gef-file-to-map 0.1.1__tar.gz → 0.2.0__tar.gz

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/CHANGELOG.md

@@ -2,13 +2,44 @@
 
 All notable changes to this project will be documented in this file.
 
+## [0.2.0] - 2025-03-10
+
+### Bug Fixes
+
+#### Deps
+
+- Update nom to 8
+
+
+### Refactor
+
+#### Deps
+
+- Migrate nom to winnow 0.3
+
+#### Parsing
+
+- [**breaking**] Improve parsing of headers and add better error reporting
+
+
 ## [0.1.1] - 2025-03-10
 
+### Bug Fixes
+
+#### Ci
+
+- Fix publishing CI
+
+
 ### Documentation
 
 
 ### Miscellaneous Tasks
 
+#### Ci
+
+- Only run release CI on tags
+
 
 ## [0.1.0] - 2023-02-22
 

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/Cargo.lock

@@ -22,12 +22,12 @@ checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719"
 
 [[package]]
 name = "gef-file-to-map"
-version = "0.1.1"
+version = "0.2.0"
 dependencies = [
  "itertools",
- "nom",
  "pyo3",
  "thiserror",
+ "winnow",
 ]
 
 [[package]]
@@ -72,22 +72,6 @@ dependencies = [
  "autocfg",
 ]
 
-[[package]]
-name = "minimal-lexical"
-version = "0.2.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
-
-[[package]]
-name = "nom"
-version = "7.1.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
-dependencies = [
- "memchr",
- "minimal-lexical",
-]
-
 [[package]]
 name = "once_cell"
 version = "1.20.3"
@@ -229,3 +213,12 @@ name = "unindent"
 version = "0.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7264e107f553ccae879d21fbea1d6724ac785e8c3bfc762137959b5802826ef3"
+
+[[package]]
+name = "winnow"
+version = "0.7.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0e7f4ea97f6f78012141bcdb6a216b2609f0979ada50b20ca5b52dde2eac2bb1"
+dependencies = [
+ "memchr",
+]

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "gef-file-to-map"
-version = "0.1.1"
+version = "0.2.0"
 edition = "2024"
 rust-version = "1.85.0"
 
@@ -11,9 +11,9 @@ crate-type = ["cdylib"]
 
 [dependencies]
 itertools = "0.14.0"
-nom = "7.1.3"
 pyo3 = { version = "0.24.0", features = ["extension-module", "abi3-py37"] }
 thiserror = "2.0.12"
+winnow = "0.7.3"
 
 [profile.release]
 lto = true

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gef-file-to-map
-Version: 0.1.1
+Version: 0.2.0
 Classifier: Programming Language :: Rust
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "gef-file-to-map"
-version = "0.1.1"
+version = "0.2.0"
 description = "Utility library for parsing .gef files"
 requires-python = ">=3.7"
 readme = "README.md"

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/src/error.rs

@@ -1,6 +1,6 @@
 //! Error primitives.
 
-use pyo3::{exceptions::PyException, PyErr};
+use pyo3::{PyErr, exceptions::PyException};
 use std::convert::From;
 
 /// Alias result to always return our own error type.

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/src/header.rs

@@ -1,12 +1,14 @@
 //! Header parsing.
 
-use crate::{
-    error::{Error, Result},
-    nom::IResult,
+use winnow::{
+    ascii::*,
+    combinator::*,
+    error::{StrContext, StrContextValue},
+    prelude::*,
+    token::*,
 };
 
-/// Separator used between values in the GEF file.
-const VALUE_SEPARATOR: char = ',';
+use crate::error::{Error, Result};
 
 /// References to the strings of the file where the header is and it's values.
 #[derive(Debug, Clone, PartialEq)]
@@ -27,69 +29,71 @@ impl<'a> Header<'a> {
     ///
     /// The string shouldn't contain a '#' character at the begin nor a newline
     /// at the end.
-    fn from_str(s: &'a str) -> IResult<'a, Self> {
-        // Get the name of the header, the left hand side
-        let (s, name) = nom::error::context(
-            "the name of the header",
-            nom::character::complete::alphanumeric1,
-        )(s)?;
-
-        let (s, values) = nom::sequence::preceded(
-            // Take all whitespace between the column header name and the = symbol
-            nom::character::complete::space0,
-            // Get the values between the '=' char and the end of the line
-            nom::sequence::preceded(
-                // Take all spaces and = characters
-                nom::sequence::preceded(
-                    nom::character::complete::char('='),
-                    nom::character::complete::space0,
-                ),
-                nom::error::context(
-                    "the header values",
-                    // Get the comma-space separated values
-                    nom::multi::separated_list0(
-                        // Get until the ',' character and trim the spaces
-                        nom::sequence::delimited(
-                            nom::character::complete::space0,
-                            nom::character::complete::char(VALUE_SEPARATOR),
-                            nom::character::complete::space0,
-                        ),
-                        // Take until the end of the line or until a separator is found
-                        nom::bytes::complete::take_till(|c: char| {
-                            c == VALUE_SEPARATOR || c.is_control()
-                        }),
-                    ),
-                ),
-            ),
-        )(s)?;
-
-        // Make empty lists actually empty
-        let values = if values == [""] { Vec::new() } else { values };
-
-        Ok((s, Self { name, values }))
+    fn from_str(input: &mut &'a str) -> winnow::Result<Self> {
+        // Take the hash symbol at the beginning and the whitespace
+        ('#', space0)
+            .context(StrContext::Label("header hash"))
+            .parse_next(input)?;
+
+        // Take the name
+        let name = alphanumeric1
+            .context(StrContext::Label("header name"))
+            .parse_next(input)?;
+
+        // Ignore the '=' symbol, although it is required
+        (space0, '=', space0)
+            .context(StrContext::Label("header equality symbol"))
+            .parse_next(input)?;
+
+        // Get the comma-space separated values
+        let mut values: Vec<&str> = separated(
+            0..,
+            // Get until the ',' character and trim the spaces
+            delimited(space0, take_till(1.., (',', '\r', '\n')), space0)
+                .context(StrContext::Label("header value")),
+            ',',
+        )
+        .context(StrContext::Label("header values"))
+        .context(StrContext::Expected(StrContextValue::Description(
+            "comma separated list of values",
+        )))
+        .parse_next(input)?;
+
+        // Remove empty values
+        if values == [""] {
+            values.clear();
+        }
+
+        // Take the newline
+        (space0, take_while(0.., ('\n', '\r'))).parse_next(input)?;
+
+        Ok(Self { name, values })
     }
 }
 
 /// Parse the headers of the GEF file.
 ///
 /// Return the parsed headers and a reference to the rest of the file.
-pub(crate) fn parse_headers(gef: &'_ str) -> Result<(&'_ str, Vec<Header<'_>>)> {
-    nom::sequence::preceded(
-        // Ignore the whitespace before the first header line
-        nom::character::complete::multispace0,
-        // Loop over all sequences starting with # until the newline character
-        nom::multi::many0(nom::error::context(
-            "a header line",
-            nom::sequence::delimited(
-                nom::character::complete::char('#'),
-                Header::from_str,
-                // Allow multiple lines
-                nom::multi::many1(nom::character::complete::line_ending),
-            ),
-        )),
-    )(gef)
-    // Convert the nom error to our own error type
-    .map_err(|err| Error::Parsing(err.to_string()))
+pub(crate) fn parse_headers(mut gef: &'_ str) -> Result<(&'_ str, Vec<Header<'_>>)> {
+    parse_headers_impl
+        .parse_next(&mut gef)
+        // Convert the winnow error to our own error type
+        .map_err(|err| Error::Parsing(err.to_string()))
+}
+
+/// Parse the list of headers implementation.
+pub(crate) fn parse_headers_impl<'a>(
+    input: &mut &'a str,
+) -> winnow::Result<(&'a str, Vec<Header<'a>>)> {
+    // Skip initial whitespace
+    multispace0.parse_next(input)?;
+
+    // Loop over all sequences starting with # until the newline character
+    let (values, _) = repeat_till(0.., Header::from_str, not('#'))
+        .context(StrContext::Label("headers"))
+        .parse_next(input)?;
+
+    Ok((input, values))
 }
 
 #[cfg(test)]
@@ -99,7 +103,7 @@ mod tests {
     #[test]
     fn test_single_header() {
         assert_eq!(
-            Header::from_str("A= 1").unwrap().1,
+            Header::from_str(&mut "#A= 1").unwrap(),
             Header {
                 name: "A",
                 values: vec!["1"]
@@ -107,7 +111,7 @@ mod tests {
         );
 
         assert_eq!(
-            Header::from_str("A = 1").unwrap().1,
+            Header::from_str(&mut "#A = 1").unwrap(),
             Header {
                 name: "A",
                 values: vec!["1"]
@@ -115,7 +119,7 @@ mod tests {
         );
 
         assert_eq!(
-            Header::from_str("A= 1, 2").unwrap().1,
+            Header::from_str(&mut "#A= 1, 2").unwrap(),
             Header {
                 name: "A",
                 values: vec!["1", "2"]
@@ -126,7 +130,7 @@ mod tests {
     #[test]
     fn test_empty_header() {
         assert_eq!(
-            Header::from_str("A=").unwrap().1,
+            Header::from_str(&mut "#A=").unwrap(),
             Header {
                 name: "A",
                 values: vec![]
@@ -134,7 +138,6 @@ mod tests {
         );
 
         let (rest, headers) = parse_headers("#A=\nrest").unwrap();
-        assert_eq!(rest, "rest");
         assert_eq!(
             headers,
             vec![Header {
@@ -142,6 +145,7 @@ mod tests {
                 values: vec![]
             },]
         );
+        assert_eq!(rest, "rest");
     }
 
     #[test]

{gef_file_to_map-0.1.1 → gef_file_to_map-0.2.0}/src/lib.rs

@@ -15,7 +15,6 @@
 
 pub mod error;
 mod header;
-mod nom;
 
 use itertools::Itertools;
 use pyo3::prelude::*;

gef_file_to_map-0.1.1/src/nom.rs

@@ -1,7 +0,0 @@
-//! Tools and utilities for the nom crate.
-
-use nom::error::VerboseError;
-
-/// Remove a lot of boilerplate for nom, use the verbose error type for span
-/// information.
-pub(crate) type IResult<'a, T> = nom::IResult<&'a str, T, VerboseError<&'a str>>;