From 684c0e50607a1180b828a193bb1bfee5248ba0ee Mon Sep 17 00:00:00 2001
From: Martin Joerg <martin.joerg@gmail.com>
Date: Sun, 15 Sep 2024 08:14:42 +0000
Subject: [PATCH 1/2] pdf-parser: add updateScript

---
 pkgs/tools/misc/pdf-parser/default.nix | 35 ++++++++++++++++++++------
 1 file changed, 27 insertions(+), 8 deletions(-)

diff --git a/pkgs/tools/misc/pdf-parser/default.nix b/pkgs/tools/misc/pdf-parser/default.nix
index 593960c4d2cf..614bb9d30826 100644
--- a/pkgs/tools/misc/pdf-parser/default.nix
+++ b/pkgs/tools/misc/pdf-parser/default.nix
@@ -1,12 +1,19 @@
-{ lib, python3Packages, fetchzip }:
+{
+  lib,
+  python3Packages,
+  fetchzip,
+  writeScript,
+}:
 
-python3Packages.buildPythonApplication {
+python3Packages.buildPythonApplication rec {
   pname = "pdf-parser";
   version = "0.7.4";
 
   src = fetchzip {
-    url = "https://didierstevens.com/files/software/pdf-parser_V0_7_4.zip";
-    sha256 = "1j39yww2yl4cav8xgd4zfl5jchbbkvffnrynkamkzvz9dd5np2mh";
+    url = "https://didierstevens.com/files/software/pdf-parser_V${
+      lib.replaceStrings [ "." ] [ "_" ] version
+    }.zip";
+    hash = "sha256-sIprS2vp7z+rmtZn69yea0EmC3WftNfRVoxQLzj3acg=";
   };
 
   format = "other";
@@ -20,16 +27,28 @@ python3Packages.buildPythonApplication {
       --replace '/usr/bin/python' '${python3Packages.python}/bin/python'
   '';
 
-  meta = with lib; {
+  passthru.updateScript = writeScript "update-pdf-parser" ''
+    #!/usr/bin/env nix-shell
+    #!nix-shell -i bash -p common-updater-scripts curl pcre2
+
+    set -eu -o pipefail
+
+    version="$(curl -s https://blog.didierstevens.com/programs/pdf-tools/ |
+      pcre2grep -O '$1.$2.$3' '\bpdf-parser_V(\d+)_(\d+)_(\d+)\.zip\b.*')"
+
+    update-source-version "$UPDATE_NIX_ATTR_PATH" "$version"
+  '';
+
+  meta = {
     description = "Parse a PDF document";
     longDescription = ''
       This tool will parse a PDF document to identify the fundamental elements used in the analyzed file.
       It will not render a PDF document.
     '';
     homepage = "https://blog.didierstevens.com/programs/pdf-tools/";
-    license = licenses.publicDomain;
-    maintainers = [ maintainers.lightdiscord ];
-    platforms = platforms.all;
+    license = lib.licenses.publicDomain;
+    maintainers = [ lib.maintainers.lightdiscord ];
+    platforms = lib.platforms.all;
     mainProgram = "pdf-parser.py";
   };
 }

From f23cdfedfa6ce783fa97edfa88f63bb49e4a8cb3 Mon Sep 17 00:00:00 2001
From: Martin Joerg <martin.joerg@gmail.com>
Date: Sun, 15 Sep 2024 08:14:42 +0000
Subject: [PATCH 2/2] pdf-parser: 0.7.4 -> 0.7.9

https://github.com/DidierStevens/DidierStevensSuite/compare/5f81a8f7a8aac15b580413f6f3a2ec3d72c5d10c..91faa8f5d101e9216de0206af44d9864818ba233#diff-901accb21b87dbbf58011e6c337f4f45cd040ed7b65e61cbc266089767e3fb38
---
 pkgs/tools/misc/pdf-parser/default.nix | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/pkgs/tools/misc/pdf-parser/default.nix b/pkgs/tools/misc/pdf-parser/default.nix
index 614bb9d30826..b083ebd3e4ed 100644
--- a/pkgs/tools/misc/pdf-parser/default.nix
+++ b/pkgs/tools/misc/pdf-parser/default.nix
@@ -7,16 +7,26 @@
 
 python3Packages.buildPythonApplication rec {
   pname = "pdf-parser";
-  version = "0.7.4";
+  version = "0.7.9";
+  pyproject = false;
 
   src = fetchzip {
     url = "https://didierstevens.com/files/software/pdf-parser_V${
       lib.replaceStrings [ "." ] [ "_" ] version
     }.zip";
-    hash = "sha256-sIprS2vp7z+rmtZn69yea0EmC3WftNfRVoxQLzj3acg=";
+    hash = "sha256-1mFThtTe1LKkM/MML44RgskGv3FZborNVBsTqSKanks=";
   };
 
-  format = "other";
+  postPatch = ''
+    # quote regular expressions correctly
+    substituteInPlace pdf-parser.py \
+      --replace-fail \
+        "re.sub('" \
+        "re.sub(r'" \
+      --replace-fail \
+        "re.match('" \
+        "re.match(r'"
+  '';
 
   installPhase = ''
     install -Dm555 pdf-parser.py $out/bin/pdf-parser.py
@@ -24,7 +34,7 @@ python3Packages.buildPythonApplication rec {
 
   preFixup = ''
     substituteInPlace $out/bin/pdf-parser.py \
-      --replace '/usr/bin/python' '${python3Packages.python}/bin/python'
+      --replace-fail '/usr/bin/python' '${python3Packages.python}/bin/python'
   '';
 
   passthru.updateScript = writeScript "update-pdf-parser" ''