feat(sheet): read every Excel container in-house and drop calamine

.gitignore

@@ -12,3 +12,7 @@ __pycache__/
 # wasm-pack build output: pkg/ is the npm package, www/pkg/ the demo site copy
 /wasm/pkg/
 /wasm/www/pkg/
+# cargo-fuzz working dirs: seeds live in fuzz/seeds/ and are checked in
+/fuzz/corpus/
+/fuzz/artifacts/
+/fuzz/coverage/

Cargo.toml

@@ -7,7 +7,7 @@ exclude = ["fuzz"]
 name = "anydoc"
 version = "0.1.9"
 edition = "2024"
-# Edition 2024 needs 1.85; zip and calamine both raise it to 1.88.
+# Edition 2024 needs 1.85; zip raises it to 1.88.
 rust-version = "1.88"
 description = "Convert documents (doc, docx, odt, rtf, epub, pdf, presentations, spreadsheets, csv) to GitHub-Flavored Markdown"
 license = "MIT"
@@ -22,7 +22,6 @@ insta = "1"
 sha2 = "0.11"
 
 [dependencies]
-calamine = { version = "0.36.1", features = ["dates"] }
 cfb = "0.14.0"
 csv = "1.4.0"
 flate2 = "1"

fuzz/Cargo.toml

@@ -10,6 +10,8 @@ cargo-fuzz = true
 [dependencies]
 libfuzzer-sys = "0.4"
 anydoc = { path = ".." }
+cfb = "0.14.0"
+zip = { version = "8.6.0", default-features = false, features = ["deflate"] }
 
 [[bin]]
 name = "docx"
@@ -64,3 +66,21 @@ name = "pdf"
 path = "fuzz_targets/pdf.rs"
 test = false
 doc = false
+
+[[bin]]
+name = "xls"
+path = "fuzz_targets/xls.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "xlsb"
+path = "fuzz_targets/xlsb.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "numfmt"
+path = "fuzz_targets/numfmt.rs"
+test = false
+doc = false

fuzz/README.md

@@ -0,0 +1,12 @@
+# Fuzzing
+
+    cargo +nightly fuzz run xlsx fuzz/corpus/xlsx fuzz/seeds/xlsx
+
+`fuzz/seeds/` holds checked-in starting inputs; `fuzz/corpus/` is the working
+directory libfuzzer writes to and is not checked in.
+
+`xls`, `xlsb` and `numfmt` wrap their input in a valid container (an OLE
+compound file, an OPC package, a styles part) so mutation reaches the record
+and format-code parsers rather than dying at the container gate. `xlsx` takes
+a whole workbook, and its seeds cover all three containers because the
+frontend picks the reader from the bytes.

fuzz/fuzz_targets/numfmt.rs

@@ -0,0 +1,54 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use std::io::{Cursor, Write};
+
+// A number format code, carried into the engine on a minimal workbook. The
+// grammar is where the parsing complexity lives, and reaching it through a
+// discovered styles part would cost the fuzzer most of its budget. Only the
+// control characters XML forbids are dropped, so whitespace a code may carry
+// still reaches the parser.
+fuzz_target!(|data: &[u8]| {
+    let code = String::from_utf8_lossy(data);
+    let escaped: String = code
+        .chars()
+        .filter(|c| !c.is_control() || matches!(c, '\t' | '\n' | '\r'))
+        .map(|c| match c {
+            '&' => "&".to_string(),
+            '<' => "&lt;".to_string(),
+            '>' => "&gt;".to_string(),
+            '"' => "&quot;".to_string(),
+            c => c.to_string(),
+        })
+        .collect();
+    let styles = format!(
+        r#"<?xml version="1.0"?><styleSheet xmlns="{SML}"><numFmts count="1"><numFmt numFmtId="164" formatCode="{escaped}"/></numFmts><cellXfs count="2"><xf numFmtId="0"/><xf numFmtId="164" applyNumberFormat="1"/></cellXfs></styleSheet>"#
+    );
+
+    let mut zip = zip::ZipWriter::new(Cursor::new(Vec::new()));
+    let opts = zip::write::SimpleFileOptions::default();
+    let parts: [(&str, &str); 6] = [
+        ("[Content_Types].xml", CONTENT_TYPES),
+        ("_rels/.rels", RELS),
+        ("xl/workbook.xml", WORKBOOK),
+        ("xl/_rels/workbook.xml.rels", WORKBOOK_RELS),
+        ("xl/styles.xml", &styles),
+        ("xl/worksheets/sheet1.xml", SHEET),
+    ];
+    for (name, body) in parts {
+        if zip.start_file(name, opts).is_err() || zip.write_all(body.as_bytes()).is_err() {
+            return;
+        }
+    }
+    let Ok(bytes) = zip.finish() else {
+        return;
+    };
+    let _ = anydoc::to_markdown_bytes(bytes.into_inner().as_slice(), anydoc::Format::Excel);
+});
+
+const SML: &str = "http://schemas.openxmlformats.org/spreadsheetml/2006/main";
+const CONTENT_TYPES: &str = 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/styles.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.styles+xml"/><Override PartName="/xl/worksheets/sheet1.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/></Types>"#;
+const RELS: &str = 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>"#;
+const WORKBOOK: &str = 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="S" sheetId="1" r:id="rId1"/></sheets></workbook>"#;
+const WORKBOOK_RELS: &str = 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"/><Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/></Relationships>"#;
+const SHEET: &str = r#"<?xml version="1.0"?><worksheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main"><sheetData><row r="1"><c r="A1" s="1"><v>1234.5</v></c><c r="B1" s="1"><v>0.075</v></c><c r="C1" s="1" t="inlineStr"><is><t>text</t></is></c></row></sheetData></worksheet>"#;

fuzz/fuzz_targets/xls.rs

@@ -0,0 +1,22 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use std::io::{Cursor, Write};
+
+// The BIFF record stream, wrapped in a valid OLE container so mutation
+// reaches the reader instead of dying at the container gate.
+fuzz_target!(|data: &[u8]| {
+    let Ok(mut ole) = cfb::CompoundFile::create(Cursor::new(Vec::new())) else {
+        return;
+    };
+    match ole.create_stream("Workbook") {
+        Ok(mut stream) => {
+            if stream.write_all(data).is_err() {
+                return;
+            }
+        }
+        Err(_) => return,
+    }
+    let bytes = ole.into_inner().into_inner();
+    let _ = anydoc::to_markdown_bytes(&bytes, anydoc::Format::Excel);
+});

fuzz/fuzz_targets/xlsb.rs

@@ -0,0 +1,40 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use std::io::{Cursor, Write};
+
+// The worksheet, styles and shared-string parts, wrapped in a package whose
+// workbook part and relationships are already valid. Fuzzing the workbook
+// part instead would strand the input there: without a resolvable worksheet
+// relationship every sheet is skipped, and the cell, format and string
+// readers never run. The workbook part is parsed by the same record reader
+// this does reach.
+fuzz_target!(|data: &[u8]| {
+    let mut zip = zip::ZipWriter::new(Cursor::new(Vec::new()));
+    let opts = zip::write::SimpleFileOptions::default();
+    let parts: [(&str, &[u8]); 8] = [
+        ("[Content_Types].xml", CONTENT_TYPES.as_bytes()),
+        ("_rels/.rels", RELS.as_bytes()),
+        ("xl/workbook.bin", WORKBOOK),
+        ("xl/_rels/workbook.bin.rels", WORKBOOK_RELS.as_bytes()),
+        ("xl/worksheets/sheet1.bin", data),
+        ("xl/worksheets/sheet2.bin", data),
+        ("xl/styles.bin", data),
+        ("xl/sharedStrings.bin", data),
+    ];
+    for (name, body) in parts {
+        if zip.start_file(name, opts).is_err() || zip.write_all(body).is_err() {
+            return;
+        }
+    }
+    let Ok(bytes) = zip.finish() else {
+        return;
+    };
+    let _ = anydoc::to_markdown_bytes(bytes.into_inner().as_slice(), anydoc::Format::Excel);
+});
+
+const WORKBOOK: &[u8] = include_bytes!("../seeds/xlsb/workbook-bin");
+
+const CONTENT_TYPES: &str = 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="bin" ContentType="application/vnd.ms-excel.sheet.binary.macroEnabled.main"/></Types>"#;
+const RELS: &str = 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.bin"/></Relationships>"#;
+const WORKBOOK_RELS: &str = 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.bin"/><Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet2.bin"/><Relationship Id="rId3" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.bin"/><Relationship Id="rId4" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/sharedStrings" Target="sharedStrings.bin"/></Relationships>"#;

fuzz/seeds/numfmt/ampm-multibyte

@@ -0,0 +1 @@
+aaaaé

fuzz/seeds/numfmt/ampm-token-multibyte

@@ -0,0 +1 @@
+h:mm AM/PMé

fuzz/seeds/numfmt/conditions

@@ -0,0 +1 @@
+[>100]0.0;[<=100]0.00

fuzz/seeds/numfmt/currency

@@ -0,0 +1 @@
+"$"#,##0.00

fuzz/seeds/numfmt/datetime

@@ -0,0 +1 @@
+yyyy-mm-dd hh:mm:ss AM/PM

fuzz/seeds/numfmt/elapsed

@@ -0,0 +1 @@
+[h]:mm:ss

fuzz/seeds/numfmt/four-sections

@@ -0,0 +1 @@
+#,##0.00;[Red](#,##0.00);"nil";@

fuzz/seeds/numfmt/fraction

@@ -0,0 +1 @@
+# ??/??

fuzz/seeds/numfmt/percent

@@ -0,0 +1 @@
+0.0%

fuzz/seeds/numfmt/scientific

@@ -0,0 +1 @@
+0.00E+00