Skip to content

Add config option to limit which string literal encodings are used #356

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

Merged
LagoLunatic merged 9 commits into from
Jun 1, 2026
Merged

Add config option to limit which string literal encodings are used #356

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
bb29126
Fix annoying rust-analyzer error with a cfg
LagoLunatic May 31, 2026
eda1a1b
Refactor literal info into a struct
LagoLunatic May 31, 2026
bc2b01d
Add field hide a literal from the GUI but still use for diffing
LagoLunatic May 31, 2026
b24327a
Add config option to limit string literal diffing
LagoLunatic May 31, 2026
bf72bc5
Fix deprecation warning `float_literal_f32_fallback`
LagoLunatic May 31, 2026
4aeb586
Hide non-preferred string encodings from the GUI
LagoLunatic May 31, 2026
6492515
Fix new cargo check warning
LagoLunatic May 31, 2026
4ce12fb
Clippy again
LagoLunatic May 31, 2026
16cc211
Rename "None" -> "Auto"
LagoLunatic Jun 1, 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
4 changes: 2 additions & 2 deletions objdiff-cli/src/cmd/diff.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -198,8 +198,8 @@ fn run_oneshot(
.transpose()?;
let result =
diff::diff_objs(target.as_ref(), base.as_ref(), None, &diff_config, &mapping_config)?;
let left = target.as_ref().and_then(|o| result.left.as_ref().map(|d| (o, d)));
let right = base.as_ref().and_then(|o| result.right.as_ref().map(|d| (o, d)));
let left = target.as_ref().zip(result.left.as_ref());
let right = base.as_ref().zip(result.right.as_ref());
let diff_result = DiffResult::new(left, right, &diff_config)?;
write_output(&diff_result, Some(output), output_format)?;
Ok(())
Expand Down
46 changes: 46 additions & 0 deletions objdiff-core/config-schema.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,6 +117,51 @@
"name": "Combine text sections",
"description": "Combines all text sections into one."
},
{
"id": "preferredStringEncoding",
"type": "choice",
"default": "auto",
"name": "Preferred string encoding",
"description": "Which encoding to use when diffing string literals used in functions.",
"items": [
{
"value": "auto",
"name": "Auto"
},
{
"value": "ascii",
"name": "ASCII"
},
{
"value": "utf_8",
"name": "UTF-8"
},
{
"value": "shift_jis",
"name": "Shift JIS"
},
{
"value": "windows_1252",
"name": "Windows-1252"
},
{
"value": "euc_jp",
"name": "EUC-JP"
},
{
"value": "big5",
"name": "Big5"
},
{
"value": "utf_16be",
"name": "UTF-16BE"
},
{
"value": "utf_16le",
"name": "UTF-16LE"
}
]
},
{
"id": "arm.archVersion",
"type": "choice",
Expand Down Expand Up @@ -338,6 +383,7 @@
"id": "general",
"name": "General",
"properties": [
"preferredStringEncoding",
"functionRelocDiffs",
"demangler",
"showSymbolSizes",
Expand Down
122 changes: 85 additions & 37 deletions objdiff-core/src/arch/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ use object::Endian as _;

use crate::{
diff::{
DiffObjConfig, DiffSide,
ConfigEnum, DiffObjConfig, DiffSide, PreferredStringEncoding,
display::{ContextItem, HoverItem, InstructionPart},
},
obj::{
Expand Down Expand Up @@ -52,6 +52,33 @@ const SUPPORTED_ENCODINGS_WITH_NULL_TERM: [(&encoding_rs::Encoding, &str); 5] =
const SUPPORTED_ENCODINGS_NO_NULL_TERM: [(&encoding_rs::Encoding, &str); 2] =
[(encoding_rs::UTF_16BE, "UTF-16BE"), (encoding_rs::UTF_16LE, "UTF-16LE")];

#[derive(Debug, Clone, Default, PartialEq)]
pub struct LiteralInfo {
pub literal: String,
pub label_override: Option<String>,
pub copy_string: Option<String>,
pub hidden: bool, // Only used when the user hasn't set a preferred string encoding
pub is_string: bool,
}

impl LiteralInfo {
pub fn hidden(&self, diff_config: Option<&DiffObjConfig>) -> bool {
let Some(diff_config) = diff_config else {
return self.hidden;
};
if !self.is_string {
return self.hidden;
}
if diff_config.preferred_string_encoding == PreferredStringEncoding::Auto {
return self.hidden;
}
let Some(ref label) = self.label_override else {
return self.hidden;
};
*label != diff_config.preferred_string_encoding.name()
}
}

/// Represents the type of data associated with an instruction
#[derive(PartialEq)]
pub enum DataType {
Expand Down Expand Up @@ -83,18 +110,14 @@ impl fmt::Display for DataType {
impl DataType {
pub fn display_labels(&self, endian: object::Endianness, bytes: &[u8]) -> Vec<String> {
let mut strs = Vec::new();
for (literal, label_override, _) in self.display_literals(endian, bytes) {
let label = label_override.unwrap_or_else(|| self.to_string());
strs.push(format!("{label}: {literal:?}"))
for lit_info in self.display_literals(endian, bytes) {
let label = lit_info.label_override.unwrap_or_else(|| self.to_string());
strs.push(format!("{}: {:?}", label, lit_info.literal))
}
strs
}

pub fn display_literals(
&self,
endian: object::Endianness,
bytes: &[u8],
) -> Vec<(String, Option<String>, Option<String>)> {
pub fn display_literals(&self, endian: object::Endianness, bytes: &[u8]) -> Vec<LiteralInfo> {
let mut strs = Vec::new();
if self.required_len().is_some_and(|l| bytes.len() < l) {
log::warn!(
Expand All @@ -118,60 +141,70 @@ impl DataType {
match self {
DataType::Int8 => {
let i = i8::from_ne_bytes(bytes.try_into().unwrap());
strs.push((format!("{i:#x}"), None, None));
strs.push(LiteralInfo { literal: format!("{i:#x}"), ..Default::default() });

if i < 0 {
strs.push((format!("{:#x}", ReallySigned(i)), None, None));
strs.push(LiteralInfo {
literal: format!("{:#x}", ReallySigned(i)),
..Default::default()
});
}
}
DataType::Int16 => {
let i = endian.read_i16_bytes(bytes.try_into().unwrap());
strs.push((format!("{i:#x}"), None, None));
strs.push(LiteralInfo { literal: format!("{i:#x}"), ..Default::default() });

if i < 0 {
strs.push((format!("{:#x}", ReallySigned(i)), None, None));
strs.push(LiteralInfo {
literal: format!("{:#x}", ReallySigned(i)),
..Default::default()
});
}
}
DataType::Int32 => {
let i = endian.read_i32_bytes(bytes.try_into().unwrap());
strs.push((format!("{i:#x}"), None, None));
strs.push(LiteralInfo { literal: format!("{i:#x}"), ..Default::default() });

if i < 0 {
strs.push((format!("{:#x}", ReallySigned(i)), None, None));
strs.push(LiteralInfo {
literal: format!("{:#x}", ReallySigned(i)),
..Default::default()
});
}
}
DataType::Int64 => {
let i = endian.read_i64_bytes(bytes.try_into().unwrap());
strs.push((format!("{i:#x}"), None, None));
strs.push(LiteralInfo { literal: format!("{i:#x}"), ..Default::default() });

if i < 0 {
strs.push((format!("{:#x}", ReallySigned(i)), None, None));
strs.push(LiteralInfo {
literal: format!("{:#x}", ReallySigned(i)),
..Default::default()
});
}
}
DataType::Float => {
let bytes: [u8; 4] = bytes.try_into().unwrap();
strs.push((
format!("{:?}f", match endian {
strs.push(LiteralInfo {
literal: format!("{:?}f", match endian {
object::Endianness::Little => f32::from_le_bytes(bytes),
object::Endianness::Big => f32::from_be_bytes(bytes),
}),
None,
None,
));
..Default::default()
});
}
DataType::Double => {
let bytes: [u8; 8] = bytes.try_into().unwrap();
strs.push((
format!("{:?}", match endian {
strs.push(LiteralInfo {
literal: format!("{:?}", match endian {
object::Endianness::Little => f64::from_le_bytes(bytes),
object::Endianness::Big => f64::from_be_bytes(bytes),
}),
None,
None,
));
..Default::default()
});
}
DataType::Bytes => {
strs.push((format!("{bytes:#?}"), None, None));
strs.push(LiteralInfo { literal: format!("{bytes:#?}"), ..Default::default() });
}
DataType::String => {
if let Some(nul_idx) = bytes.iter().position(|&c| c == b'\0') {
Expand All @@ -181,15 +214,28 @@ impl DataType {
if !had_errors && cow.is_ascii() {
let string = format!("{cow}");
let copy_string = escape_special_ascii_characters(&string);
strs.push((string, Some("ASCII".into()), Some(copy_string)));
strs.push(LiteralInfo {
literal: string,
label_override: Some("ASCII".into()),
copy_string: Some(copy_string),
hidden: false,
is_string: true,
});
}
for (encoding, encoding_name) in SUPPORTED_ENCODINGS_WITH_NULL_TERM {
let (cow, _, had_errors) = encoding.decode(str_bytes);
// Avoid showing ASCII-only strings more than once if the encoding is ASCII-compatible.
if !had_errors && (!encoding.is_ascii_compatible() || !cow.is_ascii()) {
if !had_errors {
let string = format!("{cow}");
let copy_string = escape_special_ascii_characters(&string);
strs.push((string, Some(encoding_name.into()), Some(copy_string)));
// Avoid showing ASCII-only strings more than once if the encoding is ASCII-compatible.
let hidden = encoding.is_ascii_compatible() && cow.is_ascii();
strs.push(LiteralInfo {
literal: string,
label_override: Some(encoding_name.into()),
copy_string: Some(copy_string),
hidden,
is_string: true,
});
}
}
}
Expand All @@ -202,11 +248,13 @@ impl DataType {
let trimmed = cow.trim_end_matches('\0');
if !trimmed.is_empty() {
let copy_string = escape_special_ascii_characters(trimmed);
strs.push((
trimmed.to_string(),
Some(encoding_name.into()),
Some(copy_string),
));
strs.push(LiteralInfo {
literal: trimmed.to_string(),
label_override: Some(encoding_name.into()),
copy_string: Some(copy_string),
hidden: false,
is_string: true,
});
}
}
}
Expand Down
27 changes: 23 additions & 4 deletions objdiff-core/src/diff/code.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@ use anyhow::{Context, Result, anyhow, ensure};

use super::{
DiffObjConfig, FunctionRelocDiffs, InstructionArgDiffIndex, InstructionBranchFrom,
InstructionBranchTo, InstructionDiffKind, InstructionDiffRow, SymbolDiff,
display::display_ins_data_literals,
InstructionBranchTo, InstructionDiffKind, InstructionDiffRow, PreferredStringEncoding,
SymbolDiff, display::display_ins_data_literals,
};
use crate::obj::{
InstructionArg, InstructionArgValue, InstructionRef, Object, ResolvedInstructionRef,
Expand Down Expand Up @@ -296,6 +296,26 @@ pub(crate) fn section_name_eq(
})
}

fn ins_data_literals_eq(
left_obj: &Object,
right_obj: &Object,
left_ins: ResolvedInstructionRef,
right_ins: ResolvedInstructionRef,
diff_config: &DiffObjConfig,
) -> bool {
let mut left_literals = display_ins_data_literals(left_obj, left_ins);
let mut right_literals = display_ins_data_literals(right_obj, right_ins);
if left_literals == right_literals {
return true;
}
if diff_config.preferred_string_encoding == PreferredStringEncoding::Auto {
return left_literals == right_literals;
}
left_literals.retain(|lit_info| !lit_info.hidden(Some(diff_config)));
right_literals.retain(|lit_info| !lit_info.hidden(Some(diff_config)));
left_literals == right_literals
}

fn reloc_eq(
left_obj: &Object,
right_obj: &Object,
Expand Down Expand Up @@ -330,8 +350,7 @@ fn reloc_eq(
&& (diff_config.function_reloc_diffs == FunctionRelocDiffs::NameAddress
|| left_reloc.symbol.kind != SymbolKind::Object
|| right_reloc.symbol.size == 0 // Likely a pool symbol like ...data, don't treat this as a diff
|| display_ins_data_literals(left_obj, left_ins)
== display_ins_data_literals(right_obj, right_ins))
|| ins_data_literals_eq(left_obj, right_obj, left_ins, right_ins, diff_config))
}
(Some(_), None) | (None, Some(_)) | (None, None) => symbol_name_addend_matches,
}
Expand Down
Loading
Loading