fix: skip leading bom if exist

fmt/src/document/factory.rs

@@ -19,7 +19,7 @@ use std::path::Path;
 use std::path::PathBuf;
 
 use anyhow::Context;
-use gix::date::time::format;
+use gix::date::time::CustomFormat;
 
 use crate::config::Mapping;
 use crate::document::Document;
@@ -79,15 +79,14 @@ impl DocumentFactory {
         properties.insert("hawkeye.core.filename".to_string(), filename);
 
         if let Some(attrs) = self.git_file_attrs.get(filepath) {
+            const YEAR_FORMAT: CustomFormat = CustomFormat::new("%Y");
             properties.insert(
                 "hawkeye.git.fileCreatedYear".to_string(),
-                // TODO(tisonkun): hack until git-date provide external constructor for
-                //  CustomFormat.
-                attrs.created_time.format(format::SHORT)[0..4].to_string(),
+                attrs.created_time.format(YEAR_FORMAT),
             );
             properties.insert(
                 "hawkeye.git.fileModifiedYear".to_string(),
-                attrs.modified_time.format(format::SHORT)[0..4].to_string(),
+                attrs.modified_time.format(YEAR_FORMAT),
             );
         }
 

fmt/src/header/parser.rs

@@ -75,7 +75,19 @@ fn find_first_position(
     file_content: &mut FileContent,
     header_def: &HeaderDef,
 ) -> usize {
+    const UTF8_BOM: [u8; 3] = [0xEF, 0xBB, 0xBF];
+
     let mut begin_pos = 0;
+
+    if let Some(l) = line.as_ref() {
+        // skip UTF-8 BOM if exists
+        if l.as_bytes().starts_with(&UTF8_BOM) {
+            log::debug!("Detected UTF-8 BOM for {}; skip", file_content);
+            begin_pos = 3;
+            file_content.reset_to(3);
+        }
+    }
+
     if header_def.skip_line_pattern.is_some() {
         // the format expect to find lines to be skipped
         while line
@@ -104,8 +116,17 @@ fn find_first_position(
             begin_pos = 0;
             file_content.reset();
             *line = file_content.next_line();
+
+            // recheck for UTF-8 BOM
+            if let Some(l) = line.as_ref() {
+                if l.as_bytes().starts_with(&UTF8_BOM) {
+                    begin_pos = 3;
+                    file_content.reset_to(3);
+                }
+            }
         }
     }
+
     begin_pos
 }
 

licenserc.toml

@@ -22,6 +22,7 @@ excludes = [
     "fmt/tests/content/**",
     "tests/load_header_path/**",
     "tests/regression_blank_line/**",
+    "tests/bom_issue/**",
 
     # Generated files
     ".github/workflows/release.yml",

tests/bom_issue/headerless_bom.cs

@@ -0,0 +1,11 @@
+using System;
+
+namespace Some.Fake.Namespace;
+
+public class SomeFakeClass
+{
+    public void SomeFakeMethod()
+    {
+        Console.WriteLine("Hello, World!");
+    }
+}

tests/bom_issue/headerless_bom.cs.expected

@@ -0,0 +1,15 @@
+// -------------------------------------------------------
+// Static Copyright
+// -------------------------------------------------------
+
+using System;
+
+namespace Some.Fake.Namespace;
+
+public class SomeFakeClass
+{
+    public void SomeFakeMethod()
+    {
+        Console.WriteLine("Hello, World!");
+    }
+}

tests/bom_issue/license.txt

@@ -0,0 +1 @@
+Static Copyright 

tests/bom_issue/licenserc.toml

@@ -0,0 +1,11 @@
+baseDir = "."
+headerPath = "license.txt"
+
+excludes = ["*.toml", "*.expected"]
+
+additionalHeaders = [
+    "style.toml"
+]
+
+[mapping.HASH_SOURCE_STYLE]
+extensions = ["cs"]

tests/bom_issue/style.toml

@@ -0,0 +1,9 @@
+[HASH_SOURCE_STYLE]
+firstLine = "// -------------------------------------------------------"
+endLine = "// -------------------------------------------------------\n"
+skipLinePattern = "^#!.*$"
+allowBlankLines = false
+multipleLines = false
+beforeEachLine = "// "
+firstLineDetectionPattern = "//\\s+\\-+"
+lastLineDetectionPattern = "//\\s+\\-+"

tests/it.py

@@ -38,3 +38,6 @@ def diff_files(file1, file2):
 
 subprocess.run([hawkeye, "format", "--fail-if-unknown", "--fail-if-updated=false", "--dry-run"], cwd=(basedir / "regression_blank_line"), check=True)
 diff_files(basedir / "regression_blank_line" / "main.rs.expected", basedir / "regression_blank_line" / "main.rs.formatted")
+
+subprocess.run([hawkeye, "format", "--fail-if-unknown", "--fail-if-updated=false", "--dry-run"], cwd=(basedir / "bom_issue"), check=True)
+diff_files(basedir / "bom_issue" / "headerless_bom.cs.expected", basedir / "bom_issue" / "headerless_bom.cs.formatted")