removed unnecessary test

Reimplement size-based filters using FeroxFilter trait

.github/pull_request_template.md

@@ -11,7 +11,7 @@ Long form explanations of most of the items below can be found in the [CONTRIBUT
 
 ## Static analysis checks
 - [ ] All rust files are formatted using `cargo fmt`
-- [ ] All `clippy` checks pass when running `cargo clippy --all-targets --all-features -- -D warnings -A clippy::unnecessary_unwrap`
+- [ ] All `clippy` checks pass when running `cargo clippy --all-targets --all-features -- -D warnings -A clippy::deref_addrof`
 - [ ] All existing tests pass
 
 ## Documentation

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "feroxbuster"
-version = "1.6.0"
+version = "1.6.3"
 authors = ["Ben 'epi' Risher <epibar052@gmail.com>"]
 license = "MIT"
 edition = "2018"

src/client.rs

@@ -32,31 +32,33 @@ pub fn initialize(
         .default_headers(header_map)
         .redirect(policy);
 
-    let client = if proxy.is_some() && !proxy.unwrap().is_empty() {
-        match Proxy::all(proxy.unwrap()) {
-            Ok(proxy_obj) => client.proxy(proxy_obj),
-            Err(e) => {
-                eprintln!(
-                    "{} {} Could not add proxy ({:?}) to Client configuration",
-                    status_colorizer("ERROR"),
-                    module_colorizer("Client::initialize"),
-                    proxy
-                );
-                eprintln!(
-                    "{} {} {}",
-                    status_colorizer("ERROR"),
-                    module_colorizer("Client::initialize"),
-                    e
-                );
+    let client = match proxy {
+        // a proxy is specified, need to add it to the client
+        Some(some_proxy) => {
+            if !some_proxy.is_empty() {
+                // it's not an empty string
+                match Proxy::all(some_proxy) {
+                    Ok(proxy_obj) => client.proxy(proxy_obj),
+                    Err(e) => {
+                        eprintln!(
+                            "{} {} {}",
+                            status_colorizer("ERROR"),
+                            module_colorizer("Client::initialize"),
+                            e
+                        );
 
-                #[cfg(test)]
-                panic!();
-                #[cfg(not(test))]
-                exit(1);
+                        #[cfg(test)]
+                        panic!();
+                        #[cfg(not(test))]
+                        exit(1);
+                    }
+                }
+            } else {
+                client // Some("") was used?
             }
         }
-    } else {
-        client
+        // no proxy specified
+        None => client,
     };
 
     match client.build() {

src/filters.rs

@@ -53,7 +53,7 @@ impl FeroxFilter for WildcardFilter {
     /// Examine size, dynamic, and content_len to determine whether or not the response received
     /// is a wildcard response and therefore should be filtered out
     fn should_filter_response(&self, response: &FeroxResponse) -> bool {
-        log::trace!("enter: should_filter_response({:?} {:?})", self, response);
+        log::trace!("enter: should_filter_response({:?} {})", self, response);
 
         // quick return if dont_filter is set
         if CONFIGURATION.dont_filter {
@@ -114,7 +114,7 @@ pub struct StatusCodeFilter {
 impl FeroxFilter for StatusCodeFilter {
     /// Check `filter_code` against what was passed in via -C|--filter-status
     fn should_filter_response(&self, response: &FeroxResponse) -> bool {
-        log::trace!("enter: should_filter_response({:?} {:?})", self, response);
+        log::trace!("enter: should_filter_response({:?} {})", self, response);
 
         if response.status().as_u16() == self.filter_code {
             log::debug!(
@@ -140,3 +140,152 @@ impl FeroxFilter for StatusCodeFilter {
         self
     }
 }
+
+/// Simple implementor of FeroxFilter; used to filter out responses based on the number of lines
+/// in a Response body; specified using -N|--filter-lines
+#[derive(Default, Debug, PartialEq)]
+pub struct LinesFilter {
+    /// Number of lines in a Response's body that should be filtered
+    pub line_count: usize,
+}
+
+/// implementation of FeroxFilter for LinesFilter
+impl FeroxFilter for LinesFilter {
+    /// Check `line_count` against what was passed in via -N|--filter-lines
+    fn should_filter_response(&self, response: &FeroxResponse) -> bool {
+        log::trace!("enter: should_filter_response({:?} {})", self, response);
+
+        let result = response.line_count() == self.line_count;
+
+        log::trace!("exit: should_filter_response -> {}", result);
+
+        result
+    }
+
+    /// Compare one LinesFilter to another
+    fn box_eq(&self, other: &dyn Any) -> bool {
+        other.downcast_ref::<Self>().map_or(false, |a| self == a)
+    }
+
+    /// Return self as Any for dynamic dispatch purposes
+    fn as_any(&self) -> &dyn Any {
+        self
+    }
+}
+
+/// Simple implementor of FeroxFilter; used to filter out responses based on the number of words
+/// in a Response body; specified using -W|--filter-words
+#[derive(Default, Debug, PartialEq)]
+pub struct WordsFilter {
+    /// Number of words in a Response's body that should be filtered
+    pub word_count: usize,
+}
+
+/// implementation of FeroxFilter for WordsFilter
+impl FeroxFilter for WordsFilter {
+    /// Check `word_count` against what was passed in via -W|--filter-words
+    fn should_filter_response(&self, response: &FeroxResponse) -> bool {
+        log::trace!("enter: should_filter_response({:?} {})", self, response);
+
+        let result = response.word_count() == self.word_count;
+
+        log::trace!("exit: should_filter_response -> {}", result);
+
+        result
+    }
+
+    /// Compare one WordsFilter to another
+    fn box_eq(&self, other: &dyn Any) -> bool {
+        other.downcast_ref::<Self>().map_or(false, |a| self == a)
+    }
+
+    /// Return self as Any for dynamic dispatch purposes
+    fn as_any(&self) -> &dyn Any {
+        self
+    }
+}
+
+/// Simple implementor of FeroxFilter; used to filter out responses based on the length of a
+/// Response body; specified using -S|--filter-size
+#[derive(Default, Debug, PartialEq)]
+pub struct SizeFilter {
+    /// Overall length of a Response's body that should be filtered
+    pub content_length: u64,
+}
+
+/// implementation of FeroxFilter for SizeFilter
+impl FeroxFilter for SizeFilter {
+    /// Check `content_length` against what was passed in via -S|--filter-size
+    fn should_filter_response(&self, response: &FeroxResponse) -> bool {
+        log::trace!("enter: should_filter_response({:?} {})", self, response);
+
+        let result = response.content_length() == self.content_length;
+
+        log::trace!("exit: should_filter_response -> {}", result);
+
+        result
+    }
+
+    /// Compare one SizeFilter to another
+    fn box_eq(&self, other: &dyn Any) -> bool {
+        other.downcast_ref::<Self>().map_or(false, |a| self == a)
+    }
+
+    /// Return self as Any for dynamic dispatch purposes
+    fn as_any(&self) -> &dyn Any {
+        self
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    /// just a simple test to increase code coverage by hitting as_any and the inner value
+    fn lines_filter_as_any() {
+        let filter = LinesFilter { line_count: 1 };
+
+        assert_eq!(filter.line_count, 1);
+        assert_eq!(
+            *filter.as_any().downcast_ref::<LinesFilter>().unwrap(),
+            filter
+        );
+    }
+
+    #[test]
+    /// just a simple test to increase code coverage by hitting as_any and the inner value
+    fn words_filter_as_any() {
+        let filter = WordsFilter { word_count: 1 };
+
+        assert_eq!(filter.word_count, 1);
+        assert_eq!(
+            *filter.as_any().downcast_ref::<WordsFilter>().unwrap(),
+            filter
+        );
+    }
+
+    #[test]
+    /// just a simple test to increase code coverage by hitting as_any and the inner value
+    fn size_filter_as_any() {
+        let filter = SizeFilter { content_length: 1 };
+
+        assert_eq!(filter.content_length, 1);
+        assert_eq!(
+            *filter.as_any().downcast_ref::<SizeFilter>().unwrap(),
+            filter
+        );
+    }
+
+    #[test]
+    /// just a simple test to increase code coverage by hitting as_any and the inner value
+    fn status_code_filter_as_any() {
+        let filter = StatusCodeFilter { filter_code: 200 };
+
+        assert_eq!(filter.filter_code, 200);
+        assert_eq!(
+            *filter.as_any().downcast_ref::<StatusCodeFilter>().unwrap(),
+            filter
+        );
+    }
+}

src/heuristics.rs

@@ -89,8 +89,10 @@ pub async fn wildcard_test(
 
                 if !CONFIGURATION.quiet {
                     let msg = format!(
-                            "{} {:>10} Wildcard response is dynamic; {} ({} + url length) responses; toggle this behavior by using {}\n",
+                            "{} {:>8}l {:>8}w {:>8}c Wildcard response is dynamic; {} ({} + url length) responses; toggle this behavior by using {}\n",
                             status_colorizer("WLD"),
+                            ferox_response.line_count(),
+                            ferox_response.word_count(),
                             wildcard.dynamic,
                             style("auto-filtering").yellow(),
                             style(wc_length - url_len).cyan(),
@@ -110,8 +112,10 @@ pub async fn wildcard_test(
 
                 if !CONFIGURATION.quiet {
                     let msg = format!(
-                        "{} {:>10} Wildcard response is static; {} {} responses; toggle this behavior by using {}\n",
+                        "{} {:>8}l {:>8}w {:>8}c Wildcard response is static; {} {} responses; toggle this behavior by using {}\n",
                         status_colorizer("WLD"),
+                        ferox_response.line_count(),
+                        ferox_response.word_count(),
                         wc_length,
                         style("auto-filtering").yellow(),
                         style(wc_length).cyan(),
@@ -235,7 +239,7 @@ async fn make_wildcard_request(
                         }
                     }
                 }
-                log::trace!("exit: make_wildcard_request -> {:?}", ferox_response);
+                log::trace!("exit: make_wildcard_request -> {}", ferox_response);
                 return Some(ferox_response);
             }
         }

src/lib.rs

@@ -8,13 +8,11 @@ pub mod logger;
 pub mod parser;
 pub mod progress;
 pub mod reporter;
+pub mod scan_manager;
 pub mod scanner;
 pub mod utils;
 
-use reqwest::{
-    header::HeaderMap,
-    {Response, StatusCode, Url},
-};
+use reqwest::{header::HeaderMap, Response, StatusCode, Url};
 use std::{error, fmt};
 use tokio::sync::mpsc::{UnboundedReceiver, UnboundedSender};
 
@@ -103,6 +101,19 @@ pub struct FeroxResponse {
     headers: HeaderMap,
 }
 
+/// Implement Display for FeroxResponse
+impl fmt::Display for FeroxResponse {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(
+            f,
+            "FeroxResponse {{ url: {}, status: {}, content-length: {} }}",
+            self.url(),
+            self.status(),
+            self.content_length()
+        )
+    }
+}
+
 /// `FeroxResponse` implementation
 impl FeroxResponse {
     /// Get the `StatusCode` of this `FeroxResponse`

src/logger.rs

@@ -19,8 +19,8 @@ pub fn initialize(verbosity: u8) {
                 0 => (),
                 1 => env::set_var("RUST_LOG", "warn"),
                 2 => env::set_var("RUST_LOG", "info"),
-                3 => env::set_var("RUST_LOG", "debug,hyper=info,reqwest=info"),
-                _ => env::set_var("RUST_LOG", "trace,hyper=info,reqwest=info"),
+                3 => env::set_var("RUST_LOG", "feroxbuster=debug,info"),
+                _ => env::set_var("RUST_LOG", "feroxbuster=trace,info"),
             }
         }
     }
@@ -55,9 +55,10 @@ pub fn initialize(verbosity: u8) {
             };
 
             let msg = format!(
-                "{} {:10.03} {}\n",
+                "{} {:10.03} {} {}\n",
                 style(level_name).bg(level_color).black(),
                 style(t).dim(),
+                record.target(),
                 style(record.args()).dim(),
             );
 

src/main.rs

@@ -3,7 +3,8 @@ use feroxbuster::{
     banner,
     config::{CONFIGURATION, PROGRESS_BAR, PROGRESS_PRINTER},
     heuristics, logger, reporter,
-    scanner::{scan_url, PAUSE_SCAN},
+    scan_manager::PAUSE_SCAN,
+    scanner::{self, scan_url},
     utils::{ferox_print, get_current_depth, module_colorizer, status_colorizer},
     FeroxError, FeroxResponse, FeroxResult, SLEEP_DURATION, VERSION,
 };
@@ -112,6 +113,16 @@ async fn scan(
         return Err(Box::new(err));
     }
 
+    scanner::initialize(
+        words.len(),
+        CONFIGURATION.scan_limit,
+        &CONFIGURATION.extensions,
+        &CONFIGURATION.filter_status,
+        &CONFIGURATION.filter_line_count,
+        &CONFIGURATION.filter_word_count,
+        &CONFIGURATION.filter_size,
+    );
+
     let mut tasks = vec![];
 
     for target in targets {
@@ -243,7 +254,7 @@ async fn clean_up(
     save_output: bool,
 ) {
     log::trace!(
-        "enter: clean_up({:?}, {:?}, {:?}, {:?}, {}",
+        "enter: clean_up({:?}, {:?}, {:?}, {:?}, {})",
         tx_term,
         term_handle,
         tx_file,

src/parser.rs

@@ -55,7 +55,7 @@ pub fn initialize() -> App<'static, 'static> {
                 .long("verbosity")
                 .takes_value(false)
                 .multiple(true)
-                .help("Increase verbosity level (use -vv or more for greater effect)"),
+                .help("Increase verbosity level (use -vv or more for greater effect. [CAUTION] 4 -v's is probably too much)"),
         )
         .arg(
             Arg::with_name("proxy")

src/reporter.rs

@@ -1,5 +1,5 @@
 use crate::config::{CONFIGURATION, PROGRESS_PRINTER};
-use crate::utils::{ferox_print, make_request, status_colorizer};
+use crate::utils::{create_report_string, ferox_print, make_request};
 use crate::{FeroxChannel, FeroxResponse};
 use console::strip_ansi_codes;
 use std::io::Write;
@@ -95,23 +95,13 @@ async fn spawn_terminal_reporter(
         log::trace!("received {} on reporting channel", resp.url());
 
         if CONFIGURATION.status_codes.contains(&resp.status().as_u16()) {
-            let report = if CONFIGURATION.quiet {
-                // -q used, just need the url
-                format!("{}\n", resp.url())
-            } else {
-                // normal printing with status and size
-                let status = status_colorizer(&resp.status().as_str());
-                format!(
-                    // example output
-                    // 200       3280 https://localhost.com/FAQ
-                    "{} {:>8}l {:>8}w {:>8}c {}\n",
-                    status,
-                    resp.line_count(),
-                    resp.word_count(),
-                    resp.content_length(),
-                    resp.url()
-                )
-            };
+            let report = create_report_string(
+                resp.status().as_str(),
+                &resp.line_count().to_string(),
+                &resp.word_count().to_string(),
+                &resp.content_length().to_string(),
+                &resp.url().to_string(),
+            );
 
             // print to stdout
             ferox_print(&report, &PROGRESS_PRINTER);

src/scan_manager.rs

@@ -0,0 +1,531 @@
+use crate::{config::PROGRESS_PRINTER, progress, scanner::NUMBER_OF_REQUESTS, SLEEP_DURATION};
+use console::style;
+use indicatif::{ProgressBar, ProgressStyle};
+use lazy_static::lazy_static;
+use std::{
+    cmp::PartialEq,
+    fmt,
+    sync::{Arc, Mutex, RwLock},
+};
+use std::{
+    io::{stderr, Write},
+    sync::atomic::{AtomicBool, AtomicUsize, Ordering},
+};
+use tokio::{task::JoinHandle, time};
+use uuid::Uuid;
+
+lazy_static! {
+    /// A clock spinner protected with a RwLock to allow for a single thread to use at a time
+    // todo remove this when issue #107 is resolved
+    static ref SINGLE_SPINNER: RwLock<ProgressBar> = RwLock::new(get_single_spinner());
+}
+
+/// Single atomic number that gets incremented once, used to track first thread to interact with
+/// when pausing a scan
+static INTERACTIVE_BARRIER: AtomicUsize = AtomicUsize::new(0);
+
+/// Atomic boolean flag, used to determine whether or not a scan should pause or resume
+pub static PAUSE_SCAN: AtomicBool = AtomicBool::new(false);
+
+/// Simple enum used to flag a `FeroxScan` as likely a directory or file
+#[derive(Debug)]
+pub enum ScanType {
+    File,
+    Directory,
+}
+
+/// Struct to hold scan-related state
+///
+/// The purpose of this container is to open up the pathway to aborting currently running tasks and
+/// serialization of all scan state into a state file in order to resume scans that were cut short
+#[derive(Debug)]
+pub struct FeroxScan {
+    /// UUID that uniquely ID's the scan
+    pub id: String,
+
+    /// The URL that to be scanned
+    pub url: String,
+
+    /// The type of scan
+    pub scan_type: ScanType,
+
+    /// Whether or not this scan has completed
+    pub complete: bool,
+
+    /// The spawned tokio task performing this scan
+    pub task: Option<JoinHandle<()>>,
+
+    /// The progress bar associated with this scan
+    pub progress_bar: Option<ProgressBar>,
+}
+
+/// Implementation of FeroxScan
+impl FeroxScan {
+    /// Stop a currently running scan
+    pub fn abort(&self) {
+        self.stop_progress_bar();
+
+        if let Some(_task) = &self.task {
+            // task.abort();  todo uncomment once upgraded to tokio 0.3
+        }
+    }
+
+    /// Create a default FeroxScan, populates ID with a new UUID
+    fn default() -> Self {
+        let new_id = Uuid::new_v4().to_simple().to_string();
+
+        FeroxScan {
+            id: new_id,
+            task: None,
+            complete: false,
+            url: String::new(),
+            progress_bar: None,
+            scan_type: ScanType::File,
+        }
+    }
+
+    /// Simple helper to call .finish on the scan's progress bar
+    fn stop_progress_bar(&self) {
+        if let Some(pb) = &self.progress_bar {
+            pb.finish();
+        }
+    }
+
+    /// Simple helper get a progress bar
+    pub fn progress_bar(&mut self) -> ProgressBar {
+        if let Some(pb) = &self.progress_bar {
+            pb.clone()
+        } else {
+            let num_requests = NUMBER_OF_REQUESTS.load(Ordering::Relaxed);
+            let pb = progress::add_bar(&self.url, num_requests, false);
+
+            pb.reset_elapsed();
+
+            self.progress_bar = Some(pb.clone());
+
+            pb
+        }
+    }
+
+    /// Given a URL and ProgressBar, create a new FeroxScan, wrap it in an Arc and return it
+    pub fn new(url: &str, scan_type: ScanType, pb: Option<ProgressBar>) -> Arc<Mutex<Self>> {
+        let mut me = Self::default();
+
+        me.url = url.to_string();
+        me.scan_type = scan_type;
+        me.progress_bar = pb;
+
+        Arc::new(Mutex::new(me))
+    }
+
+    /// Mark the scan as complete and stop the scan's progress bar
+    pub fn finish(&mut self) {
+        self.complete = true;
+        self.stop_progress_bar();
+    }
+}
+
+/// Display implementation
+impl fmt::Display for FeroxScan {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        let complete = if self.complete {
+            style("complete").green()
+        } else {
+            style("incomplete").red()
+        };
+
+        write!(f, "{:10} {}", complete, self.url)
+    }
+}
+
+/// PartialEq implementation; uses FeroxScan.id for comparison
+impl PartialEq for FeroxScan {
+    fn eq(&self, other: &Self) -> bool {
+        self.id == other.id
+    }
+}
+
+/// Container around a locked hashset of `FeroxScan`s, adds wrappers for insertion and searching
+#[derive(Debug, Default)]
+pub struct FeroxScans {
+    /// Internal structure: locked hashset of `FeroxScan`s
+    pub scans: Mutex<Vec<Arc<Mutex<FeroxScan>>>>,
+}
+
+/// Implementation of `FeroxScans`
+impl FeroxScans {
+    /// Add a `FeroxScan` to the internal container
+    ///
+    /// If the internal container did NOT contain the scan, true is returned; else false
+    pub fn insert(&self, scan: Arc<Mutex<FeroxScan>>) -> bool {
+        let sentry = match scan.lock() {
+            Ok(locked_scan) => {
+                // If the container did contain the scan, set sentry to false
+                // If the container did not contain the scan, set sentry to true
+                !self.contains(&locked_scan.url)
+            }
+            Err(e) => {
+                // poisoned lock
+                log::error!("FeroxScan's ({:?}) mutex is poisoned: {}", self, e);
+                false
+            }
+        };
+
+        if sentry {
+            // can't update the internal container while the scan itself is locked, so first
+            // lock the scan and check the container for the scan's presence, then add if
+            // not found
+            match self.scans.lock() {
+                Ok(mut scans) => {
+                    scans.push(scan);
+                }
+                Err(e) => {
+                    log::error!("FeroxScans' container's mutex is poisoned: {}", e);
+                    return false;
+                }
+            }
+        }
+
+        sentry
+    }
+
+    /// Simple check for whether or not a FeroxScan is contained within the inner container based
+    /// on the given URL
+    pub fn contains(&self, url: &str) -> bool {
+        match self.scans.lock() {
+            Ok(scans) => {
+                for scan in scans.iter() {
+                    if let Ok(locked_scan) = scan.lock() {
+                        if locked_scan.url == url {
+                            return true;
+                        }
+                    }
+                }
+            }
+            Err(e) => {
+                log::error!("FeroxScans' container's mutex is poisoned: {}", e);
+            }
+        }
+        false
+    }
+
+    /// Find and return a `FeroxScan` based on the given URL
+    pub fn get_scan_by_url(&self, url: &str) -> Option<Arc<Mutex<FeroxScan>>> {
+        if let Ok(scans) = self.scans.lock() {
+            for scan in scans.iter() {
+                if let Ok(locked_scan) = scan.lock() {
+                    if locked_scan.url == url {
+                        return Some(scan.clone());
+                    }
+                }
+            }
+        }
+        None
+    }
+
+    /// Print all FeroxScans of type Directory
+    ///
+    /// Example:
+    ///   0: complete   https://10.129.45.20
+    ///   9: complete   https://10.129.45.20/images
+    ///  10: complete   https://10.129.45.20/assets
+    pub fn display_scans(&self) {
+        if let Ok(scans) = self.scans.lock() {
+            for (i, scan) in scans.iter().enumerate() {
+                if let Ok(unlocked_scan) = scan.lock() {
+                    match unlocked_scan.scan_type {
+                        ScanType::Directory => {
+                            PROGRESS_PRINTER.println(format!("{:3}: {}", i, unlocked_scan));
+                        }
+                        ScanType::File => {
+                            // we're only interested in displaying directory scans, as those are
+                            // the only ones that make sense to be stopped
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+    /// Forced the calling thread into a busy loop
+    ///
+    /// Every `SLEEP_DURATION` milliseconds, the function examines the result stored in `PAUSE_SCAN`
+    ///
+    /// When the value stored in `PAUSE_SCAN` becomes `false`, the function returns, exiting the busy
+    /// loop
+    pub async fn pause(&self, get_user_input: bool) {
+        // function uses tokio::time, not std
+
+        // local testing showed a pretty slow increase (less than linear) in CPU usage as # of
+        // concurrent scans rose when SLEEP_DURATION was set to 500, using that as the default for now
+        let mut interval = time::interval(time::Duration::from_millis(SLEEP_DURATION));
+
+        // ignore any error returned
+        let _ = stderr().flush();
+
+        if INTERACTIVE_BARRIER.load(Ordering::Relaxed) == 0 {
+            INTERACTIVE_BARRIER.fetch_add(1, Ordering::Relaxed);
+
+            if get_user_input {
+                self.display_scans();
+
+                let mut user_input = String::new();
+                std::io::stdin().read_line(&mut user_input).unwrap();
+                // todo actual logic for parsing user input in a way that allows for
+                // calling .abort on the scan retrieved based on the input (issue #107)
+            }
+        }
+
+        if SINGLE_SPINNER.read().unwrap().is_finished() {
+            // todo remove this when issue #107 is resolved
+
+            // in order to not leave draw artifacts laying around in the terminal, we call
+            // finish_and_clear on the progress bar when resuming scans. For this reason, we need to
+            // check if the spinner is finished, and repopulate the RwLock with a new spinner if
+            // necessary
+            if let Ok(mut guard) = SINGLE_SPINNER.write() {
+                *guard = get_single_spinner();
+            }
+        }
+
+        if let Ok(spinner) = SINGLE_SPINNER.write() {
+            spinner.enable_steady_tick(120);
+        }
+
+        loop {
+            // first tick happens immediately, all others wait the specified duration
+            interval.tick().await;
+
+            if !PAUSE_SCAN.load(Ordering::Acquire) {
+                // PAUSE_SCAN is false, so we can exit the busy loop
+
+                if INTERACTIVE_BARRIER.load(Ordering::Relaxed) == 1 {
+                    INTERACTIVE_BARRIER.fetch_sub(1, Ordering::Relaxed);
+                }
+
+                if let Ok(spinner) = SINGLE_SPINNER.write() {
+                    // todo remove this when issue #107 is resolved
+                    spinner.finish_and_clear();
+                }
+
+                let _ = stderr().flush();
+
+                log::trace!("exit: pause_scan");
+                return;
+            }
+        }
+    }
+
+    /// Given a url, create a new `FeroxScan` and add it to `FeroxScans`
+    ///
+    /// If `FeroxScans` did not already contain the scan, return true; otherwise return false
+    ///
+    /// Also return a reference to the new `FeroxScan`
+    fn add_scan(&self, url: &str, scan_type: ScanType) -> (bool, Arc<Mutex<FeroxScan>>) {
+        let bar = match scan_type {
+            ScanType::Directory => {
+                let progress_bar =
+                    progress::add_bar(&url, NUMBER_OF_REQUESTS.load(Ordering::Relaxed), false);
+
+                progress_bar.reset_elapsed();
+
+                Some(progress_bar)
+            }
+            ScanType::File => None,
+        };
+
+        let ferox_scan = FeroxScan::new(&url, scan_type, bar);
+
+        // If the set did not contain the scan, true is returned.
+        // If the set did contain the scan, false is returned.
+        let response = self.insert(ferox_scan.clone());
+
+        (response, ferox_scan)
+    }
+
+    /// Given a url, create a new `FeroxScan` and add it to `FeroxScans` as a Directory Scan
+    ///
+    /// If `FeroxScans` did not already contain the scan, return true; otherwise return false
+    ///
+    /// Also return a reference to the new `FeroxScan`
+    pub fn add_directory_scan(&self, url: &str) -> (bool, Arc<Mutex<FeroxScan>>) {
+        self.add_scan(&url, ScanType::Directory)
+    }
+
+    /// Given a url, create a new `FeroxScan` and add it to `FeroxScans` as a File Scan
+    ///
+    /// If `FeroxScans` did not already contain the scan, return true; otherwise return false
+    ///
+    /// Also return a reference to the new `FeroxScan`
+    pub fn add_file_scan(&self, url: &str) -> (bool, Arc<Mutex<FeroxScan>>) {
+        self.add_scan(&url, ScanType::File)
+    }
+}
+
+/// Return a clock spinner, used when scans are paused
+// todo remove this when issue #107 is resolved
+fn get_single_spinner() -> ProgressBar {
+    log::trace!("enter: get_single_spinner");
+
+    let spinner = ProgressBar::new_spinner().with_style(
+        ProgressStyle::default_spinner()
+            .tick_strings(&[
+                "🕛", "🕐", "🕑", "🕒", "🕓", "🕔", "🕕", "🕖", "🕗", "🕘", "🕙", "🕚",
+            ])
+            .template(&format!(
+                "\t-= All Scans {{spinner}} {} =-",
+                style("Paused").red()
+            )),
+    );
+
+    log::trace!("exit: get_single_spinner -> {:?}", spinner);
+    spinner
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    /// test that get_single_spinner returns the correct spinner
+    // todo remove this when issue #107 is resolved
+    fn scanner_get_single_spinner_returns_spinner() {
+        let spinner = get_single_spinner();
+        assert!(!spinner.is_finished());
+    }
+
+    #[tokio::test(core_threads = 1)]
+    /// tests that pause_scan pauses execution and releases execution when PAUSE_SCAN is toggled
+    /// the spinner used during the test has had .finish_and_clear called on it, meaning that
+    /// a new one will be created, taking the if branch within the function
+    async fn scanner_pause_scan_with_finished_spinner() {
+        let now = time::Instant::now();
+        let urls = FeroxScans::default();
+
+        PAUSE_SCAN.store(true, Ordering::Relaxed);
+
+        let expected = time::Duration::from_secs(2);
+
+        tokio::spawn(async move {
+            time::delay_for(expected).await;
+            PAUSE_SCAN.store(false, Ordering::Relaxed);
+        });
+
+        urls.pause(false).await;
+
+        assert!(now.elapsed() > expected);
+    }
+
+    #[test]
+    /// add an unknown url to the hashset, expect true
+    fn add_url_to_list_of_scanned_urls_with_unknown_url() {
+        let urls = FeroxScans::default();
+        let url = "http://unknown_url";
+        let (result, _scan) = urls.add_scan(url, ScanType::Directory);
+        assert_eq!(result, true);
+    }
+
+    #[test]
+    /// add a known url to the hashset, with a trailing slash, expect false
+    fn add_url_to_list_of_scanned_urls_with_known_url() {
+        let urls = FeroxScans::default();
+        let pb = ProgressBar::new(1);
+        let url = "http://unknown_url/";
+        let scan = FeroxScan::new(url, ScanType::Directory, Some(pb));
+
+        assert_eq!(urls.insert(scan), true);
+
+        let (result, _scan) = urls.add_scan(url, ScanType::Directory);
+
+        assert_eq!(result, false);
+    }
+
+    #[test]
+    /// abort should call stop_progress_bar, marking it as finished
+    fn abort_stops_progress_bar() {
+        let pb = ProgressBar::new(1);
+        let url = "http://unknown_url/";
+        let scan = FeroxScan::new(url, ScanType::Directory, Some(pb));
+
+        assert_eq!(
+            scan.lock()
+                .unwrap()
+                .progress_bar
+                .as_ref()
+                .unwrap()
+                .is_finished(),
+            false
+        );
+
+        scan.lock().unwrap().abort();
+
+        assert_eq!(
+            scan.lock()
+                .unwrap()
+                .progress_bar
+                .as_ref()
+                .unwrap()
+                .is_finished(),
+            true
+        );
+    }
+
+    #[test]
+    /// add a known url to the hashset, without a trailing slash, expect false
+    fn add_url_to_list_of_scanned_urls_with_known_url_without_slash() {
+        let urls = FeroxScans::default();
+        let url = "http://unknown_url";
+        let scan = FeroxScan::new(url, ScanType::File, None);
+
+        assert_eq!(urls.insert(scan), true);
+
+        let (result, _scan) = urls.add_scan(url, ScanType::File);
+
+        assert_eq!(result, false);
+    }
+
+    #[test]
+    /// just increasing coverage, no real expectations
+    fn call_display_scans() {
+        let urls = FeroxScans::default();
+        let pb = ProgressBar::new(1);
+        let pb_two = ProgressBar::new(2);
+        let url = "http://unknown_url/";
+        let url_two = "http://unknown_url/fa";
+        let scan = FeroxScan::new(url, ScanType::Directory, Some(pb));
+        let scan_two = FeroxScan::new(url_two, ScanType::Directory, Some(pb_two));
+
+        scan_two.lock().unwrap().finish(); // one complete, one incomplete
+
+        assert_eq!(urls.insert(scan), true);
+
+        urls.display_scans();
+    }
+
+    #[test]
+    /// ensure that PartialEq compares FeroxScan.id fields
+    fn partial_eq_compares_the_id_field() {
+        let url = "http://unknown_url/";
+        let scan = FeroxScan::new(url, ScanType::Directory, None);
+        let scan_two = FeroxScan::new(url, ScanType::Directory, None);
+
+        assert!(!scan.lock().unwrap().eq(&scan_two.lock().unwrap()));
+
+        scan_two.lock().unwrap().id = scan.lock().unwrap().id.clone();
+
+        assert!(scan.lock().unwrap().eq(&scan_two.lock().unwrap()));
+    }
+
+    #[test]
+    /// show that a new progress bar is created if one doesn't exist
+    fn ferox_scan_get_progress_bar_when_none_is_set() {
+        let mut scan = FeroxScan::default();
+
+        assert!(scan.progress_bar.is_none()); // no pb exists
+
+        let pb = scan.progress_bar();
+
+        assert!(scan.progress_bar.is_some()); // new pb created
+        assert!(!pb.is_finished()) // not finished
+    }
+}

src/scanner.rs

@@ -1,25 +1,25 @@
 use crate::{
     config::CONFIGURATION,
     extractor::get_links,
-    filters::{FeroxFilter, StatusCodeFilter, WildcardFilter},
-    heuristics, progress,
+    filters::{
+        FeroxFilter, LinesFilter, SizeFilter, StatusCodeFilter, WildcardFilter, WordsFilter,
+    },
+    heuristics,
+    scan_manager::{FeroxScans, PAUSE_SCAN},
     utils::{format_url, get_current_depth, make_request},
-    FeroxChannel, FeroxResponse, SLEEP_DURATION,
+    FeroxChannel, FeroxResponse,
 };
-use console::style;
 use futures::{
     future::{BoxFuture, FutureExt},
     stream, StreamExt,
 };
-use indicatif::{ProgressBar, ProgressStyle};
 use lazy_static::lazy_static;
 use reqwest::Url;
 use std::{
     collections::HashSet,
     convert::TryInto,
-    io::{stderr, Write},
     ops::Deref,
-    sync::atomic::{AtomicBool, AtomicUsize, Ordering},
+    sync::atomic::{AtomicU64, AtomicUsize, Ordering},
     sync::{Arc, RwLock},
 };
 use tokio::{
@@ -28,21 +28,17 @@ use tokio::{
         Semaphore,
     },
     task::JoinHandle,
-    time,
 };
 
 /// Single atomic number that gets incremented once, used to track first scan vs. all others
 static CALL_COUNT: AtomicUsize = AtomicUsize::new(0);
 
-/// Atomic boolean flag, used to determine whether or not a scan should pause or resume
-pub static PAUSE_SCAN: AtomicBool = AtomicBool::new(false);
+/// Single atomic number that gets holds the number of requests to be sent per directory scanned
+pub static NUMBER_OF_REQUESTS: AtomicU64 = AtomicU64::new(0);
 
 lazy_static! {
     /// Set of urls that have been sent to [scan_url](fn.scan_url.html), used for deduplication
-    static ref SCANNED_URLS: RwLock<HashSet<String>> = RwLock::new(HashSet::new());
-
-    /// A clock spinner protected with a RwLock to allow for a single thread to use at a time
-    static ref SINGLE_SPINNER: RwLock<ProgressBar> = RwLock::new(get_single_spinner());
+    pub static ref SCANNED_URLS: FeroxScans = FeroxScans::default();
 
     /// Vector of implementors of the FeroxFilter trait
     static ref FILTERS: Arc<RwLock<Vec<Box<dyn FeroxFilter>>>> = Arc::new(RwLock::new(Vec::<Box<dyn FeroxFilter>>::new()));
@@ -51,101 +47,6 @@ lazy_static! {
     static ref SCAN_LIMITER: Semaphore = Semaphore::new(CONFIGURATION.scan_limit);
 }
 
-/// Return a clock spinner, used when scans are paused
-fn get_single_spinner() -> ProgressBar {
-    log::trace!("enter: get_single_spinner");
-
-    let spinner = ProgressBar::new_spinner().with_style(
-        ProgressStyle::default_spinner()
-            .tick_strings(&[
-                "🕛", "🕐", "🕑", "🕒", "🕓", "🕔", "🕕", "🕖", "🕗", "🕘", "🕙", "🕚",
-            ])
-            .template(&format!(
-                "\t-= All Scans {{spinner}} {} =-",
-                style("Paused").red()
-            )),
-    );
-
-    log::trace!("exit: get_single_spinner -> {:?}", spinner);
-    spinner
-}
-
-/// Forced the calling thread into a busy loop
-///
-/// Every `SLEEP_DURATION` milliseconds, the function examines the result stored in `PAUSE_SCAN`
-///
-/// When the value stored in `PAUSE_SCAN` becomes `false`, the function returns, exiting the busy
-/// loop
-async fn pause_scan() {
-    log::trace!("enter: pause_scan");
-    // function uses tokio::time, not std
-
-    // local testing showed a pretty slow increase (less than linear) in CPU usage as # of
-    // concurrent scans rose when SLEEP_DURATION was set to 500, using that as the default for now
-    let mut interval = time::interval(time::Duration::from_millis(SLEEP_DURATION));
-
-    // ignore any error returned
-    let _ = stderr().flush();
-
-    if SINGLE_SPINNER.read().unwrap().is_finished() {
-        // in order to not leave draw artifacts laying around in the terminal, we call
-        // finish_and_clear on the progress bar when resuming scans. For this reason, we need to
-        // check if the spinner is finished, and repopulate the RwLock with a new spinner if
-        // necessary
-        if let Ok(mut guard) = SINGLE_SPINNER.write() {
-            *guard = get_single_spinner();
-        }
-    }
-
-    if let Ok(spinner) = SINGLE_SPINNER.write() {
-        spinner.enable_steady_tick(120);
-    }
-
-    loop {
-        // first tick happens immediately, all others wait the specified duration
-        interval.tick().await;
-
-        if !PAUSE_SCAN.load(Ordering::Acquire) {
-            // PAUSE_SCAN is false, so we can exit the busy loop
-            if let Ok(spinner) = SINGLE_SPINNER.write() {
-                spinner.finish_and_clear();
-            }
-            let _ = stderr().flush();
-            log::trace!("exit: pause_scan");
-            return;
-        }
-    }
-}
-
-/// Adds the given url to `SCANNED_URLS`
-///
-/// If `SCANNED_URLS` did not already contain the url, return true; otherwise return false
-fn add_url_to_list_of_scanned_urls(resp: &str, scanned_urls: &RwLock<HashSet<String>>) -> bool {
-    log::trace!(
-        "enter: add_url_to_list_of_scanned_urls({}, {:?})",
-        resp,
-        scanned_urls
-    );
-
-    match scanned_urls.write() {
-        // check new url against what's already been scanned
-        Ok(mut urls) => {
-            // If the set did not contain resp, true is returned.
-            // If the set did contain resp, false is returned.
-            let response = urls.insert(resp.to_string());
-
-            log::trace!("exit: add_url_to_list_of_scanned_urls -> {}", response);
-            response
-        }
-        Err(e) => {
-            // poisoned lock
-            log::error!("Set of scanned urls poisoned: {}", e);
-            log::trace!("exit: add_url_to_list_of_scanned_urls -> false");
-            false
-        }
-    }
-}
-
 /// Adds the given FeroxFilter to the given list of FeroxFilter implementors
 ///
 /// If the given list did not already contain the filter, return true; otherwise return false
@@ -205,7 +106,7 @@ fn spawn_recursion_handler(
         let mut scans = vec![];
 
         while let Some(resp) = recursion_channel.recv().await {
-            let unknown = add_url_to_list_of_scanned_urls(&resp, &SCANNED_URLS);
+            let (unknown, _) = SCANNED_URLS.add_directory_scan(&resp);
 
             if !unknown {
                 // not unknown, i.e. we've seen the url before and don't need to scan again
@@ -219,7 +120,7 @@ fn spawn_recursion_handler(
             let resp_clone = resp.clone();
             let list_clone = wordlist.clone();
 
-            scans.push(tokio::spawn(async move {
+            let future = tokio::spawn(async move {
                 scan_url(
                     resp_clone.to_owned().as_str(),
                     list_clone,
@@ -228,7 +129,9 @@ fn spawn_recursion_handler(
                     file_clone,
                 )
                 .await
-            }));
+            });
+
+            scans.push(future);
         }
         scans
     }
@@ -285,7 +188,7 @@ fn create_urls(target_url: &str, word: &str, extensions: &[String]) -> Vec<Url>
 /// handles 2xx and 3xx responses by either checking if the url ends with a / (2xx)
 /// or if the Location header is present and matches the base url + / (3xx)
 fn response_is_directory(response: &FeroxResponse) -> bool {
-    log::trace!("enter: is_directory({:?})", response);
+    log::trace!("enter: is_directory({})", response);
 
     if response.status().is_redirection() {
         // status code is 3xx
@@ -311,10 +214,7 @@ fn response_is_directory(response: &FeroxResponse) -> bool {
                 }
             }
             None => {
-                log::debug!(
-                    "expected Location header, but none was found: {:?}",
-                    response
-                );
+                log::debug!("expected Location header, but none was found: {}", response);
                 log::trace!("exit: is_directory -> false");
                 return false;
             }
@@ -370,7 +270,7 @@ async fn try_recursion(
     transmitter: UnboundedSender<String>,
 ) {
     log::trace!(
-        "enter: try_recursion({:?}, {}, {:?})",
+        "enter: try_recursion({}, {}, {:?})",
         response,
         base_depth,
         transmitter
@@ -418,22 +318,6 @@ async fn try_recursion(
 /// Simple helper to stay DRY; determines whether or not a given `FeroxResponse` should be reported
 /// to the user or not.
 pub fn should_filter_response(response: &FeroxResponse) -> bool {
-    if CONFIGURATION
-        .filter_size
-        .contains(&response.content_length())
-        || CONFIGURATION
-            .filter_line_count
-            .contains(&response.line_count())
-        || CONFIGURATION
-            .filter_word_count
-            .contains(&response.word_count())
-    {
-        // filtered value from --filter-size, size filters and wildcards are two separate filters
-        // and are applied independently
-        log::debug!("size filter: filtered out {}", response.url());
-        return true;
-    }
-
     match FILTERS.read() {
         Ok(filters) => {
             for filter in filters.iter() {
@@ -476,7 +360,7 @@ async fn make_requests(
     for url in urls {
         if let Ok(response) = make_request(&CONFIGURATION.client, &url).await {
             // response came back without error, convert it to FeroxResponse
-            let ferox_response = FeroxResponse::from(response, CONFIGURATION.extract_links).await;
+            let ferox_response = FeroxResponse::from(response, true).await;
 
             // do recursion if appropriate
             if !CONFIGURATION.no_recursion {
@@ -494,13 +378,6 @@ async fn make_requests(
                 let new_links = get_links(&ferox_response).await;
 
                 for new_link in new_links {
-                    let unknown = add_url_to_list_of_scanned_urls(&new_link, &SCANNED_URLS);
-
-                    if !unknown {
-                        // not unknown, i.e. we've seen the url before and don't need to scan again
-                        continue;
-                    }
-
                     // create a url based on the given command line options, continue on error
                     let new_url = match format_url(
                         &new_link,
@@ -513,14 +390,18 @@ async fn make_requests(
                         Err(_) => continue,
                     };
 
+                    if SCANNED_URLS.get_scan_by_url(&new_url.to_string()).is_some() {
+                        //we've seen the url before and don't need to scan again
+                        continue;
+                    }
+
                     // make the request and store the response
                     let new_response = match make_request(&CONFIGURATION.client, &new_url).await {
                         Ok(resp) => resp,
                         Err(_) => continue,
                     };
 
-                    let mut new_ferox_response =
-                        FeroxResponse::from(new_response, CONFIGURATION.extract_links).await;
+                    let mut new_ferox_response = FeroxResponse::from(new_response, true).await;
 
                     // filter if necessary
                     if should_filter_response(&new_ferox_response) {
@@ -529,11 +410,9 @@ async fn make_requests(
 
                     if new_ferox_response.is_file() {
                         // very likely a file, simply request and report
-                        log::debug!(
-                            "Singular extraction: {} ({})",
-                            new_ferox_response.url(),
-                            new_ferox_response.status().as_str(),
-                        );
+                        log::debug!("Singular extraction: {}", new_ferox_response);
+
+                        SCANNED_URLS.add_file_scan(&new_url.to_string());
 
                         send_report(report_chan.clone(), new_ferox_response);
 
@@ -541,11 +420,7 @@ async fn make_requests(
                     }
 
                     if !CONFIGURATION.no_recursion {
-                        log::debug!(
-                            "Recursive extraction: {} ({})",
-                            new_ferox_response.url(),
-                            new_ferox_response.status().as_str()
-                        );
+                        log::debug!("Recursive extraction: {}", new_ferox_response);
 
                         if new_ferox_response.status().is_success()
                             && !new_ferox_response.url().as_str().ends_with('/')
@@ -571,7 +446,7 @@ async fn make_requests(
 
 /// Simple helper to send a `FeroxResponse` over the tx side of an `mpsc::unbounded_channel`
 fn send_report(report_sender: UnboundedSender<FeroxResponse>, response: FeroxResponse) {
-    log::trace!("enter: send_report({:?}, {:?}", report_sender, response);
+    log::trace!("enter: send_report({:?}, {}", report_sender, response);
 
     match report_sender.send(response) {
         Ok(_) => {}
@@ -606,31 +481,33 @@ pub async fn scan_url(
 
     let (tx_dir, rx_dir): FeroxChannel<String> = mpsc::unbounded_channel();
 
-    let num_reqs_expected: u64 = if CONFIGURATION.extensions.is_empty() {
-        wordlist.len().try_into().unwrap()
-    } else {
-        let total = wordlist.len() * (CONFIGURATION.extensions.len() + 1);
-        total.try_into().unwrap()
-    };
-
-    let progress_bar = progress::add_bar(&target_url, num_reqs_expected, false);
-    progress_bar.reset_elapsed();
-
     if CALL_COUNT.load(Ordering::Relaxed) == 0 {
         CALL_COUNT.fetch_add(1, Ordering::Relaxed);
 
         // this protection allows us to add the first scanned url to SCANNED_URLS
         // from within the scan_url function instead of the recursion handler
-        add_url_to_list_of_scanned_urls(&target_url, &SCANNED_URLS);
-
-        if CONFIGURATION.scan_limit == 0 {
-            // scan_limit == 0 means no limit should be imposed... however, scoping the Semaphore
-            // permit is tricky, so as a workaround, we'll add a ridiculous number of permits to
-            // the semaphore (1,152,921,504,606,846,975 to be exact) and call that 'unlimited'
-            SCAN_LIMITER.add_permits(usize::MAX >> 4);
-        }
+        SCANNED_URLS.add_directory_scan(&target_url);
     }
 
+    let ferox_scan = match SCANNED_URLS.get_scan_by_url(&target_url) {
+        Some(scan) => scan,
+        None => {
+            log::error!(
+                "Could not find FeroxScan associated with {}; this shouldn't happen... exiting",
+                target_url
+            );
+            return;
+        }
+    };
+
+    let progress_bar = match ferox_scan.lock() {
+        Ok(mut scan) => scan.progress_bar(),
+        Err(e) => {
+            log::error!("FeroxScan's ({:?}) mutex is poisoned: {}", ferox_scan, e);
+            return;
+        }
+    };
+
     // When acquire is called and the semaphore has remaining permits, the function immediately
     // returns a permit. However, if no remaining permits are available, acquire (asynchronously)
     // waits until an outstanding permit is dropped. At this point, the freed permit is assigned
@@ -665,15 +542,6 @@ pub async fn scan_url(
 
     add_filter_to_list_of_ferox_filters(filter, FILTERS.clone());
 
-    // add any status code filters to `FILTERS`
-    for code_filter in &CONFIGURATION.filter_status {
-        let filter = StatusCodeFilter {
-            filter_code: *code_filter,
-        };
-        let boxed_filter = Box::new(filter);
-        add_filter_to_list_of_ferox_filters(boxed_filter, FILTERS.clone());
-    }
-
     // producer tasks (mp of mpsc); responsible for making requests
     let producers = stream::iter(looping_words.deref().to_owned())
         .map(|word| {
@@ -687,7 +555,9 @@ pub async fn scan_url(
                         // for every word in the wordlist, check to see if PAUSE_SCAN is set to true
                         // when true; enter a busy loop that only exits by setting PAUSE_SCAN back
                         // to false
-                        pause_scan().await;
+
+                        // todo change to true when issue #107 is resolved
+                        SCANNED_URLS.pause(false).await;
                     }
                     make_requests(&tgt, &word, base_depth, txd, txr).await
                 }),
@@ -713,7 +583,9 @@ pub async fn scan_url(
     // drop the current permit so the semaphore will allow another scan to proceed
     drop(permit);
 
-    progress_bar.finish();
+    if let Ok(mut scan) = ferox_scan.lock() {
+        scan.finish();
+    }
 
     // manually drop tx in order for the rx task's while loops to eval to false
     log::trace!("dropped recursion handler's transmitter");
@@ -727,6 +599,84 @@ pub async fn scan_url(
     log::trace!("exit: scan_url");
 }
 
+/// Perform steps necessary to run scans that only need to be performed once (warming up the
+/// engine, as it were)
+pub fn initialize(
+    num_words: usize,
+    scan_limit: usize,
+    extensions: &[String],
+    status_code_filters: &[u16],
+    lines_filters: &[usize],
+    words_filters: &[usize],
+    size_filters: &[u64],
+) {
+    log::trace!(
+        "enter: initialize({}, {}, {:?}, {:?}, {:?}, {:?}, {:?})",
+        num_words,
+        scan_limit,
+        extensions,
+        status_code_filters,
+        lines_filters,
+        words_filters,
+        size_filters,
+    );
+
+    // number of requests only needs to be calculated once, and then can be reused
+    let num_reqs_expected: u64 = if extensions.is_empty() {
+        num_words.try_into().unwrap()
+    } else {
+        let total = num_words * (extensions.len() + 1);
+        total.try_into().unwrap()
+    };
+
+    NUMBER_OF_REQUESTS.store(num_reqs_expected, Ordering::Relaxed);
+
+    // add any status code filters to `FILTERS` (-C|--filter-status)
+    for code_filter in status_code_filters {
+        let filter = StatusCodeFilter {
+            filter_code: *code_filter,
+        };
+        let boxed_filter = Box::new(filter);
+        add_filter_to_list_of_ferox_filters(boxed_filter, FILTERS.clone());
+    }
+
+    // add any line count filters to `FILTERS` (-N|--filter-lines)
+    for lines_filter in lines_filters {
+        let filter = LinesFilter {
+            line_count: *lines_filter,
+        };
+        let boxed_filter = Box::new(filter);
+        add_filter_to_list_of_ferox_filters(boxed_filter, FILTERS.clone());
+    }
+
+    // add any line count filters to `FILTERS` (-W|--filter-words)
+    for words_filter in words_filters {
+        let filter = WordsFilter {
+            word_count: *words_filter,
+        };
+        let boxed_filter = Box::new(filter);
+        add_filter_to_list_of_ferox_filters(boxed_filter, FILTERS.clone());
+    }
+
+    // add any line count filters to `FILTERS` (-S|--filter-size)
+    for size_filter in size_filters {
+        let filter = SizeFilter {
+            content_length: *size_filter,
+        };
+        let boxed_filter = Box::new(filter);
+        add_filter_to_list_of_ferox_filters(boxed_filter, FILTERS.clone());
+    }
+
+    if scan_limit == 0 {
+        // scan_limit == 0 means no limit should be imposed... however, scoping the Semaphore
+        // permit is tricky, so as a workaround, we'll add a ridiculous number of permits to
+        // the semaphore (1,152,921,504,606,846,975 to be exact) and call that 'unlimited'
+        SCAN_LIMITER.add_permits(usize::MAX >> 4);
+    }
+
+    log::trace!("exit: initialize");
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;
@@ -824,68 +774,4 @@ mod tests {
         let result = reached_max_depth(&url, 0, 2);
         assert!(result);
     }
-
-    #[test]
-    /// add an unknown url to the hashset, expect true
-    fn add_url_to_list_of_scanned_urls_with_unknown_url() {
-        let urls = RwLock::new(HashSet::<String>::new());
-        let url = "http://unknown_url";
-        assert_eq!(add_url_to_list_of_scanned_urls(url, &urls), true);
-    }
-
-    #[test]
-    /// add a known url to the hashset, with a trailing slash, expect false
-    fn add_url_to_list_of_scanned_urls_with_known_url() {
-        let urls = RwLock::new(HashSet::<String>::new());
-        let url = "http://unknown_url/";
-
-        assert_eq!(urls.write().unwrap().insert(url.to_string()), true);
-
-        assert_eq!(add_url_to_list_of_scanned_urls(url, &urls), false);
-    }
-
-    #[test]
-    /// add a known url to the hashset, without a trailing slash, expect false
-    fn add_url_to_list_of_scanned_urls_with_known_url_without_slash() {
-        let urls = RwLock::new(HashSet::<String>::new());
-        let url = "http://unknown_url";
-
-        assert_eq!(
-            urls.write()
-                .unwrap()
-                .insert("http://unknown_url".to_string()),
-            true
-        );
-
-        assert_eq!(add_url_to_list_of_scanned_urls(url, &urls), false);
-    }
-
-    #[test]
-    /// test that get_single_spinner returns the correct spinner
-    fn scanner_get_single_spinner_returns_spinner() {
-        let spinner = get_single_spinner();
-        assert!(!spinner.is_finished());
-    }
-
-    #[tokio::test(core_threads = 1)]
-    /// tests that pause_scan pauses execution and releases execution when PAUSE_SCAN is toggled
-    /// the spinner used during the test has had .finish_and_clear called on it, meaning that
-    /// a new one will be created, taking the if branch within the function
-    async fn scanner_pause_scan_with_finished_spinner() {
-        let now = time::Instant::now();
-
-        PAUSE_SCAN.store(true, Ordering::Relaxed);
-        SINGLE_SPINNER.write().unwrap().finish_and_clear();
-
-        let expected = time::Duration::from_secs(2);
-
-        tokio::spawn(async move {
-            time::delay_for(expected).await;
-            PAUSE_SCAN.store(false, Ordering::Relaxed);
-        });
-
-        pause_scan().await;
-
-        assert!(now.elapsed() > expected);
-    }
 }

src/utils.rs

@@ -1,8 +1,10 @@
-use crate::{FeroxError, FeroxResult};
+use crate::{
+    config::{CONFIGURATION, PROGRESS_PRINTER},
+    FeroxError, FeroxResult,
+};
 use console::{strip_ansi_codes, style, user_attended};
 use indicatif::ProgressBar;
-use reqwest::Url;
-use reqwest::{Client, Response};
+use reqwest::{Client, Response, Url};
 #[cfg(not(target_os = "windows"))]
 use rlimit::{getrlimit, setrlimit, Resource, Rlim};
 use std::convert::TryInto;
@@ -21,13 +23,7 @@ use std::convert::TryInto;
 pub fn get_current_depth(target: &str) -> usize {
     log::trace!("enter: get_current_depth({})", target);
 
-    let target = if !target.ends_with('/') {
-        // target url doesn't end with a /, for the purposes of determining depth, we'll normalize
-        // all urls to end in a / and then calculate accordingly
-        format!("{}/", target)
-    } else {
-        String::from(target)
-    };
+    let target = normalize_url(target);
 
     match Url::parse(&target) {
         Ok(url) => {
@@ -90,8 +86,8 @@ pub fn get_url_path_length(url: &Url) -> u64 {
 
     let path = url.path();
 
-    let segments = if path.starts_with('/') {
-        path[1..].split_terminator('/')
+    let segments = if let Some(split) = path.strip_prefix('/') {
+        split.split_terminator('/')
     } else {
         log::trace!("exit: get_url_path_length -> 0");
         return 0;
@@ -244,7 +240,6 @@ pub async fn make_request(client: &Client, url: &Url) -> FeroxResult<Response> {
 
     match client.get(url.to_owned()).send().await {
         Ok(resp) => {
-            log::debug!("requested Url: {}", resp.url());
             log::trace!("exit: make_request -> {:?}", resp);
             Ok(resp)
         }
@@ -253,6 +248,19 @@ pub async fn make_request(client: &Client, url: &Url) -> FeroxResult<Response> {
             if e.to_string().contains("operation timed out") {
                 // only warn for timeouts, while actual errors are still left as errors
                 log::warn!("Error while making request: {}", e);
+            } else if e.is_redirect() {
+                if let Some(last_redirect) = e.url() {
+                    // get where we were headed (last_redirect) and where we came from (url)
+                    let fancy_message = format!("{} !=> {}", url, last_redirect);
+
+                    let report = if let Some(msg_status) = e.status() {
+                        create_report_string(msg_status.as_str(), "-1", "-1", "-1", &fancy_message)
+                    } else {
+                        create_report_string("UNK", "-1", "-1", "-1", &fancy_message)
+                    };
+
+                    ferox_print(&report, &PROGRESS_PRINTER)
+                };
             } else {
                 log::error!("Error while making request: {}", e);
             }
@@ -261,6 +269,30 @@ pub async fn make_request(client: &Client, url: &Url) -> FeroxResult<Response> {
     }
 }
 
+/// Helper to create the standard line for output to file/terminal
+///
+/// example output:
+/// 200      127l      283w     4134c http://localhost/faq
+pub fn create_report_string(
+    status: &str,
+    line_count: &str,
+    word_count: &str,
+    content_length: &str,
+    url: &str,
+) -> String {
+    if CONFIGURATION.quiet {
+        // -q used, just need the url
+        format!("{}\n", url)
+    } else {
+        // normal printing with status and sizes
+        let color_status = status_colorizer(status);
+        format!(
+            "{} {:>8}l {:>8}w {:>8}c {}\n",
+            color_status, line_count, word_count, content_length, url
+        )
+    }
+}
+
 /// Attempts to set the soft limit for the RLIMIT_NOFILE resource
 ///
 /// RLIMIT_NOFILE is the maximum number of file descriptors that can be opened by this process
@@ -312,6 +344,22 @@ pub fn set_open_file_limit(limit: usize) -> bool {
     false
 }
 
+/// Simple helper to abstract away adding a forward-slash to a url if not present
+///
+/// used mostly for deduplication purposes and url state tracking
+pub fn normalize_url(url: &str) -> String {
+    log::trace!("enter: normalize_url({})", url);
+
+    let normalized = if url.ends_with('/') {
+        url.to_string()
+    } else {
+        format!("{}/", url)
+    };
+
+    log::trace!("exit: normalize_url -> {}", normalized);
+    normalized
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;

tests/test_filters.rs

@@ -55,3 +55,150 @@ fn filters_status_code_should_filter_response() {
     assert_eq!(mock_two.times_called(), 1);
     teardown_tmp_directory(tmp_dir);
 }
+
+#[test]
+/// create a FeroxResponse that should elicit a true from
+/// LinesFilter::should_filter_response
+fn filters_lines_should_filter_response() {
+    let srv = MockServer::start();
+    let (tmp_dir, file) =
+        setup_tmp_directory(&["LICENSE".to_string(), "file.js".to_string()], "wordlist").unwrap();
+
+    let mock = Mock::new()
+        .expect_method(GET)
+        .expect_path("/LICENSE")
+        .return_status(302)
+        .return_body("this is a test")
+        .create_on(&srv);
+
+    let mock_two = Mock::new()
+        .expect_method(GET)
+        .expect_path("/file.js")
+        .return_status(200)
+        .return_body("this is also a test of some import\nwith 2 lines, no less")
+        .create_on(&srv);
+
+    let cmd = Command::cargo_bin("feroxbuster")
+        .unwrap()
+        .arg("--url")
+        .arg(srv.url("/"))
+        .arg("--wordlist")
+        .arg(file.as_os_str())
+        .arg("--filter-lines")
+        .arg("2")
+        .unwrap();
+
+    cmd.assert().success().stdout(
+        predicate::str::contains("/LICENSE")
+            .and(predicate::str::contains("302"))
+            .and(predicate::str::contains("14"))
+            .and(predicate::str::contains("/file.js"))
+            .not()
+            .and(predicate::str::contains("200"))
+            .not()
+            .and(predicate::str::contains("2l"))
+            .not(),
+    );
+
+    assert_eq!(mock.times_called(), 1);
+    assert_eq!(mock_two.times_called(), 1);
+    teardown_tmp_directory(tmp_dir);
+}
+
+#[test]
+/// create a FeroxResponse that should elicit a true from
+/// WordsFilter::should_filter_response
+fn filters_words_should_filter_response() {
+    let srv = MockServer::start();
+    let (tmp_dir, file) =
+        setup_tmp_directory(&["LICENSE".to_string(), "file.js".to_string()], "wordlist").unwrap();
+
+    let mock = Mock::new()
+        .expect_method(GET)
+        .expect_path("/LICENSE")
+        .return_status(302)
+        .return_body("this is a test")
+        .create_on(&srv);
+
+    let mock_two = Mock::new()
+        .expect_method(GET)
+        .expect_path("/file.js")
+        .return_status(200)
+        .return_body("this is also a test of some import\nwith 2 lines, no less")
+        .create_on(&srv);
+
+    let cmd = Command::cargo_bin("feroxbuster")
+        .unwrap()
+        .arg("--url")
+        .arg(srv.url("/"))
+        .arg("--wordlist")
+        .arg(file.as_os_str())
+        .arg("--filter-words")
+        .arg("13")
+        .unwrap();
+
+    cmd.assert().success().stdout(
+        predicate::str::contains("/LICENSE")
+            .and(predicate::str::contains("302"))
+            .and(predicate::str::contains("14"))
+            .and(predicate::str::contains("/file.js"))
+            .not()
+            .and(predicate::str::contains("200"))
+            .not()
+            .and(predicate::str::contains("13w"))
+            .not(),
+    );
+
+    assert_eq!(mock.times_called(), 1);
+    assert_eq!(mock_two.times_called(), 1);
+    teardown_tmp_directory(tmp_dir);
+}
+
+#[test]
+/// create a FeroxResponse that should elicit a true from
+/// SizeFilter::should_filter_response
+fn filters_size_should_filter_response() {
+    let srv = MockServer::start();
+    let (tmp_dir, file) =
+        setup_tmp_directory(&["LICENSE".to_string(), "file.js".to_string()], "wordlist").unwrap();
+
+    let mock = Mock::new()
+        .expect_method(GET)
+        .expect_path("/LICENSE")
+        .return_status(302)
+        .return_body("this is a test")
+        .create_on(&srv);
+
+    let mock_two = Mock::new()
+        .expect_method(GET)
+        .expect_path("/file.js")
+        .return_status(200)
+        .return_body("this is also a test of some import\nwith 2 lines, no less")
+        .create_on(&srv);
+
+    let cmd = Command::cargo_bin("feroxbuster")
+        .unwrap()
+        .arg("--url")
+        .arg(srv.url("/"))
+        .arg("--wordlist")
+        .arg(file.as_os_str())
+        .arg("--filter-size")
+        .arg("56")
+        .unwrap();
+
+    cmd.assert().success().stdout(
+        predicate::str::contains("/LICENSE")
+            .and(predicate::str::contains("302"))
+            .and(predicate::str::contains("14"))
+            .and(predicate::str::contains("/file.js"))
+            .not()
+            .and(predicate::str::contains("200"))
+            .not()
+            .and(predicate::str::contains("56c"))
+            .not(),
+    );
+
+    assert_eq!(mock.times_called(), 1);
+    assert_eq!(mock_two.times_called(), 1);
+    teardown_tmp_directory(tmp_dir);
+}