Skip to content

Avoid panicking on invalid shared string indexes #24

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
carl-distill wants to merge 2 commits into
base: main
Choose a base branch
from
Draft

Avoid panicking on invalid shared string indexes #24

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
93b8bc0
Return errors for invalid style indexes
carl-distill Aug 7, 2026
f5b73ac
Return errors for invalid shared string indexes
carl-distill Aug 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions src/extension/writer.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,15 +3,16 @@
use crate::database::column::Column;
use crate::database::column::ColumnType;
use crate::error::RustySheetError;
use crate::spreadsheet::SpreadsheetError;
use crate::spreadsheet::cell::Cell;
use crate::spreadsheet::cell::CellType;
use crate::spreadsheet::resolve_loaded_shared_string;
use crate::spreadsheet::sheet::Sheet;
use duckdb::core::FlatVector;
use duckdb::core::Inserter;
use libduckdb_sys::duckdb_date;
use libduckdb_sys::duckdb_time;
use libduckdb_sys::duckdb_timestamp;
use crate::spreadsheet::sheet::Sheet;
use crate::spreadsheet::SpreadsheetError;

/// Writes a cell value to a DuckDB vector based on column type.
/// Handles type conversion and error mapping for different data types.
Expand All @@ -25,8 +26,9 @@ pub(super) fn write_to_vector(sheet: &Sheet, column: &Column, cell: &Cell, vecto
)
};
let cell = if cell.kind == CellType::SharedString {
let index = cell.value.parse::<usize>()?;
if let Some(shared_string) = &shared_strings[index] {
if let Some(shared_string) =
resolve_loaded_shared_string(shared_strings, &sheet.file_name, &sheet.name, cell)?
{
&Cell {
row: cell.row,
col: cell.col,
Expand Down
168 changes: 131 additions & 37 deletions src/spreadsheet/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,77 @@ pub(crate) mod xlsx;
pub(crate) mod criteria;
pub(crate) mod sheet;

pub(crate) fn resolve_number_format(
number_formats: &[CellType],
file_name: &str,
sheet_name: &str,
row: usize,
col: usize,
style_index: usize,
) -> Result<CellType, SpreadsheetError> {
number_formats.get(style_index).copied().ok_or_else(|| {
SpreadsheetError::CellStyleIndexError(
file_name.to_owned(),
sheet_name.to_owned(),
crate::spreadsheet::reference::index_to_reference(row, col),
style_index,
number_formats.len(),
)
})
}

pub(crate) fn resolve_shared_string<'a>(
shared_strings: &'a [String],
mappings: &HashMap<usize, usize>,
file_name: &str,
sheet_name: &str,
cell: &Cell,
) -> Result<&'a str, RustySheetError> {
let shared_string_index = cell.value.parse::<usize>()?;
let mapped_index = mappings.get(&shared_string_index).copied().ok_or_else(|| {
SpreadsheetError::CellSharedStringIndexError(
file_name.to_owned(),
sheet_name.to_owned(),
cell.reference(),
shared_string_index,
)
})?;

shared_strings
.get(mapped_index)
.map(String::as_str)
.ok_or_else(|| {
SpreadsheetError::CellSharedStringIndexError(
file_name.to_owned(),
sheet_name.to_owned(),
cell.reference(),
shared_string_index,
)
.into()
})
}

pub(crate) fn resolve_loaded_shared_string<'a>(
shared_strings: &'a [Option<String>],
file_name: &str,
sheet_name: &str,
cell: &Cell,
) -> Result<Option<&'a str>, RustySheetError> {
let shared_string_index = cell.value.parse::<usize>()?;
shared_strings
.get(shared_string_index)
.map(Option::as_deref)
.ok_or_else(|| {
SpreadsheetError::CellSharedStringIndexError(
file_name.to_owned(),
sheet_name.to_owned(),
cell.reference(),
shared_string_index,
)
.into()
})
}

#[derive(Error, Debug)]
pub(crate) enum SpreadsheetError {
/// Error indicating the spreadsheet format is not supported
Expand All @@ -48,6 +119,14 @@ pub(crate) enum SpreadsheetError {
/// Error indicating a specific cell value is invalid
#[error("Cell '[{0}]{1}!{2}': {3}")]
CellValueError(String, String, String, String),

/// Error indicating a cell style index is outside the workbook's style table.
#[error("Cell '[{0}]{1}!{2}': invalid style index {3}; workbook defines {4} styles")]
CellStyleIndexError(String, String, String, usize, usize),

/// Error indicating a shared string index is outside the workbook's shared string table.
#[error("Cell '[{0}]{1}!{2}': invalid shared string index {3}")]
CellSharedStringIndexError(String, String, String, usize),
}

pub(crate) trait Spreadsheet {
Expand Down Expand Up @@ -109,56 +188,71 @@ pub(crate) trait Spreadsheet {
let (shared_strings, mappings) = self.load_shared_strings(Some(shared_indexes))?;

let mut tables = Vec::<Table>::new();
for (name, header, data, row_lower_bound, col_lower_bound, col_upper_bound) in sheets.into_iter() {
let names = (col_lower_bound..=col_upper_bound).map(|col| {
let index = col - col_lower_bound;
if let Some(cell) = &header[index] {
let value = if cell.kind == CellType::SharedString {
let id = cell.value.parse::<usize>().expect("Shared string index");
let index = mappings[&id];
shared_strings[index].to_owned()
} else {
cell.to_string()
};
if !criteria.nulls.contains(&value) {
value
let file_name = self.name();
for (sheet_name, header, data, row_lower_bound, col_lower_bound, col_upper_bound) in sheets.into_iter() {
let names = (col_lower_bound..=col_upper_bound)
.map(|col| -> Result<String, RustySheetError> {
let index = col - col_lower_bound;
if let Some(cell) = &header[index] {
let value = if cell.kind == CellType::SharedString {
resolve_shared_string(
&shared_strings,
&mappings,
&file_name,
&sheet_name,
cell,
)?.to_owned()
} else {
cell.to_string()
};
Ok(if !criteria.nulls.contains(&value) {
value
} else {
index_to_col(col).to_owned()
})
} else {
index_to_col(col).to_owned()
Ok(index_to_col(col).to_owned())
}
} else {
index_to_col(col).to_owned()
}
}).collect::<Vec<_>>();
})
.collect::<Result<Vec<_>, _>>()?;

let columns = names.iter().zip(data)
.map(|(name, cells)| {
.map(|(column_name, cells)| -> Result<Column, RustySheetError> {
let types = cells.iter()
.map(|cell| {
.map(|cell| -> Result<Option<ColumnType>, RustySheetError> {
if cell.kind == CellType::SharedString {
let id = cell.value.parse::<usize>().expect("Shared string index");
let index = mappings[&id];
let value = shared_strings[index].as_str();
ColumnType::from(if criteria.nulls.contains(value) {
&CellType::Empty
} else {
&cell.kind
}, value)
let value = resolve_shared_string(
&shared_strings,
&mappings,
&file_name,
&sheet_name,
cell,
)?;
Ok(ColumnType::from(
if criteria.nulls.contains(value) {
&CellType::Empty
} else {
&cell.kind
},
value,
))
} else {
ColumnType::from(&cell.kind, &cell.value)
Ok(ColumnType::from(&cell.kind, &cell.value))
}
})
.collect::<Vec<_>>();
Column {
name: name.to_owned(),
kind: presets.iter()
.find(|(pattern, _)| pattern.matches(name))
.collect::<Result<Vec<_>, _>>()?;
Ok(Column {
name: column_name.to_owned(),
kind: presets
.iter()
.find(|(pattern, _)| pattern.matches(column_name))
.map(|(_, kind)| kind.to_owned())
.unwrap_or(ColumnType::detect(types)),
}
})
})
.collect::<Vec<_>>();
.collect::<Result<Vec<_>, _>>()?;
tables.push(Table {
name,
name: sheet_name,
columns,
row_lower_bound,
col_lower_bound,
Expand Down
19 changes: 17 additions & 2 deletions src/spreadsheet/xls.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@ use crate::spreadsheet::cell::CellType;
use crate::spreadsheet::criteria::Criteria;
use crate::spreadsheet::excel::load_number_formats;
use crate::spreadsheet::reference::index_to_reference;
use crate::spreadsheet::resolve_number_format;
use crate::spreadsheet::sheet::Sheet;
use crate::spreadsheet::Spreadsheet;
use crate::spreadsheet::SpreadsheetError;
Expand Down Expand Up @@ -201,7 +202,14 @@ impl Spreadsheet for XlsSpreadsheet {
}
last_row = Some(row);
let index = self.reader.read_u16()? as usize;
let kind = self.number_formats[index];
let kind = resolve_number_format(
&self.number_formats,
&sheet.file_name,
&sheet.name,
row,
col,
index,
)?;
let value = self.reader.read_rk_number()?;
sheet.push(Cell {
row,
Expand Down Expand Up @@ -234,7 +242,14 @@ impl Spreadsheet for XlsSpreadsheet {
};
let kind = match either {
Either::Left(kind) => kind,
Either::Right(index) => self.number_formats[index],
Either::Right(index) => resolve_number_format(
&self.number_formats,
&sheet.file_name,
&sheet.name,
row,
col,
index,
)?,
};
if kind != CellType::Error {
if !criteria.nulls.contains(&value) {
Expand Down
12 changes: 10 additions & 2 deletions src/spreadsheet/xlsb.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@ use crate::spreadsheet::criteria::Criteria;
use crate::spreadsheet::excel;
use crate::spreadsheet::excel::load_relationships;
use crate::spreadsheet::reference::index_to_reference;
use crate::spreadsheet::resolve_number_format;
use crate::spreadsheet::sheet::Sheet;
use crate::spreadsheet::Spreadsheet;
use crate::spreadsheet::SpreadsheetError;
Expand Down Expand Up @@ -221,7 +222,14 @@ impl Spreadsheet for XlsbSpreadsheet {
};
let kind = match either {
Either::Left(kind) => kind,
Either::Right(index) => (self.number_formats)[index],
Either::Right(index) => resolve_number_format(
&self.number_formats,
&sheet.file_name,
&sheet.name,
row,
col,
index,
)?,
};
if kind != CellType::Error {
if !criteria.nulls.contains(&value) {
Expand Down Expand Up @@ -451,4 +459,4 @@ fn read_rk_cell(reader: &mut Biff12Reader<BufReader<ZipFile<UnifiedReader>>>) ->
};

(Either::Right(index), value)
}
}
Loading