Utilities: Add strings

3 files changed, 180 insertions, 1 deletions

diff --git a/Base/usr/share/man/man1/strings.md b/Base/usr/share/man/man1/strings.md
new file mode 100644
index 0000000000..abcaf5284b
--- /dev/null
+++ b/Base/usr/share/man/man1/strings.md
@@ -0,0 +1,39 @@
+## Name
+
+strings - find printable strings in files
+
+## Synopsis
+
+```**sh
+$ strings [-n NUMBER] [-p] [-t FORMAT] [PATHS...]  
+```
+
+## Description
+
+`strings` looks for printable strings in each file specified in `PATHS` and writes them to standard output. If `PATHS` is not specified, input is read from standard input.
+
+## Options
+
+* `-n NUMBER`: Specify the minimum string length (4 is default).
+* `-p`: Write the pathname for each file specified in `PATHS` to standard output.
+* `-t FORMAT`: Write each string preceded by its byte offset from the start of the file in the specified `FORMAT`, where `FORMAT` matches one of the following: `d` (decimal), `o` (octal), or `x` (hexidecimal).
+
+## Examples
+
+Display the printable strings in /bin/strings with a minimum length of 8 characters:
+
+```sh
+$ strings -n 8 /bin/strings
+```
+
+Display the printable strings in a binary file, preceded by their byte offset in hexadecimal format:
+
+```sh
+$ strings -t x ~/Videos/test.webm
+```
+
+Display the printable strings in all .txt files in the current directory, preceded by their pathname:
+
+```sh
+$ strings -p *.txt
+```
diff --git a/Userland/Utilities/CMakeLists.txt b/Userland/Utilities/CMakeLists.txt
index c9f1226d9a..f342ad292b 100644
--- a/Userland/Utilities/CMakeLists.txt
+++ b/Userland/Utilities/CMakeLists.txt
@@ -8,7 +8,7 @@ list(APPEND REQUIRED_TARGETS
 )
 list(APPEND RECOMMENDED_TARGETS
     adjtime aplay abench asctl bt checksum chres cksum copy fortune gunzip gzip init install keymap lsirq lsof lspci man mknod mktemp
-    nc netstat notify ntpquery open passwd pls printf pro shot tar tt unzip wallpaper zip
+    nc netstat notify ntpquery open passwd pls printf pro shot strings tar tt unzip wallpaper zip
 )
 
 # FIXME: Support specifying component dependencies for utilities (e.g. WebSocket for telws)
diff --git a/Userland/Utilities/strings.cpp b/Userland/Utilities/strings.cpp
new file mode 100644
index 0000000000..531db31838
--- /dev/null
+++ b/Userland/Utilities/strings.cpp
@@ -0,0 +1,140 @@
+/*
+ * Copyright (c) 2022, the SerenityOS developers.
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <AK/CharacterTypes.h>
+#include <AK/Forward.h>
+#include <LibCore/ArgsParser.h>
+#include <LibCore/Stream.h>
+#include <LibCore/System.h>
+#include <LibMain/Main.h>
+#include <unistd.h>
+
+enum class StringOffsetFormat {
+    None = 0,
+    Decimal,
+    Octal,
+    Hexadecimal
+};
+
+// NOTE: This is similar to how the cat utility works in the sense of aggregating
+// data in 32K buffer.
+static constexpr size_t buffer_read_size = 32768;
+
+static bool should_print_characters(Vector<u8> const& characters)
+{
+    for (u8 ch : characters) {
+        if (is_ascii_printable(ch) && !is_ascii_space(ch))
+            return true;
+    }
+    return false;
+}
+
+static void print_characters(Vector<u8> const& characters, StringOffsetFormat string_offset_format, size_t string_offset_position)
+{
+    switch (string_offset_format) {
+    case StringOffsetFormat::Decimal:
+        out("{:>7d} ", string_offset_position);
+        break;
+    case StringOffsetFormat::Octal:
+        out("{:>7o} ", string_offset_position);
+        break;
+    case StringOffsetFormat::Hexadecimal:
+        out("{:>7x} ", string_offset_position);
+        break;
+    default:
+        break;
+    }
+    outln("{:s}", characters.span());
+}
+
+static int process_characters_in_span(Vector<u8>& characters, ReadonlyBytes span)
+{
+    int processed_characters = 0;
+    for (u8 ch : span) {
+        ++processed_characters;
+        if (is_ascii_printable(ch) || ch == '\t')
+            characters.append(ch);
+        else
+            break;
+    }
+    return processed_characters;
+}
+
+static ErrorOr<void> process_strings_in_file(StringView path, bool show_paths, StringOffsetFormat string_offset_format, size_t minimum_string_length)
+{
+    Array<u8, buffer_read_size> buffer;
+    Vector<u8> output_characters;
+    auto file = TRY(Core::Stream::File::open_file_or_standard_stream(path, Core::Stream::OpenMode::Read));
+    size_t processed_characters = 0;
+    size_t string_offset_position = 0;
+    bool did_show_path = false;
+    while (!file->is_eof()) {
+        auto buffer_span = TRY(file->read(buffer));
+        while (!buffer_span.is_empty()) {
+            string_offset_position += processed_characters;
+            processed_characters = process_characters_in_span(output_characters, buffer_span);
+            if (show_paths && !did_show_path) {
+                outln("path {}:", path);
+                did_show_path = true;
+            }
+            if (output_characters.size() >= minimum_string_length && should_print_characters(output_characters)) {
+                print_characters(output_characters, string_offset_format, string_offset_position);
+            }
+            buffer_span = buffer_span.slice(processed_characters);
+            output_characters.clear();
+        }
+    }
+    return {};
+}
+
+ErrorOr<int> serenity_main(Main::Arguments arguments)
+{
+    TRY(Core::System::pledge("stdio rpath"));
+
+    Vector<StringView> paths;
+    size_t minimum_string_length = 4;
+    bool show_paths = false;
+
+    StringOffsetFormat string_offset_format { StringOffsetFormat::None };
+
+    Core::ArgsParser args_parser;
+    args_parser.add_option(minimum_string_length, "Specify the minimum string length.", nullptr, 'n', "number");
+    args_parser.add_option(show_paths, "Display the path for each matched file.", nullptr, 'p');
+    args_parser.add_option({ Core::ArgsParser::OptionArgumentMode::Required,
+        "Write offset relative to start of each file in (d)ec, (o)ct, or he(x) format.",
+        nullptr,
+        't',
+        "format",
+        [&string_offset_format](char const* s) {
+            StringView value = { s, strlen(s) };
+            if (value == "d") {
+                string_offset_format = StringOffsetFormat::Decimal;
+            } else if (value == "o") {
+                string_offset_format = StringOffsetFormat::Octal;
+            } else if (value == "x") {
+                string_offset_format = StringOffsetFormat::Hexadecimal;
+            } else {
+                return false;
+            }
+            return true;
+        } });
+    args_parser.set_general_help("Write the sequences of printable characters in files or pipes to stdout.");
+    args_parser.add_positional_argument(paths, "File path", "path", Core::ArgsParser::Required::No);
+    args_parser.parse(arguments);
+
+    if (minimum_string_length < 1) {
+        warnln("Invalid minimum string length {}", minimum_string_length);
+        return 1;
+    }
+
+    if (paths.is_empty())
+        paths.append("-"sv);
+
+    for (auto const& path : paths)
+        TRY(process_strings_in_file(path, show_paths, string_offset_format, minimum_string_length));
+
+    return 0;
+}