From efad31dac17c3462b328a390f6c6b66d52a371b6 Mon Sep 17 00:00:00 2001
From: Nico Weber <thakis@chromium.org>
Date: Mon, 26 Jun 2023 20:42:10 -0400
Subject: [PATCH] Utilities: Add a `pdf` utility

This utility will learn tricks such as extracting images from PDFs and
dumping tables from PDFs so that we can create code from specs.

It also allows testing LibPDF things in lagom, and allows testing
reading large amounts of PDFs using a shell script.
---
 Meta/Lagom/CMakeLists.txt         |  3 ++
 Userland/Utilities/CMakeLists.txt |  1 +
 Userland/Utilities/pdf.cpp        | 47 +++++++++++++++++++++++++++++++
 3 files changed, 51 insertions(+)
 create mode 100644 Userland/Utilities/pdf.cpp

diff --git a/Meta/Lagom/CMakeLists.txt b/Meta/Lagom/CMakeLists.txt
index b3261e85a5..9894b5d327 100644
--- a/Meta/Lagom/CMakeLists.txt
+++ b/Meta/Lagom/CMakeLists.txt
@@ -555,6 +555,9 @@ if (BUILD_LAGOM)
             target_link_libraries(ntpquery LibCore LibMain)
         endif()
 
+        add_executable(pdf ../../Userland/Utilities/pdf.cpp)
+        target_link_libraries(pdf LibCore LibPDF LibMain)
+
         add_executable(sql ../../Userland/Utilities/sql.cpp)
         target_link_libraries(sql LibCore LibFileSystem LibIPC LibLine LibMain LibSQL)
 
diff --git a/Userland/Utilities/CMakeLists.txt b/Userland/Utilities/CMakeLists.txt
index f6c490f644..73f89eacc3 100644
--- a/Userland/Utilities/CMakeLists.txt
+++ b/Userland/Utilities/CMakeLists.txt
@@ -123,6 +123,7 @@ target_link_libraries(notify PRIVATE LibGfx LibGUI)
 target_link_libraries(open PRIVATE LibDesktop LibFileSystem)
 target_link_libraries(passwd PRIVATE LibCrypt)
 target_link_libraries(paste PRIVATE LibGUI)
+target_link_libraries(pdf PRIVATE LibPDF)
 target_link_libraries(pgrep PRIVATE LibRegex)
 target_link_libraries(pixelflut PRIVATE LibImageDecoderClient LibIPC LibGfx)
 target_link_libraries(pkill PRIVATE LibRegex)
diff --git a/Userland/Utilities/pdf.cpp b/Userland/Utilities/pdf.cpp
new file mode 100644
index 0000000000..b53cb7d668
--- /dev/null
+++ b/Userland/Utilities/pdf.cpp
@@ -0,0 +1,47 @@
+/*
+ * Copyright (c) 2023, Nico Weber <thakis@chromium.org>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <LibCore/ArgsParser.h>
+#include <LibCore/File.h>
+#include <LibCore/MappedFile.h>
+#include <LibPDF/Document.h>
+
+static PDF::PDFErrorOr<int> pdf_main(Main::Arguments arguments)
+{
+    Core::ArgsParser args_parser;
+
+    StringView in_path;
+    args_parser.add_positional_argument(in_path, "Path to input image file", "FILE");
+
+    args_parser.parse(arguments);
+
+    auto file = TRY(Core::MappedFile::map(in_path));
+
+    auto document = TRY(PDF::Document::create(file->bytes()));
+
+    if (auto handler = document->security_handler(); handler && !handler->has_user_password()) {
+        // FIXME: Add a flag for passing in a password and suggest it in this diag.
+        warnln("PDF requires password");
+        return 1;
+    }
+
+    TRY(document->initialize());
+
+    outln("{} pages", document->get_page_count());
+
+    return 0;
+}
+
+ErrorOr<int> serenity_main(Main::Arguments arguments)
+{
+    auto maybe_error = pdf_main(move(arguments));
+    if (maybe_error.is_error()) {
+        auto error = maybe_error.release_error();
+        warnln("{}", error.message());
+        return 1;
+    }
+    return 0;
+}