fix(sheet): skip hidden and very-hidden worksheets

src/formats/sheet/mod.rs

@@ -4,7 +4,7 @@ use crate::error::ConvertError;
 use crate::model::{Block, Cell, Document, GridBuilder, Inline, TableKind};
 use crate::shared::header::resolve_header_rows;
 use crate::shared::text::clean_text;
-use calamine::{Data, Dimensions, Reader, Sheets, open_workbook_auto_from_rs};
+use calamine::{Data, Dimensions, Reader, SheetVisible, Sheets, open_workbook_auto_from_rs};
 use std::collections::{HashMap, HashSet};
 use std::io::Cursor;
 
@@ -25,12 +25,33 @@ pub fn parse(bytes: &[u8]) -> Result<Document, ConvertError> {
         contained("workbook open", || open_workbook_auto_from_rs(Cursor::new(bytes)))?
             .map_err(map_open_error)?;
     let sheet_names = contained("sheet listing", || workbook.sheet_names().to_owned())?;
-    let multi_sheet = sheet_names.len() > 1;
+    // `sheets_metadata()` shares one backing `Vec<Sheet>` with `sheet_names()`,
+    // so the Nth metadata entry describes the Nth name - a name lookup would
+    // reintroduce the bug under re-ordering, only positional access is sound.
+    let metadata = contained("sheet metadata", || workbook.sheets_metadata().to_owned())?;
+    // The multi-sheet heading ("## <name>") is only useful when more than one
+    // sheet is actually shown: a single visible sheet inside a workbook full
+    // of hidden ones must not gain (or lose) its heading. Count visible sheets
+    // rather than all sheets so hidden ones stay invisible to this decision.
+    let multi_sheet = sheet_names
+        .iter()
+        .zip(metadata.iter())
+        .filter(|(_, m)| m.visible == SheetVisible::Visible)
+        .count()
+        > 1;
     let merged = merged_regions(&mut workbook, &sheet_names)?;
 
     let mut doc = Document::default();
     let mut failed = 0usize;
-    for name in &sheet_names {
+    for (name, meta) in sheet_names.iter().zip(metadata.iter()) {
+        // A hidden or veryHidden sheet is not visible to an end user opening
+        // the workbook, so it must contribute no block at all - skip it before
+        // any heading or table is emitted. This runs before the read attempt,
+        // so a hidden sheet never counts as "unreadable" (hidden != unreadable)
+        // and an all-hidden workbook degrades to an empty Document, not an error.
+        if meta.visible != SheetVisible::Visible {
+            continue;
+        }
         let range = match contained("worksheet read", || workbook.worksheet_range(name))? {
             Ok(r) => r,
             Err(e) => {

tests/gen_fixtures.py

@@ -914,6 +914,71 @@ def merged_xlsx():
     ])
 
 
+# ---------------------------------------------------------------------------
+# R-hidden-sheet: hidden / veryHidden worksheets must not render as visible
+# content (sheet-level visibility, the half of #9 calamine exposes). A visible
+# sheet keeps its table; a hidden and a veryHidden sheet drop out entirely.
+
+def hidden_xlsx():
+    ct = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types">
+<Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>
+<Default Extension="xml" ContentType="application/xml"/>
+<Override PartName="/xl/workbook.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet.main+xml"/>
+<Override PartName="/xl/worksheets/sheet1.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/>
+<Override PartName="/xl/worksheets/sheet2.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/>
+<Override PartName="/xl/worksheets/sheet3.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/>
+</Types>"""
+    root_rels = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+<Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="xl/workbook.xml"/>
+</Relationships>"""
+    workbook = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<workbook xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main"
+ xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships">
+<sheets>
+<sheet name="Visible" sheetId="1" r:id="rId1"/>
+<sheet name="Hidden" sheetId="2" r:id="rId2" state="hidden"/>
+<sheet name="VeryHidden" sheetId="3" r:id="rId3" state="veryHidden"/>
+</sheets></workbook>"""
+    wb_rels = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+<Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet1.xml"/>
+<Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet2.xml"/>
+<Relationship Id="rId3" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet3.xml"/>
+</Relationships>"""
+    sheet1 = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<worksheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
+<sheetData><row r="1">
+<c r="A1" t="inlineStr"><is><t>visible cell</t></is></c>
+<c r="B1" t="inlineStr"><is><t>shown</t></is></c>
+</row></sheetData>
+</worksheet>"""
+    sheet2 = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<worksheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
+<sheetData><row r="1">
+<c r="A1" t="inlineStr"><is><t>hidden cell</t></is></c>
+<c r="B1" t="inlineStr"><is><t>must not appear</t></is></c>
+</row></sheetData>
+</worksheet>"""
+    sheet3 = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<worksheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
+<sheetData><row r="1">
+<c r="A1" t="inlineStr"><is><t>very hidden cell</t></is></c>
+<c r="B1" t="inlineStr"><is><t>must not appear either</t></is></c>
+</row></sheetData>
+</worksheet>"""
+    write_zip(OUT / "xlsx" / "handmade-hidden.xlsx", [
+        ("[Content_Types].xml", ct),
+        ("_rels/.rels", root_rels),
+        ("xl/workbook.xml", workbook),
+        ("xl/_rels/workbook.xml.rels", wb_rels),
+        ("xl/worksheets/sheet1.xml", sheet1),
+        ("xl/worksheets/sheet2.xml", sheet2),
+        ("xl/worksheets/sheet3.xml", sheet3),
+    ])
+
+
 # ---------------------------------------------------------------------------
 # R16: ODF style:default-style beneath named chains; full ISO durations
 
@@ -1748,6 +1813,7 @@ def main():
     manyrefs_docx()
     defaults_odf()
     merged_xlsx()
+    hidden_xlsx()
     features_epub()
     bin_rtf()
     csvs()

tests/snapshots.rs

@@ -11,6 +11,7 @@ mod common;
 
 use common::{fixture_root, walk};
 use std::fmt::Write as _;
+use std::io::Write as _;
 use std::path::Path;
 
 /// Convert one file, capturing panics so a bad parser records a baseline
@@ -166,6 +167,74 @@ fn doc_inline_picture_is_retained() {
     );
 }
 
+/// Hidden and veryHidden worksheets contribute no heading and no table; only
+/// the visible sheet renders. This is the sheet-level half of #9 (calamine
+/// exposes `state="hidden"`/`state="veryHidden"`; row/column visibility is a
+/// separate, out-of-scope half). The committed fixture holds one visible sheet
+/// plus a hidden and a veryHidden sheet, each with identifying cell content.
+#[test]
+fn hidden_worksheets_do_not_render() {
+    let path = fixture_root().join("xlsx").join("handmade-hidden.xlsx");
+    let bytes = std::fs::read(&path).unwrap();
+    let doc = anydoc::to_document(&bytes, anydoc::Format::Excel).unwrap();
+    // Exactly one table - the visible sheet's; no heading is emitted because a
+    // single visible sheet needs no "## <name>" disambiguator.
+    assert_eq!(doc.blocks.len(), 1, "only the visible sheet should render, got {:?}", doc.blocks);
+    assert!(matches!(doc.blocks[0], anydoc::model::Block::Table(_)));
+
+    // The hidden sheets' cell content and names must be absent from the
+    // rendered Markdown (no heading, no rendered table).
+    let md = anydoc::to_markdown(&path).unwrap();
+    assert!(md.contains("visible cell"), "visible content should render:\n{md}");
+    for needle in ["Hidden", "VeryHidden", "must not appear", "very hidden cell"] {
+        assert!(!md.contains(needle), "hidden content leaked into output ({needle}):\n{md}");
+    }
+}
+
+/// A workbook whose every sheet is hidden degrades to an empty Document, not a
+/// panic and not a misleading "no sheet could be read" error (hidden is a valid
+/// state, not unreadable). Built inline so no committed fixture is all-hidden.
+#[test]
+fn all_hidden_workbook_yields_empty_document() {
+    let bytes = all_hidden_xlsx();
+    let doc = anydoc::to_document(&bytes, anydoc::Format::Excel).unwrap();
+    assert!(doc.blocks.is_empty(), "an all-hidden workbook must render no blocks");
+    let md = anydoc::to_markdown_bytes(&bytes, anydoc::Format::Excel).unwrap();
+    assert!(md.trim().is_empty(), "an all-hidden workbook must render no markdown");
+}
+
+/// Minimal xlsx whose single sheet is `state="hidden"`, so every sheet is
+/// non-visible. Mirrors the inline xlsx builder the sheet unit tests use.
+fn all_hidden_xlsx() -> Vec<u8> {
+    let parts: &[(&str, &str)] = &[
+        (
+            "[Content_Types].xml",
+            r#"<?xml version="1.0"?><Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types"><Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/><Default Extension="xml" ContentType="application/xml"/><Override PartName="/xl/workbook.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet.main+xml"/><Override PartName="/xl/worksheets/sheet1.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/></Types>"#,
+        ),
+        (
+            "_rels/.rels",
+            r#"<?xml version="1.0"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="xl/workbook.xml"/></Relationships>"#,
+        ),
+        (
+            "xl/workbook.xml",
+            r#"<?xml version="1.0"?><workbook xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"><sheets><sheet name="Only" sheetId="1" r:id="rId1" state="hidden"/></sheets></workbook>"#,
+        ),
+        (
+            "xl/_rels/workbook.xml.rels",
+            r#"<?xml version="1.0"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet1.xml"/></Relationships>"#,
+        ),
+    ];
+    let sheet = r#"<?xml version="1.0"?><worksheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main"><sheetData><row r="1"><c r="A1" t="inlineStr"><is><t>hidden only</t></is></c></row></sheetData></worksheet>"#;
+    let mut w = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
+    for (name, body) in parts {
+        w.start_file(*name, zip::write::SimpleFileOptions::default()).unwrap();
+        w.write_all(body.as_bytes()).unwrap();
+    }
+    w.start_file("xl/worksheets/sheet1.xml", zip::write::SimpleFileOptions::default()).unwrap();
+    w.write_all(sheet.as_bytes()).unwrap();
+    w.finish().unwrap().into_inner()
+}
+
 /// The RTF `\pict` payload is retained as an asset (the Markdown output
 /// shows only the alt text, which is empty for pictures without one).
 #[test]

tests/snapshots/snapshots__xlsx__handmade-hidden.xlsx.snap

@@ -0,0 +1,7 @@
+---
+source: tests/snapshots.rs
+expression: output
+---
+|  |  |
+| --- | --- |
+| visible cell | shown |