bhf: complete extraction

README.md

@@ -38,6 +38,7 @@ FLAGS:
 SUBCOMMANDS:
     bhd     Extracts BHD/BDT contents
     bhds    Extracts all BHD/BDT content (alphabetically) in a folder
+    bhf     Extracts BHF/BDT contents
     bnd     Extracts BND contents
     dcx     Extracts and decompress DCX data
     hash    Calculates hash for a string
@@ -52,6 +53,7 @@ Features
 - BHD5 / BDT: extraction from disk to disk.
 - DCX: decompression from disk to disk.
 - BND (v3): extraction from disk/memory to disk/memory.
+- BHF (v3): extraction from disk/memory to disk/memory.
 
 Repacking is not supported, maybe one day. It is not that useful when using
 [UDSFM][udsfm] and [Yabber][yabber], but if you really need it you can check out

src/bin/ironring.rs

@@ -73,6 +73,21 @@ fn main() {
                 .long("force")
                 .takes_value(false)
                 .required(false)))
+        .subcommand(SubCommand::with_name("bhf")
+            .about("Extracts BHF/BDT contents")
+            .arg(Arg::with_name("file")
+                .takes_value(true)
+                .required(true))
+            .arg(Arg::with_name("output")
+                .short("o")
+                .long("output")
+                .takes_value(true)
+                .required(false))
+            .arg(Arg::with_name("overwrite")
+                .short("f")
+                .long("force")
+                .takes_value(false)
+                .required(false)))
         .get_matches();
 
     process::exit(match matches.subcommand() {
@@ -81,6 +96,7 @@ fn main() {
         ("hash", Some(s)) => { cmd_hash(s) }
         ("dcx", Some(s)) => { cmd_dcx(s) }
         ("bnd", Some(s)) => { cmd_bnd(s) }
+        ("bhf", Some(s)) => { cmd_bhf(s) }
         _ => { 0 }
     })
 }
@@ -215,6 +231,10 @@ fn cmd_bnd(args: &ArgMatches) -> i32 {
 
 fn cmd_bhf(args: &ArgMatches) -> i32 {
     let file_path: &str = args.value_of("file").unwrap();
-    let output_path: &str = args.value_of("output").unwrap();
-    0
+    let output_path: Option<&str> = args.value_of("output");
+    let overwrite: bool = args.is_present("overwrite");
+    match unpackers::bhf::extract_bhf_file(file_path, output_path, overwrite) {
+        Err(e) => { eprintln!("Failed to extract BHF: {:?}", e); return 1 }
+        _ => { 0 }
+    }
 }

src/unpackers/bhd.rs

@@ -20,6 +20,7 @@ pub fn extract_bhd(
     names: &HashMap<String, String>,
     output_path: &str
 ) -> Result<(), UnpackError> {
+    let bhd_path = path::Path::new(bhd_path);
     let bhd_data = utils_fs::open_file_to_vec(bhd_path)?;
     let bhd = match bhd::parse(&bhd_data) {
         Ok((_, bhd)) => bhd,
@@ -27,7 +28,7 @@ pub fn extract_bhd(
         e => return Err(UnpackError::Unknown(format!("Unknown error: {:?}", e)))
     };
 
-    let bdt_path = path::PathBuf::from(bhd_path).with_extension("bdt");
+    let bdt_path = bhd_path.to_path_buf().with_extension("bdt");
     let mut bdt_file = fs::File::open(bdt_path.to_str().unwrap())?;
 
     extract_files(&bhd, &mut bdt_file, &names, &output_path)?;

src/unpackers/bhf.rs

@@ -1,5 +1,5 @@
 use std::fs;
-use std::io::Read;
+use std::io::Write;
 use std::path;
 
 use nom::Err::{Error as NomError, Failure as NomFailure};
@@ -8,24 +8,49 @@ use crate::parsers::bhf;
 use crate::unpackers::errors::UnpackError;
 use crate::utils::fs as utils_fs;
 
-pub fn extract_bhf(bhf_path: &str) -> Result<(), UnpackError> {
-    let bhf_data = utils_fs::open_file_to_vec(bhf_path)?;
-    let bhf = match bhf::parse(&bhf_data) {
-        Ok((_, bhf)) => { bhf }
-        Err(NomError(e)) | Err(NomFailure(e)) => return Err(UnpackError::parsing_err("BHF", e.1)),
-        e => return Err(UnpackError::Unknown(format!("Unknown error: {:?}", e))),
-    };
+/// Extract BHF file and corresponding BDT contents to disk.
+///
+/// Wraps around `extract_bhf` to load the BHF file from disk.
+/// If output_dir is none, entries will be extracted relatively to the
+/// BHF path.
+pub fn extract_bhf_file(
+    bhf_path: &str,
+    output_dir: Option<&str>,
+    overwrite: bool
+) -> Result<(), UnpackError> {
+    let bhf = load_bhf_file(bhf_path)?;
 
-    let bdt_path = get_bdt_for_bhf(bhf_path);
-    if bdt_path.is_none() {
+    let bdt_path: path::PathBuf = if let Some(path) = get_bdt_for_bhf(bhf_path) {
+        if !path.exists() {
+            return Err(UnpackError::Naming(format!("Can't find BDT: {:?}", path)))
+        }
+        path
+    } else {
         return Err(UnpackError::Naming(format!("Can't find BDT for BHF: {}", bhf_path)))
-    }
-    
+    };
+    let bdt_data = utils_fs::open_file_to_vec(&bdt_path)?;
 
+    let output_dir: &str = if output_dir.is_none() {
+        let parent = path::Path::new(bhf_path).parent();
+        if parent.is_none() {
+            return Err(UnpackError::Naming(format!("Can't find BHF parent dir: {:?}", bhf_path)))
+        }
+        parent.unwrap().to_str().unwrap()  // Conversion should not fail as bhf_path is valid.
+    } else {
+        output_dir.unwrap()
+    };
+    extract_bhf(&bhf, &bdt_data, output_dir, overwrite)?;
     Ok(())
 }
 
-fn get_bdt_for_bhf(bhf_path: &str) -> Option<path::PathBuf> {
+/// Return corresponding BDT path for a BHF path.
+///
+/// Replaces the "bhd" suffix in extension wiuth "bdt". If the
+/// extension does not ends with "bhd", a warning is printed but does
+/// the extension is still appended with "bdt". If the path does not
+/// have an extension or is overall invalid, it returns None.
+/// It does not check if the file exists either.
+pub fn get_bdt_for_bhf(bhf_path: &str) -> Option<path::PathBuf> {
     let mut path = path::PathBuf::from(bhf_path);
     let ext = path.extension()?.to_str()?;
     if !ext.ends_with("bhd") {
@@ -36,6 +61,77 @@ fn get_bdt_for_bhf(bhf_path: &str) -> Option<path::PathBuf> {
     Some(path)
 }
 
+/// Extract BHF+BDT contents to disk.
+///
+/// Files are written in output_dir, creating it if needed, without
+/// preserving directory structure.
+pub fn extract_bhf(
+    bhf: &bhf::Bhf,
+    bdt_data: &Vec<u8>,
+    output_dir: &str,
+    overwrite: bool,
+) -> Result<(), UnpackError> {
+    let output_dir = path::Path::new(output_dir);
+    utils_fs::ensure_dir_exists(output_dir)?;
+    for file_info in &bhf.file_infos {
+        // Extract all entries, print but ignore path errors.
+        match extract_bhf_entry(file_info, bdt_data, output_dir, overwrite) {
+            Err(UnpackError::Naming(e)) => { eprintln!("{}", e) }
+            _ => {}
+        }
+    }
+    Ok(())
+}
+
+
+/// Extract a file contained in a BHF+BDT using its BhfFileInfo.
+///
+/// The info struct must have a valid internal path.
+pub fn extract_bhf_entry(
+    file_info: &bhf::BhfFileInfo,
+    bdt_data: &Vec<u8>,
+    output_dir: &path::Path,
+    overwrite: bool,
+) -> Result<(), UnpackError> {
+    if file_info.path.is_none() {
+        return Err(UnpackError::Naming("No path for BHF entry.".to_owned()))
+    }
+
+    let ofs_start = file_info.ofs_data as usize;
+    let ofs_end = ofs_start + file_info.size as usize;
+    let data = &bdt_data[ofs_start..ofs_end];
+
+    let internal_path = file_info.path.to_owned().unwrap();
+    let file_name = internal_path.trim_start_matches('\\');
+    let mut file_path = output_dir.to_path_buf();
+    file_path.push(file_name);
+    if !overwrite && file_path.exists() {
+        let existing = file_path.to_string_lossy();
+        return Err(UnpackError::Naming(format!("File already exists: {}", existing)))
+    }
+
+    let mut output_file = fs::File::create(file_path)?;
+    output_file.write_all(&data)?;
+    Ok(())
+}
+
+/// Load a BHF file from disk.
+///
+/// Wraps around `load_bhf` to load the BHF from disk.
+pub fn load_bhf_file(bhf_path: &str) -> Result<bhf::Bhf, UnpackError> {
+    let bhf_data = utils_fs::open_file_to_vec(path::Path::new(bhf_path))?;
+    load_bhf(&bhf_data)
+}
+
+/// Load a BHF file from a bytes slice.
+pub fn load_bhf(bhf_data: &[u8]) -> Result<bhf::Bhf, UnpackError> {
+    match bhf::parse(&bhf_data) {
+        Ok((_, bhf)) => Ok(bhf),
+        Err(NomError(e)) | Err(NomFailure(e)) => Err(UnpackError::parsing_err("BHF", e.1)),
+        e => Err(UnpackError::Unknown(format!("Unknown error: {:?}", e))),
+    }
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;

src/unpackers/bnd.rs

@@ -46,18 +46,18 @@ pub fn extract_bnd(
 /// Extract a file contained in a BND using its BndFileInfo.
 ///
 /// The info struct must have a valid internal path.
-fn extract_bnd_entry(
+pub fn extract_bnd_entry(
     file_info: &bnd::BndFileInfo,
     bnd_data: &Vec<u8>,
     output_dir: &path::Path,
     overwrite: bool,
 ) -> Result<(), UnpackError> {
     if file_info.path.is_none() {
-        return Err(UnpackError::Naming("No path for BND entry.".to_owned()));
+        return Err(UnpackError::Naming("No path for BND entry.".to_owned()))
     }
 
     let ofs_start = file_info.ofs_data as usize;
-    let ofs_end = (file_info.ofs_data + file_info.size) as usize;
+    let ofs_end = ofs_start + file_info.size as usize;
     let data = &bnd_data[ofs_start..ofs_end];
 
     let internal_path = file_info.path.to_owned().unwrap();
@@ -69,10 +69,11 @@ fn extract_bnd_entry(
     };
     let mut file_path = output_dir.to_path_buf();
     file_path.push(file_name);
-
     if !overwrite && file_path.exists() {
-        return Err(UnpackError::Naming(format!("File already exists: {:?}", file_path)));
+        let existing = file_path.to_string_lossy();
+        return Err(UnpackError::Naming(format!("File already exists: {}", existing)))
     }
+
     let mut output_file = fs::File::create(file_path)?;
     output_file.write_all(&data)?;
     Ok(())
@@ -83,16 +84,15 @@ fn extract_bnd_entry(
 /// Wraps around `load_bnd` to load the BND from disk. It returns the
 /// parsed BND metadata and the whole file as a byte vector.
 pub fn load_bnd_file(bnd_path: &str) -> Result<(bnd::Bnd, Vec<u8>), UnpackError> {
-    let bnd_data = utils_fs::open_file_to_vec(bnd_path)?;
+    let bnd_data = utils_fs::open_file_to_vec(path::Path::new(bnd_path))?;
     Ok((load_bnd(&bnd_data)?, bnd_data))
 }
 
 /// Load a BND file from a bytes slice.
 pub fn load_bnd(bnd_data: &[u8]) -> Result<bnd::Bnd, UnpackError> {
-    let (_, bnd) = match bnd::parse(bnd_data) {
-        Ok(result) => result,
-        Err(NomError(e)) | Err(NomFailure(e)) => return Err(UnpackError::parsing_err("BND", e.1)),
-        e => return Err(UnpackError::Unknown(format!("Unknown error: {:?}", e))),
-    };
-    Ok(bnd)
+    match bnd::parse(bnd_data) {
+        Ok((_, result)) => Ok(result),
+        Err(NomError(e)) | Err(NomFailure(e)) => Err(UnpackError::parsing_err("BND", e.1)),
+        e => Err(UnpackError::Unknown(format!("Unknown error: {:?}", e))),
+    }
 }

src/unpackers/dcx.rs

@@ -1,5 +1,6 @@
 use std::fs;
 use std::io::{Read, Write};
+use std::path;
 
 use flate2::read::ZlibDecoder;
 use nom::Err::{Error as NomError, Failure as NomFailure};
@@ -18,6 +19,7 @@ pub fn extract_dcx(dcx_path: &str, output_path: &str) -> Result<(), UnpackError>
 
 /// Load a DCX file in memory along with its decompressed content.
 pub fn load_dcx(dcx_path: &str) -> Result<(dcx::Dcx, Vec<u8>), UnpackError> {
+    let dcx_path = path::Path::new(dcx_path);
     let dcx_data = utils_fs::open_file_to_vec(dcx_path)?;
     let (data, dcx) = match dcx::parse(&dcx_data) {
         Ok(result) => result,

src/utils/fs.rs

@@ -25,7 +25,7 @@ pub fn strip_extension(path: &path::PathBuf) -> Option<path::PathBuf> {
 }
 
 /// Open a binary file and read it to the end in a byte vector.
-pub fn open_file_to_vec(path: &str) -> Result<Vec<u8>, io::Error> {
+pub fn open_file_to_vec(path: &path::Path) -> Result<Vec<u8>, io::Error> {
     let mut file = fs::File::open(path)?;
     let file_len = file.metadata()?.len() as usize;
     let mut data = vec![0u8; file_len];