit is not available for .{ext}.") |
| 122 | |
| 123 | changed = updated != before |
| 124 | updated_doc = ( |
| 125 | document_store.replace_document_bytes( |
| 126 | doc["file_id"], |
| 127 | updated, |
| 128 | actor="office_artifact:edit", |
| 129 | invalidate_sessions=invalidate_sessions, |
| 130 | ) |
| 131 | if changed |
| 132 | else doc |
| 133 | ) |
| 134 | if changed: |
| 135 | _refresh_open_office_sessions(updated_doc["file_id"]) |
| 136 | preview = read_artifact(updated_doc, max_chars=int(kwargs.get("preview_chars") or 4000)) |
| 137 | payload = { |
| 138 | "ok": True, |
| 139 | "action": "edit", |
| 140 | "operation": op, |
| 141 | "changed": changed, |
| 142 | **details, |
| 143 | "preview": preview, |
| 144 | } |
| 145 | return updated_doc, payload |
| 146 | |
| 147 | |
| 148 | def _normalize_edit_inputs( |
| 149 | *, |
| 150 | operation: str = "", |
| 151 | content: str = "", |
| 152 | find: str = "", |
| 153 | replace: str = "", |
| 154 | cells: Any = None, |
| 155 | rows: Any = None, |
| 156 | chart: Any = None, |
| 157 | slides: Any = None, |
| 158 | kwargs: dict[str, Any] | None = None, |
| 159 | ) -> tuple[str, str, str, str, Any, Any, Any, Any, dict[str, Any]]: |
| 160 | kwargs = dict(kwargs or {}) |
| 161 | edit_spec = _edit_spec_from_kwargs(kwargs) |
| 162 | |
| 163 | operation = _first_text( |
| 164 | operation, |
| 165 | edit_spec.get("operation"), |
| 166 | edit_spec.get("op"), |
| 167 | edit_spec.get("edit"), |
| 168 | edit_spec.get("type"), |
| 169 | ) |
| 170 | |
| 171 | lines = _first_present( |
| 172 | edit_spec.get("add_lines"), |
| 173 | edit_spec.get("append_lines"), |
| 174 | edit_spec.get("lines"), |
| 175 | kwargs.get("add_lines"), |
| 176 | kwargs.get("append_lines"), |
| 177 | ) |
| 178 | if not operation and lines is not None: |
| 179 | operation = "append_text" |
| 180 | |
| 181 | if content == "": |
| 182 | content = _text_from_lines(lines) |
| 183 | if content == "": |
| 184 | content = _first_text( |
| 185 | edit_spec.get("content"), |
| 186 | edit_spec.get("text"), |
| 187 | edit_spec.get("value"), |
| 188 | edit_spec.get("body"), |
| 189 | kwargs.get("content"), |
| 190 | kwargs.get("text"), |
| 191 | kwargs.get("value"), |
| 192 | kwargs.get("body"), |
| 193 | ) |
| 194 | |
| 195 | if find == "": |
| 196 | find = _first_text( |
| 197 | edit_spec.get("find"), |
| 198 | edit_spec.get("old_text"), |
| 199 | edit_spec.get("old"), |
| 200 | kwargs.get("old_text"), |
| 201 | kwargs.get("old"), |
| 202 | ) |
| 203 | |
| 204 | if replace == "": |
| 205 | replace = _first_text( |
| 206 | edit_spec.get("replace"), |
| 207 | edit_spec.get("replacement"), |
| 208 | edit_spec.get("new_text"), |
| 209 | edit_spec.get("new"), |
| 210 | kwargs.get("replacement"), |
| 211 | kwargs.get("new_text"), |
| 212 | kwargs.get("new"), |
| 213 | ) |
| 214 | |
| 215 | if replace == "" and _looks_like_replace_operation(operation): |
| 216 | replace = _first_text(edit_spec.get("value"), kwargs.get("value")) |
| 217 | |
| 218 | cells = cells if cells is not None else _first_present(edit_spec.get("cells"), kwargs.get("cells")) |
| 219 | rows = rows if rows is not None else _first_present(edit_spec.get("rows"), kwargs.get("rows")) |
| 220 | chart = chart if chart is not None else _first_present(edit_spec.get("chart"), kwargs.get("chart")) |
| 221 | slides = slides if slides is not None else _first_present(edit_spec.get("slides"), kwargs.get("slides")) |
| 222 | |
| 223 | return operation, content, find, replace, cells, rows, chart, slides, kwargs |
| 224 | |
| 225 | |
| 226 | def _edit_spec_from_kwargs(kwargs: dict[str, Any]) -> dict[str, Any]: |
| 227 | for key in ("edit", "edits", "update", "patch"): |
| 228 | spec = _first_mapping(kwargs.get(key)) |
| 229 | if spec: |
| 230 | return spec |
| 231 | return {} |
| 232 | |
| 233 | |
| 234 | def _first_mapping(value: Any) -> dict[str, Any]: |
| 235 | if isinstance(value, dict): |
| 236 | return value |
| 237 | if isinstance(value, list): |
| 238 | for item in value: |
| 239 | if isinstance(item, dict): |
| 240 | return item |
| 241 | return {} |
| 242 | |
| 243 | |
| 244 | def _first_present(*values: Any) -> Any: |
| 245 | for value in values: |
| 246 | if value is not None: |
| 247 | return value |
| 248 | return None |
| 249 | |
| 250 | |
| 251 | def _first_text(*values: Any) -> str: |
| 252 | for value in values: |
| 253 | text = _text_from_lines(value) |
| 254 | if text != "": |
| 255 | return text |
| 256 | return "" |
| 257 | |
| 258 | |
| 259 | def _text_from_lines(value: Any) -> str: |
| 260 | if value is None: |
| 261 | return "" |
| 262 | if isinstance(value, list): |
| 263 | return "\n".join(str(item) for item in value) |
| 264 | return str(value) |
| 265 | |
| 266 | |
| 267 | def _looks_like_replace_operation(operation: str = "") -> bool: |
| 268 | op = str(operation or "").strip().lower().replace("-", "_") |
| 269 | return op in {"replace", "replace_text", "patch", "update"} |
| 270 | |
| 271 | |
| 272 | def _refresh_open_office_sessions(file_id: str) -> None: |
| 273 | try: |
| 274 | from plugins._desktop.helpers import desktop_session |
| 275 | |
| 276 | desktop_session.get_manager().refresh_document(file_id) |
| 277 | except Exception: |
| 278 | # Direct artifact edits should never fail just because no Office surface is open. |
| 279 | pass |
| 280 | |
| 281 | |
| 282 | def normalize_operation( |
| 283 | operation: str, |
| 284 | *, |
| 285 | content: str = "", |
| 286 | find: str = "", |
| 287 | cells: Any = None, |
| 288 | rows: Any = None, |
| 289 | chart: Any = None, |
| 290 | slides: Any = None, |
| 291 | ) -> str: |
| 292 | op = str(operation or "").strip().lower().replace("-", "_") |
| 293 | operation_map = { |
| 294 | "patch": "replace_text" if find else "set_text", |
| 295 | "update": "replace_text" if find else "set_text", |
| 296 | "replace": "replace_text", |
| 297 | "append": "append_text", |
| 298 | "append_line": "append_text", |
| 299 | "append_lines": "append_text", |
| 300 | "add_line": "append_text", |
| 301 | "add_lines": "append_text", |
| 302 | "prepend": "prepend_text", |
| 303 | "prepend_line": "prepend_text", |
| 304 | "prepend_lines": "prepend_text", |
| 305 | "write": "set_text", |
| 306 | "set": "set_text", |
| 307 | "set_content": "set_text", |
| 308 | "set_sheet": "set_rows", |
| 309 | "write_sheet": "set_rows", |
| 310 | "add_rows": "append_rows", |
| 311 | "add_chart": "create_chart", |
| 312 | "chart": "create_chart", |
| 313 | "insert_chart": "create_chart", |
| 314 | "set_chart": "create_chart", |
| 315 | "add_slide": "append_slide", |
| 316 | "set_deck": "set_slides", |
| 317 | } |
| 318 | op = operation_map.get(op, op) |
| 319 | if op: |
| 320 | return op |
| 321 | if cells: |
| 322 | return "set_cells" |
| 323 | if rows: |
| 324 | return "append_rows" |
| 325 | if chart: |
| 326 | return "create_chart" |
| 327 | if slides: |
| 328 | return "set_slides" |
| 329 | if find: |
| 330 | return "replace_text" |
| 331 | if content: |
| 332 | return "set_text" |
| 333 | raise ValueError("operation is required") |
| 334 | |
| 335 | |
| 336 | def _read_odt(path: Path) -> dict[str, Any]: |
| 337 | root = _odf_content_root(path) |
| 338 | paragraphs = _odf_text_lines(root) |
| 339 | headings = [ |
| 340 | "".join(node.itertext()).strip() |
| 341 | for node in root.iter(qn(ODF_TEXT_NS, "h")) |
| 342 | if "".join(node.itertext()).strip() |
| 343 | ] |
| 344 | return { |
| 345 | "kind": "document", |
| 346 | "format": "odt", |
| 347 | "paragraph_count": len(paragraphs), |
| 348 | "headings": headings[:40], |
| 349 | "text": "\n".join(paragraphs), |
| 350 | "paragraphs": paragraphs[:80], |
| 351 | } |
| 352 | |
| 353 | |
| 354 | def _read_ods(path: Path) -> dict[str, Any]: |
| 355 | sheets = _ods_sheets_from_bytes( |
| 356 | path.read_bytes(), |
| 357 | max_rows=ODS_DIRECT_EDIT_ROW_LIMIT, |
| 358 | max_cols=ODS_DIRECT_EDIT_COLUMN_LIMIT, |
| 359 | ) |
| 360 | return { |
| 361 | "kind": "spreadsheet", |
| 362 | "format": "ods", |
| 363 | "sheet_count": len(sheets), |
| 364 | "sheets": [ |
| 365 | { |
| 366 | "name": sheet["name"], |
| 367 | "max_row": len(sheet["rows"]), |
| 368 | "max_column": max((len(row) for row in sheet["rows"]), default=0), |
| 369 | "chart_count": 0, |
| 370 | "charts": [], |
| 371 | "preview_rows": sheet["rows"][:80], |
| 372 | } |
| 373 | for sheet in sheets[:8] |
| 374 | ], |
| 375 | } |
| 376 | |
| 377 | |
| 378 | def _read_odp(path: Path) -> dict[str, Any]: |
| 379 | slides = _odp_text_slides(path.read_bytes()) |
| 380 | return { |
| 381 | "kind": "presentation", |
| 382 | "format": "odp", |
| 383 | "slide_count": len(slides), |
| 384 | "slides": [ |
| 385 | { |
| 386 | "index": index + 1, |
| 387 | "title": slide.get("title", ""), |
| 388 | "lines": [slide.get("title", ""), *slide.get("bullets", [])], |
| 389 | } |
| 390 | for index, slide in enumerate(slides[:40]) |
| 391 | ], |
| 392 | } |
| 393 | |
| 394 | |
| 395 | def _read_docx(path: Path) -> dict[str, Any]: |
| 396 | with zipfile.ZipFile(path) as archive: |
| 397 | xml = archive.read("word/document.xml") |
| 398 | root = ET.fromstring(xml) |
| 399 | paragraphs = [] |
| 400 | for paragraph in root.iter(qn(W_NS, "p")): |
| 401 | text = "".join(node.text or "" for node in paragraph.iter(qn(W_NS, "t"))) |
| 402 | if text.strip(): |
| 403 | paragraphs.append(text) |
| 404 | return { |
| 405 | "kind": "document", |
| 406 | "paragraph_count": len(paragraphs), |
| 407 | "text": "\n".join(paragraphs), |
| 408 | "paragraphs": paragraphs[:80], |
| 409 | } |
| 410 | |
| 411 | |
| 412 | def _read_xlsx(path: Path) -> dict[str, Any]: |
| 413 | openpyxl = _require_openpyxl() |
| 414 | workbook = openpyxl.load_workbook(path, data_only=False) |
| 415 | sheets = [] |
| 416 | for worksheet in workbook.worksheets[:8]: |
| 417 | rows = [] |
| 418 | max_row = min(worksheet.max_row or 0, 80) |
| 419 | max_col = min(worksheet.max_column or 0, 30) |
| 420 | for row in worksheet.iter_rows(min_row=1, max_row=max_row, max_col=max_col, values_only=True): |
| 421 | values = ["" if value is None else value for value in row] |
| 422 | if any(str(value).strip() for value in values): |
| 423 | rows.append(values) |
| 424 | charts = [_chart_summary(chart) for chart in getattr(worksheet, "_charts", [])[:20]] |
| 425 | sheets.append({ |
| 426 | "name": worksheet.title, |
| 427 | "max_row": worksheet.max_row, |
| 428 | "max_column": worksheet.max_column, |
| 429 | "chart_count": len(getattr(worksheet, "_charts", [])), |
| 430 | "charts": charts, |
| 431 | "preview_rows": rows, |
| 432 | }) |
| 433 | return { |
| 434 | "kind": "spreadsheet", |
| 435 | "sheet_count": len(workbook.worksheets), |
| 436 | "sheets": sheets, |
| 437 | } |
| 438 | |
| 439 | |
| 440 | def _read_pptx(path: Path) -> dict[str, Any]: |
| 441 | slides = [] |
| 442 | with zipfile.ZipFile(path) as archive: |
| 443 | for name in _slide_names(archive): |
| 444 | root = ET.fromstring(archive.read(name)) |
| 445 | lines = [] |
| 446 | for paragraph in root.iter(qn(A_NS, "p")): |
| 447 | text = "".join(node.text or "" for node in paragraph.iter(qn(A_NS, "t"))).strip() |
| 448 | if text: |
| 449 | lines.append(text) |
| 450 | slides.append({ |
| 451 | "index": len(slides) + 1, |
| 452 | "title": lines[0] if lines else "", |
| 453 | "lines": lines, |
| 454 | }) |
| 455 | return { |
| 456 | "kind": "presentation", |
| 457 | "slide_count": len(slides), |
| 458 | "slides": slides[:40], |
| 459 | } |
| 460 | |
| 461 | |
| 462 | def _edit_odt(before: bytes, op: str, *, content: str = "", find: str = "", replace: str = "", **kwargs: Any) -> tuple[bytes, dict[str, Any]]: |
| 463 | if op not in {"set_text", "append_text", "prepend_text", "replace_text", "delete_text"}: |
| 464 | raise ValueError(f"Unsupported ODT operation: {op}") |
| 465 | |
| 466 | paragraphs = _odf_text_lines(ET.fromstring(_zip_member(before, "content.xml"))) |
| 467 | if op == "set_text": |
| 468 | lines = _text_lines(content) |
| 469 | return document_store.odt_bytes_from_paragraphs(lines), {"paragraphs_written": len(lines)} |
| 470 | if op == "append_text": |
| 471 | lines = [*paragraphs, *_text_lines(content)] |
| 472 | return document_store.odt_bytes_from_paragraphs(lines), {"paragraphs_written": len(lines)} |
| 473 | if op == "prepend_text": |
| 474 | lines = [*_text_lines(content), *paragraphs] |
| 475 | return document_store.odt_bytes_from_paragraphs(lines), {"paragraphs_written": len(lines)} |
| 476 | |
| 477 | if not find: |
| 478 | raise ValueError("find is required for replace_text") |
| 479 | replacement = "" if op == "delete_text" else replace |
| 480 | joined, count = _replace_limited("\n".join(paragraphs), find, replacement, _int_or_none(kwargs.get("count"))) |
| 481 | if count == 0: |
| 482 | return before, {"replacements": count} |
| 483 | return document_store.odt_bytes_from_paragraphs(joined.splitlines()), {"replacements": count} |
| 484 | |
| 485 | |
| 486 | def _edit_docx(before: bytes, op: str, *, content: str = "", find: str = "", replace: str = "", **kwargs: Any) -> tuple[bytes, dict[str, Any]]: |
| 487 | if op not in {"set_text", "append_text", "prepend_text", "replace_text", "delete_text"}: |
| 488 | raise ValueError(f"Unsupported DOCX operation: {op}") |
| 489 | |
| 490 | with zipfile.ZipFile(io.BytesIO(before)) as archive: |
| 491 | files = {info.filename: archive.read(info.filename) for info in archive.infolist()} |
| 492 | root = ET.fromstring(files["word/document.xml"]) |
| 493 | |
| 494 | if op == "replace_text" or op == "delete_text": |
| 495 | if not find: |
| 496 | raise ValueError("find is required for replace_text") |
| 497 | replacement = "" if op == "delete_text" else replace |
| 498 | count = _replace_text_in_paragraphs( |
| 499 | root, |
| 500 | paragraph_tag=qn(W_NS, "p"), |
| 501 | text_tag=qn(W_NS, "t"), |
| 502 | set_text=_set_word_paragraph_text, |
| 503 | find=find, |
| 504 | replacement=replacement, |
| 505 | limit=_int_or_none(kwargs.get("count")), |
| 506 | ) |
| 507 | details = {"replacements": count} |
| 508 | if count == 0: |
| 509 | return before, details |
| 510 | else: |
| 511 | lines = _text_lines(content) |
| 512 | body = root.find(f".//{qn(W_NS, 'body')}") |
| 513 | if body is None: |
| 514 | raise ValueError("DOCX document body not found") |
| 515 | paragraphs = [_word_paragraph(line) for line in lines] |
| 516 | if op == "set_text": |
| 517 | sect_pr = [child for child in list(body) if child.tag == qn(W_NS, "sectPr")] |
| 518 | for child in list(body): |
| 519 | body.remove(child) |
| 520 | for paragraph in paragraphs: |
| 521 | body.append(paragraph) |
| 522 | for child in sect_pr: |
| 523 | body.append(child) |
| 524 | elif op == "append_text": |
| 525 | insert_at = len(body) |
| 526 | for idx, child in enumerate(list(body)): |
| 527 | if child.tag == qn(W_NS, "sectPr"): |
| 528 | insert_at = idx |
| 529 | break |
| 530 | for paragraph in reversed(paragraphs): |
| 531 | body.insert(insert_at, paragraph) |
| 532 | elif op == "prepend_text": |
| 533 | for paragraph in reversed(paragraphs): |
| 534 | body.insert(0, paragraph) |
| 535 | details = {"paragraphs_written": len(paragraphs)} |
| 536 | |
| 537 | files["word/document.xml"] = _xml_bytes(root) |
| 538 | return _zip_from_existing(files), details |
| 539 | |
| 540 | |
| 541 | def _edit_ods( |
| 542 | before: bytes, |
| 543 | op: str, |
| 544 | *, |
| 545 | content: str = "", |
| 546 | find: str = "", |
| 547 | replace: str = "", |
| 548 | sheet: str = "", |
| 549 | cells: Any = None, |
| 550 | rows: Any = None, |
| 551 | **kwargs: Any, |
| 552 | ) -> tuple[bytes, dict[str, Any]]: |
| 553 | if op not in {"set_text", "set_rows", "append_text", "append_rows", "set_cells", "replace_text", "delete_text"}: |
| 554 | raise ValueError(f"Unsupported ODS operation: {op}") |
| 555 | |
| 556 | sheets = _ods_sheets_from_bytes( |
| 557 | before, |
| 558 | max_rows=ODS_DIRECT_EDIT_ROW_LIMIT, |
| 559 | max_cols=ODS_DIRECT_EDIT_COLUMN_LIMIT, |
| 560 | strict_limits=True, |
| 561 | ) |
| 562 | if not sheets: |
| 563 | sheets = [{"name": "Sheet1", "rows": []}] |
| 564 | worksheet = _ods_sheet(sheets, sheet) |
| 565 | details: dict[str, Any] = {"sheet": worksheet["name"]} |
| 566 | |
| 567 | if op in {"set_text", "set_rows"}: |
| 568 | parsed_rows = _normalize_rows(rows if rows is not None else content) |
| 569 | worksheet["rows"] = parsed_rows |
| 570 | details["rows_written"] = len(parsed_rows) |
| 571 | elif op in {"append_text", "append_rows"}: |
| 572 | parsed_rows = _normalize_rows(rows if rows is not None else content) |
| 573 | worksheet["rows"].extend(parsed_rows) |
| 574 | details["rows_appended"] = len(parsed_rows) |
| 575 | details["start_row"] = max(len(worksheet["rows"]) - len(parsed_rows) + 1, 1) |
| 576 | elif op == "set_cells": |
| 577 | assignments = _normalize_cells(cells, default_sheet=worksheet["name"]) |
| 578 | for sheet_name, cell, value in assignments: |
| 579 | target = _ods_sheet(sheets, sheet_name) |
| 580 | row_idx, col_idx = _cell_indices(cell) |
| 581 | _set_matrix_value(target["rows"], row_idx, col_idx, value) |
| 582 | details["cells_written"] = len(assignments) |
| 583 | else: |
| 584 | if not find: |
| 585 | raise ValueError("find is required for replace_text") |
| 586 | replacement = "" if op == "delete_text" else replace |
| 587 | count = 0 |
| 588 | limit = _int_or_none(kwargs.get("count")) |
| 589 | for item in sheets: |
| 590 | for row_idx, row in enumerate(item["rows"]): |
| 591 | for col_idx, value in enumerate(row): |
| 592 | if not isinstance(value, str) or find not in value: |
| 593 | continue |
| 594 | remaining = None if limit is None else max(limit - count, 0) |
| 595 | if remaining == 0: |
| 596 | break |
| 597 | row[col_idx], replaced = _replace_limited(value, find, replacement, remaining) |
| 598 | count += replaced |
| 599 | if limit is not None and count >= limit: |
| 600 | break |
| 601 | if limit is not None and count >= limit: |
| 602 | break |
| 603 | details["replacements"] = count |
| 604 | if count == 0: |
| 605 | return before, details |
| 606 | |
| 607 | return document_store.ods_bytes_from_sheets(sheets), details |
| 608 | |
| 609 | |
| 610 | def _edit_xlsx( |
| 611 | path: Path, |
| 612 | op: str, |
| 613 | *, |
| 614 | content: str = "", |
| 615 | find: str = "", |
| 616 | replace: str = "", |
| 617 | sheet: str = "", |
| 618 | cells: Any = None, |
| 619 | rows: Any = None, |
| 620 | chart: Any = None, |
| 621 | **kwargs: Any, |
| 622 | ) -> tuple[bytes, dict[str, Any]]: |
| 623 | if op not in {"set_text", "set_rows", "append_text", "append_rows", "set_cells", "replace_text", "delete_text", "create_chart"}: |
| 624 | raise ValueError(f"Unsupported XLSX operation: {op}") |
| 625 | openpyxl = _require_openpyxl() |
| 626 | workbook = openpyxl.load_workbook(path) |
| 627 | worksheet = _worksheet(workbook, sheet) |
| 628 | |
| 629 | details: dict[str, Any] = {"sheet": worksheet.title} |
| 630 | if op in {"set_text", "set_rows"}: |
| 631 | parsed_rows = _normalize_rows(rows if rows is not None else content) |
| 632 | _clear_worksheet(worksheet) |
| 633 | _write_rows(worksheet, parsed_rows, start_row=1) |
| 634 | details["rows_written"] = len(parsed_rows) |
| 635 | elif op in {"append_text", "append_rows"}: |
| 636 | parsed_rows = _normalize_rows(rows if rows is not None else content) |
| 637 | start_row = max((worksheet.max_row or 0) + 1, 1) |
| 638 | _write_rows(worksheet, parsed_rows, start_row=start_row) |
| 639 | details["rows_appended"] = len(parsed_rows) |
| 640 | details["start_row"] = start_row |
| 641 | elif op == "set_cells": |
| 642 | assignments = _normalize_cells(cells, default_sheet=worksheet.title) |
| 643 | for sheet_name, cell, value in assignments: |
| 644 | target = _worksheet(workbook, sheet_name) |
| 645 | target[cell] = value |
| 646 | details["cells_written"] = len(assignments) |
| 647 | elif op in {"replace_text", "delete_text"}: |
| 648 | if not find: |
| 649 | raise ValueError("find is required for replace_text") |
| 650 | replacement = "" if op == "delete_text" else replace |
| 651 | count = 0 |
| 652 | limit = _int_or_none(kwargs.get("count")) |
| 653 | for target in workbook.worksheets: |
| 654 | for row in target.iter_rows(): |
| 655 | for cell in row: |
| 656 | if not isinstance(cell.value, str) or find not in cell.value: |
| 657 | continue |
| 658 | remaining = None if limit is None else max(limit - count, 0) |
| 659 | if remaining == 0: |
| 660 | break |
| 661 | cell.value, replaced = _replace_limited(cell.value, find, replacement, remaining) |
| 662 | count += replaced |
| 663 | if limit is not None and count >= limit: |
| 664 | break |
| 665 | if limit is not None and count >= limit: |
| 666 | break |
| 667 | details["replacements"] = count |
| 668 | if count == 0: |
| 669 | return path.read_bytes(), details |
| 670 | elif op == "create_chart": |
| 671 | chart_details = [] |
| 672 | for chart_spec in _normalize_chart_specs(chart, kwargs): |
| 673 | chart_details.append(_create_xlsx_chart(workbook, worksheet, chart_spec)) |
| 674 | details["charts_created"] = len(chart_details) |
| 675 | details["charts"] = chart_details |
| 676 | |
| 677 | buffer = io.BytesIO() |
| 678 | workbook.save(buffer) |
| 679 | return buffer.getvalue(), details |
| 680 | |
| 681 | |
| 682 | _CHART_SPEC_KEYS = { |
| 683 | "anchor", |
| 684 | "categories", |
| 685 | "chart_type", |
| 686 | "close", |
| 687 | "data_range", |
| 688 | "fields", |
| 689 | "from_rows", |
| 690 | "height", |
| 691 | "high", |
| 692 | "include_headers", |
| 693 | "labels", |
| 694 | "legend", |
| 695 | "low", |
| 696 | "open", |
| 697 | "position", |
| 698 | "replace_existing", |
| 699 | "series", |
| 700 | "sheet", |
| 701 | "style", |
| 702 | "title", |
| 703 | "titles_from_data", |
| 704 | "type", |
| 705 | "values", |
| 706 | "width", |
| 707 | "x_axis_title", |
| 708 | "xvalues", |
| 709 | "y_axis_title", |
| 710 | "yvalues", |
| 711 | } |
| 712 | |
| 713 | _CHART_TYPE_NORMALIZATIONS = { |
| 714 | "area": "area", |
| 715 | "bar": "bar", |
| 716 | "candlestick": "stock", |
| 717 | "col": "column", |
| 718 | "column": "column", |
| 719 | "columns": "column", |
| 720 | "line": "line", |
| 721 | "ohlc": "stock", |
| 722 | "pie": "pie", |
| 723 | "scatter": "scatter", |
| 724 | "stock": "stock", |
| 725 | } |
| 726 | |
| 727 | |
| 728 | def _normalize_chart_specs(chart: Any, kwargs: dict[str, Any]) -> list[dict[str, Any]]: |
| 729 | parsed = _parse_chart_value(chart) |
| 730 | if isinstance(parsed, list): |
| 731 | if not parsed: |
| 732 | raise ValueError("chart list must include at least one chart spec") |
| 733 | return [_normalize_chart_spec(item, {}) for item in parsed] |
| 734 | if parsed is None: |
| 735 | parsed = {} |
| 736 | if not isinstance(parsed, dict): |
| 737 | raise ValueError("chart must be an object, JSON object, or list of chart objects") |
| 738 | return [_normalize_chart_spec(parsed, kwargs)] |
| 739 | |
| 740 | |
| 741 | def _parse_chart_value(value: Any) -> Any: |
| 742 | if value is None or value == "": |
| 743 | return None |
| 744 | if isinstance(value, str): |
| 745 | stripped = value.strip() |
| 746 | if not stripped: |
| 747 | return None |
| 748 | if stripped.startswith("{") or stripped.startswith("["): |
| 749 | return json.loads(stripped) |
| 750 | return {"type": stripped} |
| 751 | return value |
| 752 | |
| 753 | |
| 754 | def _normalize_chart_spec(value: Any, kwargs: dict[str, Any]) -> dict[str, Any]: |
| 755 | if isinstance(value, str): |
| 756 | value = _parse_chart_value(value) |
| 757 | if value is None: |
| 758 | value = {} |
| 759 | if not isinstance(value, dict): |
| 760 | raise ValueError("each chart spec must be an object") |
| 761 | |
| 762 | spec = dict(value) |
| 763 | explicit_include_headers = "include_headers" in spec or "titles_from_data" in spec |
| 764 | for key in _CHART_SPEC_KEYS: |
| 765 | if key in kwargs and kwargs[key] not in (None, ""): |
| 766 | spec[key] = kwargs[key] |
| 767 | if key in {"include_headers", "titles_from_data"}: |
| 768 | explicit_include_headers = True |
| 769 | |
| 770 | explicit_type = bool(spec.get("type") or spec.get("chart_type")) |
| 771 | chart_type = str(spec.get("type") or spec.get("chart_type") or "").strip().lower().replace("-", "_") |
| 772 | if chart_type: |
| 773 | chart_type = _CHART_TYPE_NORMALIZATIONS.get(chart_type, chart_type) |
| 774 | spec["type"] = chart_type |
| 775 | spec["_explicit_type"] = explicit_type |
| 776 | spec["position"] = str(spec.get("position") or spec.get("anchor") or "H2") |
| 777 | spec["include_headers"] = _bool_value(spec.get("include_headers", spec.get("titles_from_data")), default=True) |
| 778 | spec["_include_headers_explicit"] = explicit_include_headers |
| 779 | spec["from_rows"] = _bool_value(spec.get("from_rows"), default=False) |
| 780 | spec["replace_existing"] = _bool_value(spec.get("replace_existing"), default=False) |
| 781 | spec["width"] = _float_or_default(spec.get("width"), 18.0) |
| 782 | spec["height"] = _float_or_default(spec.get("height"), 10.0) |
| 783 | return spec |
| 784 | |
| 785 | |
| 786 | def _create_xlsx_chart(workbook: Any, default_worksheet: Any, spec: dict[str, Any]) -> dict[str, Any]: |
| 787 | openpyxl = _require_openpyxl() |
| 788 | worksheet = _worksheet(workbook, str(spec.get("sheet") or default_worksheet.title)) |
| 789 | chart_type = spec["type"] or _infer_default_chart_type(worksheet) |
| 790 | if chart_type not in _CHART_TYPE_NORMALIZATIONS.values(): |
| 791 | raise ValueError(f"Unsupported XLSX chart type: {chart_type}") |
| 792 | |
| 793 | if spec["replace_existing"]: |
| 794 | charts_removed = len(getattr(worksheet, "_charts", [])) |
| 795 | worksheet._charts = [] |
| 796 | else: |
| 797 | charts_removed = 0 |
| 798 | |
| 799 | if chart_type == "stock": |
| 800 | chart, data_range, categories = _stock_chart(openpyxl, workbook, worksheet, spec) |
| 801 | elif chart_type == "scatter": |
| 802 | chart, data_range, categories = _scatter_chart(openpyxl, workbook, worksheet, spec) |
| 803 | else: |
| 804 | chart, data_range, categories = _standard_chart(openpyxl, workbook, worksheet, spec, chart_type) |
| 805 | |
| 806 | _apply_chart_options(chart, spec) |
| 807 | worksheet.add_chart(chart, spec["position"]) |
| 808 | return { |
| 809 | "type": chart_type, |
| 810 | "title": str(spec.get("title") or ""), |
| 811 | "sheet": worksheet.title, |
| 812 | "position": spec["position"], |
| 813 | "data_range": data_range, |
| 814 | "categories": categories, |
| 815 | "series_count": len(getattr(chart, "series", [])), |
| 816 | "charts_removed": charts_removed, |
| 817 | } |
| 818 | |
| 819 | |
| 820 | def _standard_chart(openpyxl: Any, workbook: Any, worksheet: Any, spec: dict[str, Any], chart_type: str) -> tuple[Any, str, str]: |
| 821 | chart_classes = { |
| 822 | "area": openpyxl.chart.AreaChart, |
| 823 | "bar": openpyxl.chart.BarChart, |
| 824 | "column": openpyxl.chart.BarChart, |
| 825 | "line": openpyxl.chart.LineChart, |
| 826 | "pie": openpyxl.chart.PieChart, |
| 827 | } |
| 828 | chart = chart_classes[chart_type]() |
| 829 | if chart_type == "bar": |
| 830 | chart.type = "bar" |
| 831 | elif chart_type == "column": |
| 832 | chart.type = "col" |
| 833 | |
| 834 | include_headers = bool(spec["include_headers"]) |
| 835 | categories = str(spec.get("categories") or spec.get("labels") or "") |
| 836 | if spec.get("series"): |
| 837 | data_range = _add_explicit_series(openpyxl, chart, workbook, worksheet, spec, validate_numeric=True) |
| 838 | else: |
| 839 | range_value = spec.get("values") or spec.get("data_range") or _default_data_range(worksheet, chart_type, include_headers) |
| 840 | include_headers = _include_headers_for_range(spec, range_value) |
| 841 | data_ref, data_sheet, bounds, data_range = _reference_from_range(openpyxl, workbook, worksheet, range_value) |
| 842 | _validate_numeric_series(data_sheet, bounds, include_headers=include_headers, label=data_range) |
| 843 | chart.add_data(data_ref, titles_from_data=include_headers, from_rows=bool(spec["from_rows"])) |
| 844 | |
| 845 | if not categories: |
| 846 | categories = _default_category_range(worksheet, include_headers=include_headers) |
| 847 | if categories: |
| 848 | categories_ref, _, _, categories = _reference_from_range(openpyxl, workbook, worksheet, categories) |
| 849 | chart.set_categories(categories_ref) |
| 850 | return chart, data_range, categories |
| 851 | |
| 852 | |
| 853 | def _stock_chart(openpyxl: Any, workbook: Any, worksheet: Any, spec: dict[str, Any]) -> tuple[Any, str, str]: |
| 854 | chart = openpyxl.chart.StockChart() |
| 855 | field_ranges = _stock_field_ranges(spec) |
| 856 | |
| 857 | if field_ranges: |
| 858 | data_labels = [] |
| 859 | for label in ("open", "high", "low", "close"): |
| 860 | include_headers = _include_headers_for_range(spec, field_ranges[label]) |
| 861 | series_ref, data_sheet, bounds, label_range = _reference_from_range(openpyxl, workbook, worksheet, field_ranges[label]) |
| 862 | _validate_numeric_series(data_sheet, bounds, include_headers=include_headers, label=label) |
| 863 | chart.series.append(openpyxl.chart.Series(series_ref, title_from_data=include_headers)) |
| 864 | data_labels.append(label_range) |
| 865 | data_range = ", ".join(data_labels) |
| 866 | elif spec.get("series"): |
| 867 | include_headers = bool(spec["include_headers"]) |
| 868 | data_range = _add_explicit_series(openpyxl, chart, workbook, worksheet, spec, expected_count=4, validate_numeric=True) |
| 869 | else: |
| 870 | include_headers = bool(spec["include_headers"]) |
| 871 | range_value = spec.get("data_range") or _default_data_range(worksheet, "stock", include_headers) |
| 872 | include_headers = _include_headers_for_range(spec, range_value) |
| 873 | _, data_sheet, bounds, range_label = _reference_from_range(openpyxl, workbook, worksheet, range_value) |
| 874 | min_col, min_row, max_col, max_row = bounds |
| 875 | columns = list(range(min_col, max_col + 1)) |
| 876 | if len(columns) > 4 and _looks_like_category_header(data_sheet.cell(row=min_row, column=min_col).value): |
| 877 | columns = columns[1:5] |
| 878 | else: |
| 879 | columns = columns[:4] |
| 880 | if len(columns) != 4: |
| 881 | raise ValueError("stock charts require exactly four Open, High, Low, Close data series") |
| 882 | _validate_stock_headers(data_sheet, columns, min_row, include_headers=include_headers) |
| 883 | for column in columns: |
| 884 | _validate_numeric_series(data_sheet, (column, min_row, column, max_row), include_headers=include_headers, label=data_sheet.cell(row=min_row, column=column).value or _column_letter(column)) |
| 885 | series_ref = openpyxl.chart.Reference(data_sheet, min_col=column, min_row=min_row, max_col=column, max_row=max_row) |
| 886 | chart.series.append(openpyxl.chart.Series(series_ref, title_from_data=include_headers)) |
| 887 | data_range = range_label |
| 888 | |
| 889 | categories = str(spec.get("categories") or spec.get("labels") or _default_category_range(worksheet, include_headers=bool(spec["include_headers"]))) |
| 890 | if categories: |
| 891 | categories_ref, _, _, categories = _reference_from_range(openpyxl, workbook, worksheet, categories) |
| 892 | chart.set_categories(categories_ref) |
| 893 | chart.hiLowLines = openpyxl.chart.axis.ChartLines() |
| 894 | chart.upDownBars = openpyxl.chart.updown_bars.UpDownBars() |
| 895 | return chart, data_range, categories |
| 896 | |
| 897 | |
| 898 | def _scatter_chart(openpyxl: Any, workbook: Any, worksheet: Any, spec: dict[str, Any]) -> tuple[Any, str, str]: |
| 899 | chart = openpyxl.chart.ScatterChart() |
| 900 | include_headers = bool(spec["include_headers"]) |
| 901 | categories = str(spec.get("xvalues") or spec.get("categories") or _default_category_range(worksheet, include_headers=include_headers)) |
| 902 | x_ref, x_sheet, x_bounds, categories = _reference_from_range(openpyxl, workbook, worksheet, categories) |
| 903 | if include_headers and x_bounds[1] == 1 and x_bounds[3] > 1: |
| 904 | x_ref = openpyxl.chart.Reference(x_sheet, min_col=x_bounds[0], min_row=2, max_col=x_bounds[2], max_row=x_bounds[3]) |
| 905 | |
| 906 | data_ranges = [] |
| 907 | series_items = _series_items(spec) |
| 908 | if series_items: |
| 909 | for item in series_items: |
| 910 | values_ref, title, data_range = _series_values_reference(openpyxl, workbook, worksheet, item, include_headers=include_headers, validate_numeric=True) |
| 911 | xvalues = item.get("xvalues") or item.get("x") or item.get("categories") |
| 912 | if xvalues: |
| 913 | item_x_ref, item_x_sheet, item_x_bounds, _ = _reference_from_range(openpyxl, workbook, worksheet, xvalues) |
| 914 | if include_headers and item_x_bounds[1] == 1 and item_x_bounds[3] > 1: |
| 915 | item_x_ref = openpyxl.chart.Reference(item_x_sheet, min_col=item_x_bounds[0], min_row=2, max_col=item_x_bounds[2], max_row=item_x_bounds[3]) |
| 916 | else: |
| 917 | item_x_ref = x_ref |
| 918 | chart.series.append(openpyxl.chart.Series(values_ref, xvalues=item_x_ref, title=title)) |
| 919 | data_ranges.append(data_range) |
| 920 | else: |
| 921 | range_value = spec.get("yvalues") or spec.get("values") or spec.get("data_range") or _default_data_range(worksheet, "scatter", include_headers) |
| 922 | _, data_sheet, bounds, data_range = _reference_from_range(openpyxl, workbook, worksheet, range_value) |
| 923 | min_col, min_row, max_col, max_row = bounds |
| 924 | first_row = min_row + 1 if include_headers and min_row == 1 and max_row > 1 else min_row |
| 925 | for column in range(min_col, max_col + 1): |
| 926 | title = data_sheet.cell(row=min_row, column=column).value if first_row > min_row else None |
| 927 | _validate_numeric_series(data_sheet, (column, min_row, column, max_row), include_headers=include_headers, label=title or _column_letter(column)) |
| 928 | y_ref = openpyxl.chart.Reference(data_sheet, min_col=column, min_row=first_row, max_col=column, max_row=max_row) |
| 929 | chart.series.append(openpyxl.chart.Series(y_ref, xvalues=x_ref, title=str(title) if title is not None else None)) |
| 930 | return chart, ", ".join(data_ranges) if data_ranges else data_range, categories |
| 931 | |
| 932 | |
| 933 | def _add_explicit_series( |
| 934 | openpyxl: Any, |
| 935 | chart: Any, |
| 936 | workbook: Any, |
| 937 | worksheet: Any, |
| 938 | spec: dict[str, Any], |
| 939 | expected_count: int | None = None, |
| 940 | validate_numeric: bool = False, |
| 941 | ) -> str: |
| 942 | include_headers = bool(spec["include_headers"]) |
| 943 | ranges = [] |
| 944 | for item in _series_items(spec): |
| 945 | values_ref, title, label = _series_values_reference(openpyxl, workbook, worksheet, item, include_headers=include_headers, validate_numeric=validate_numeric) |
| 946 | if title: |
| 947 | chart.series.append(openpyxl.chart.Series(values_ref, title=title)) |
| 948 | else: |
| 949 | chart.series.append(openpyxl.chart.Series(values_ref, title_from_data=include_headers)) |
| 950 | ranges.append(label) |
| 951 | if expected_count is not None and len(ranges) != expected_count: |
| 952 | raise ValueError(f"chart requires exactly {expected_count} series") |
| 953 | return ", ".join(ranges) |
| 954 | |
| 955 | |
| 956 | def _series_items(spec: dict[str, Any]) -> list[dict[str, Any]]: |
| 957 | raw = spec.get("series") or [] |
| 958 | if isinstance(raw, str): |
| 959 | raw = json.loads(raw) |
| 960 | if not isinstance(raw, list): |
| 961 | raise ValueError("chart series must be a list") |
| 962 | items = [] |
| 963 | for item in raw: |
| 964 | if isinstance(item, str): |
| 965 | items.append({"values": item}) |
| 966 | elif isinstance(item, dict): |
| 967 | items.append(item) |
| 968 | else: |
| 969 | raise ValueError("chart series entries must be objects or range strings") |
| 970 | return items |
| 971 | |
| 972 | |
| 973 | def _series_values_reference( |
| 974 | openpyxl: Any, |
| 975 | workbook: Any, |
| 976 | worksheet: Any, |
| 977 | item: dict[str, Any], |
| 978 | *, |
| 979 | include_headers: bool, |
| 980 | validate_numeric: bool = False, |
| 981 | ) -> tuple[Any, str | None, str]: |
| 982 | values = item.get("values") or item.get("range") or item.get("yvalues") or item.get("y") |
| 983 | if not values: |
| 984 | raise ValueError("chart series entries require values or range") |
| 985 | ref, data_sheet, bounds, label = _reference_from_range(openpyxl, workbook, worksheet, values) |
| 986 | title = item.get("title") or item.get("name") |
| 987 | min_col, min_row, max_col, max_row = bounds |
| 988 | if include_headers and min_row == 1 and max_col == min_col and max_row > min_row: |
| 989 | title = title if title is not None else data_sheet.cell(row=min_row, column=min_col).value |
| 990 | ref = openpyxl.chart.Reference(data_sheet, min_col=min_col, min_row=min_row + 1, max_col=max_col, max_row=max_row) |
| 991 | if validate_numeric: |
| 992 | _validate_numeric_series(data_sheet, bounds, include_headers=include_headers, label=title or label) |
| 993 | return ref, str(title) if title is not None else None, label |
| 994 | |
| 995 | |
| 996 | def _stock_field_ranges(spec: dict[str, Any]) -> dict[str, Any]: |
| 997 | fields = spec.get("fields") or {} |
| 998 | if isinstance(fields, str): |
| 999 | fields = json.loads(fields) |
| 1000 | if not isinstance(fields, dict): |
| 1001 | raise ValueError("stock chart fields must be an object") |
| 1002 | result = {} |
| 1003 | for label in ("open", "high", "low", "close"): |
| 1004 | value = spec.get(label) or fields.get(label) |
| 1005 | if value: |
| 1006 | result[label] = value |
| 1007 | if result and set(result) != {"open", "high", "low", "close"}: |
| 1008 | raise ValueError("stock chart fields must include open, high, low, and close") |
| 1009 | return result |
| 1010 | |
| 1011 | |
| 1012 | def _reference_from_range(openpyxl: Any, workbook: Any, default_worksheet: Any, value: Any) -> tuple[Any, Any, tuple[int, int, int, int], str]: |
| 1013 | sheet_name, cell_range = _split_range_ref(value, default_worksheet.title) |
| 1014 | worksheet = _worksheet(workbook, sheet_name) |
| 1015 | min_col, min_row, max_col, max_row = openpyxl.utils.cell.range_boundaries(cell_range) |
| 1016 | reference = openpyxl.chart.Reference(worksheet, min_col=min_col, min_row=min_row, max_col=max_col, max_row=max_row) |
| 1017 | return reference, worksheet, (min_col, min_row, max_col, max_row), _range_label(openpyxl, worksheet.title, min_col, min_row, max_col, max_row) |
| 1018 | |
| 1019 | |
| 1020 | def _split_range_ref(value: Any, default_sheet: str) -> tuple[str, str]: |
| 1021 | if isinstance(value, dict): |
| 1022 | sheet = str(value.get("sheet") or default_sheet) |
| 1023 | min_cell = value.get("range") or value.get("ref") |
| 1024 | if min_cell: |
| 1025 | return _split_range_ref(str(min_cell), sheet) |
| 1026 | min_col = value.get("min_col") |
| 1027 | min_row = value.get("min_row") |
| 1028 | max_col = value.get("max_col", min_col) |
| 1029 | max_row = value.get("max_row", min_row) |
| 1030 | if min_col is None or min_row is None: |
| 1031 | raise ValueError("range objects require range/ref or min_col and min_row") |
| 1032 | return sheet, f"{_cell_ref(min_col, min_row)}:{_cell_ref(max_col, max_row)}" |
| 1033 | ref = str(value or "").strip() |
| 1034 | if not ref: |
| 1035 | raise ValueError("chart range is required") |
| 1036 | if "!" not in ref: |
| 1037 | return default_sheet, ref |
| 1038 | sheet, cell_range = ref.rsplit("!", 1) |
| 1039 | return sheet.strip().strip("'") or default_sheet, cell_range |
| 1040 | |
| 1041 | |
| 1042 | def _range_label(openpyxl: Any, sheet_title: str, min_col: int, min_row: int, max_col: int, max_row: int) -> str: |
| 1043 | start = f"{openpyxl.utils.cell.get_column_letter(min_col)}{min_row}" |
| 1044 | end = f"{openpyxl.utils.cell.get_column_letter(max_col)}{max_row}" |
| 1045 | sheet = sheet_title if re.fullmatch(r"[A-Za-z_][A-Za-z0-9_]*", sheet_title) else f"'{sheet_title}'" |
| 1046 | return f"{sheet}!{start}:{end}" |
| 1047 | |
| 1048 | |
| 1049 | def _include_headers_for_range(spec: dict[str, Any], range_value: Any) -> bool: |
| 1050 | include_headers = bool(spec["include_headers"]) |
| 1051 | if spec.get("_include_headers_explicit"): |
| 1052 | return include_headers |
| 1053 | try: |
| 1054 | _, cell_range = _split_range_ref(range_value, "") |
| 1055 | _, min_row, _, _ = __import__("openpyxl").utils.cell.range_boundaries(cell_range) |
| 1056 | except Exception: |
| 1057 | return include_headers |
| 1058 | return include_headers and min_row == 1 |
| 1059 | |
| 1060 | |
| 1061 | def _validate_stock_headers(data_sheet: Any, columns: list[int], min_row: int, *, include_headers: bool) -> None: |
| 1062 | if not include_headers or min_row != 1: |
| 1063 | return |
| 1064 | expected = ["open", "high", "low", "close"] |
| 1065 | found = [str(data_sheet.cell(row=min_row, column=column).value or "").strip().lower() for column in columns] |
| 1066 | if found != expected: |
| 1067 | raise ValueError(f"stock charts require Open, High, Low, Close columns in order; found {found}") |
| 1068 | |
| 1069 | |
| 1070 | def _validate_numeric_series(data_sheet: Any, bounds: tuple[int, int, int, int], *, include_headers: bool, label: Any) -> None: |
| 1071 | min_col, min_row, max_col, max_row = bounds |
| 1072 | start_row = min_row + 1 if include_headers and min_row == 1 and max_row > min_row else min_row |
| 1073 | values = [ |
| 1074 | data_sheet.cell(row=row, column=column).value |
| 1075 | for column in range(min_col, max_col + 1) |
| 1076 | for row in range(start_row, max_row + 1) |
| 1077 | ] |
| 1078 | numeric_count = sum(1 for value in values if isinstance(value, (int, float)) and not isinstance(value, bool)) |
| 1079 | if numeric_count == 0: |
| 1080 | name = str(label or _range_label(__import__("openpyxl"), data_sheet.title, min_col, min_row, max_col, max_row)) |
| 1081 | raise ValueError(f"chart series '{name}' has no numeric data") |
| 1082 | |
| 1083 | |
| 1084 | def _default_category_range(worksheet: Any, *, include_headers: bool) -> str: |
| 1085 | if (worksheet.max_column or 0) < 2 or (worksheet.max_row or 0) < 2: |
| 1086 | return "" |
| 1087 | first_row_is_header = _looks_like_category_header(worksheet.cell(row=1, column=1).value) |
| 1088 | return f"A{2 if include_headers or first_row_is_header else 1}:A{worksheet.max_row}" |
| 1089 | |
| 1090 | |
| 1091 | def _default_data_range(worksheet: Any, chart_type: str, include_headers: bool) -> str: |
| 1092 | max_row = worksheet.max_row or 1 |
| 1093 | max_col = worksheet.max_column or 1 |
| 1094 | if chart_type == "stock": |
| 1095 | if max_col < 5 or max_row < 2: |
| 1096 | raise ValueError("stock charts need Date, Open, High, Low, Close columns or explicit ranges") |
| 1097 | return f"B{1 if include_headers else 2}:E{max_row}" |
| 1098 | if chart_type == "pie" and max_col >= 2: |
| 1099 | return f"B{1 if include_headers else 2}:B{max_row}" |
| 1100 | start_col = 2 if max_col >= 2 else 1 |
| 1101 | return f"{_column_letter(start_col)}{1 if include_headers else 2}:{_column_letter(max_col)}{max_row}" |
| 1102 | |
| 1103 | |
| 1104 | def _cell_ref(column: Any, row: Any) -> str: |
| 1105 | return f"{_column_letter(column)}{int(row)}" |
| 1106 | |
| 1107 | |
| 1108 | def _column_letter(column: Any) -> str: |
| 1109 | if isinstance(column, str) and column.isalpha(): |
| 1110 | return column.upper() |
| 1111 | return __import__("openpyxl").utils.cell.get_column_letter(int(column)) |
| 1112 | |
| 1113 | |
| 1114 | def _infer_default_chart_type(worksheet: Any) -> str: |
| 1115 | headers = [str(worksheet.cell(row=1, column=column).value or "").strip().lower() for column in range(1, (worksheet.max_column or 0) + 1)] |
| 1116 | if {"open", "high", "low", "close"}.issubset(set(headers)): |
| 1117 | return "stock" |
| 1118 | return "line" |
| 1119 | |
| 1120 | |
| 1121 | def _looks_like_category_header(value: Any) -> bool: |
| 1122 | return str(value or "").strip().lower() in {"date", "time", "category", "label", "month", "year"} |
| 1123 | |
| 1124 | |
| 1125 | def _apply_chart_options(chart: Any, spec: dict[str, Any]) -> None: |
| 1126 | if spec.get("title"): |
| 1127 | chart.title = str(spec["title"]) |
| 1128 | if spec.get("style") not in (None, ""): |
| 1129 | chart.style = int(spec["style"]) |
| 1130 | chart.width = spec["width"] |
| 1131 | chart.height = spec["height"] |
| 1132 | if _bool_value(spec.get("legend"), default=True) is False: |
| 1133 | chart.legend = None |
| 1134 | if spec.get("x_axis_title") and hasattr(chart, "x_axis"): |
| 1135 | chart.x_axis.title = str(spec["x_axis_title"]) |
| 1136 | if spec.get("y_axis_title") and hasattr(chart, "y_axis"): |
| 1137 | chart.y_axis.title = str(spec["y_axis_title"]) |
| 1138 | |
| 1139 | |
| 1140 | def _chart_summary(chart: Any) -> dict[str, Any]: |
| 1141 | return { |
| 1142 | "type": _chart_kind(chart), |
| 1143 | "title": _chart_title(chart), |
| 1144 | "anchor": _chart_anchor(chart), |
| 1145 | "series_count": len(getattr(chart, "series", [])), |
| 1146 | } |
| 1147 | |
| 1148 | |
| 1149 | def _chart_kind(chart: Any) -> str: |
| 1150 | name = chart.__class__.__name__.replace("Chart", "").lower() |
| 1151 | return {"bar": "bar_or_column", "stock": "stock"}.get(name, name) |
| 1152 | |
| 1153 | |
| 1154 | def _chart_title(chart: Any) -> str: |
| 1155 | title = getattr(chart, "title", None) |
| 1156 | if title is None or isinstance(title, str): |
| 1157 | return title or "" |
| 1158 | try: |
| 1159 | paragraphs = title.tx.rich.p |
| 1160 | parts = [] |
| 1161 | for paragraph in paragraphs: |
| 1162 | for run in paragraph.r: |
| 1163 | if run.t: |
| 1164 | parts.append(run.t) |
| 1165 | return "".join(parts) |
| 1166 | except Exception: |
| 1167 | return "" |
| 1168 | |
| 1169 | |
| 1170 | def _chart_anchor(chart: Any) -> str: |
| 1171 | openpyxl = _require_openpyxl() |
| 1172 | anchor = getattr(chart, "anchor", "") |
| 1173 | if isinstance(anchor, str): |
| 1174 | return anchor |
| 1175 | marker = getattr(anchor, "_from", None) |
| 1176 | if marker is None: |
| 1177 | return "" |
| 1178 | return f"{openpyxl.utils.cell.get_column_letter(marker.col + 1)}{marker.row + 1}" |
| 1179 | |
| 1180 | |
| 1181 | def _bool_value(value: Any, default: bool = False) -> bool: |
| 1182 | if value in (None, ""): |
| 1183 | return default |
| 1184 | if isinstance(value, bool): |
| 1185 | return value |
| 1186 | if isinstance(value, (int, float)): |
| 1187 | return bool(value) |
| 1188 | return str(value).strip().lower() not in {"0", "false", "no", "off", "none"} |
| 1189 | |
| 1190 | |
| 1191 | def _float_or_default(value: Any, default: float) -> float: |
| 1192 | if value in (None, ""): |
| 1193 | return default |
| 1194 | try: |
| 1195 | return float(value) |
| 1196 | except (TypeError, ValueError): |
| 1197 | return default |
| 1198 | |
| 1199 | |
| 1200 | def _edit_pptx(before: bytes, op: str, *, content: str = "", find: str = "", replace: str = "", slides: Any = None, **kwargs: Any) -> tuple[bytes, dict[str, Any]]: |
| 1201 | if op not in {"set_text", "set_slides", "append_text", "append_slide", "replace_text", "delete_text"}: |
| 1202 | raise ValueError(f"Unsupported PPTX operation: {op}") |
| 1203 | |
| 1204 | if op in {"set_text", "set_slides"}: |
| 1205 | parsed_slides = _normalize_slides(slides if slides is not None else content) |
| 1206 | return _pptx_from_slides(parsed_slides), {"slides_written": len(parsed_slides)} |
| 1207 | |
| 1208 | if op in {"append_text", "append_slide"}: |
| 1209 | existing = _pptx_text_slides(before) |
| 1210 | existing.extend(_normalize_slides(slides if slides is not None else content)) |
| 1211 | return _pptx_from_slides(existing), {"slides_written": len(existing)} |
| 1212 | |
| 1213 | with zipfile.ZipFile(io.BytesIO(before)) as archive: |
| 1214 | files = {info.filename: archive.read(info.filename) for info in archive.infolist()} |
| 1215 | if not find: |
| 1216 | raise ValueError("find is required for replace_text") |
| 1217 | replacement = "" if op == "delete_text" else replace |
| 1218 | count = 0 |
| 1219 | limit = _int_or_none(kwargs.get("count")) |
| 1220 | for name in sorted([name for name in files if name.startswith("ppt/slides/slide") and name.endswith(".xml")], key=_natural_key): |
| 1221 | root = ET.fromstring(files[name]) |
| 1222 | count += _replace_text_in_paragraphs( |
| 1223 | root, |
| 1224 | paragraph_tag=qn(A_NS, "p"), |
| 1225 | text_tag=qn(A_NS, "t"), |
| 1226 | set_text=_set_drawing_paragraph_text, |
| 1227 | find=find, |
| 1228 | replacement=replacement, |
| 1229 | limit=None if limit is None else max(limit - count, 0), |
| 1230 | ) |
| 1231 | files[name] = _xml_bytes(root) |
| 1232 | if limit is not None and count >= limit: |
| 1233 | break |
| 1234 | if count == 0: |
| 1235 | return before, {"replacements": count} |
| 1236 | return _zip_from_existing(files), {"replacements": count} |
| 1237 | |
| 1238 | |
| 1239 | def _edit_odp(before: bytes, op: str, *, content: str = "", find: str = "", replace: str = "", slides: Any = None, **kwargs: Any) -> tuple[bytes, dict[str, Any]]: |
| 1240 | if op not in {"set_text", "set_slides", "append_text", "append_slide", "replace_text", "delete_text"}: |
| 1241 | raise ValueError(f"Unsupported ODP operation: {op}") |
| 1242 | |
| 1243 | if op in {"set_text", "set_slides"}: |
| 1244 | parsed_slides = _normalize_slides(slides if slides is not None else content) |
| 1245 | return document_store.odp_bytes_from_slides(parsed_slides), {"slides_written": len(parsed_slides)} |
| 1246 | |
| 1247 | existing = _odp_text_slides(before) |
| 1248 | if op in {"append_text", "append_slide"}: |
| 1249 | existing.extend(_normalize_slides(slides if slides is not None else content)) |
| 1250 | return document_store.odp_bytes_from_slides(existing), {"slides_written": len(existing)} |
| 1251 | |
| 1252 | if not find: |
| 1253 | raise ValueError("find is required for replace_text") |
| 1254 | replacement = "" if op == "delete_text" else replace |
| 1255 | count = 0 |
| 1256 | limit = _int_or_none(kwargs.get("count")) |
| 1257 | for slide in existing: |
| 1258 | title, title_count = _replace_limited( |
| 1259 | str(slide.get("title") or ""), |
| 1260 | find, |
| 1261 | replacement, |
| 1262 | None if limit is None else max(limit - count, 0), |
| 1263 | ) |
| 1264 | if title_count: |
| 1265 | slide["title"] = title |
| 1266 | count += title_count |
| 1267 | if limit is not None and count >= limit: |
| 1268 | break |
| 1269 | bullets = [] |
| 1270 | for bullet in slide.get("bullets") or []: |
| 1271 | updated, replaced = _replace_limited( |
| 1272 | str(bullet), |
| 1273 | find, |
| 1274 | replacement, |
| 1275 | None if limit is None else max(limit - count, 0), |
| 1276 | ) |
| 1277 | bullets.append(updated) |
| 1278 | count += replaced |
| 1279 | if limit is not None and count >= limit: |
| 1280 | bullets.extend(slide.get("bullets", [])[len(bullets):]) |
| 1281 | break |
| 1282 | slide["bullets"] = bullets |
| 1283 | if limit is not None and count >= limit: |
| 1284 | break |
| 1285 | if count == 0: |
| 1286 | return before, {"replacements": count} |
| 1287 | return document_store.odp_bytes_from_slides(existing), {"replacements": count} |
| 1288 | |
| 1289 | |
| 1290 | def _replace_text_in_paragraphs( |
| 1291 | root: ET.Element, |
| 1292 | *, |
| 1293 | paragraph_tag: str, |
| 1294 | text_tag: str, |
| 1295 | set_text: Any, |
| 1296 | find: str, |
| 1297 | replacement: str, |
| 1298 | limit: int | None, |
| 1299 | ) -> int: |
| 1300 | count = 0 |
| 1301 | for paragraph in root.iter(paragraph_tag): |
| 1302 | texts = list(paragraph.iter(text_tag)) |
| 1303 | if not texts: |
| 1304 | continue |
| 1305 | current = "".join(node.text or "" for node in texts) |
| 1306 | if find not in current: |
| 1307 | continue |
| 1308 | remaining = None if limit is None else max(limit - count, 0) |
| 1309 | if remaining == 0: |
| 1310 | break |
| 1311 | updated, replaced = _replace_limited(current, find, replacement, remaining) |
| 1312 | if replaced: |
| 1313 | set_text(paragraph, updated) |
| 1314 | count += replaced |
| 1315 | return count |
| 1316 | |
| 1317 | |
| 1318 | def _replace_limited(value: str, find: str, replacement: str, limit: int | None) -> tuple[str, int]: |
| 1319 | if limit is None: |
| 1320 | return value.replace(find, replacement), value.count(find) |
| 1321 | return value.replace(find, replacement, limit), min(value.count(find), limit) |
| 1322 | |
| 1323 | |
| 1324 | def _set_word_paragraph_text(paragraph: ET.Element, text: str) -> None: |
| 1325 | keep = [child for child in list(paragraph) if child.tag == qn(W_NS, "pPr")] |
| 1326 | for child in list(paragraph): |
| 1327 | paragraph.remove(child) |
| 1328 | for child in keep: |
| 1329 | paragraph.append(child) |
| 1330 | paragraph.append(_word_run(text)) |
| 1331 | |
| 1332 | |
| 1333 | def _word_paragraph(text: str) -> ET.Element: |
| 1334 | paragraph = ET.Element(qn(W_NS, "p")) |
| 1335 | paragraph.append(_word_run(text)) |
| 1336 | return paragraph |
| 1337 | |
| 1338 | |
| 1339 | def _word_run(text: str) -> ET.Element: |
| 1340 | run = ET.Element(qn(W_NS, "r")) |
| 1341 | text_node = ET.SubElement(run, qn(W_NS, "t")) |
| 1342 | if text.startswith(" ") or text.endswith(" "): |
| 1343 | text_node.set(qn(XML_NS, "space"), "preserve") |
| 1344 | text_node.text = text |
| 1345 | return run |
| 1346 | |
| 1347 | |
| 1348 | def _set_drawing_paragraph_text(paragraph: ET.Element, text: str) -> None: |
| 1349 | keep = [child for child in list(paragraph) if child.tag == qn(A_NS, "pPr")] |
| 1350 | for child in list(paragraph): |
| 1351 | paragraph.remove(child) |
| 1352 | for child in keep: |
| 1353 | paragraph.append(child) |
| 1354 | run = ET.SubElement(paragraph, qn(A_NS, "r")) |
| 1355 | text_node = ET.SubElement(run, qn(A_NS, "t")) |
| 1356 | text_node.text = text |
| 1357 | |
| 1358 | |
| 1359 | def _require_openpyxl() -> Any: |
| 1360 | try: |
| 1361 | import openpyxl |
| 1362 | except ImportError as exc: |
| 1363 | raise RuntimeError("openpyxl is required for spreadsheet edits") from exc |
| 1364 | return openpyxl |
| 1365 | |
| 1366 | |
| 1367 | def _worksheet(workbook: Any, sheet: str = "") -> Any: |
| 1368 | if sheet: |
| 1369 | if sheet not in workbook.sheetnames: |
| 1370 | return workbook.create_sheet(sheet) |
| 1371 | return workbook[sheet] |
| 1372 | return workbook.active |
| 1373 | |
| 1374 | |
| 1375 | def _clear_worksheet(worksheet: Any) -> None: |
| 1376 | if worksheet.max_row: |
| 1377 | worksheet.delete_rows(1, worksheet.max_row) |
| 1378 | |
| 1379 | |
| 1380 | def _write_rows(worksheet: Any, rows: list[list[Any]], start_row: int) -> None: |
| 1381 | for row_offset, row in enumerate(rows): |
| 1382 | for col_offset, value in enumerate(row): |
| 1383 | worksheet.cell(row=start_row + row_offset, column=1 + col_offset, value=_cell_value(value)) |
| 1384 | |
| 1385 | |
| 1386 | def _normalize_rows(value: Any) -> list[list[Any]]: |
| 1387 | if value is None: |
| 1388 | return [] |
| 1389 | if isinstance(value, list): |
| 1390 | rows = value |
| 1391 | elif isinstance(value, str): |
| 1392 | rows = _rows_from_text(value) |
| 1393 | else: |
| 1394 | rows = [[value]] |
| 1395 | normalized = [] |
| 1396 | for row in rows: |
| 1397 | if isinstance(row, (list, tuple)): |
| 1398 | normalized.append([_cell_value(value) for value in row]) |
| 1399 | else: |
| 1400 | normalized.append([_cell_value(row)]) |
| 1401 | return normalized |
| 1402 | |
| 1403 | |
| 1404 | def _normalize_cells(cells: Any, default_sheet: str) -> list[tuple[str, str, Any]]: |
| 1405 | if isinstance(cells, str): |
| 1406 | parsed = json.loads(cells) |
| 1407 | else: |
| 1408 | parsed = cells |
| 1409 | if not parsed: |
| 1410 | raise ValueError("cells is required for set_cells") |
| 1411 | |
| 1412 | result: list[tuple[str, str, Any]] = [] |
| 1413 | if isinstance(parsed, dict): |
| 1414 | for ref, value in parsed.items(): |
| 1415 | sheet, cell = _split_cell_ref(str(ref), default_sheet) |
| 1416 | result.append((sheet, cell, _cell_value(value))) |
| 1417 | elif isinstance(parsed, list): |
| 1418 | for item in parsed: |
| 1419 | if not isinstance(item, dict): |
| 1420 | raise ValueError("cells list entries must be objects") |
| 1421 | ref = str(item.get("cell") or item.get("ref") or "") |
| 1422 | sheet = str(item.get("sheet") or default_sheet) |
| 1423 | if "!" in ref: |
| 1424 | sheet, ref = _split_cell_ref(ref, default_sheet) |
| 1425 | if not ref: |
| 1426 | raise ValueError("cell is required for each cells entry") |
| 1427 | result.append((sheet, ref, _cell_value(item.get("value")))) |
| 1428 | else: |
| 1429 | raise ValueError("cells must be an object or list") |
| 1430 | return result |
| 1431 | |
| 1432 | |
| 1433 | def _split_cell_ref(ref: str, default_sheet: str) -> tuple[str, str]: |
| 1434 | if "!" not in ref: |
| 1435 | return default_sheet, ref |
| 1436 | sheet, cell = ref.split("!", 1) |
| 1437 | return sheet.strip("'") or default_sheet, cell |
| 1438 | |
| 1439 | |
| 1440 | def _cell_value(value: Any) -> Any: |
| 1441 | if isinstance(value, (dict, list)): |
| 1442 | return json.dumps(value, ensure_ascii=False) |
| 1443 | return value |
| 1444 | |
| 1445 | |
| 1446 | def _rows_from_text(content: str) -> list[list[str]]: |
| 1447 | text = str(content or "").strip("\n") |
| 1448 | if not text.strip(): |
| 1449 | return [] |
| 1450 | lines = [line for line in text.splitlines() if line.strip()] |
| 1451 | markdown_rows = _markdown_table_rows(lines) |
| 1452 | if markdown_rows: |
| 1453 | return markdown_rows |
| 1454 | |
| 1455 | delimiter = "\t" if any("\t" in line for line in lines) else ("," if any("," in line for line in lines) else None) |
| 1456 | if delimiter: |
| 1457 | return [row for row in csv.reader(io.StringIO("\n".join(lines)), delimiter=delimiter)] |
| 1458 | return [[line] for line in lines] |
| 1459 | |
| 1460 | |
| 1461 | def _markdown_table_rows(lines: list[str]) -> list[list[str]]: |
| 1462 | table_lines = [line.strip() for line in lines if line.strip().startswith("|") and line.strip().endswith("|")] |
| 1463 | if len(table_lines) < 2: |
| 1464 | return [] |
| 1465 | rows = [] |
| 1466 | for line in table_lines: |
| 1467 | cells = [cell.strip() for cell in line.strip("|").split("|")] |
| 1468 | if all(re.fullmatch(r":?-{3,}:?", cell or "") for cell in cells): |
| 1469 | continue |
| 1470 | rows.append(cells) |
| 1471 | return rows |
| 1472 | |
| 1473 | |
| 1474 | def _zip_member(data: bytes, name: str) -> bytes: |
| 1475 | with zipfile.ZipFile(io.BytesIO(data)) as archive: |
| 1476 | return archive.read(name) |
| 1477 | |
| 1478 | |
| 1479 | def _odf_content_root(path: Path) -> ET.Element: |
| 1480 | with zipfile.ZipFile(path) as archive: |
| 1481 | return ET.fromstring(archive.read("content.xml")) |
| 1482 | |
| 1483 | |
| 1484 | def _odf_text_lines(root: ET.Element) -> list[str]: |
| 1485 | lines = [] |
| 1486 | for node in root.iter(): |
| 1487 | if node.tag not in {qn(ODF_TEXT_NS, "h"), qn(ODF_TEXT_NS, "p")}: |
| 1488 | continue |
| 1489 | text = "".join(node.itertext()).strip() |
| 1490 | if text: |
| 1491 | lines.append(text) |
| 1492 | return lines |
| 1493 | |
| 1494 | |
| 1495 | def _ods_sheets_from_bytes( |
| 1496 | data: bytes, |
| 1497 | *, |
| 1498 | max_rows: int | None = None, |
| 1499 | max_cols: int | None = None, |
| 1500 | strict_limits: bool = False, |
| 1501 | ) -> list[dict[str, Any]]: |
| 1502 | root = ET.fromstring(_zip_member(data, "content.xml")) |
| 1503 | sheets = [] |
| 1504 | for index, table in enumerate(root.iter(qn(ODF_TABLE_NS, "table")), start=1): |
| 1505 | name = table.get(qn(ODF_TABLE_NS, "name")) or table.get("name") or f"Sheet{index}" |
| 1506 | rows = [] |
| 1507 | for row in table: |
| 1508 | if row.tag != qn(ODF_TABLE_NS, "table-row"): |
| 1509 | continue |
| 1510 | values = _ods_row_values(row, max_cols=max_cols, strict_limits=strict_limits) |
| 1511 | repeat_rows = _repeat_count(row.get(qn(ODF_TABLE_NS, "number-rows-repeated"))) |
| 1512 | append_count = repeat_rows |
| 1513 | if max_rows is not None: |
| 1514 | remaining = max(max_rows - len(rows), 0) |
| 1515 | append_count = min(repeat_rows, remaining) |
| 1516 | if strict_limits and repeat_rows > append_count and _row_has_content(values): |
| 1517 | raise ValueError(f"ODS direct editing is limited to {max_rows} populated rows per sheet.") |
| 1518 | if remaining == 0: |
| 1519 | if not strict_limits: |
| 1520 | break |
| 1521 | continue |
| 1522 | for _ in range(append_count): |
| 1523 | rows.append(values.copy()) |
| 1524 | if max_rows is not None and len(rows) >= max_rows and not strict_limits: |
| 1525 | break |
| 1526 | sheets.append({"name": name, "rows": _trim_blank_edges(rows)}) |
| 1527 | return sheets |
| 1528 | |
| 1529 | |
| 1530 | def _ods_row_values( |
| 1531 | row: ET.Element, |
| 1532 | *, |
| 1533 | max_cols: int | None = None, |
| 1534 | strict_limits: bool = False, |
| 1535 | ) -> list[Any]: |
| 1536 | values = [] |
| 1537 | for cell in row: |
| 1538 | if cell.tag not in {qn(ODF_TABLE_NS, "table-cell"), qn(ODF_TABLE_NS, "covered-table-cell")}: |
| 1539 | continue |
| 1540 | value = _ods_cell_value(cell) |
| 1541 | repeat = _repeat_count(cell.get(qn(ODF_TABLE_NS, "number-columns-repeated"))) |
| 1542 | append_count = repeat |
| 1543 | if max_cols is not None: |
| 1544 | remaining = max(max_cols - len(values), 0) |
| 1545 | append_count = min(repeat, remaining) |
| 1546 | if strict_limits and repeat > append_count and _cell_has_content(value): |
| 1547 | raise ValueError(f"ODS direct editing is limited to {max_cols} populated columns per sheet.") |
| 1548 | if remaining == 0: |
| 1549 | if not strict_limits: |
| 1550 | break |
| 1551 | continue |
| 1552 | for _ in range(append_count): |
| 1553 | values.append(value) |
| 1554 | if max_cols is not None and len(values) >= max_cols and not strict_limits: |
| 1555 | break |
| 1556 | return values |
| 1557 | |
| 1558 | |
| 1559 | def _ods_cell_value(cell: ET.Element) -> Any: |
| 1560 | value_type = str(cell.get(qn(ODF_OFFICE_NS, "value-type")) or "").lower() |
| 1561 | if value_type in {"float", "currency", "percentage"}: |
| 1562 | raw = cell.get(qn(ODF_OFFICE_NS, "value")) |
| 1563 | if raw not in (None, ""): |
| 1564 | try: |
| 1565 | number = float(raw) |
| 1566 | return int(number) if number.is_integer() else number |
| 1567 | except ValueError: |
| 1568 | pass |
| 1569 | if value_type == "boolean": |
| 1570 | raw = str(cell.get(qn(ODF_OFFICE_NS, "boolean-value")) or "").lower() |
| 1571 | if raw in {"true", "false"}: |
| 1572 | return raw == "true" |
| 1573 | text = "\n".join("".join(node.itertext()).strip() for node in cell.iter(qn(ODF_TEXT_NS, "p"))) |
| 1574 | return text.strip() |
| 1575 | |
| 1576 | |
| 1577 | def _repeat_count(value: Any) -> int: |
| 1578 | try: |
| 1579 | count = int(value or 1) |
| 1580 | except (TypeError, ValueError): |
| 1581 | count = 1 |
| 1582 | return max(1, count) |
| 1583 | |
| 1584 | |
| 1585 | def _trim_blank_edges(rows: list[list[Any]]) -> list[list[Any]]: |
| 1586 | trimmed = [] |
| 1587 | for row in rows: |
| 1588 | next_row = list(row) |
| 1589 | while next_row and not _cell_has_content(next_row[-1]): |
| 1590 | next_row.pop() |
| 1591 | trimmed.append(next_row) |
| 1592 | while trimmed and not _row_has_content(trimmed[-1]): |
| 1593 | trimmed.pop() |
| 1594 | return trimmed |
| 1595 | |
| 1596 | |
| 1597 | def _cell_has_content(value: Any) -> bool: |
| 1598 | if value is None: |
| 1599 | return False |
| 1600 | if isinstance(value, str): |
| 1601 | return bool(value.strip()) |
| 1602 | return True |
| 1603 | |
| 1604 | |
| 1605 | def _row_has_content(row: list[Any]) -> bool: |
| 1606 | return any(_cell_has_content(value) for value in row) |
| 1607 | |
| 1608 | |
| 1609 | def _ods_sheet(sheets: list[dict[str, Any]], name: str = "") -> dict[str, Any]: |
| 1610 | normalized = str(name or "").strip() |
| 1611 | if normalized: |
| 1612 | for sheet in sheets: |
| 1613 | if str(sheet["name"]).casefold() == normalized.casefold(): |
| 1614 | return sheet |
| 1615 | sheet = {"name": normalized, "rows": []} |
| 1616 | sheets.append(sheet) |
| 1617 | return sheet |
| 1618 | return sheets[0] |
| 1619 | |
| 1620 | |
| 1621 | def _cell_indices(cell: str) -> tuple[int, int]: |
| 1622 | match = re.fullmatch(r"\$?([A-Za-z]{1,4})\$?([1-9][0-9]*)", str(cell or "").strip()) |
| 1623 | if not match: |
| 1624 | raise ValueError(f"Invalid cell reference: {cell}") |
| 1625 | col = 0 |
| 1626 | for char in match.group(1).upper(): |
| 1627 | col = col * 26 + (ord(char) - 64) |
| 1628 | return int(match.group(2)), col |
| 1629 | |
| 1630 | |
| 1631 | def _set_matrix_value(rows: list[list[Any]], row_idx: int, col_idx: int, value: Any) -> None: |
| 1632 | while len(rows) < row_idx: |
| 1633 | rows.append([]) |
| 1634 | row = rows[row_idx - 1] |
| 1635 | while len(row) < col_idx: |
| 1636 | row.append("") |
| 1637 | row[col_idx - 1] = value |
| 1638 | |
| 1639 | |
| 1640 | def _odp_text_slides(data: bytes) -> list[dict[str, Any]]: |
| 1641 | root = ET.fromstring(_zip_member(data, "content.xml")) |
| 1642 | slides = [] |
| 1643 | for page in root.iter(qn(ODF_DRAW_NS, "page")): |
| 1644 | lines = [] |
| 1645 | for node in page.iter(): |
| 1646 | if node.tag not in {qn(ODF_TEXT_NS, "h"), qn(ODF_TEXT_NS, "p")}: |
| 1647 | continue |
| 1648 | text = "".join(node.itertext()).strip() |
| 1649 | if text: |
| 1650 | lines.append(text) |
| 1651 | if lines: |
| 1652 | slides.append({"title": lines[0], "bullets": lines[1:]}) |
| 1653 | return slides |
| 1654 | |
| 1655 | |
| 1656 | def _pptx_text_slides(data: bytes) -> list[dict[str, Any]]: |
| 1657 | slides = [] |
| 1658 | with zipfile.ZipFile(io.BytesIO(data)) as archive: |
| 1659 | for name in _slide_names(archive): |
| 1660 | root = ET.fromstring(archive.read(name)) |
| 1661 | lines = [] |
| 1662 | for paragraph in root.iter(qn(A_NS, "p")): |
| 1663 | text = "".join(node.text or "" for node in paragraph.iter(qn(A_NS, "t"))).strip() |
| 1664 | if text: |
| 1665 | lines.append(text) |
| 1666 | if lines: |
| 1667 | slides.append({"title": lines[0], "bullets": lines[1:]}) |
| 1668 | return slides |
| 1669 | |
| 1670 | |
| 1671 | def _normalize_slides(value: Any) -> list[dict[str, Any]]: |
| 1672 | if value is None: |
| 1673 | return [] |
| 1674 | if isinstance(value, str): |
| 1675 | stripped = value.strip() |
| 1676 | if not stripped: |
| 1677 | return [] |
| 1678 | if stripped.startswith("[") or stripped.startswith("{"): |
| 1679 | return _normalize_slides(json.loads(stripped)) |
| 1680 | chunks = re.split(r"(?m)^\s*---+\s*$", stripped) |
| 1681 | result = [] |
| 1682 | for chunk in chunks: |
| 1683 | lines = [line.strip(" -\t") for line in chunk.splitlines() if line.strip()] |
| 1684 | if not lines: |
| 1685 | continue |
| 1686 | result.append({"title": lines[0], "bullets": lines[1:]}) |
| 1687 | return result |
| 1688 | if isinstance(value, dict): |
| 1689 | return [_slide_from_mapping(value)] |
| 1690 | if isinstance(value, list): |
| 1691 | result = [] |
| 1692 | for item in value: |
| 1693 | if isinstance(item, dict): |
| 1694 | result.append(_slide_from_mapping(item)) |
| 1695 | elif isinstance(item, str): |
| 1696 | result.extend(_normalize_slides(item)) |
| 1697 | elif isinstance(item, (list, tuple)): |
| 1698 | lines = [str(part) for part in item if str(part).strip()] |
| 1699 | if lines: |
| 1700 | result.append({"title": lines[0], "bullets": lines[1:]}) |
| 1701 | else: |
| 1702 | result.append({"title": str(item), "bullets": []}) |
| 1703 | return result |
| 1704 | return [{"title": str(value), "bullets": []}] |
| 1705 | |
| 1706 | |
| 1707 | def _slide_from_mapping(value: dict[str, Any]) -> dict[str, Any]: |
| 1708 | title = str(value.get("title") or value.get("heading") or "Slide") |
| 1709 | bullets = value.get("bullets") |
| 1710 | if bullets is None: |
| 1711 | body = value.get("body") or value.get("content") or "" |
| 1712 | bullets = [line.strip(" -\t") for line in str(body).splitlines() if line.strip()] |
| 1713 | elif isinstance(bullets, str): |
| 1714 | bullets = [line.strip(" -\t") for line in bullets.splitlines() if line.strip()] |
| 1715 | else: |
| 1716 | bullets = [str(item) for item in bullets] |
| 1717 | return {"title": title, "bullets": bullets} |
| 1718 | |
| 1719 | |
| 1720 | def _pptx_from_slides(slides: list[dict[str, Any]]) -> bytes: |
| 1721 | return pptx_writer.pptx_from_slides(slides) |
| 1722 | |
| 1723 | |
| 1724 | def _pptx_content_types(count: int) -> str: |
| 1725 | overrides = [ |
| 1726 | '<Override PartName="/ppt/presentation.xml" ' |
| 1727 | 'ContentType="application/vnd.openxmlformats-officedocument.presentationml.presentation.main+xml"/>' |
| 1728 | ] |
| 1729 | for index in range(1, count + 1): |
| 1730 | overrides.append( |
| 1731 | f'<Override PartName="/ppt/slides/slide{index}.xml" ' |
| 1732 | 'ContentType="application/vnd.openxmlformats-officedocument.presentationml.slide+xml"/>' |
| 1733 | ) |
| 1734 | return ( |
| 1735 | '<?xml version="1.0" encoding="UTF-8"?>' |
| 1736 | f'<Types xmlns="{CT_NS}">' |
| 1737 | '<Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>' |
| 1738 | '<Default Extension="xml" ContentType="application/xml"/>' |
| 1739 | + "".join(overrides) |
| 1740 | + "</Types>" |
| 1741 | ) |
| 1742 | |
| 1743 | |
| 1744 | def _pptx_presentation_rels(count: int) -> str: |
| 1745 | rels = [] |
| 1746 | for index in range(1, count + 1): |
| 1747 | rels.append( |
| 1748 | f'<Relationship Id="rId{index}" ' |
| 1749 | 'Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/slide" ' |
| 1750 | f'Target="slides/slide{index}.xml"/>' |
| 1751 | ) |
| 1752 | return '<?xml version="1.0" encoding="UTF-8"?>' + f'<Relationships xmlns="{REL_NS}">' + "".join(rels) + "</Relationships>" |
| 1753 | |
| 1754 | |
| 1755 | def _pptx_presentation_xml(count: int) -> str: |
| 1756 | slide_ids = "".join(f'<p:sldId id="{255 + index}" r:id="rId{index}"/>' for index in range(1, count + 1)) |
| 1757 | return ( |
| 1758 | '<?xml version="1.0" encoding="UTF-8"?>' |
| 1759 | f'<p:presentation xmlns:p="{P_NS}" xmlns:r="{R_NS}">' |
| 1760 | f"<p:sldIdLst>{slide_ids}</p:sldIdLst>" |
| 1761 | '<p:sldSz cx="9144000" cy="5143500"/>' |
| 1762 | "</p:presentation>" |
| 1763 | ) |
| 1764 | |
| 1765 | |
| 1766 | def _pptx_slide_xml(slide: dict[str, Any]) -> str: |
| 1767 | title = str(slide.get("title") or "Slide") |
| 1768 | bullets = [str(item) for item in slide.get("bullets") or []] |
| 1769 | paragraphs = [title, *bullets] |
| 1770 | text = "".join(f"<a:p><a:r><a:t>{escape(item)}</a:t></a:r></a:p>" for item in paragraphs) |
| 1771 | return ( |
| 1772 | '<?xml version="1.0" encoding="UTF-8"?>' |
| 1773 | f'<p:sld xmlns:a="{A_NS}" xmlns:p="{P_NS}">' |
| 1774 | "<p:cSld><p:spTree>" |
| 1775 | '<p:nvGrpSpPr><p:cNvPr id="1" name=""/><p:cNvGrpSpPr/><p:nvPr/></p:nvGrpSpPr>' |
| 1776 | "<p:grpSpPr/>" |
| 1777 | '<p:sp><p:nvSpPr><p:cNvPr id="2" name="Content"/><p:cNvSpPr/><p:nvPr/></p:nvSpPr>' |
| 1778 | f"<p:txBody><a:bodyPr/><a:lstStyle/>{text}</p:txBody>" |
| 1779 | "</p:sp>" |
| 1780 | "</p:spTree></p:cSld>" |
| 1781 | "</p:sld>" |
| 1782 | ) |
| 1783 | |
| 1784 | |
| 1785 | def _slide_names(archive: zipfile.ZipFile) -> list[str]: |
| 1786 | return sorted( |
| 1787 | [name for name in archive.namelist() if name.startswith("ppt/slides/slide") and name.endswith(".xml")], |
| 1788 | key=_natural_key, |
| 1789 | ) |
| 1790 | |
| 1791 | |
| 1792 | def _natural_key(value: str) -> list[Any]: |
| 1793 | return [int(part) if part.isdigit() else part for part in re.split(r"(\d+)", value)] |
| 1794 | |
| 1795 | |
| 1796 | def _text_lines(content: str) -> list[str]: |
| 1797 | lines = [line.rstrip() for line in str(content or "").splitlines()] |
| 1798 | return lines or [""] |
| 1799 | |
| 1800 | |
| 1801 | def _int_or_none(value: Any) -> int | None: |
| 1802 | if value in (None, ""): |
| 1803 | return None |
| 1804 | try: |
| 1805 | number = int(value) |
| 1806 | except (TypeError, ValueError): |
| 1807 | return None |
| 1808 | return number if number > 0 else None |
| 1809 | |
| 1810 | |
| 1811 | def _xml_bytes(root: ET.Element) -> bytes: |
| 1812 | return ET.tostring(root, encoding="utf-8", xml_declaration=True) |
| 1813 | |
| 1814 | |
| 1815 | def _zip_from_existing(files: dict[str, bytes]) -> bytes: |
| 1816 | buffer = io.BytesIO() |
| 1817 | with zipfile.ZipFile(buffer, "w", compression=zipfile.ZIP_DEFLATED) as archive: |
| 1818 | for name, data in files.items(): |
| 1819 | archive.writestr(name, data) |
| 1820 | return buffer.getvalue() |
| 1821 | |
| 1822 | |
| 1823 | def _zip_map(files: dict[str, str | bytes]) -> bytes: |
| 1824 | buffer = io.BytesIO() |
| 1825 | with zipfile.ZipFile(buffer, "w", compression=zipfile.ZIP_DEFLATED) as archive: |
| 1826 | for name, value in files.items(): |
| 1827 | archive.writestr(name, value.encode("utf-8") if isinstance(value, str) else value) |
| 1828 | return buffer.getvalue() |
| 1829 | |
| 1830 | |
| 1831 | def _trim_payload(payload: dict[str, Any], max_chars: int) -> dict[str, Any]: |
| 1832 | text = json.dumps(payload, ensure_ascii=False, default=str) |
| 1833 | if len(text) <= max_chars: |
| 1834 | return payload |
| 1835 | trimmed = dict(payload) |
| 1836 | if "paragraphs" in trimmed: |
| 1837 | trimmed["paragraphs"] = trimmed["paragraphs"][:20] |
| 1838 | if "sheets" in trimmed: |
| 1839 | trimmed["sheets"] = [ |
| 1840 | {**sheet, "preview_rows": sheet.get("preview_rows", [])[:20]} |
| 1841 | for sheet in trimmed["sheets"][:4] |
| 1842 | ] |
| 1843 | if "slides" in trimmed: |
| 1844 | trimmed["slides"] = trimmed["slides"][:12] |
| 1845 | if "text" in trimmed and isinstance(trimmed["text"], str): |
| 1846 | trimmed["text"] = trimmed["text"][:max_chars] + "\n... [truncated]" |
| 1847 | trimmed["truncated"] = True |
| 1848 | return trimmed |