any-file to PDF (Phase 1): converters/ dispatcher, images/text/md/csv/json/xml/svg/pdf-passthrough, 112 tests green

converters/csv_converter.py

@@ -0,0 +1,71 @@
+"""CSV / TSV -> PDF (typed table).
+
+The first row is rendered as the header (<th>); every row becomes a
+<table> row — the table rules from app.BASE_CSS are injected at the top of
+<head> so they under-layer any user CSS. Delimiter: forced tab for .tsv;
+for .csv, csv.Sniffer on a sample with a comma fallback (single-line files
+that the sniffer rejects still render).
+
+Empty / whitespace-only input is a job error ("empty file: ...").
+"""
+from __future__ import annotations
+
+import csv
+import io
+from pathlib import Path
+
+from . import common
+from .text_converter import decode_bytes
+
+
+def _rows(text: str, delimiter: str) -> list[list[str]]:
+    rows = list(csv.reader(io.StringIO(text), delimiter=delimiter))
+    return [row for row in rows if any(cell.strip() for cell in row)]
+
+
+def _table_html(rows: list[list[str]]) -> str:
+    if not rows:
+        return ""
+    header, body = rows[0], rows[1:]
+    head = "".join(f"<th>{common.escape(c)}</th>" for c in header)
+    out = [f"<table><tr>{head}</tr>"]
+    for row in body:
+        # pad ragged rows to header width so borders line up
+        cells = row + [""] * (len(header) - len(row))
+        out.append("<tr>" + "".join(f"<td>{common.escape(c)}</td>" for c in cells) + "</tr>")
+    out.append("</table>")
+    return "\n".join(out)
+
+
+def convert(path: Path, job_id: str, progress_cb=None, output_dir=None,
+            source_name=None) -> str:
+    """Render a CSV or TSV file to a tabular PDF."""
+    import app  # lazy: BASE_CSS lives with the EPUB pipeline
+
+    raw = path.read_bytes()
+    if not raw.strip():
+        raise ValueError("empty file: nothing to convert")
+    text = decode_bytes(raw)
+
+    if path.suffix.lower() == ".tsv":
+        delimiter = "\t"
+    else:
+        try:
+            delimiter = csv.Sniffer().sniff(text[:8192], delimiters=",;\t|").delimiter
+        except csv.Error:
+            delimiter = ","
+    rows = _rows(text, delimiter)
+    table = _table_html(rows)
+    html_doc = (
+        "<!DOCTYPE html><html><head><style>\n" + app.BASE_CSS + "\n</style>"
+        f"</head><body>{table}</body></html>"
+    )
+
+    def _build():
+        return html_doc, path.name
+
+    return common.convert_single(
+        path, job_id, build_html=_build, css="",
+        progress_cb=progress_cb, output_dir=output_dir,
+        name_src=source_name or path.name,
+    )