fif/src/formats.rs

//! The various formats that [fif](crate) can output to.

use std::ffi::OsStr;
use std::io::{self, Write};
#[cfg(unix)]
use std::os::unix::ffi::OsStrExt;
use std::path::Path;

use cfg_if::cfg_if;
use snailquote::escape;

use crate::findings::ScanError;
use crate::parameters::VERSION;
use crate::{Findings, BACKEND};
use itertools::Itertools;

/// A macro for creating an array of `Writable`s without needing to pepper your code with `into()`s.
/// # Usage
/// ```
/// let f = std::io::stdout();
/// // Instead of...
/// smart_write(f, &["hello".into(), Writable::Newline]);
/// // ...just use:
/// smart_write(f, writables!["hello", Newline]);
/// ```
#[macro_export]
macro_rules! writables {
	[$($args:tt),+] => {
		&[$(writables!(@do $args),)*]
	};

	(@do Newline) => {
		$crate::formats::Writable::Newline
	};

	(@do Space) => {
		$crate::formats::Writable::Space
	};

	(@do $arg:expr) => {
		$arg.into()
	}
}

#[macro_export]
/// Does the same thing as [writables], but adds a Newline to the end.
macro_rules! writablesln {
	[$($args:tt),+] => {
		&[$(writables!(@do $args),)* writables!(@do Newline)]
	};
}

#[doc(hidden)]
type Entries<'a> = [Result<Findings<'a>, ScanError<'a>>];

#[derive(Debug, PartialEq)]
pub enum Writable<'a> {
	String(&'a str),
	Path(&'a Path),
	Space,
	Newline,
}

// the lifetime of a lifetime
impl<'a> From<&'a str> for Writable<'a> {
	fn from(s: &'a str) -> Writable<'a> { Writable::String(s) }
}

impl<'a> From<&'a Path> for Writable<'a> {
	fn from(p: &'a Path) -> Writable<'a> { Writable::Path(p) }
}

impl<'a> From<&'a OsStr> for Writable<'a> {
	fn from(p: &'a OsStr) -> Writable<'a> { Writable::Path(p.as_ref()) }
}

fn generated_by() -> String { format!("Generated by fif {} ({} backend)", VERSION.unwrap_or("???"), BACKEND) }

fn smart_write<W: Write>(f: &mut W, writeables: &[Writable]) -> io::Result<()> {
	// ehhhh
	for writeable in writeables {
		match writeable {
			Writable::Space => write!(f, " ")?,
			Writable::Newline => {
				cfg_if! {
					if #[cfg(windows)] {
						write!(f, "\r\n")?;
					} else {
						writeln!(f,)?;
					}
				}
			}
			Writable::String(s) => write!(f, "{}", s)?,
			Writable::Path(path) => {
				if let Some(path_str) = path.to_str() {
					let escaped = escape(path_str);
					if escaped.as_ref() == path_str {
						// the escaped string is the same as the input - this will occur for inputs like "file.txt" which don't
						// need to be escaped. however, it's Best Practice™ to escape such strings anyway, so we prefix/suffix the
						// escaped string with single quotes.
						write!(f, "'{}'", escaped)?;
					} else {
						write!(f, "{}", escaped)?;
					}
				} else {
					write!(f, "'")?;
					cfg_if! {
						if #[cfg(windows)] {
							// TODO: implement bonked strings for windows
							// something like:
							// f.write_all(&*path.as_os_str().encode_wide().collect::<Vec<u16>>())?;
							write!(f, "{}", path.as_os_str().to_string_lossy())?;
						} else {
							f.write_all(&*path.as_os_str().as_bytes())?;
						}
					}
					write!(f, "'")?;
				}
			}
		}
	}
	Ok(())
}

// TODO: this might need a restructure.
// it would be nice if i didn't have to write a case for every OutputFormat variant that looked like
// OutputFormat::PowerShell => PowerShell::new().write_all(...)
// also, JSON's implementation differs vastly from PowerShell and Shell's implementations. Maybe they shouldn't be
// treated as implementing the same trait, since in that case, the format trait is more of a concept rather than an
// actual definition of behaviour.
// structuring code is *hard*
pub trait Format {
	fn new() -> Self;
	fn rename<W: Write>(&self, _f: &mut W, _from: &Path, _to: &Path) -> io::Result<()> { unreachable!() }
	fn no_known_extension<W: Write>(&self, _f: &mut W, _path: &Path) -> io::Result<()> { unreachable!() }
	fn unreadable<W: Write>(&self, _f: &mut W, _path: &Path) -> io::Result<()> { unreachable!() }
	fn unknown_type<W: Write>(&self, _f: &mut W, _path: &Path) -> io::Result<()> { unreachable!() }
	fn header<W: Write>(&self, _f: &mut W, _entries: &Entries) -> io::Result<()> { unreachable!() }
	fn footer<W: Write>(&self, _f: &mut W, _entries: &Entries) -> io::Result<()> { unreachable!() }

	fn write_all<W: Write>(&self, f: &mut W, entries: &Entries) -> io::Result<()> {
		// TODO: clean this up - it's kinda messy
		self.header(f, entries)?;

		// output will be generated in the order:
		// - files that couldn't be read
		// - files with no known mime type
		// - files with no known extension
		// - files with a known extension
		// files that already have a correct extension won't be represented in the output.

		// sort errors so unreadable files appear before files with unknown mimetypes - ScanError impls Ord such that
		// ScanError::File > ScanError::Mime
		let errors = entries.iter().filter_map(|e| e.as_ref().err()).sorted();
		// sort files so that files with no known extension come before those with known extensions - None > Some("jpg")
		let findings = entries
			.iter()
			.filter_map(|e| e.as_ref().ok())
			.sorted_by(|a, b| b.recommended_extension().cmp(&a.recommended_extension()).reverse());

		for error in errors {
			match error {
				// failed to read the file
				ScanError::File(path) => self.unreadable(f, path)?,
				// file was read successfully, but we couldn't determine a mimetype
				ScanError::Mime(path) => self.unknown_type(f, path)?,
			}
		}

		for finding in findings {
			if let Some(ext) = finding.recommended_extension() {
				self.rename(f, finding.file, &finding.file.with_extension(ext.as_str()))?;
			} else {
				self.no_known_extension(f, finding.file)?;
			}
		}

		self.footer(f, entries)
	}
}

/// Bourne-Shell compatible script.
pub struct Shell {}

impl Format for Shell {
	fn new() -> Self { Self {} }

	fn rename<W: Write>(&self, f: &mut W, from: &Path, to: &Path) -> io::Result<()> {
		smart_write(f, writablesln!("mv -v -i -- ", from, Space, to))
	}

	fn no_known_extension<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["echo No known extension for ", path])
	}

	fn unreadable<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["# Failed to read", path])
	}

	fn unknown_type<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["# Failed to detect mime type for ", path])
	}

	fn header<W: Write>(&self, f: &mut W, _: &Entries) -> io::Result<()> {
		smart_write(
			f,
			writablesln!["#!/usr/bin/env sh", Newline, "# ", (generated_by().as_str())],
		)?;

		if let Ok(working_directory) = std::env::current_dir() {
			smart_write(f, writablesln!["# Run from ", (working_directory.as_path())])?;
		}

		smart_write(f, writablesln![Newline, "set -e", Newline])
	}

	fn footer<W: Write>(&self, f: &mut W, _: &Entries) -> io::Result<()> {
		smart_write(f, writablesln![Newline, "echo 'Done.'"])
	}
}

// PowerShell is a noun, not a type
#[allow(clippy::doc_markdown)]
/// PowerShell script.
pub struct PowerShell {}

impl Format for PowerShell {
	fn new() -> Self { Self {} }

	fn rename<W: Write>(&self, f: &mut W, from: &Path, to: &Path) -> io::Result<()> {
		// unfortunately there doesn't seem to be an equivalent of sh's `mv -i` -- passing the '-Confirm' flag will prompt
		// the user to confirm every single rename, and using Move-Item -Force will always overwrite without prompting.
		// there doesn't seem to be a way to rename the file, prompting only if the target already exists.
		smart_write(
			f,
			writablesln!["Rename-Item -Path ", from, " -NewName ", (to.file_name().unwrap())],
		)
	}

	fn no_known_extension<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(
			f,
			writablesln![
				"Write-Output @'",
				Newline,
				"No known extension for ",
				path,
				Newline,
				"'@"
			],
		)
	}

	fn unreadable<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(
			f,
			writablesln!["Write-Output @'", Newline, "Failed to read ", path, Newline, "'@"],
		)
	}

	fn unknown_type<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["<# Failed to detect mime type for ", path, " #>"])
	}

	fn header<W: Write>(&self, f: &mut W, _: &Entries) -> io::Result<()> {
		smart_write(
			f,
			writablesln!["#!/usr/bin/env pwsh", Newline, "<# ", (generated_by().as_str()), " #>"],
		)?;

		if let Ok(working_directory) = std::env::current_dir() {
			smart_write(f, writablesln!["<# Run from ", (working_directory.as_path()), " #>"])?;
		}

		smart_write(f, writables![Newline])
	}

	fn footer<W: Write>(&self, f: &mut W, _: &Entries) -> io::Result<()> {
		smart_write(f, writablesln![Newline, "Write-Output 'Done!'"])
	}
}

pub struct Text;
impl Format for Text {
	fn new() -> Self { Self {} }
	fn rename<W: Write>(&self, f: &mut W, from: &Path, to: &Path) -> io::Result<()> {
		smart_write(f, writablesln![from, " should be renamed to ", to])
	}

	fn no_known_extension<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["No known extension for ", path])
	}

	fn unreadable<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["Encountered IO error while accessing ", path])
	}

	fn unknown_type<W: Write>(&self, f: &mut W, path: &Path) -> io::Result<()> {
		smart_write(f, writablesln!["Couldn't determine type for ", path])
	}

	fn header<W: Write>(&self, f: &mut W, _entries: &Entries) -> io::Result<()> {
		smart_write(f, writablesln![(generated_by().as_str()), Newline])
	}

	fn footer<W: Write>(&self, f: &mut W, entries: &Entries) -> io::Result<()> {
		smart_write(
			f,
			writablesln![Newline, "Processed ", (entries.len().to_string().as_str()), " files"],
		)
	}
}

#[cfg(feature = "json")]
pub struct Json;

#[cfg(feature = "json")]
impl Format for Json {
	fn new() -> Self { Self {} }

	fn write_all<W: Write>(&self, f: &mut W, entries: &Entries) -> io::Result<()> {
		#[derive(serde::Serialize)]
		struct SerdeEntries<'a> {
			errors: &'a Vec<&'a ScanError<'a>>,
			findings: &'a Vec<&'a Findings<'a>>,
		}

		let result = serde_json::to_writer_pretty(
			f,
			&SerdeEntries {
				errors: &entries.iter().filter_map(|e| e.as_ref().err()).sorted().collect(),
				findings: &entries.iter().filter_map(|f| f.as_ref().ok()).sorted().collect(),
			},
		);

		if let Err(err) = result {
			log::error!("Error while serialising: {}", err);
			return Err(err.into());
		}

		Ok(())
	}
}