pub mod custom; pub mod decompress; pub mod ffmpeg; pub mod fns; // pub mod pdfpages; pub mod spawning; pub mod sqlite; // pub mod tar; // pub mod tesseract; pub mod writing; pub mod zip; use crate::{config::RgaConfig, matching::*}; use anyhow::*; use custom::builtin_spawning_adapters; use custom::CustomAdapterConfig; use log::*; use std::borrow::Cow; use std::collections::HashMap; use std::io::prelude::*; use std::iter::Iterator; use std::path::{Path, PathBuf}; use std::rc::Rc; pub type ReadBox<'a> = Box; pub struct AdapterMeta { /// unique short name of this adapter (a-z0-9 only) pub name: String, /// version identifier. used to key cache entries, change if your output format changes pub version: i32, pub description: String, /// indicates whether this adapter can descend (=call rga_preproc again). if true, the cache key needs to include the list of active adapters pub recurses: bool, /// list of matchers (interpreted as a OR b OR ...) pub fast_matchers: Vec, /// list of matchers when we have mime type detection active (interpreted as ORed) /// warning: this *overrides* the fast matchers pub slow_matchers: Option>, /// if true, slow_matchers is merged with fast matchers if accurate is enabled /// for example, in sqlite you want this disabled since the db extension can mean other things and the mime type matching is very accurate for sqlite. /// but for tar you want it enabled, since the tar extension is very accurate but the tar mime matcher can have false negatives pub keep_fast_matchers_if_accurate: bool, // if true, adapter is only used when user lists it in `--rga-adapters` pub disabled_by_default: bool, } impl AdapterMeta { // todo: this is pretty ugly pub fn get_matchers<'a>( &'a self, slow: bool, ) -> Box> + 'a> { match ( slow, self.keep_fast_matchers_if_accurate, &self.slow_matchers, ) { (true, false, Some(ref sm)) => Box::new(sm.iter().map(|e| Cow::Borrowed(e))), (true, true, Some(ref sm)) => Box::new( sm.iter().map(|e| Cow::Borrowed(e)).chain( self.fast_matchers .iter() .map(|e| Cow::Owned(FileMatcher::Fast(e.clone()))), ), ), // don't have slow matchers or slow matching disabled (true, _, None) | (false, _, _) => Box::new( self.fast_matchers .iter() .map(|e| Cow::Owned(FileMatcher::Fast(e.clone()))), ), } } } pub trait GetMetadata { fn metadata(&self) -> &AdapterMeta; } pub trait FileAdapter: GetMetadata { /// adapt a file. /// /// detection_reason is the Matcher that was used to identify this file. Unless --rga-accurate was given, it is always a FastMatcher fn adapt<'a>(&self, a: AdaptInfo<'a>, detection_reason: &FileMatcher) -> Result>; } pub struct AdaptInfo<'a> { /// file path. May not be an actual file on the file system (e.g. in an archive). Used for matching file extensions. pub filepath_hint: PathBuf, /// true if filepath_hint is an actual file on the file system pub is_real_file: bool, /// depth at which this file is in archives. 0 for real filesystem pub archive_recursion_depth: i32, /// stream to read the file from. can be from a file or from some decoder pub inp: ReadBox<'a>, /// prefix every output line with this string to better indicate the file's location if it is in some archive pub line_prefix: String, pub config: RgaConfig, } /// (enabledAdapters, disabledAdapters) type AdaptersTuple = (Vec>, Vec>); pub fn get_all_adapters(custom_adapters: Option>) -> AdaptersTuple { // order in descending priority let mut adapters: Vec> = vec![]; if let Some(custom_adapters) = custom_adapters { for adapter_config in custom_adapters { adapters.push(Rc::new(adapter_config.to_adapter())); } } let internal_adapters: Vec> = vec![ Rc::new(ffmpeg::FFmpegAdapter::new()), Rc::new(zip::ZipAdapter::new()), Rc::new(decompress::DecompressAdapter::new()), // Rc::new(tar::TarAdapter::new()), Rc::new(sqlite::SqliteAdapter::new()), // Rc::new(pdfpages::PdfPagesAdapter::new()), // Rc::new(tesseract::TesseractAdapter::new()), ]; adapters.extend( builtin_spawning_adapters .iter() .map(|e| -> Rc { Rc::new(e.clone().to_adapter()) }), ); adapters.extend(internal_adapters); adapters .into_iter() .partition(|e| !e.metadata().disabled_by_default) } /** * filter adapters by given names: * * - "" means use default enabled adapter list * - "a,b" means use adapters a,b * - "-a,b" means use default list except for a and b * - "+a,b" means use default list but also a and b (a,b will be prepended to the list so given higher priority) */ pub fn get_adapters_filtered>( custom_adapters: Option>, adapter_names: &Vec, ) -> Result>> { let (def_enabled_adapters, def_disabled_adapters) = get_all_adapters(custom_adapters); let adapters = if !adapter_names.is_empty() { let adapters_map: HashMap<_, _> = def_enabled_adapters .iter() .chain(def_disabled_adapters.iter()) .map(|e| (e.metadata().name.clone(), e.clone())) .collect(); let mut adapters = vec![]; let mut subtractive = false; let mut additive = false; for (i, name) in adapter_names.iter().enumerate() { let mut name = name.as_ref(); if i == 0 && (name.starts_with('-')) { subtractive = true; name = &name[1..]; adapters = def_enabled_adapters.clone(); } else if i == 0 && (name.starts_with('+')) { name = &name[1..]; adapters = def_enabled_adapters.clone(); additive = true; } if subtractive { let inx = adapters .iter() .position(|a| a.metadata().name == name) .ok_or_else(|| format_err!("Could not remove {}: Not in list", name))?; adapters.remove(inx); } else { let adapter = adapters_map .get(name) .ok_or_else(|| format_err!("Unknown adapter: \"{}\"", name))? .clone(); if additive { adapters.insert(0, adapter); } else { adapters.push(adapter); } } } adapters } else { def_enabled_adapters }; debug!( "Chosen available adapters: {}", adapters .iter() .map(|a| a.metadata().name.clone()) .collect::>() .join(",") ); Ok(adapters) }